Apology not required from this fan.

Your father died. I'm amazed that you're standing up right now, much less writing a novel.

Thank you, anyway, I'm sure I'll enjoy it. Meanwhile I recommend getting drunk and hugging your wife and not worrying about us all for a while.

I see that 666 squadron appear to be flying & what looks suspicously like a major US building or two appearing on the cover ....

Given the life-history of Dragonflies & their predatory habits ( Large nymphs will bite human fingers! ) I find the presence of one on the cover a little disturbing, as well.

Honestly don't know how you were only a few months late. Would have wiped out my year. You're a champion, Charlie, and we should all be grateful for your dedication to your work. Cheers.

It is normal for book covers to be, shall we say, only tangentially related to the contents, and to contain huge and glaring inaccuracies.

The huge and glaring error on the Tor.com cover for "The Labyrinth Index" is that there are no Eurofighter Typhoon II's chasing the Concorde in the book. (F-15s and F-16s, yes, but no Eurofighters.)

Also: dragonfly? It's a parasitic wasp in the book! I must bring western civilization to a screeching halt until the art director personally deserts her Thanksgiving dinner and fixes this heinous error.

Just wanted to say, of the authors that I follow, your dedication and rate of creation always amazes. On behalf of us fans, at least, I think no apology is owed! As Shigeru Miyamoto once said of video games "A delayed game is eventually good, but a rushed game is forever bad."

My condolences on the loss of your father. And I'm thankful to you for all the laughs, surprises and eyebrow-raising ideas you've given us.

Book preordered on Amazon. Sorry for your loss, and don't be sorry for doing what was best.

My condolences, and take your time with the writing.

I definitely agree. No apologies are necessary. When my father dies I expect to be f**ked for a year, at least.

I #1 this: Just wanted to say, of the authors that I follow, your dedication and rate of creation always amazes. On behalf of us fans, at least, I think no apology is owed!

I'm so sorry for your loss, Charlie. I certainly don't begrudge you for being all quiet on the blogging front. Besides, Hugh Hancock has contributed (plus comments) some very thought provoking stuff on the state of VR.

Another one saying: no apology needed.

I've had more than my fair share of deaths this year. Even when it's reasonably expected (93 is a damn good innings), it's still a sharp body blow, especially if the family ties are strong.

You have to take care of yourself first. If that means that deadlines are missed, then deadlines are missed. No sane, compassionate individual would begrudge you that under the circumstances.

My deepest condolences. We'll be here, ready to snap up the new books when they're ready. In the meantime, take the time you need, and remember to tell those you care about how much they mean to you. We don't do enough of that.

As others have said, no apologies needed.

For about a year after my Mum died, the only creative thing I managed was finding new places to burst into tears for no clearly identifiable reason.

One thing to remember is that it gets better. I'm not 100% ok of course, but rather than it feeling like your entire chest has been ripped open, it's just that dicky knee that twinges when you move wrong. That doesn't mean that they're less important, or that their loss isn't still a loss, it just sort of scabs over. Their loss isn't as immediate, and it sort of stops outshining the joy that their life gave you. If you're like me, after a while you'll stop thinking about them not being there, and start thinking about how great it was when they were there. And I try to remember that they wouldn't have wanted me to be sad, nor to remember them more for their absence, than their life.

If you feel like throwing yourself into work, do that. If you feel like not working, do that. Do what works for you. We'll all be here for you when you're ready. We owe you, not the other way around.

I'm very very excited about the Concorde flight. That gun has been sitting in the mantelpiece for a long long time (first novel?)

The Bookdepository sent me a confirmation email after I preordered the book. They say that "Your order will be dispatched from our warehouse soon". I rather doubt that, or their definition of "soon" is different from mine.

"Laundry Files series in particular is a hell of a lot of fun."

I think that Scalzi quote has been somewhat off-tone since the fifth book, at last... (Not that I'm complaining, mind you.)

Not an entomologist (or any other type of biologist, either) but I think it's just the position of the wings that makes the bug look dragonfly-ish. The wings look like wasp's wings to me, it's just that they seem to be posed like those of a preserved specimen, not like those of a living wasp.

It's cunning mimicry. They pose as specimens in entomologists collections and science museums, then when the victim get's close enough...

Ditto, but I have just looked through my insect book, and it's also the outline. But - mirabile dictu! - they have the venation and wing linkage roughly right for a wasp :-)

Ah man, Charlie, I feel like I need to apologise on behalf of all of us for making you feel that you need to apologise to us.

I don't know about your publisher, though. It might be that apologising to them might save you a kneecap or two. Maybe bake them a cake, just to be safe.

You know what I would do? (Note: whatever I would do, it is a safe bet that you should probably not do that.) Go for something completely different. Write a kid's book. A travel diary. Go be a Tibetan monk for a while. Just lose the obligation to do something familiar, and do it well. You can try something utterly new and be awful at it.

I mean, I love your work, but you're not a work-hose.

Afterthought: can you imagine an illustrated Laundry Files kid's book?

There's the alphabet book "Baby's First Mythos". It was more fun for us parents than the kids.

A is for Algorithm, which must not be too complex B is for Bureaucrat, who starts each form with name, age, sex C is for Cultist, a villain armed with knives D is for Deep One, under the sea they thrives

E is for Eater, devouring your brain F is for Fractal, to drive you insane

My condolences. When my dad died, it took significantly longer than a year to bounce back. As slamble says @ 11: "You have to take care of yourself first".

can you imagine an illustrated Laundry Files kid's book?

Couldn't be any wierder than the Deadpool comics. Or in my era, 2000AD and either "Slaine" or "Nemesis"...

Firstborn is now 15, started reading the Laundry books a couple of years ago, and was a tad miffed that I didn't bring him to Charlie's launch of "The Delirium Brief" [1] seeing as how I'd taken him to the launch of "The Annihilation Score"...

[1] Although it would have meant that I couldn't go to the pub afterwards...

Re. the image of Washington DC, I have always thought the "hair" of the current president may be the sensorial antenna array of a telepathic alien parasite flattening itself to the skull, which is running the host body through remote control. The odd speech pattern of the host reflects an imperfect understanding of human speech and culture.

-This is completely OT, but I just want to inject some good news in case great discoveries may cheer you up: "ALMA discovers cold dust around nearest star, implying Proxima Centauri has a planetary system" (and not just a single world). https://phys.org/news/2017-11-alma-cold-nearest-star.html BTW Ross 128 also has a rocky planet that looks better than Proxima b.

When my father died, it was as if I had emigrated to a new country, one to which I felt I could become accustomed but could never feel native.

G is for Gospel that Vicars won't like H for the hackers' half-tracked motorbike

I is for incision, long and deep J is for jugular, leaking life's blood

Here's a hint for how to tell dragonflies:

Look for the four (4) wings, the long thin BODY, and the lack of long antennae. There's also details like gynormous eyes that don't show up in all dragonflies.

That is, indeed, a wasp of some sort, most likely a parasitic one (just by the odds). The two wings and long antennae give it away.

I agree that J. Random Artist likely took the pin out of the photoshop they used to make the cover, but unless you want the wasp ovipositing into one of those jet fighters, it's the image that you're likely to get.

All I could easily see were the very large eyes, which are a give away for Dragon/Damselflies - like the sapphire-blue ones that breed in my pond ..... Some bodies are not that long & can be quite wide Like THIS ONE

^^^^^^^^^^^^^^^^^^^^^^^^^^^^ K is for Killer, Of kinds so many L is for Lurker, in Shadows so deep

Re. childrens books ... Some ideas for illustrators:

The Tjong-Khing, who diid a great and very spooky book based on Hieronymous Bosch. An image from which:

Alberti Innocenti, who's versiopn of little red riding hood also has a very paranoid feel:

Nikolaus Heidelbach, who made wonderfully weird illustrations for aset of Grimms' tales (he wlose picked a few of the weirder ones, like the one with a sausage as protagonist). Heidelbachs art is awesome - look at this self portraits with eldritch abominations:

I only have on Moomins book, but from that I'd say that Tove Jansson does spooky moods and eldritch locations quite well:

There are other great illuatrators of childrens books, where I'm not sure that they could get the dark mood right - but their take on the Laundry files or the Mythos might be interesting nonetheless - Carson Ellis, Axel Scheffler, Ciara Flood ...

Looking at these, I think Heidelbach would be the most obvious candidate. But is obvious best here? What do you think?

Off topic, but for those who need a different take on stuff to recharge their creative batteries:

real chthonians. And yes, whatever its inhuman morality, the deep biosphere, like the surface, runs on symbiosis. Looks like on this planet we're all symmers, born in sym, and that liberation from this state is only a fantasy.

Arthur Rackham Fasolt & Fafnir kidnap Freya Odinn commands Loge to encircle Brunnhilde in a Ring of Fire The doom of Fafnir Twa Corbies

Ian Banks "life continues to suprise" So says a Culture Mind ......

Errr ... ROBERTO Innocenti, please ... had to go round Google a couple of times, there, when re-directed to a 17/18th Cent artist (!) HERE - though you will need Google translate

Were he alive then Edward Gorey would be my choice.

The Gashlycrumb Tinies is already a classic of the alphabet book genre.

I note that dragonflies have four separated wings, whilst the insect on the cover has two (or possibly four wings linked together. It is unclear)

In any case, it's not THAT far off... consider this picture off Wikipedia (not the same species, of course. I don't have the skills to ID that. :) )

https://youtu.be/9CO6M2HsoIA

This video got some recent buzz. Like the curious yellow killer bots. Very plausible near future extrapolation of drone and AI tech for highly targeted killing.

I'm not sure what the campaign against autonomous weapons is supposed to achieve mind.

Apart from the explosives all those capabilities will probably be within the reach of the n+2 generation arduino hacker. The tech is nearly there, the idea is obvious and the will exists.

On the manufactured, commercial side the idea of an arms company giving up profits or a military giving up magic bullets for the public good is somewhat optimistic.

My prediction: In the future everyone will wear an armoured burqa or one of these.

And explosives are a piece of piss really.

Um.

Before we go off on AI drones, it's worth checking the history of the XREP 12-gauge shotgun shell.

These debuted around 2010. The XREP's a really cool little idea: a miniature taser sized to be shot out of a shotgun. The problem with the design is that the shells cost $160 each. They never caught on, and now they're extremely rare and collectable.

In contrast, 750 rounds of AK74 ammo can be had online for $185.00. This is the reason why rebel armies in failed states are more apt to kidnap children, brainwash and drug them, and equip them with machetes and AKs, than they are to buy drones.

The cost of drones is so high that they currently make sense in situations where humans can't do the job (e.g. flying for 24 hours nonstop, or going into lethal environments, as in hostage situations). Obviously they'll get cheaper, but I suspect that we'll see octo-copters mounting shotguns long before we see one-off suicide AI bombers.

I'll take your word for that - not my field and I know how annoying it is when software people blithely assume that everyone elses engineering problems are trivial :)

The cost of drones mostly comes from the high capacity batteries and fancy motors. For short range micro drones that only need to fly for a few minutes you can skimp on both of those.

I was watching a nice video a few days ago that showed that you can compensate for almost any amount of hardware crapness in software. The limit was controlled (if jerky) flight with a drone that amounted to a single propeller mounted on a stick.

On the manufactured, commercial side the idea of an arms company giving up profits or a military giving up magic bullets for the public good is somewhat optimistic.

Bullshit.

We have mostly managed to enforce an international convention on not using chemical weapons — the only substantial breaches have been the use of less-lethal weapons (e.g. tear gas, the Russian botched use of an opiate knock-out gas during the theater siege) and during particularly vicious civil wars (e.g. Syria) which have already escalated to genocidal levels of violence.

Similarly, we have a convention against the use of laser blinding weapons which has stood since the late 1980s and has prevented them from being deployed. Laser targeting devices are exempted, as long as they're guidance for PGMs; laser beam weapons that directly incinerate a target are also kinda-sorta exempted but aren't really deployed yet. The key point being that we could easily build cheap weapons that cause traumatic and permanent eye injury on a massive scale ... but we haven't done so, even though high power LED lasers are now well within the reach of folks like Da'esh.

Nuclear weapons ... are frighteningly expensive, escalate the jeopardy like crazy, and nobody who can afford them really wants to see what happens next after they pop one. Nor are they really very useful any more, in actual practical military terms. So the folks who've got them don't need them (PGMs do the job cheaper and without the controversy) and the folks who don't have them are up against the NPT.

These agreements matter because they serve notice on any assclowns planning on going down that route that they'd better be damned confident they're going to triumph in the face of global public opinion, or end up in the Hague facing a life sentence for crimes against humanity.

So we do need a convention against killer AI drones. If only to define a baseline. And I believe while it'll be breached some of the time, if it's established early enough most people will stick to it.

Which means that guvmints are going to have to clamp down really hard on drone use - any drone use at all, other than their own, of course. Anyone really wanting to use one will have to have a licence & probably a government "inspector" watching the action, just in case .... Because drones are already fairly cheap & they are going to get a lot cheaper.

That's kind of not the point. People have been goofing with the idea of mounting guns on quadrotors since they first came out (there's an online hoax from 2012). People already mount guns on robots of various sorts, and it's a much more flexible weapon than what they proposed.

The problem with loading a miniature quadrotor with explosive is that basically you're turning it into a bullet. To make the cost effective for warefare (as opposed to targeted assassination), you need to get the cost down around $10 per suicide drone. Packing propellers, hardware, explosive, and power into $10 is kind of hard. For example (to pick something I was just reading about batteries for cars and houses), there's some argument (hopefully wrong) that the floor price for Li-Ion batteries is around $100/kWh due to materials costs. If this is correct, if you have a $1-2 battery, that gives you a 10-20 Watt-hour battery to fly your microdrone. That's not a lot to play with.

Yes, you're right of course!

I actually looked up the correct name (in my bookshelf) to google the image, but somehow still wrote the wrong name into my comment ...

Already happening (because of risk to civil aviation) (link is to UK government announcement). Note that licensing exempts drones weighing under 250 grams, meaning anything with a payload of less than a couple of olde-worlde ounces.

Nice! Too bad a tarantula hawk wouldn't fit on the cover--Most of them have black heads. You can watch someone get stung by one (probably Pepsis grossa) here. Perhaps OGH could write a mythos-based short story about someone who does something similar as a palate cleanser or something.

Might be NSFW, although there is no sexuality or nudity involved. Some people trigger on stupidity, I think.

I hope you are correct, but in the case of things like gas and land mines the fact that they are dumb and indiscriminate was a major factor in the bans. When there's a computer in the mix you can waffle about "safeguards" forever and then blame anything bad on rogue actors.

Hobbyists can't really do a lot with chemicals and nukes, but I have encountered people having a go with lasers on 3 occasions in the last 10 years when I have been driving or cycling. Cheap low power pointers fortunately but I'm sure the kind of clown who will try lighting up a moving car at night would be happy to use a class IV if they could afford it.

Most bullets fired in conflict never hit anyone or anything (important). The hit rate is in the 0.1% region or even lower. The rest is suppressive fire -- in WW1 a British infantry company with four belt-fed machine-guns fired a million rounds in 24 hours over No Mans Land to deny the Germans a line of advance. They never hit anyone, which was sort of the point of the exercise.

A drone released against a target within range and in good conditions (i.e. not flying into a 30km/h wind) has a decent chance of reaching that target and doing some damage to them, if not killing them outright. That's worth spending a thousand bucks on hardware for the higher probability of success compared to a thousand bucks worth of bullets fired from guns, and absent the drone being a suicide device it may even be recoverable for re-use.

However there's no need for a military drone to have an AI or be controlled by an AI to do the job -- current drone attacks have a human in the loop to fire the weapons they carry which are often drones of another sort, precision-guided munitions.

Quite true, and you're right to equate precision munitions to bomb-drones.

The only use for AI in drones is the proliferation of things like the DroneShield DroneGun. It looks like there's going to be a Red Queen race between drones and anti-drone technologies. Indeed, I see a good market for quadrotors with proprietary control technology (or even AI) that deploy anti-drone measures to exclude remote controlled drones from areas.

That is what trained snipers are for.

Suicide drones, deployed in ngorno-karabakh.

Nice sales vid, it chases fleeing vehicles.

2kg explosive.

http://foreignpolicy.com/2017/08/30/israel-freezes-export-of-suicide-drone-to-azerbaijan-after-allegation-of-abuse-azerbaijan-armenia/

What I was getting at was that out of all the technologies required for killer drones, it is only explosives that were well-established and well-understood 100 years ago - as WW1 so comprehensively demonstrated. (To be sure, there have been plenty of developments since then, but they're not "required" in this context.) Consequently there is a huge amount of information about their synthesis, handling, and application, which is all readily and freely available. If you knew nothing about any of the technologies, that would probably be the easiest one to get up to speed on, since all you have to do is look stuff up and follow recipes; there's no innovation involved, nor anything resembling it.

As far as the chemistry is concerned, explosive syntheses generally tend to be easier than clandestine drug syntheses, many are actually less dangerous to perform (though some definitely aren't), and the materials are generally easier to get hold of.

WW1-level poison gases are even easier to make (which was a significant part of their attraction at the time), although it's rather more of an effort to survive making them than it is for explosives.

I'm not claiming that it's completely plain sailing, but I do say that the perceived difficulty is much greater than the actual difficulty, and the actual difficulty is quite a bit less than the difficulty of some other aspects, like, say, for instance, writing the guidance software.

That battery has an output roughly equivalent to 20-40 minutes flying time for a wood pigeon, giving say 15-30 miles range. Which is pretty useful really. The thing is, can you make your drone as efficient as a wood pigeon?

I guess the saving grace is that that kind of clown doesn't realise you can put together a class IV for less money than pretty well any power of laser pointer...

Hobbyists can't really do a lot with chemicals and nukes, ... Nukes, sure. Many lethal[1] chemicals, OTOH, are pretty easy for hobbyists to produce. Not sure what to do about that. Kitchens, glassware, hardware stores and etc exist. Recipes and ideas on the internet make it much easier for people who would otherwise not think of these things or know how to implement/reinvent them.

Anyway, to the broader point, Charlie (quoted below) is absolutely correct, the public arguments battle needs to be fought with arms merchants and their allies, now and without respite, to prevent normalization of lethal autonomous drones. So we do need a convention against killer AI drones. If only to define a baseline. And I believe while it'll be breached some of the time, if it's established early enough most people will stick to it.

[1] Or otherwise interesting. :-) Perhaps that's the solution; legalize certain forms of amateur chemistry.

In not-good-news, but interesting in part for the straightforward mentions of policy/regulatory concerns (small quote below). New science of climate change impacts on agriculture implies higher social cost of carbon (20 November 2017) Updating the damage functions in the agriculture sector alone increases the SCC from $8.6 to $14.8 (AgMIP) and $19.7 (meta-analysis) ton−1 CO2, increases of 72 and 129%, respectively. ... Governments are currently relying on IAMs to evaluate climate and energy policy and these models have already come under legal scrutiny as a result31, 32, 5. It is therefore important, from both a regulatory and an academic perspective, that the representation of damages reflect current scientific consensus on impacts in a timely and transparent manner. Here we have shown that improving the empirical basis of just one sector, agriculture, results in a large increase in the SCC.

M is for Mhari, our queen of the night. Or is it M is for Mo, musician of fright? N is for neurons. Hope you have some left.

https://youtu.be/9CO6M2HsoIA

This video got some recent buzz. Like the curious yellow killer bots. Very plausible near future extrapolation of drone and AI tech for highly targeted killing.

http://hexacopters.com/

I ran across this site about 10 - 12 years ago right after I got home from Iraq. Looks like it's been updated a few times (the videos date from 2010 - 2012 and I got home at the beginning of 2005), and the domain may have been abandoned.

Still, this bit's been there ever since I first found it:

“No doubt that these will eventually be used for military applications. That said, they might actually level the battle field quite a bit. Citizens fighting against better-funded oppressive regimes could certainly pester their enemy with this technology. (Imagine a swarm of these things descending on your location with five pounds of explosives strapped to each one of them!!) ...More than 10,000 radio / GPS controlled hexacopters could be purchased for the price of just one "conventional" attack helicopter.”

That is an interesting point about the laser blinding weapons. Regarding poison gas in the light, it seems like it's just not worth the trouble of using. Laser weapons may not be as effective as we think? Terrorists aren't going to care about the Geneva convention but they won't want to waste their time on weapons and tactics that are ineffective.

Good point above about how few bullets hit targets. I think in Vietnam it was estimated 25k rounds for every dead soldier. But there's a lot more to consider about the value of killing the right people. You wouldn't pay for a $10k bullet to shoot an infantryman but one that could take out a general in his tent reliably.... for a terror group, the calculus is probably a matter of cost to mount attack and loss of suicide expendable assets and non-expendable support assets against casualties and terror inflicted. I'm sure it's expensive to train people and set up networks so if upgrading an attack from guns and car bombs to terror drones increases the body count sufficiently they may think it acceptable. Currently drones can bypass any checkpoint, traffic barriers, etc. And I think countermeasures would be expensive and another inconvenience. I see our current airport security theater as a monument to the success of Al-Qaeda. Every time we take off our shoes the terrorists win.

History is full of wonder weapons that never quite work as well as the old stuff but everything we consider conventional today was once newfangled. Guns were inferior to bows for a long time and the only saving Grace was a shorter training time. It took a long time for guns to beat bows in every way.

Drones could well amount to nothing but I'm feeling we are at the equivalent of the start of military aviation. We are just getting started.

Since this appears to be sort of an open thread, recently read How We Talk: The Inner Workings of Conversation by N.J. Enfield, 14 Nov 2017. (Via a review in New Scientist) It covers turn-taking[1], repair and other things, and it's a quick read, with plenty of current references. Interesting for those of us who have spent time attempting to reverse engineer "normal" human conversation, how conversation timing works, etc. (I had already figured out most of this over the past several decades. Getting conversational timing in particular to be "normal" can be tough.) [1] Universals and cultural variation in turn-taking in conversation (2009; Fig. 2 interesting)

Anyway (for rationalists like Greg), that book had a reference to "Spider Divination", which is far more interesting than say contemporary aleuromancy, though less tasty. (One presumes.) Professor Garfinkel Visits the Soothsayers Ethnomethodology and Mambila Divination Mambila spider divination entails the posing of questions couched as binary alternatives, often requiring yes/no answers.

A hole in the ground inhabited by a spider is covered by an enclosure, usually an inverted pot. A stick and a stone are placed within this enclosure, near the entrance to the spider's hole. A set of marked leaves is placed over the entrance to the burrow. When questions are posed the pot is tapped; in response to the knocking the spider emerges from its hole. In doing so it disturbs the leaves. The resulting pattern of the leaves in relation to the stick and to the stone is interpreted as an answer to a question. Questions allow one of two responses, one is explicitly associated with the stick and the other with the stone.

Several different spiders may be consulted simultaneously. This enables a faster rate of questioning since some twenty minutes elapse before the diviner can check whether the spider has responded to a question. It also allows a consistency check to be made by asking the same question of different spiders. Diviners admit that ambiguous or unintelligible answers are possible but few such instances were observed.

No particular point; just came to mind because a young niece at a recent holiday dinner indicated that she was afraid of spiders. But not insects. (or snakes) "So 6 legs good, 8 legs bad?" "Yes [smiled]" "How about octopuses?" "Those are tentacles!"

Instead of guns, I'd wonder about drone-mounting something based on a hobbyist model-rocket engine. For those who haven't used one, a model rocket engine is a solid-rocket engine built into a cardboard tube. When the engine is done burning it fires a small charge to blow the parachute out of the nosecone of the rocket.

One could take a small engine, insert a few grams of any real explosive, and pack the rest with ball bearings. Or just pack the rest of the rocket engine with explosives. Or explosives plus a bell-shaped bit of copper. Or white phosphorus. Or just make the end of your missile pointy!

When I was a kid I used to fire them off with a firecracker inserted into the top of the tube. They'd land and then blow up! I have no idea whether the idea actually has any military utility, but I'll bet someone is experimenting with it.

"(Imagine a swarm of these things descending on your location with five pounds of explosives strapped to each one of them!!)"

Or imagine a swarm of them descending on a big, glass office building with five pounds of explosives strapped to each of them. Each drone stops outside a window and blows up. The over-pressure probably kills everyone in the building, even without the flying glass!

It occurs to me that model rockets have known ranges and burn times for their engines. A drone could be equipped with the software to calculate the right burn distance to a target for maximum utility when the engine fires off the actual explosive charge.

Just happened across the 2015 movie Eye in the Sky, a forensic examination of a drone strike in Nairobi, flipping between the Whitehall COBRA meeting, (Alan Rickman's last movie appearance as the general in charge) to Helen Mirren running the show in Northwood HQ to the US airbase and the team flying the drone. Down the road from the target compound, the Kenyan undercover is supplying eyes in ceiling with a remote control beetle. A literal bug. This is not, you will notice, billed as SF.

O is for "oh no!" - we say it quite often P is for PHANGS; they've come out of the coffin

Q is for quantum entanglement doom R is for rainbow of god games quite soon

AND ... This moning's news, the gumint/Parliament are to introduce a new law/regulations covering drones. "Not above 400 ft & not near airports & should have licence & training" according to what I heard ... ]

It's ridiculously easy to make something that will go "BANG" - what is much more difficult is to make sure it will only go "bang" when you want it to, not at other times & also won't fail when you do want it to go off. Thiis something lots of people forget, or don't know, or choose to ignore - if the latter, they usually blow themsleves up, how sad. My father spent from the middle of 1941 until about the 8th of May 1945, doing exactly that.

Actually, making something go "BANG" is actually quite difficult. Making it go "fizzzz" is less hard. High-energy chemicals like acids and nitrates are difficult for the layman to get hold of these days and working your way up the ladder from simpler and more readily available precursors is also a tricky path to follow -- try making something like potassium nitrate from scratch to see what I mean.

It depends what level of layman you mean. I have a variety of stuff, and the licence to own them, including potassium nitrate, without any real difficulty beyond a bit of a search online and in real life. It does require getting hold of some old books and the willingness to get your hands dirty, but I've had no training in anything from anyone and could still make a lot of them work if I was stupid enough to think it a good idea.

If you have the licence to own such chemicals you aren't a "layman" and if you believe that reading some old books will get stuff you cook up yourself to explode with sufficient brisance to do significant damage then I simply smile and point to this Wikipedia article:

https://en.wikipedia.org/wiki/Dunning%E2%80%93Kruger_effect

It's not difficult to learn about this sort of stuff but actually doing it is another matter and doing it well enough is even more difficult.

Terrorists in general seem have approximately zero imagination. Apparently, the overlap between the ability to think up new ways to do things and the inclination to randomly kill civilians in a bloody stupid attempt to achieve whatever political goal you have is very nearly an empty set.

Not completely empty, or everyone would still be trying to high-jack planes, but in general the fact that terrorists have not started using some specific type of mayhem is not evidence it is not possible, it is just evidence that the overwhelming majority of terrorists pick their method of attack based on what they saw on tv.

.. Uhm. So, probably, do not come up with too detailed terror plots if you ever get a job writing for televison..

Apparently, the overlap between the ability to think up new ways to do things and the inclination to randomly kill civilians in a bloody stupid attempt to achieve whatever political goal you have is very nearly an empty set.

Yes, and we're very lucky.

Terrorists feel very strongly about some political/religious ideology, but the ones with imagination usually figure out how to engage with existing political frameworks to negotiate for the change they want.

The time we see serious innovation in terror techniques is when there's a totalitarian government in play, applying selection pressure for smarts against a population with a genuine grievance. Let's remember the French Resistance were "terrorists" (and treated as such by the Nazis) circa 1940-44. Let's remember that Da'esh is the best-of-breed pathogen that emerged from Abu Ghraib when it was used as a dumping-ground for sunni resistance fighters during the US occupation of Iraq.

When I was a teenager in the 1960sI had a group of friends, all doing ‘O’ level and then ‘A’ level chemistry who liked to make explosives and rockets. Our first “bomb” using about thirty penny bangers and exploded using a fuse from one of the bangers had all the front doors for about a quarter of a mile round opened by worried householders. After that we did all our experiments in the countryside. We then moved on to homemade rockets after experimenting with cheap 4d rockets combined into multi stage launchers - three was the optimum. There were lots of ways we could make “fizzy” fireworks especially after I discovered that a railway level crossing in a local industrial area was lined with yellow nodules which turned out to be sulphur. We couldn’t buy potassium nitrate but if we has been serious and less picky we knew how to make it from manure heaps. Charcoal was no problem. Other people our age who we knew were doing similar things. We were all doing this because we liked fireworks, bangs and rockets.In the modern world we’d have been picked up by the police quickly and they would put a stop it. However if we’d been ‘terrorists’ we would have been able to make inprovised bombs fairly easily - if we had survived the trials. Luckily we mainly wanted fireworks and rockets.

Well before 300, but I saw an article on Tor about the Culture novels of Iain M. Banks.

It made me sad the number of people who value two-fisted libertarian utopias over the Culture. One commenter even noted that: See, it is only ever libertarianism for them, it is doing what they are told for everyone else.

It's also worth pointing out that, for many systems, the chemistry might be "simple," but the infrastructure is not. Two examples:

Many decades ago, my uncle gave me the blueprints for a nuclear warhead as a gag gift. Since my father had worked on them, I asked him if it was real. He said "nope," after looking at it for about two seconds. When I asked him why, he replied, "Can't tell you." So I had it up in my office for awhile. The circuit diagram was the most masterful short circuit I've so far seen, and I presume every other bit of it was similarly boobytrapped. Still, the poster made the point that a plan for a nuke is fairly simple to make. Acquiring the elements (want to get isotopically pure uranium from uranium yellow paint? Have fun, especially with the part where you centrifuge the UF6, so I guess you need a lot of fluorine too. Similarly, making heavy water from tap water takes a long time and a lot of water) is a bit trickier. Then there's the finicky task of milling a large block of plutonium safely, shaping the explosive lens, and so on... It's the massive infrastructure you'd need to build such a beast keeps us safe from every idiot building a nuke.

The second example is poison gases. A bunch of them have a fluorine in the molecule, meaning that you've got to play with fluorine in making it. Aside from the little side-issue of toxicity and keeping your chemists alive, often this isn't something that can be accommodated in a garage-sized space. Aum Shinrikyo, notorious for the sarin attack, actually built a three-story chemical factory to synthesize that sarin. They did it under false pretenses, but still. WWI may have run on chemistry, but it didn't run on guys doing desktop synthesis and pooling the results.

Actually, it's not innovation in the deadly arts that chafes my nether regions, it's proliferation. We got into that game with WWII and arming the resistance, but it really spun up in the Cold War with all those proxy insurgencies.

The problem is that everything from AK rifle designs being sold everywhere to improvised explosive and guerrilla manuals proliferated around the globe for the last 50 years, as all the "idealists" fought to either make the world safe for communism or capitalism. While the totalitarians did come up with their own spins on stuff, a lot of what we see now seems to be the street finding its own uses for stuff the CIA, KGB, Chinese intelligence, the Green Berets, and so on taught groups to use. It's intellectual hazardous waste of the worst sort.

I suspect that, in the US, part of the problem was the experience in WWII with the OSS, where there was the appearance that after the war, everybody who'd been trained to kill in various and nefarious ways simply abandoned their training as they went back to civilian life. I suppose most of the Americans did, anyway, but I think that a lot of the subsequent CIA and Green Beret leaders (mostly OSS alums for decades) got the mistaken idea that they could teach people to be assassins and guerrillas, and the knowledge would only be used for approved purposes and a limited time, then forgotten during the ensuing peace. That turned out not to be the case, and now their nasty tricks are pretty much everywhere. Probably the same thing happened on the communist side.

Actually thinking about it more, I'm not surprised. For example, a bunch of the OSS street fighting techniques were picked up from 1920s Chinese gangs, so it's not surprising that later gangs might be interested in using similar stuff.

Fireworks are not making the materials from scratch, that's repurposing stuff you can buy over-the-counter. The Boston Marathon bombers used fireworks to make their device go boom but they knew enough to put it into a pressure cooker to increase the brisance and provide a good fragmentation cloud for extra effect. Pile up gunpowder in a heap and it burns slowly if at all, it takes some knowledge and usually practice to make an effective explosive device capable of doing a lot of damage. See, for example the Glasgow Airport "bombers" of Nutcracker fame for Dunning-Kruger exemplars and "I read it in a book" terrorism fails.

You think that just mixing sulphur, charcoal and saltpetre makes gunpowder? Ummm... I knew some re-enactment folks who made their own gunpowder, it took a bit more than simply mixing it to make stuff that would go boom! in their 19th century cannons. When they put the shed they used and the wooden corning mill they ran in it on a Guy Fawkes bonfire it apparently burned very prettily, lots of odd colours and sparkles.

S is for the szyzygies of stars-come-right T 's tea's bitter comfort 'gainst the Endless Night

U for unitary transforms with fixed points in Dis V for vampires who, we know, do not exist

The chemistry is fairly simple, and well documented (for explosives, at least). Getting the ingredients is rather less trivial (you can no longer get Potassium Nitrate over the counter at the local chemist because people no longer preserve their own meat), and getting them in quantity will also attract Official Attention. The crunch comes when you get to deal with the actual process. Gunpowder mills used to explode with some regularity, and that was when they were operated by experts (using remote control). Some of the production knowledge has also been lost over the years as well, modern black powder is considerably less dense than the early 20th century variety, and probably less powerful as a result (someone discovered it was impossible to duplicate the .455 Webley Mk.1 ammunition because the powder charge alone over-filled the cartridge case, leaving no room for the card wad, let alone the bullet).

High explosives production also had its risks, and the factories were designed to mitigate those risks as far as was possible. A lot of that information will not be in any of the things you find online, and older works (Fordham, Weingart, Davis, etc.) pre-dated a lot of the safety improvements. You'd also have difficulty in disposing safely of some of the by-products without getting noticed - dissolving your drainage system is likely to annoy people.

Improvised (and obsolete) explosives will be even worse: the essential safety data will not exist (or in the case of "Armstrongs Mixture" can be summed up as "DON'T" (with a side order of "Are you completely mad?).) There's a very good reason for one of them being known as "Mother of Satan". (By now all of the precursors should be known to the authorities and a watch being kept for suspicious purchases - which makes me wonder what on earth nitrobenzene was used for in the 1960s/1970s such that the IRA could get it in quantity for use in bombs? Farms could be burgled for ammonium nitrate or sodium chlorate in quantity, but I can't think of a use for nitrobenzene that would lead to it being readily obtainable... it's inflammable and toxic just for a start.)

Burning the corning mill and its shed may well have been the safest way of disposing of it - and also prevented anyone else from experimenting with gunpowder production.

I thought most turn-of-the-century pistol ammo had gone to cordite or other nitropropellants because of the corrosiveness of gunpowder residue aided and abetted by mercuric percussion caps (Down Styphon!) Was .455 S&W the same? I had a 1917 S&W .45 Long Colt that had been converted to .455 for British military use with a mix of Kynoch and S&W brass for reloading. I used small loads of modern nitropropellant (Hercules IIRC), shot OK but I wasn't really using it for target shooting per se.

Nitrobenzene was, IIRC, an additive used to increase the octane rating of lower-grade petrol for performance engines so it was probably available for racing teams and the like in 55-gallon barrels. Lots of Irish folk were into motorbike racing then and now.

Burning the corning mill and its shed may well have been the safest way of disposing of it

The corning mill was made entirely of wood, deliberately so (no nails or other sparky metals). As it was explained to me it was driven by an electric motor outside the shed with a leather drivebelt (no static risk as from nylon or rubber drivebelts) on a wooden pulley. It was always intended the shed and mill would be burned after sufficient use. Getting it to produce decent-grade Fg powder took some experimentation -- apparently it could be weather dependent, humidity mostly.

Actually you need hydrofluoric acid, not elemental fluorine to make UF6. Not QUITE as nasty but damn close.

It was one of the things during the run-up to George Bush's Excellent Iraqi Adventure when we were fed the Dodgy Dossier stuff about Saddam's secret nuclear bomb/chemical weapons/bioweapons factories, where were all the chemical precursors, the electrical power plants, the cooling water supplies etc. etc.? Why couldn't the inspectors find a janitor or low-level security guard who could lead them to a door of one of these facilities in exchange for a big wadge of cash?

Actually yes, all the .455 Mk.II ammunition was cordite or nitrocellulose propellant. The earlier .455 Mk.I was a longer case to contain the required amount of blackpowder - and 1980's commercial powder simply wouldn't fit if you tried to cram the full military load into it. The Royal Gunpowder Factory knew what it was about. :)

I used to reload a pair of Webleys (Mk.V and Mk.VI) using (I think) Nobel 82, with a card wad to keep the small amount of propellant in place by the primer. It worked very well, though the range did get a lot of .45 card disks cut from Kelloggs corn flakes packets scattered around. (Indoors, so easily swept up.)

We were aware of the need for confinement. Note that our first attempt had a satisfactory bang. We were also aware of a lot more explosive and pyrotechnic mixtures which we tried. One of the 'others' I mentioned spent months trying to make TNT, another made home made guns with the help of the school metalwork lessons and also injured one of his eyes with another very easy to make explosive. I don't intend to go into details. Nobody was trying to damage anything apart from the gun maker who tried to shoot pigeons but never to my knowledge ever hurt one.

I think the OP was referring to chemical agents, and I'm fairly sure they don't involve fluorine. (Vague memory of a somewhat specialised phosphorus compound with an odd triple-bond being involved - very few (if any) civilian applications so is bound to be on a watch list for export controls everywhere.)

Mustard gas also has odd precursors, used in toothpaste and some industrial detergents apparently, large quantities of which were supplied to Iraq during the war with Iran (also apparently). The historic use of chemical warfare (and supply of precursors by the West) was an open secret, but Saddam had disarmed and disposed of the production facilities as demanded and there was nothing left apart from 'old stock' by the time of the 'dodgy dossier' (which is almost certainly why they needed it "sexed-up").

(I've no idea how I managed to learn all this odd stuff, or why it stuck, probably growing up during the Cold War and wondering how I was likely to be killed off.)

You really are letting your egotistical know it all out to play, aren't you? The licence I refer to is one I only reluctantly got because the government changed the law, as they often do; I have had to warn a few people about said licence because it wasn't well publicised at all and they were potentially going to be breaking the law. I only found out about it by accident.

Also your definitio of layman seems to change rather too easily, the personal anecdotes upthread indicate you need to define things better, earlier, next time.

I might have mentioned this on here before, but one of the boys in my dad's year at school in the 1960's killed himself making home made explosives, tamping it down when holding it near his body. Unsurprisingly everyone else suddenly realised that it was a bad idea to do this sort of thing.

Whilst we are at it, I thought part of the problem nowadays was that techniques deliberately taught to the military in various countries have been passed into the 'civilian sector', deliberately from the military trainers, or else by the extremists simply hiring the trained folk, or them defecting to the extremists. Wasn't that also why Iraq proved so hard to deal with, because Saddam had a whiole lot of well trained military folk who then used their knowledge to kill the occupiers?

As for the 'Techniques', there were various 'survivalist' publishers reprinting U.S. military manuals, plus things like "The Anarchist's Cookbook", "Poor man's James Bond", etc., so the basic know-how was available.

The real idiocy was the complete lack of planning (and political objectives set) for once Iraq had been defeated and occupied - the main one being disbanding the entire army and leaving everything unguarded while they did 'important' things like switching the Oil currency from the Euro to the Dollar.

They ended up with a lot of unemployed, unpaid, troops (still armed) and apparently stores of explosives et cetera looted by all and sundry.

It was "All about the Oil", with no thought being given to how to stabilise the place and get the people on their (or a neutral) side, with the result that everybody got pissed-off at the 'invaders'.

"(Imagine a swarm of these things descending on your location with five pounds of explosives strapped to each one of them!!)"

Or imagine a swarm of them descending on a big, glass office building with five pounds of explosives strapped to each of them. Each drone stops outside a window and blows up. The over-pressure probably kills everyone in the building, even without the flying glass!

I'm NOT endorsing terrorists using swarms of drones as weapons, I just wanted to point out that the idea has already been around for a few years.

But just to further ruin everyone's sleep, it just occurred to me to look up the M18 Claymore mine.

In its currently deployed version it weighs just 3.5 lb (1.6 kg).

do not come up with too detailed terror plots if you ever get a job writing for televison

When John Barnes wrote Payback City he deliberately put small technical errors into the terrorists' method so that anyone using it as an instruction manual wouldn't get the anticipated fire/explosion. (He also put in errors in the fire forensics so that if they were stupid enough to use a novel as a lab manual they could be tracked.)

A recommended book, BTW. Interesting discussion on why it was cancelled (TLDR: post 9/11 no American publisher would touch a book with a terrorist attack on the US) and what he would do differently if he was writing it now, with 15+ years more writing experience.

Just happened across the 2015 movie Eye in the Sky, a forensic examination of a drone strike in Nairobi, flipping between the Whitehall COBRA meeting, (Alan Rickman's last movie appearance as the general in charge) to Helen Mirren running the show in Northwood HQ to the US airbase and the team flying the drone. Down the road from the target compound, the Kenyan undercover is supplying eyes in ceiling with a remote control beetle. A literal bug. This is not, you will notice, billed as SF.

Probably because the US DoD already has working prototypes of miniature surveillance drones. They're already "state of the art" rather than speculation.

It's a great film BTW.

Actually, there are a bunch of anti-drone measures already deploying (reference), and I suspect we'll see a brisk Red Queen Race between drone tech and anti-drone tech in the years ahead.

Anti-drone measures currently include military lasers (finally, a use for them!), guns, net guns, jammers, and spoofers. The last are probably the most devious, as they appear to be gun-formatted directional antennas that blast the drone with a wide variety of commands to either land or return home. Since every model of drone uses a different set of signals for this, the "anti-drone guns" rapidly cycle through the list of commands.

Now, imagine a swarm of suicide drones carrying five kg of explosives, all targeted on someone. Imagine all those drones getting a very loud "system compromised, return home immediately" message. Chaos ensues...

Not that I have any interest in assassinations (other than perhaps writing bad SFF around them), but it seems to me that attacking a defended public figure with a weaponized civilian quadcopter right now could end very, very badly for whoever is piloting the drone. There will likely be small windows of time when things like signal switching and encrypted communications allow drones to have the upper hand, but I suspect that remote-controlled drones will never consistently be as dangerous as their potential suggests.

As for autonomous drones, they could be bad. Of course, if they have any link to the outside world, they can always be reprogrammed to think they are washing machines about to enter the spin cycle. The only truly "safe" ones would run on inertial guidance systems that could not be called off once launched. I think my Mom worked on those guidance systems as an engineer back in the early 1960s, too... And they're a bit inaccurate compared with GPS.

The "Anarchists Cookbook" is probably eligible for a lifetime Special Darwin Award due to the numbnuts who have blown off appendages, manipulative and/or reproductive by thinking that reading the book makes them a super-Genius when it comes to making explosives and explosive devices.

The arms storage facility at Al Qa'qaa in Iraq was ignored by the Coalition forces as they raced to secure the filing cabinets in the Oil Ministry in Baghdad. Al Qa'qaa was a complex several miles square containing thousands of tonnes of weapons, explosives, artillery shells etc. which was looted by the truckload by locals after the invasion was complete. Some of the explosives on the site were under IAEA control including HDX and RDX, useful for making effective implosion lenses for nuclear weapons. When the Iraqi guards boogied a lot of this stuff disappeared. The US Army turned up several days later and even broke IAEA seals on the bunkers containing such materials, but did little or nothing to secure the entire complex. Later the term "IED" became popular...

What happened at Tuwaitha, another place the Coalition should have had on their priority list to secure is even worse but, hey, water under the bridge...

The question is, drones carrying explosives where predicted (for non state actors!) for years. We haven't seen them yet, of have we? Why?

Is there a solid reason, or are we witnessing the hype cycle where everyone talks about an idea, then it get'S silent, and then the idea gets implemented at scale?

p.s. One idea that would fit neatly into a near-future story or current techno thriller - mini claymore hidden behind mirror, cheap smartphone looking through tiny hole in mirror with face recognition software. Could be hidden in a pub the target frequents. When the target washes their hands and looks in the mirror, boom.

Only question, does commonly availabel face rec. software run locally on a smartphone? And if not, would it matter that after the fact we find someone bought a little face-rec working time on a cloud server?

Another idea that would fit into a techno thriller - laser guided spear, arrow, or other primitive missile. A primitive four-quadrant system, like on the first guided bombs, could suffice. One person handles the bow or atlatel, the other points the laser. Pointless when guns or proper military hardware are available, but a group that want s to kill one person or sabotage something, and does not want to go near guns.

It depends what level of layman you mean. I have a variety of stuff, and the licence to own them, including potassium nitrate, without any real difficulty beyond a bit of a search online and in real life. It does require getting hold of some old books and the willingness to get your hands dirty, but I've had no training in anything from anyone and could still make a lot of them work if I was stupid enough to think it a good idea.

Avoid "The Anarchists Cookbook". It was written by anarchists & they thought it was funny to include recipes with major flaws that will bite you in the ass. That's what happened to the Weathermen's Greenwich Village (NYC) bomb factory.

https://en.wikipedia.org/wiki/Greenwich_Village_townhouse_explosion

You wouldn't use inertial navigation these days. It's all about matching features and point clouds.

The question is, drones carrying explosives where predicted (for non state actors!) for years. We haven't seen them yet, of have we? Why?

We have. Various factions in libya have hd a go at using commercial drones to drop small bombs on people. They suffer from the sort of inaccuracy you would expect from home made dumb bombs.

They don't blow the drones themselves up because they are currently too expensive and useful for other things.

drones carrying explosives where predicted (for non state actors!) for years. We haven't seen them yet, of have we?

I'm going to take this report as plausible evidence that drones built by Da'esh and used for recon and attack are enough of a concern that the RAF is targeting the factories/workshops that make them.

Face rec software ranges from piss-poor (those Japanese beer vending machines that have a camera to detect an "adult" face and can be fooled by a 100 yen banknote) to terrifyingly good (the iPhone X, which uses depth mapping and 3D to confirm it's looking at a real face, not just a mask, never mind a flat photograph). I don't think we're at the stage of being able to train it to recognize an arbitrary target from publicly available data yet, although who the hell knows what the likes of Mossad can get up to.

Wrt. laser-guided spear/crossbow: yes, that's a possibility. Advantage: it's immune to counter-sniper systems that rely on microphones to detect a gunshot, or infrared to detect heat bloom. And you could hypothetically use it with a crossbow that uses an electric or compressed gas motor to draw, with a draw in the thousand-pound range (i.e. armor-piercing), long range, or for rapid reload. A rapid-reloading CO2-cocked crossbow with laser guidance might make a very effective silent gadget for poking holes in body armour. On the other hand? Radar. Until we get the first generation of stealthed laser-guided crossbow bolts ...

AIUI, the "Roosians" are using drones in serious quantities against the Ukrainian forces in the bits of Ukr that they ( oops", not them, of course, "local militias" ha, ha, with tanks, yet ) have illegally siezed & are having a distressing level of success, to the point where a Brit genaral (last week?) said in public that our armed forces are not up to dealing with this threat-level. So the future is unevenly distributed, again, it would seem.

Unrelated question (for anyone): am looking for a recommendation for a free quick botany course, roughly undergraduate level but no fluff. (Need to fill in a bunch of gaps.) Any suggestions?

Which aspects? Botany is a HUGE topic.

The basic questions are: --What do you mean by botany? General botany 101, plant taxonomy, something else? --Where roughly do you live? With a few hundred thousand plant species in the world, nobody knows them all.

Quick is relative: most college level courses make you learn 200-300 terms, whether they're teaching general botany or taxonomy. Figure 500 hours to really learn the stuff. Of course you can take a survival class in a weekend if that's what you're after (that would be an introduction to a couple of dozen useful plants), but basically, most botany classes have terminology lists as long as most foreign language classes, and need a comparable level of work.

I mean botany 101, including some plant taxonomy/systematics. N US. Have very little trouble with biology papers (when term lookup is available), FWIW. Sounds like some time and effort is required to learn some complexities; thanks. Started poking through this on wikibooks just because it was available, but it's pretty sketchy.

Also want some (more) herbal lore (not meant as a pejorative, just distinguished from science), but that's easy to pick up. (Ancient lore less so because the species involved are often not clear from writings.)

Well, the three general options are: --Master gardener courses (I don't know if you have a master gardener program in your state, but that's one way to get a fair amount of knowledge at a not-too-technical level).
--There are at least three online general botany classes, including a MOOC. --Check your local colleges and see which ones have extension or community college. Quality varies widely, but some community college profs are extremely good.

Note that general botany has very little species-level taxonomy. It's mostly an introduction to evolution, anatomy, physiology, and the like. Plant taxonomy/systematics is generally (and best) taught at the family level. These are typically taught with a few hundred species, but that's only a few per family. The reason for focusing on families is that these tend to be fairly stable, and if you can eyeball something to family, it gives you a huge leg up on figuring out the species identification.

One thing to point out is that herbalism and horticulture often work at the population/clone level (that's the realm of cultivars and varieties), so learning botany doesn't necessarily equip you for being a great gardener or plantsman. And, if you're really paying attention, good herbalists work at this level too, because wild plants vary quite a bit in chemistry, and knowing which individual plants are the best drug sources can be important.

"Pile up gunpowder in a heap and it burns slowly if at all"

Actually it burns very-nicely-thank-you, and it doesn't have to be all that big a heap before the heap itself provides enough confinement to make it go bang. There have been innumerable accidents over the centuries from loose gunpowder being ignited by stray sparks, people dicking about, etc. Your re-enactors knew this. Although it's only a low explosive and not that powerful, it's still one of the more dangerous ones. On the other hand, it doesn't shatter gun barrels, you can make it from actual shit, and you can make it using actual cookery techniques by pure recipe-following with no understanding of the combustion physics or chemistry, which is why it's been around so much longer than any other explosive.

"it takes some knowledge and usually practice to make an effective explosive device capable of doing a lot of damage. See, for example the Glasgow Airport "bombers" of Nutcracker fame for Dunning-Kruger exemplars and "I read it in a book" terrorism fails."

See also goodness only knows how many other terrorist/asymmetrical-warfare type events where people have blown things up with conspicuous success.

I find it quite astonishing that this thread has taken the direction of arguing that the biggest obstacle to making killer drones is not any of the novel and experimental aspects, but the one aspect that people have been doing successfully since before some/all of the other aspects even existed.

The V1 got called the "flying bomb", not the "exploding roboplane"...

Heck, even the Ming Chinese at least tried flying bombs, whether or not they made them fly.

I agree though. This is the usual problem of SFF weapons: it's tricky to make an expensive "smart" munition for anti-personnel (or worse, non-lethal) uses. Tasers are arguably the last one in the real world to gain widespread use. Yes militaries are always testing novel bullets, but so far as I can tell, the number that make their way to common use are vanishingly small.

Thank you, that was a very helpful answer. Link for that MOOC?

...good herbalists work at this level too, because wild plants vary quite a bit in chemistry, and knowing which individual plants are the best drug sources can be important. Indeed. In very small part an amusing thing with a stunted specimen of a Rubus species drove this aspect of the question.

It's from Saylor. I don't know anything about it. Rather I just googled for botany MOOC and followed the links. There are also Coursera courses that, again, I don't know much about. I'm not linking because I don't want to recommend something that I don't know anything about, but if you google "botany MOOC," that's what you get.

"there were various 'survivalist' publishers reprinting U.S. military manuals"

I found it a somewhat peculiar experience reading the US Army Improvised Warfare Manual, because it gives such an impression of being written in anticipation of a second US civil war. It is apparently unaware of quite how US-centric are many of its assumptions about what sort of stuff you are likely to be able to find around the place. Chemicals of the "cor, you can get that in America?" variety, US-specific commercial products, ironmongery that no other country specifies in such a bent system of units, all crop up quite frequently.

It also seems to assume that no matter what else you're having to make from scratch instead of the army issuing it you, you'll always have plenty of blasting caps on hand. It's for the second US civil war, being fought in a quarry where the quarry workers didn't have time to grab the blasting caps as well as the explosives before they ran away...

I can't remember who it was, but it was a few hundred years ago, some time between the Norman conquest and the Act of Union, someone tried to use an off-the-shelf solution for the self-guided airframe to attack London. But because (a) they don't really fly where you want them to, (b) they don't fly very far, and (c) they tend to set fire to themselves in mid-flight, the incendiary chicken never caught on.

(Taser bullets, of course, were invented by Jules Verne, and became commonplace when someone first had the idea of throwing a charged capacitor at someone across the lab and shouting "Catch!")

Don't forget the bats

IF you are looking at genuine herbal plants-for-cooking-&-medicine, most of them are in a very few specific groups. Huge numbers in the "mints" Labatiae & the "potatoes" Solonaceae & the "wormwoods" Artemisiae. ( Excuse any sp-mistakes, please! ) because those families have evolved chemical defenses against, usually, insect-&-slug attack & those same internal compounds are useful &/or tasty to humans. Careful with the Solonaceae though ... many plants in that family are both tasty & edible & also poisonous, in either different parts of the plant - like the potato iteslf - or at different times of the year: "Kangaroo Apple"Solanum lacinatium is edible when fully ripe .... [ Mine out the front has gone mad, this year ....] As always, look at the flowers & growth-habits of these plants, as once you have "got your eye in" the similarities will visually shriek at you. The "mints" have square stems, leaves in opposite pairs, alternating in orientation & lipped flowers, often blue or white/pink. Solonaceous .. think giant tomato or chili pepper or deadly nightshade - esp the very distinctinve flower-shape & "branching into 3" growing habit. Etc .....

And good luck with that, too. You DO realise that the Rubus interbreed & vary enormously, quite naturally? For them the strict definition of "species" can get very plastic indeed. Hence artificial "new" species being fairly-regularly introduced by breeders, seeking to make "better" ( i.e. more saleable ) fruits. Loganberries are an entirely artificial species that can, more-or-less breed true etc, etc .... One of my yellow raspberries produced a "sport", or the bees naturally crossed it with one of my red ones, to produce a new, grown-from-seed plant, still with yellow fruits, but noticeably larger, that when really fully ripe, had a very slight orange tint? [ And, yes, I have kept it, obviously ]

Actually it burns very-nicely-thank-you, and it doesn't have to be all that big a heap before the heap itself provides enough confinement to make it go bang.

By "slow" I mean in terms of explosions -- oxidation takes minutes, combustion takes seconds, explosions takes milliseconds and detonation takes microseconds (nuclear isn't a chemical explosion and happens at close to the speed of light). If you can see it happening it isn't an explosion and that's the lower limit to do real percussive and shattering damage. Detonating material like the plastiques and poured explosive compounds with accelerators such as Torpex are a lot more devastating at short range as all the chemical energy expresses itself much faster since the pressure pulse is travelling through a solid material not through a loose agglomeration of particles. They're a LOT more difficult to cook up in a kitchen or a bathtub though.

Good-quality gunpowder using purified materials rather than dunghill scrapings and semi-burnt wood, properly compounded, wetted down, caked, corned to uniform shape and dimensions etc. will perform a lot better than kitchen chemistry mixes. Fireworks gunpowder is actually made to quite high standards since it has to meet rigorous safety standards and perform properly when the customer lights the blue touchpaper.

See also goodness only knows how many other terrorist/asymmetrical-warfare type events where people have blown things up with conspicuous success.

The Glasgow airport "bombers" burnt a lot of stuff but they thought that setting fire to gas cylinders would cause a rupturing explosion and it didn't, not surprisingly -- gas bottles have safety "fuses" using very-low temp melting alloy (Woods metal and similar) that lets the gas out in a fire situation. TV and movies set things up so a gas bottle fire looks a lot more spectacular.

As for the successes they tend to be carried out by folks who either practice and run tests before their spectacular(s) (The ANFO bomb that blew up the Federal building in Oklahoma City wasn't the first McVey and co. had fired off) or they were experienced pros with access to boosters like Semtex (the IRA) or military-grade materials (the Iraqi resistance) or they had pro help in cooking their explosives and building their devices (The London 7/7 bombers).

I'd be very hesitant to use laser guided anything if the other side is a first world army. I've seen how the Apache responds to being tagged by a laser and I doubt the guys on the ground lived to tell the tale. An modern army sends troops with armored support and their laser warners will identify and respond to somebody being tagged by a laser crossbow.

A crossbow bolt is probably big enough to add modern SOC, super cap, and pinhole camera, have the bow do target selection and pass a computed feature to a Haar detector in the arrow head. No AI here, so no worries.

That's a very parochial viewpoint. Most of those familiar to us are in a small number of families, but that's largely because of the small number available in the ecosystems that have influenced us (fertile crescent, mediterranean and boreal). The more fertile areas of the subtropics and tropics are MUCH less limited, and so are the plants used there, and the ones traditionally used in the less fertile areas are often very different from those we use.

Even learning about boreal culinary and medicinal plants is a fair task, and that's a pretty restricted ecology. My hypothesis is that there isn't the energy for plants to produce complex chemicals just for defence, so boreal ones tend to use low-energy solutions like thorns and cyanide - I don't know enough about the biochemistry to know why they also produce tannins and phenols.

Reading about Rubus fruticosus (blackberry) reproduction alone is enough to make one's brain hurt. Citrus (is it a species? is it a genus?) is another good one, too, as is Ipomoea indica. But that's just ones that I can think of off the top of my head. Plant sex is WEIRD.

Actually it burns very-nicely-thank-you, and it doesn't have to be all that big a heap before the heap itself provides enough confinement to make it go bang.

As a young student, I spent some time in the UOTC artillery sub-unit (then, briefly using the L5 Pack Howitzer, presumably because it allowed them to use up their remaining ammunition stocks). In order to manage the range of a shell, they used charge bags within the case - essentially, seven numbered pouches of propellant on a cord attached to the brass case; so, to fire on Charge 2 you would just cut the cord after the bag marked "2", remove the five unneeded propellant bags, and fit the shell to the case ready for loading into the gun.

Because we were firing within the UK on shorter ranges, this meant we normally had a lot of charge bags left over at the end of a day's firing. The disposal system was simple - burn it. Build a pyramid of the excess charge bags, then a trail of charge bags leading to it, then cut a few open at the end of the trail. Set fire to some of the propellant, and retire to a safe distance.

I still vividly remember the amount of heat that came from the blaze, and the sight of a rather impressive vertical flame. In my memory, you could feel the heat on your face and hands from well over 50m away.

Nojay@116 or they were experienced pros with access to boosters like Semtex (the IRA)

One of our platoon went off to the Regular Army; while on a rural patrol in Northern Ireland, his patrol stopped briefly, and so took their rucksacks off their shoulders. His turned out to be resting on top of an IED; when the man from PIRA fired it, the booster went off but the main charge didn't, his rucksack absorbed the blast, and he got thrown across the road but survived. Lucky...

I have never done that to more than the contents of a single shotgun cartridge :-) But I find this debate surprisingly blinkered. The converse is that you don't need anything normally classified as explosive to make an impressive explosion. If I recall, you can do it even with baking power (just), and quite impressively with a suitable base and acid. All that is required is to confine a suitable self-sustaining reaction in a suitable container. Such approaches are of little military use, as they produce relatively little bang for the hundredweight or cubic foot (the scale on which they start to be effective), and it's difficult to predict exactly when they will go off. But I can think of many uses in terrorism :-(

And then there are also 'firebombs', which ARE used by the military, but which are more effective against non-military targets, and occasionally by terrorists. They don't go bang, either.

Modern compound cross-bows have much lower draw-weight than their medieval cousins. But they have a much (much!) longer draw distance, so can spend much more time shuffling energy into the bolt. This leads to something that is easily cocked by hand, while providing a frightening amount of pointy kinetic energy down the range.

Just had a quick look at a random modern example, 150 pound draw weight, 330 FPS bolt speed. I am not sure how similar it is to a model I've seen demonstrated, that happily went through bullet-resistant material designed for pistol/revolver protection.

I was under the impression that ANFO is a reliable explosive* and relatively easy to make. Am I wrong?

*Technically "blasting agent", I know, but it still explodes when triggered.

Even a longbow shaft can do that. Making something that will go through even light vehicle armour (e.g. on helicopters) is trickier.

As far as drones are concerned, I would expect that Da'esh factories are modifying commodity drones for warfare, rather than actually making them from scratch or even parts. It's only recently that even the leading industrial powers have been able to make reasonably effective ones.

ANFO is used commercially in quarrying and other extractive industries. It's a liquid(ish) slurry and needs to be well-mixed before being pumped down a hole. It is very inflammable and requires careful handling during mixing and pumping but it generally not that shock-sensitive from what I remember. The booster charge is usually gelignite, a stable low-medium explosive initiated by a detonator or six (detonators are cheap, recharging the hole costs money). It's not easy to set off a barrel of ANFO slurry, it requires more than a cartoon-style fuze spluttering up to the lid while Wil. E. Coyote hides round the corner, fingers in his ears.

Modern fertiliser mixes which contain ammonium nitrate have suppressor additives to prevent its easy use in ANFO charges and the authorities tend to keep track of mass quantities of the stuff these days for obvious reasons, just as they track other high-energy chemicals and some precursors.

The Oklahoma City bomb was over a tonne of ANFO if I recall correctly but it did a lot less damage than a "real" bomb of similar mass and better explosive content would have. This is generally true of most "asymmetric" home-made bombs. The Iraqi resistance had access to top-quality materials like 155mm artillery shells to make their IEDs; a typical charge in one of those is 10kg of TNT in a fragmentation case which will ruin anyone's day.

TNT? THat suprises me - I would have thought that Ammonal or its modern equivalent would have been better - after all, RN explosive shells usually contained said mixture in WW II ( didn't they? )

And totally off-topic, but as I know there are a few gamers here (and a few Scots):

https://www.kickstarter.com/projects/ragnarbrothers/darien-apocalypse

NOT that "parochial" I mentioned the Solonaceae didn't I? Most of this family come from the "New World" & are tropical, sub-tropical or tropical-montane in origin. [ Potato, "normal" chilis, tomato, Physalis sp, etc ] I admit that Atropa belladonna is European, but ....

My point was that THIS sentence was parochial: "IF you are looking at genuine herbal plants-for-cooking-&-medicine, most of them are in a very few specific groups."

There are a zillion such plants used in other countries belonging to other groups, but we don't use them much, not least because they don't grow well here (or often at all). To mention just five families that are important sources of starch in some other places: dioscoreaceae, convolvulace, araceae, chenopodiacea and musaceae.

Chenopdia were much more important here, once. However if you are looking at food in bulk, the number of species, never mind families is 'orribly small: Grasses ( Wheat-to rice ), spuds, beans ....

In bulk, I would agree with you. Something less than 1% of species commonly eaten (let alone used for medicine) have been commercialised on a larger scale than families of farmers growing them 'by hand'. But such farming is still very important.

OK, this is (or rather, might be) fun. Will keep it short. Scientists Have Built a 'Hallucination Machine' For a Drug-Free Brain Trip A Deep-Dream Virtual Reality Platform for Studying Altered Perceptual Phenomenology (open, 22 November 2017) So UK people; I'm presuming that this would be allowed because it is not substance-based. Would UK lawmakers go out of their way to ban it and things like it? First, we show that the system induces visual phenomenology qualitatively similar to classical psychedelics. In a second experiment, we find that simulated hallucinations do not evoke the temporal distortion commonly associated with altered states. Overall, the Hallucination Machine offers a valuable new technique for simulating altered phenomenology without directly altering the underlying neurophysiology. ... The Hallucination Machine enables systematic and parameterizable manipulation of distinct aspects of altered states of consciousness (ASCs), specifically visual hallucinations, without involving the widespread systemic effects caused by pharmacological manipulations.

Plant sex is WEIRD. Why qualify it? Sex is weird. :-)

You may think that animal sex is weird, but you will find that it becomes positively straightforward when you learn about plant sex :-)

My condolences on the death of your father. Mine died almost 20 years ago and I still have moments when I see something I would like to share and then remember that I can't. I am glad, in some ways, for these moments since they keep memory from fading.

Some questions, I am interested in others' views

• Why are A.I. controlled military offensive drones worse than person controlled drones? I can see that such drones could be seen as mobile mines, able to injure civilians long after a conflict has finished, but I would be interested to hear what I have missed.
• Are human-controlled military offensive drones a "bad thing"? Some clearly believe that offensive drones are intrinsically bad - indeed at least one film has been made with that point of view. I don't share this belief but would like to understand why it is held to see if I change my mind.
• Is it just offensive drones or is any use of drones by the military undesirable?

Looking forward to reading the new book and have it pre-ordered.

My condolences on the passing of your father. Please take as much time as you need, we'll still be here when you're ready to resume regular blogging. Lucky for you (but not for the human race), there's oodles of completely asinine nonsense that's going on for you to blog about.

Just pre-ordered Dark State. Nice to have a good read to look forward to post-holiday season. (Great review btw!)

Re: Oops! Boom!

Not sure whether Charlie has already written about this or if anyone else here has already mentioned it.

https://en.wikipedia.org/wiki/Main_Page

'1944 – Between 3,500 and 4,000 tonnes of ordnance exploded at the RAF Fauld underground munitions storage depot in the largest non-nuclear explosion in the United Kingdom.'

Phooey, there are plenty of hybrids on the animal side: humans, for instance. If you believe that the paleonanthropologist's Homo species weren't mutually interbreeding, then why are there Neanderthal (and Denisovan, and other) genes floating around in human DNA? Dogs are up there too (ever try to figure out why dogs can cross with coyotes, wolves, and/or jackals?). I agree this has nothing on hexaploid bread wheat (genomes from two Aegilops species and one Triticum species), but it's not that shabby.

That's not to say that plant sex isn't weird, but the most fundamental weirdness (the sporic lifecycle) is something most people don't get. Which reminds me, I really want to set another story in Nysus, the world I built where the animal analogs have a sporic lifecycle. It's fun to play with, and to my knowledge, only one author has ever used it in a story.

Re: Drones - biologicals

Since no one's mentioned biologicals and chemical agents as the drone payload - this should also be up for discussion.

Yes, biologicals and chemicals are illegal, but we're talking terrorists whose mantra is 'death to everyone who is not a member of my sect'. My understanding is that current war tech seems solely built to detect regular (boom!) type payloads and not biologicals/chemicals. Since the weapons industry is always looking for new ways to make $$$, I'm guessing this is already in their R&D pipeline both as offense and defense/detection.

Of course, these drones could also be used for peaceful, humanitarian purposes such as delivering urgently needed vaccines to some remote village that's experiencing a deadly outbreak.

"there were various 'survivalist' publishers reprinting U.S. military manuals"

I found it a somewhat peculiar experience reading the US Army Improvised Warfare Manual, because it gives such an impression of being written in anticipation of a second US civil war. It is apparently unaware of quite how US-centric are many of its assumptions about what sort of stuff you are likely to be able to find around the place. Chemicals of the "cor, you can get that in America?" variety, US-specific commercial products, ironmongery that no other country specifies in such a bent system of units, all crop up quite frequently.

Most of the US Army material was derived from OSS "lessons learned" during the Second World War, which were in turn informed by the experience of the British SOE, adapted for the different materials that were available in Cold War USA. They're based on an out-dated assumption there could be a communist invasion of the US mainland.

They were meant to be used by small groups of US soldiers who would be staying behind in occupied areas to lead a covert resistance mainly made up of civilians. Those manuals were to "train the trainer".

It also seems to assume that no matter what else you're having to make from scratch instead of the army issuing it you, you'll always have plenty of blasting caps on hand. It's for the second US civil war, being fought in a quarry where the quarry workers didn't have time to grab the blasting caps as well as the explosives before they ran away...

The manuals were originally written in the 50s and early 60s when it wasn't that hard for US civilians to obtain explosives. And while they do assume the ready availability of commercial explosives and blasting caps, they also included instructions how to improvise them if you didn't have them.

By the early 70s those manuals were considered obsolete by the US Military, because conditions had changed and the assumption that civilians could readily access the required materials was already being widely challenged, as was the idea that Russians/Cubans/Chinese would actually mount an invasion.

I was under the impression that ANFO is a reliable explosive* and relatively easy to make. Am I wrong?

No, you're correct. But the base ingredient, Ammonium Nitrate fertilizer, is somewhat more tightly regulated since the Oklahoma City bombing.

Presumably based on research done by M Persinger & others of/on the so-called "god helmet" & inducing religious experiences through electro-magnetic brain stimulation?

There's a very large crater, which is, only now, begfinning to be overgrown. Should be in the bottom-centre of this map link: https://www.bing.com/maps?osid=69b6b629-e29a-4c3e-a406-cf9510c23a24&cp=52.85519~-1.733538&lvl=15&style=s&v=2&sV=2&form=S00027

Why on earth did you think I was talking about mere hybridisation?

Rubus fruticosus reproduces vegetatively, apomyctically, sexually (with itself) and sexually (with other plants) - that's rather more modes than any animal I know of can do. Citrus is sometimes 'interestingly' polyembryonic. Ipomoea indica is self-infertile, but will cross with other species.

Nojay is your actual expert on certain parts of this topic, but...

WW2 (and WW1) both suffered from severe material shortages, particularly explosives. In WW1 all manner of compounds were packed into shells (and later, bombs), from standard black powder (shrapnel shells where you need a 'push' rather than a detonation) to semi-improvised mixtures of nitroglycerine and nitrated sawdust. The Germans ran low on glycerine and started up a "corpse conversion unit" to produce it (they actually used animal remains but that didn't stop tha Allied side making use of it as propaganda).

By WW2 most of the ineffective explosives had been abandoned, along with the "really a bit too sensitive for military use" things like picric acid, but there was still a shortage of decent high explosives. TNT could be mixed with ammonium nitrate to make Amatol which is nearly as good as straight TNT for most purposes. Fragmentation grenades were filled with ammonal (ammonium nitrate, aluminium powder, and charcoal) as you want reasonable-sized fragments and not powdered cast iron when it goes bang. Some things required the very best available: RDX, PETN and TNT for antitank shells and mines because you need the maximum possible energy and detonation velocity to attack the armour with. Different explosives have different characteristics and it's a matter of matching the attack to the target.

Modern HE shells ("common" shell) are filled with a mixture of RDX and TNT to get the best bang at a reasonable cost. HEAT (shaped charge anti-armour) shells, warheads and mines tend to use RDX or HMX (High Molecular Weight RDX, I think) which is rather more expensive to manufacture but considerably more powerful.

Horses for courses and all that.

Chris (I can still remember large chunks of "Winston Churchill's Toyshop" and "The Secret War" (MD1 and DMWD respectively). Yikes!)

RDX and HMX are both cyclomers (hence the name "cyclonite" for RDX) of CH₂NNO₂ units; RDX is a trimer, HMX is a tetramer. (Also known as "octogen" for its 8-membered ring, just to confuse Discworld chemists.)

As for animals that reproduce clonally and sexually, check out corals. If you want virgin births, it's also worth checking out various sharks and komodo dragons.

Citrus polyembrony is kind of weird. I guess the parallel would be armadillos, which generally produce identical quadruplets with every litter.

I'm only acting blase--plant sex is neat. If you want some real fun, you can delve into the differences between animal hermaphrodism (where animals duel with each other to force the loser into the female role) and plants (which do not, apparently, do this). Also, there's the fun with fern gametophytes, which decide whether to become male or female depending on the genders of the gametophytes around them. They emit hormones and decide based on some sort of sensing of the hormone gradient. The human parallels are rather interesting: deciding which gender to become at a party based on who else is there is a bit harder for humans, although not quite impossible. Actually, ferns in general are fun. If humans had fern-style lifecycles, would it be better to be a spore-producing human, or a gamete-producing human? I've actually asked this question to students in class, and it's quite a useful question, actually.

Argh! A Doc Smith flashback: "If (X's) brain was solid cyclonite the detonation wouldn't even crack his skull".

You can even make a biological bomb - water, sugar and yeast in a sealed bottle. Not very powerful, but it does fling glass everywhere, along with a horrible sticky mess. My dad used to make them unintentionally by bottling his home-made wine before it had properly stopped fermenting, so we used to have random explosions at long but irregular intervals in the cellar.

"But I find this debate surprisingly blinkered."

It's bizarre. I'd have thought that aspects like the guidance software would be the obvious parts of the killer-drone concept to pick over. But it's an interesting demonstration of how deep-seated various kinds of "magical thinking" are.

My dad used to make them unintentionally by bottling his home-made wine before it had properly stopped fermenting, so we used to have random explosions at long but irregular intervals in the cellar. I used to make them intentionally as a kid. "Yeast bombs". IIRC worked out by self - very pre-internet. It was pretty silly and bad (broken glass) in retrospect.

It was fun, though, wasn't it :)

I became famous at school (and apparently remained so for several years after leaving it) for putting one in a teacher's desk. I did not actually do this - nor did anyone else, it never happened - but that didn't seem to make any difference.

...so-called "god helmet"

No, it's VR (I did include links): It comprises a novel combination of two powerful technologies: deep convolutional neural networks (DCNNs) and panoramic videos of natural scenes, viewed immersively through a head-mounted display (panoramic VR). ... We extended one such implementation to optimise the hallucinogenic properties of the video. In our extension, the optical flow of each frame is calculated by comparing the difference in the movement of all pixels between the current and previous frame. The hallucinatory patterns from areas where the optical flow was detected is merged to the current (not-yet-hallucinatory) frame based on the weighting provided by the blending ratio. The Deep Dream algorithm is then applied to this merged frame.

https://en.wikipedia.org/wiki/DeepDream https://deepdreamgenerator.com/

I'm not sure that they have thought much about the uses that this could be put to - with tweaks, both good and bad uses. A few of each are coming to mind as I type; use your imagination. e.g. triggering or treating phobias, helping users induce unusual mind states, etc etc etc.

I guess the parallel would be armadillos, which generally produce identical quadruplets with every litter. I'm stockpiling some of these to use at parties to derail odious conversations. :-) Thank you and EC for them. Just because it turned up in a search and is interesting: Polyembryony in Armadillos Odd as it may seem, the "uterine- constraint" hypothesis for polyembryony in Dasypus armadillos might bear some analogy to evolutionary conditions that apparently promoted polyembryony in parasitic wasps.

Anyone remember Harry Purvis and the Osmotic Bomb?

https://en.wikipedia.org/wiki/Moving_Spirit

Something no-one seems to have mentioned, and which I won't describe in too much detail for obvious reasons, is that the ultimate asymmetrical warfare drone wouldn't target people, but infrastructure. In doing so it wouldn't need explosives at all. A small saw or drill would be more effective. As such it could do damage, then go off and hide somewhere while recharging its batteries from solar or power lines, then when fully charged, go off again and cause more mischief. Not a suicide robot, but a sabotage robot.

The Casey Newton character in Tomorrowland does drone sabotage (with non-autonomous drones) early in the movie, so it's an idea that is out there, but agree on no details. (Tempting, tempting :-) Goofy movie but liked the relentless argument for optimism (my style :-) and the straightforward sci-fi treatment of timeline [redacted - no full spoiler].

Terrorists in general seem have approximately zero imagination. Apparently, the overlap between the ability to think up new ways to do things and the inclination to randomly kill civilians in a bloody stupid attempt to achieve whatever political goal you have is very nearly an empty set.

It is a common feature of more criminals. My brother in law in law retired cop said police work would be so much harder if 90% of the crooks were not so stupid. For that 90% policing is more of a herding issue. It's the other 10% where they have to work hard and think.

The movie "October Sky" was on TV around here recently. I kept thinking how today the FBI and other TLA's would be all over them instead of them winning a science fair.

Acquiring the elements (want to get isotopically pure uranium from uranium yellow paint? Have fun, especially with the part where you centrifuge the UF6, so I guess you need a lot of fluorine too.

Yep. My father was a production manager at a nuclear fuel plant. Back when gaseous diffusion was the way to go. Size of the plan was just plain enormous. And the power required unreal. And that was to take UF6 from natural concentrations of U235 up to 3%. Another plant created the UF6. And they shipped their output to Oak Ridge to for them to get higher concentrations. And leaks were not fun in the gas side of the lines.

Centrifuges are "easier" but still require huge plants full of precision machinery. Just not so much of it as GD. And instead of ginormous power requirements, just huge amounts.

But if you want to impoverish and starve people to pay for it and have a country to work with, yes you can do it.

Gives sabots a whole new meaning. Still, considering what we Americans do to our own infrastructure, it's probably cheaper (and a lot more legal) to, say, reward greedy politicians to do nothing about infrastructure than it is to attack infrastructure. We're at a point where neglect will do as much damage as sabotage. Why blow a bridge that gets an F structural grade due to rust?

Thematically, I think the bigger problem is that physical hacking might become as easy as on-line hacking is, with everything from the Internet of Things to whatever ends up in the massive recycling streams we'll need to keep civilization from collapses.

Terrorists and imagination: probably most terrorists with imagination are candidates for Darwin Awards.

Terrorists have shown plenty of imagination in environments well-adapted for this. Ask the Coalition of the Willing in the Iraq War, NATO in Afghanistan, and whoever has been fighting ISIS. To suggest otherwise is arrogance bordering on racism. Contrary to conservative propaganda, the West is a much more hostile environment for a terror cell than a failed state.

While there are possible new avenues that are possible, most "imaginative" avenues are pure nonsense.

I am not going to go into details, but I will give vague possible problems.

The sushi example illustrates how fraud-prone the whole system is http://newsroom.ucla.edu/releases/bait-and-switch-ucla-study-finds-fish-fraud-runs-rampant

If it's so decrepit, then how does it work? What makes the system work is that each stage in food processing automatically treats the previous stage as suspect, and probably checks it internally. I mean, the system already has to check for e-coli due to negligence or criminality, so terrorists can't just introduce e-coli to food randomly except perhaps at the end of the supply chain?

Yeah I can just imagein the christian Rethuglicans using it on their followers & any uppity wimmin that got in theor way, on the road to Gilead.

A tool that powerful will have very bad as well as very good uses, won't it?

And aren't armadildoes, oops, err ... CUTE?

It's the other 10% where they have to work hard and think. Which explains why they are often called "Plod" of course ... and why, all-too-often, when something really unusal happens, or someone who doesn't "fit" their very restricted mental map of characters appears, they pick on the wrong person. E.G The murder of Joanna Yates, & the unfortunate death of a young epileptic in Dorset very recently.

It's bizarre. I'd have thought that aspects like the guidance software would be the obvious parts of the killer-drone concept to pick over.

We all know that would just degenerate into a language flame war....

Decent open source auto pilot software which handles the basics of movement and control is available off the shelf and on github.

There's also free computer vision software out there, but last time I looked it was a few years behind the state of the art.

Photogrammetry and automatic mapping is a key part of most AR libraries. Portable hardware is just about up to the job but it is pushing it on current gen kit. MS hololens does a decent job of point cloud generation and tracking with multiple cameras and a pile of DSPs, phones do an acceptable but relatively mediocre job with one camera and a GPU.

My guess at a minimal system would be an off the shelf autopilot using GPS to get within a couple of hundred m of the target, then switch to photogrammetry to navigate around obstacles and whatever pattern matching is deemed appropriate for the final approach.

All these bits have been demonstrated separately, some of them still need expensive amounts of computer.

All these bits have been demonstrated separately, some of them still need expensive amounts of computer.

Yeah, I've been looking a bit at autonomous things, for hobby projects, and it seems to me that the easiest way to get enough computing power is to use a local thing to connect to a bigger computer and do the computations there.

Get it working on a laptop computer then wait for moores law to catch up...

TBH I think that Carmack is probably about right - he reckons it is reasonable to expect an order of magnitude speedup on mobile hardware in the next few years, but not necessarily two.

I think one order of magnitude faster than a current iphone is probably sufficient.

If I want to do something cheaply now, waiting for some years is not fun. Though I might be waiting anyway, because of the Too Many Things To Do problem.

The inescapable problem that has killed all my projects recently. There is a reason why students write all the cool demos.

Onboard Pi Zeroes seem to be the current favourite for heavy processing, talking to an Ardupilot for actual control.

Apropos of nothing. WTF?

https://www.aldi.co.uk/reka-decorative-santa-cam/p/080878177684600

You seem to be quoting what I call Not-Moore's Law. Clock rates and, effectively, serial computation speeds topped out in 2003, because the power demands got out of hand. What has happened since then is an increase in SIZES (because Moore's Law still holds), which has been used for larger caches, internal tables etc., more complicated hardware optimisation, and increased parallelism. Oh, yes, the polemic SAYS the new systems deliver higher speeds, but they really only deliver higher benchmarks, which are mostly an artifact of the increased sizes. The point here is that it depends very much on whether the relevant software can take advantage of the larger sizes, or whether it needs genuinely faster logic. In practice, it's in between, but anywhere on the scale.

Moore's Law is reaching its limit, and is slowing down. The cost of fabs is going up exponentially with each reduction in process size; it's unclear how far that will go before even Intel (effectively) calls a halt. That is why Carmack says what he does (and I agree), but those factors are sizes and may or may not map onto speed.

Power consumption/thermal performance is more important than speed in mobile stuff at the moment.

I mentiond that current phone hardware was just about adequate, but what I didn't state is that it is actually adequate for a couple of minutes before temperature sensors start to trip and clocks slow down.

Faster would be good but the ability to maintain current peak performance is probably more important right now.

I didn't know that about ferns! Another example for Bill Arnold that I discovered by accident is the yak's bizarre chromosomes. Apparently, the Y chromosome has attached to an autosome, so males have one fewer chromosomes than females; it is why dzo are sterile.

I realise that I didn't mention it so, in case you missed it, with polyembryonic citrus, that one embryo is the result of fertilisation, and the others are clones of the female parent.

And aren't armadildoes, oops, err ... CUTE?

As long as you can't smell them.>/p>

With a human operator, there's a chance, at least, that they may realize that those are civilians, not the target they were expecting, or that the target is surrounded by dozens of civilians, esp. women and children. The AI won't. Autonomous killing machines are like landmines on propellers.

And the difference between someone playing with an arduiino and a country's military is tens of thousands of time more powerful, and more of them.

I'm sorry, you completely forgot about everyone on the ground outside the building being sliced and diced, as the class falls.

There was a time in the, um, early oughts, in Chicago,when, two weeks after I left, a woman was killed in front of her 10-yr-old when a window fell out of an office building from x floors up.

Coulda been me.

Here goes my rant, again, about terrorists and generals working from exactly the same playbook, "terrorize the populace, and break their will to resist", when 100% of the time, the effect is the reverse of that. Therefore, it's worse than stupid, it's counterproductive.

And when it's terrorists, odds are that they have no way to do a full-on military attack... so what they're allegedly fighting for gets trashed. And it's far too small a scale for even decent vengeance.

They're all stupid.

Y'know, y'all are all fixated on flying drones, and how little they can carry has been pointed out.

Now, if I were a terrorist, I'd buy an old-style heavy-duty remote-controlled car from a hobby story, put 2.5-5kg of some explosive on it, watch from a window with it parked under a parked car, and drive it out as the motorcade passes.

Or 10 people would, so you could hit the whole motorcade, including the fake ones (carrying a double). Or do it via cameras and remote control boosters from a mile away....

But, like, y'know, that's not Sexy....

Well, to be sure, a lot of cops are out there looking for, and making sure they find trouble (c.f. Black Lives Matter, or, as most US blacks know, DWB (driving while black)).

But this is true. It goes with a piece I read around '79 or so, in the Philly City Paper: most of them are not smart enough to get and hold a regular job. The example the story gave was the guy who beat another guy up for a cigarette, that the second guy wouldn't give him. And this was while they were in the holding tank, with cameras on them, at a major downtown Philly police station....

There might be less... but the day labor job market is really minuscule, and the morons of HR depts (and I'll have to deal with one this afternoon, they're performance pages being designed by morons) think you need degrees and/or certifications for ditch digging....

You realise that the "terrorists are stupid" argument in #179 is the reason why "hobbyist" terrorists are likely to try this stuff even if militaries agree not to.

It's always possible that when this sort of thing is inevitably tried it turns out to be so comically ineffective and easy to counter that the idea is dropped. Not a given though.

RE: Unimaginative terrorists. Yup. I am forever grateful. Talk to a civil engineer about the sorts of vulnerabilities that a modern society is riddled with and you won't sleep at night. But actually you should feel better. The way I see it is like this: you find out the locks on your doors don't work and anyone could have walked in and murdered you in your sleep for the five years you've been in the house. Well, were you murdered in your sleep? You'd know if you had. So while it is a terrifying threat, it is not a common or likely one in your neighborhood. You're pretty safe there.

The moral of the story is that the occurrence of people with the motivation and means to wreck our shit is very low. E. coli on our spinach kills more people every year than foreign terrorists.

Re: truly getting motivated terrorists/freedom fighters. You're right, the people with imagination and smarts can usually find a way to engage with the society to get what they want. The problem comes when you have educated, bright people who want to have whatever is the the good life in the country they live in but are thwarted.

This seems to hold true: the revolutionary threat doesn't come from the underclass but from thwarted intelligentsia, the educated professionals who should be near the top. When there's not enough room at the top, when there's not enough jobs, opportunities, when these bright people who feel they've done their part to earn a seat at the table are denied, they're left with nothing but time and resentment on their hands.

We had a wonderful illustration of this with the second Iraq War. We demobilized the entire army in an effort to purge baathists. All this did was create an immense talent pool for terror groups to recruit from. Even if half these guys were naughty Saddam lovers, it would be better to have them inside the tent pissing out than outside the tent pissing in. it was pretty clear that their country had been invaded, taken over, and the government installed was a puppet of the US and there was no interest in establishing local rule. Shoe on the other foot and it happened here, any conservative, red-blooded American would fantasize about bringing war to a puppet government of a foreign state. Why would you imagine the Iraqis to be any different?

And can carry leprosy.

Oh, yes, it is! Our wonderful gummints (USA and UK) are both allowing our roads to be used for autonomous cars. As soon as they are available, just think what you will be able to do!

I really want to set another story in Nysus

Yes please. :)

Thanks!

If I do it, it's going to be a "redo the universe, save the world" situation, so the planet will be sort of the same, but the humans will be totally different, due to stuff I've learned in the last 10 years.

Actually, Nysus will be fairly different, mostly due to everything I learned writing Hot Earth Dreams. It turns out a habitable moon orbiting a gas giant with a livable-but-high eccentricity around the primary isn't perhaps the wisest place for humans to settle, even before you start thinking about things like how such a moon's orbit might precess around its primary (cf: Triton around Neptune). Still, Nysus was such a fun world that I do want to play with it some more.

Still, I can see the argument, as the colonist's ship goes into orbit around Nysus: --Colonists: "But that moon's so green, and the radiation belt around it is so high. We really should land and settle. --Astronomer: "But these orbital parameters suggest that we really need more measurements to make sure that..." --Colonists: "We heard you already. Your complaints are noted, but we've already voted to land, Astronomer. You can make your measurements once we're down. It's much safer down there. Don't worry so much. We can handle it."

Since I've had this experience a number of times now, as an environmentalist speaking up against rather stupid projects, I figure it's a scenario that will play out any number of times, if people colonize other planets.

Yessss See also Charlie's car-hack in (I think) "Rule 34" - it's going to be so (relatively) easy. More of the "internet of things" waiting to be screwed-with by greedy/random/ mischievous/stupid/ terrorist IDIOTS, killing people .... Tell us again why "smart meters" are a bad idea?

Talk to a civil engineer about the sorts of vulnerabilities that a modern society is riddled with

Wasn't there someone in the US shooting substations with a light armour-piercing rifle at one stage? A little bit of effort from someone along those lines could could cause all sorts of problems. And less armed countries shouldn't be smug, you can get 90% of that result with a big spanner and a bit of grunt work... electricity pylons are bolted down. Or at the risky end, a battery powered grinder will take out the support cables for just about anything (sulfuric acid will do the same since most cables are steel, it's quieter but slower... and readily available).

Look at what protesters do, and think about how polite and non-destructive they choose to be. Yes, sometimes there's minor damage as well as disruptions, but think about what would happen if they were actively trying to sabotage stuff. That's one reason why some cops are so paranoid, because infiltrating a protest is so easy even the cops manage it. Imagine one of those "occupy the nuclear plant" protests, except one of the backpacks is full of ANFO.

I've been to an awful lot of protests, and the stuff you could do if you wanted to is terrifying. But as someone above pointed out, no-one does those things. And often the worst events turn out to have been the cops or whoever is being protested against trying to turn public opinion.

Todays C++ whinge is that so much of the standard library throws the useless-by-design C++ exceptions. Which means that even in single-threaded applications you end up littering your code with exception handling because the powers that be decided that providing useful information with exceptions is not the C++ way. It's notable that there are a number of "extend the exception class to include a stack trace" type libraries, but none of them are useful because only your code will throw them... the exceptions you care about will be std::exception and thus uninformative "Something went wrong. No more information is available" ones.

It makes ye olde worlde C style error handling seem efficient and concise by comparison. Instead of 5 lines per std:: library call (to catch the exceptions), you have two. The call, and the "if it failed" line. What's the advantage of C++ exceptions again?

Actually, the more clever non-violent actors assume that they're being infiltrated, and often share their most of their planning with the cops. The problem is that most sorts of operational secrecy are antithetical to mass non-violent action, because trust is important. It's important for the authorities too: I suspect that most dictators and their goons would rather (if forced to choose) be overthrown by non-violent protests than by violent rebellion. With a non-violent overthrow, they and their families get to live, something that's far from guaranteed under violent overthrow. Adding a secret cabal plotting behind the non-violence makes the situation a lot scarier and more dangerous for everyone. Secrecy did work in the case of the Serbian uprising, but it seems to be a minority. AFAIK, that only worked because it kept the Serbian authorities from finding out who the chief strategist was among all the people that got arrested, because he wasn't one of the leaders.

Note that this does NOT apply to the kind of non-violent action we've seen the Kremlin promulgating (as in the last US election). This simply points out that, as with violence, there are many types of non-violence.

Getting to massive sabotage, I suspect that crippling a city with a cyberstrike isn't much above nuking it on the scale of military actions. People dying over a few days from lack of water and sanitation is a longer, slower death than getting fried*. I doubt most politicians that really think it through will ever class infrastructure-crippling attacks as anything other than an act of war.

*Yes, I'm aware of how most of the Japanese died after Hiroshima and Nagasaki. That you can produce something similar by taking out central infrastructure only makes the point that such cyberweapons are lethal weapons, not less-lethal ones.

ften share their most of their planning with the cops. The problem is that most sorts of operational secrecy are antithetical to mass non-violent action, because trust is important. It's important for the authorities too

There is a lot of amusement in watching PTBs negotiate with each other and ask us to selectively hide information to prevent the idiots in charge FUBARing an otherwise harmless situation. Also, once, trying to publicly cancel something that was already very heavily advertised. That was more disruptive than the protesters were hoping the protest would be.

In one case (J18 in Sydney) someone in the Police decided that there would be no Critical Mass ride as part of a big protest so they harassed the few hundred riders who turned up. Riders eventually got sick of it and rode onto the Sydney Harbour Bridge (the organisers promised not to... but they were targeted early on so had left the ride). Cops arrested the "leader" (ie, the one physically at the front), at which point everyone sat down. Blocking all lanes for about 30 minutes at rush hour on Friday. Normally Critical Mass crosses the bridge using one side for about 10-15 minutes (it could be done in one or two lanes but some motorists murder cyclists in that situation, so no motorists are allowed the opportunity). Eventually someone with a brain decided that un-arresting the person and asking everyone to ride back into the city to attend the party we were all wanting to attend... was better than blocking the bridge until they could arrange a truck to carry 200+ bicycles off the bridge (plus buses for the arrestees, obviously).

The amusing thing about Critical Mass in most places is that it is entirely legal. There's no rule against a mass of people all using the road at the same time (laws against rush hour?) and in Australia generally cops on bicycles turn up and actively instruct cyclists to go through red lights etc, in order to keep the mass together. Because the one thing worse for traffic flow than 200 cyclists in a big group is 10 groups of cyclists all riding round looking for each other. They did that in New York once, I recall, and it took hours for traffic to calm down.

Secret cabals can work excellently well and in fact most mass protests are more accurately large collections of secret cabals who share goals but organise independently. I've often been surprised at who turns up to events I help organise (RTS featuring... a float from the Sisters of Perpetual Indulgence? Where did they come from?). That's one thing that makes them very hard to control... whether you're the official organising cabal or merely some PTB numpty. Mass arrests have to be pretty mass, as in Russia, or they end up just expanding the pool of angry people. I suspect they do that regardless, but with enough arrests people are scared as well as angry.

Hmmm. There's an instructive difference between one-off protests and non-violent campaigns (here I'm speaking from book learning rather than experience, but as I'm finding out, non-violent actionists really don't spend much time learning strategy and developing tactics and counter-tactics, unlike, say, their counterparts in the military, politics, and law enforcement).

Basically, one-off protests do haul everyone out of the woodwork. So do riots. The basic difference is whether they're violent or not.

However, one-offs seldom accomplish anything. If you want to run a campaign to, say, throw out a corrupt government, you've got to organize it as strategically as you would a violent campaign. Secrecy is more problematic in a non-violent campaign because the advantage is that you can mobilize just about anyone to act. You don't have to be fit or violent to participate. However, you DO have to give everyone a reason to participate, and this is where good, simple, clear communication is critical. Discipline is critical too, both for optics (something the antifa need to learn) and because violence tends to attract violence, and a lot of people won't participate if the dude next to them is acting like a violent baton or bullet magnet.

Secrecy has some small use if you've got some people who are really good strategists but not good leaders: you don't want them selectively taken out, because then your movement falls apart, and you don't want them all together in the same room for the same reason. That's about it. Reassuring the police infiltrators that you're non-violent and resisting their calls for violence is apparently a fairly normal (and under-appreciated) need in most movements. Then again, most movements, violent or otherwise, don't organize well (cf: the late Charles Manson, the Weathermen, etc.), so activists generally don't get a lot of things.

Life-size baby dolls will become valuable weapons of nonviolent civil disobedience.

non-violent actionists really don't spend much time learning strategy and developing tactics and counter-tactics

That does validate your claim to ignorance, I think. There's such a wide range of written material available (online and off) that it's stretching credibility to suggest it's all produced by the PTB in the interests of countering protests. Starhawk, for example, strikes me as an unlikely agent provocateur, albeit a prolific one. A quick search for "NVDA training" might suggest that tactics are widely taught and understood (and define the term if you haven't seen it before).

In terms of campaign duration, my longest involvement with a single campaign was only about ten years, but that campaign is more than twenty years old. It's been remarkably successful both tactically (in the sense of single actions producing specific results) and strategically (in the sense of both infrastructure changes and political ones). I refer, of course, to Critical Mass and specifically those in Sydney. We have everything from major cycleway projects to politicians apologising to cycle activists when dumb things happen to government staff dedicated to placating us. None, of course, can be attributed solely to Critical Mass. But one example: a new lord mayor was elected on promises to wreck cycling, refused to meet with any cycling groups not matter now polite and benign, but faced with 1000 angry cyclists outside city hall he held a press conference where he used a PA system to address "members of the press" and explain his change of heart. The fact that it just happened to be on the usual day Critical Mass happened and look a lot like Critical Mass should in no way be taken to even suggest that he capitulated to anarchists who terrorise traffic every month. Ahem. Oh, look, a new cycleway, isn't it pretty.

At the other end of the scale, Native Forest Action in Aotearoa was the subject of both vigorous "cointelpro" (as the merkins so delightfully obfuscate it) and a somewhat unexpected series of victories, not least of which was the partial success of their "off the wall "demand for an end to native forest logging (...on state owned land in Te Waka O Maui). That was short term, only a couple of years, but involved occupying a subtropical rainforest through winter (not fun!) as well as media campaigns. And some reciprocal investigation of various parties by those under surveillance, shall we say :) Nicky Hagar wrote a book about it. https://en.wikipedia.org/wiki/Native_Forest_Action has some useful links.

Also, there is a big difference between "clear simple reason to support" resulting in popular support, and actual actions on the ground.

In Australia we have the lovely pejorative "doctor's wives" to describe rich white people who lend support to dangerous radicals and their campaigns for things like stopping torture, keeping whales around and preventing homeless people freezing to death. But mostly as donations of cash rather than bodies on the front lines of a "nonviolent" protest. Notoriously John Howard became the first Australian Prime Munster to lose his seat when a cabal of doctor's wives took against him. They actually changed their votes!

Anyhow, it's quite easy to get public support for many things, and often clear simple explanations are readily available. But getting public action can be nigh on impossible, and there's a lot of work been done looking at why that is. For example most Australians think torturing refugees to death is bad, but 90% of them* vote for parties that promise to keep doing it. It's important, but it doesn't actually matter as a driver of voting behaviour.

For example, the great majority of people in most countries think that climate change is a problem and we should do something about it. By "we" they mean "someone else". And they vote accordingly. Ditto racism, for example this paper (pdf) suggesting generic racism rather than specific fear of black crime etc was a driver of racist voting patterns.

• 65% of Australians are eligible to enroll, about 85% of those are enrolled, and about 90% of the enrolled vote, of which about 95% are valid votes. Of the last, about 10% vote Green or one of the other anti-torture parties, the other 90% vote .... I dunno, I'm reluctant to call torturers "mainstream". Anyhow, nearly half the population actively support killing refugees. Quick comparison of numbers: in the US the number of voters is 15%-35% depending on turnout. Obama got 65M votes of the 130M cast in 2012... from a population of 350M, so peak voting was about 1/3 of the population. Australia getting nearly 50% for anything is remarkable by democratic standards.

Moz & Hetreromeles I tend, very largely, to agree with all you say, but there are niggles: 1] The odd Plod who really is a numpty, who insosts on willy waving & arresting or otherwise seriously monstering people who are doing no harm. The classic case must be the murder/amslaughter (whatever) of a newspaper-seller a few years back. 2] The odd wankers on the "other side" who fuck it up entirely - e.g the protestors who in 2015 (I think) who were so-called "anti-capitalists" who invaded restaurants & terrified the diners - simply not on.

How one prevents these complete idiots, on either side of the action fron effing it up entirely, I'm not sure.

Britain, even though we don't have "compulsory voting" the way AUS does, is pretty good - most of the time please note the caveat. OTOH there was media total panic back in ( approx ) 2011 when a National Front ( = Nazi ) candidate won a set in a local authority by-election. When you looked at the actual figures it was something like: Turnout was under 20% of those eligible & the NF/Nazi won by about 3 votes, because most people couldn't be arsed to turn out - IIRC it was the middle of winter which didn't help. Needless to say, at the next "proper" election, he was thrown out by a large margin

Off topic (this did become a general thread for a while...)

I was pleasantly surprised twice last week.

Firstly to see Charlie retweet a comment from @epeme_ri , a Wellington based Linear B scholar (I like both ancient and future histories, but don't have a twitter a/c, so mentioning it here)

Secondly, to find a copy of the 2005's 'Nova Scotia: An Anthology of Scottish Speculative Fiction' in an Auckland charity shop, which contains a short story from OGH amongst others. Nice collection! Alas, no similar books turned up in the shop. I wonder how it ended up there?

Discipline is critical too, both for optics (something the antifa need to learn

It's worth remembering that one of the great things about an open movement and the lack of a central controlling authority is that anyone can say they're a member. And as we regularly see undercover government actors caught deliberately acting to give protests a bad name, it's always worth asking "would someone who supports the group do that" when you see something unusual. Right now I think we have quite well publicised cases in the UK "relationships scandal" and USA "undercover cop pulls gun" that leave the question of who is protesting very much open. I was present when an undercover cop in Sydney pulled out a weapon at a Reclaim the Streets event, but in that case the outcome was amusing - two uniformed cops jumped him and when he resisted they used capsicum spray to subdue him. The bought media reported that but I can't find links, it was more than 10 years ago.

Side note: that officer needed hospital treatment then was unable to work for two weeks due the spraying. Protesters, OTOH, fragile flowers that they are, can wash capsicum spray off with water and recover immediately (at least according to Police statements). Funny how it's different when a cop gets the treatment...

Like everyone else who has done NBC warfare training in the UK, I've been exposed to airborne CS (in varying quantities according to the taste of the instructor); even if it got into eyes and lungs, not a two-week issue.

I rather suspect that the two weeks off wasn't down to the capsicum - but down to the level of force that anyone would use, when confronted with a lethal weapon. Getting jumped by two fully-loaded police officers who are in full-on "this is a fight for my life" mode? Concussion is likely, along with cracked ribs, sprains, fingers hands and wrists wrenched during any effort to control the firearm.

Said undercover bloke is very lucky not to be six feet under, because "minimum force required" instantly became "lethal force permitted" the second he (was perceived to have) brandished a lethal weapon in the presence of the public, with an intent to threaten life.

The odd Plod who really is a numpty

I have seen way more problems caused by Police than by protesters, and the overwhelming majority of violence against persons has been from Police. I have way too many stories about that to ever want to start. I prefer to focus on the ones who are helpful, especially the public servants who actually serve the public and follow the law. 90% of protest is legal, and a lot of the rest is arguable (Australia has a generic "necessity" defence, for example, that extends all the way to treason and other capital crimes). But a lot of PTB response is in the unlawful-to-illegal range, from minor Police overreach to stuff like kettling and the now-revealed "black prison: in Chicago. Why bother with the law, what counts is results... I recall the British Plod have a rich and proud history of that?

FFS, you don't take hundreds of people hundreds of kilometres out into the desert on a whim, and you don't do it without serious organisation and planning (Baxter and Woomera protests, for example). But even in the city, I'm way more used to having five or six groups organising stuff like legal observers (usually lawyers or law students), medics (often actual emergency doctors and paramedics), media liason, outreach/publicity, marshalls, PTB liason, all the things. This stuff is widely known and training is readily available if you're interested (in the sense of "turn up to work" interested, not "some guy on the internet asking questions" interested). At Baxter the standout for me is the guy who took over assembling and organising the pit toilets that I've built and taken to site. I was very grateful at the time and I'm still grateful now (and he still wants to remain anonymous, otherwise I'd name him). So... if you're willing to dig and maintain pit toilets you'll be a really popular person at any protest camp.

Organising protests is hard work. It requires research, organisation and legwork. Even "flash mobs" don't just happen.

"Here goes my rant, again, about terrorists and generals working from exactly the same playbook, "terrorize the populace, and break their will to resist", when 100% of the time, the effect is the reverse of that. Therefore, it's worse than stupid, it's counterproductive."

When actions look completely stupid often the reason is that the motivations of the actors involved are nothing even remotely similar to the stated ones. I've always thought this was the case in these two examples. If the goal of generals is to bring peace and protect their country, then yeah, obviously creating enemies all over the place is counterproductive. If however the generals see their role as maintaining their organisations, keeping the flow of money and material going, then...well suddenly their actions look pretty clever. If the goal of bringing death to the infidel is the goal of the terrorist, then a series of minor annoyances that will bring a rain of random bombs down on your people look like madness. If however you wish to solidify support behind your extremist organisation and quell the real enemy (the moderates who want to open negotiations and cut you out of the action) then suddenly, it seems like a sensible course of action.

Secondly, to find a copy of the 2005's 'Nova Scotia: An Anthology of Scottish Speculative Fiction' in an Auckland charity shop

Probably the same way I got my mint, first edition hardcover of "Lanark" by Alasdair Grey in a second-hand bookshop in Sydney?

"Baxter" ? Immigration? / detention? / holding in Norfolk Island or similar?

One vote for "whatever plot you like as long as we get more Nyssian biology and DIY-scale terraforming, possibly with parrots if you want to" here. :)

No, C++ exceptions were NOT "useless by design". It's the way that they were misused (and not used when they should have been) at later stages that causes that. On the other hand, they weren't properly thought through, which opened the door for the utterly inane, self-inconsistent and just plain useless ways they are abused in standard C++.

The solution to which exceptions were provided as an answer is that of communicating errors back up through multiple levels of function call. As such, if you're putting try/catch logic directly round the actual std::calls, you're probably doing it wrong. Your RAII should revert your state to its prior state, and the exception should be handled at a sufficiently high level that a retry/abort can be sensibly decided upon.

It would be nice to have the stack trace a la Java, but I've seen way too much Java code which has to do try/finally for every level of a function call . Give me decent RAII and no stack trace rather than the opposite.

The Rust approach - no exceptions, but you always return an optional - is interesting. But the C approach of manually communicating error state all the way back up the stack is ... shudder.

See also Charlie's car-hack in (I think) "Rule 34" - it's going to be so (relatively) easy.

Halting State, I think.

I don't recall a car hack in Rule 34, but then I only read it once because I didn't really like any of the characters. (Unlike Halting State. I kept hoping to see Sgt Smith again, but no such luck.)

In THEORY, yes, sort-of. But wrapping each call isn't necessarily the wrong thing, if it happens to be the best way of testing for that case. It isn't always feasible to test propter hoc without doing so much work that you may as well replace the call with your own code.

The problem is that the C++ library's mechanism is actually an ungodly mixture of C++ exceptions, C's approach, just fouling it up (in a semi-defined way), undefined behaviour, and total obscurity in the standard (when nobody knows WHAT is allowed or required).

I was very active in the later sixties and seenties in the Movement in the US. I was never involved in negotiations for things like permits, but everyone knew we were infiltrated. In fact, my Fortran textbook from about '79, Fortran For Humans", wonderful textbook, had a cartoon (and I think a programming problem) showing a protest march, and the police infiltrator, with the shiny shoes, which no one else was wearing....

Actually, let's see: we could identify who they were infiltrating for: shiny black leather shoes were military, shiny brown leather shoes were FBI, and shiny black plastic shoes were the local cops. No, I did not just make that up, I got it from others, back then.

And at protests: some did some training, which you might or might not attend. And there were always the designated "ok to protest here, step over this line and sit down for non-violent protest (and you'll get arrested as a group)"

The ones who did intend violence, always a very, very small group, would find each other and move as a group. Most protests I were at, there was no such group... and I was at pretty much every demonstration of any size on the US east coast, between NYC and DC, from '67 through '73. And instead of a honeymoon, my first wife and I were in Chicago in '68, and that was determined by a federal commission, after the fact, to have been a police riot.

But it doesn't just bring a random dropping of bombs. It also tends to result in nasty police raids from the local unpleasant government, and also to serious hits to your sources of money.

Unless you, knowingly or unknowingly, are a proxy for a major power, who will have ways to funnel you money in spite of sanctions.

And if you do it in your own country, like the recent mosque bombings in Egypt and wherever the other one was, you piss off far more of the folks who you want to win over.

Of course, there are ignorant idiots who believe in "if we make things get worse, when it gets bad enough, The People (as defined by us) Will Rise Up...."

In THEORY, yes, sort-of. But wrapping each call isn't necessarily the wrong thing, if it happens to be the best way of testing for that case.

Part of it comes down to a minor but widespread disagreement about what exceptions are for. Normal error handling, or wildly out of control situations. Viz, where are the lines between result codes, exceptions and crashes. On the one hand it's often the case that the correct response to out of memory is termination (99% or more of all programs), and SIGSEVF is an acceptable way to terminate. But "the user entered silly values" is almost never a good time to throw an exception (that's not exceptional, that's the norm).

But with C++, the design of the exception system is such that it only covers a very small gap between "terminate, who cares" and "terminate, but with an informative message" (it doesn't cover either case).

Sadly some people disagree, so there's a huge set of noise exceptions that indicate trivially recoverable errors, and in general those need to be either pre-checked and prevented, or caught at a very fine level of detail. The correct response to a database column being NULL should not be an exception, for example, so my database code has to explicitly say "is this null? No, ok read the string. Yes? Leave the string empty" because the alternative is "try{set the string}catch..."... for every single string value I read from the database (it is rarely acceptable to blank the username and password if the nickname field is empty). After a few of those I have become much less enamoured of C++ exceptions.

Admittedly I have very rarely ever thrown an exception in any language, the exceptions being mostly in distributed languages and library code. My natural model of exceptions is fairly close to the Rust one: exceptions are for when the wheels have definitely fallen off and we are trying to salvage what we can before everything catches fire. Or as the C++ standard folk would put it "the state of the object is undefined after an exception is thrown". Put what you know in an exception, throw it and collapse in a heap.

The problem is that the C++ library's mechanism is actually an ungodly mixture of C++ exceptions, C's approach, just fouling it up (in a semi-defined way), undefined behaviour, and total obscurity in the standard (when nobody knows WHAT is allowed or required).

The problem, IMO, is an unwillingness to say "well, we certainly screwed THAT up" and have another go. Instead we get nonsense like "that's by design" and "gosh, this stuff is hard, you know". Evidence that exceptions can work well exists, just not in C++.

I am hungry to find out how this series ends!
Will Bob's possessed corpse finally turn into a tentacle demon?
Will we find out what exactly happened to Mo?
Will Nyarlathotep start a circus of horrors?
Will Cthulhu show up?
I am so excited!

Oh yes, the 'troops' are clearly morons. I just have some hope that the people in charge are doing these clearly stupid things for their own evil ends. I had a nice chat with an IRA bloke many many years ago. He was clearly several sandwiches short of a picnic and had a few roos loose in the top paddock too. It was on a tour with a couple of Aussies and Kiwis and a Canadian if I remember right. So countries that had got their independence by uniting as much as possible, then having a vote. He was rabbiting on about how the IRA had won such and such a number of provinces from the hated English and how one day they would free the rest. The ex-colonials looked at him like he was insane (which he clearly was). "Mate, if you drongos had put aside your absurd differences, made peace with everyone, got everyone on side and then gone to the Brits and said 'we're at peace, everyone is safe, we'd like to go now', they'd have packed your bloody bags for you, but if you use violence they'll never ever back down". His response was that he didn't care how many people he had to shoot to get his independence.

When we had one of the 6 main oils storage facilities in the UK went up in smoke it main explosion measured 2.4 on the reicter scale.

I happened t be in the pub after a union committee meeting and the point came up tat airwave wasn't very resilient and a sensible terrorist could nock out most of the police radios with 2 or 3 teams with an angle grinder

I want to know how the Black Chamber got caught out its implied that there is a secret occult war in the USA maybe there where other players.

I could see the founding fathers having some occult mojo what with their masonic deist background and set up a superblack version of mahogany row.

And why some one didn't drop a hint to Mossad or DGSE about Shiller being back.

Paper on Reuters Tracer, worth a skim IMO. (I'm a fan of their "theWire"; fast(er) news is fun.) Am thinking a bit about how the approach could be generalized to other domains. Reuters Tracer: Toward Automated News Production Using Large Scale Social Media Data (paper, 11 Nov 2017) (pdf) In this paper, we present Reuters Tracer, a system that automates end-to-end news production using Twitter data. It is capable of detecting, classifying, annotating, and disseminating news in real time for Reuters journalists without manual intervention. In contrast to other similar systems, Tracer is topic and domain agnostic. It has a bottom-up approach to news detection, and does not rely on a predefined set of sources or subjects. Instead, it identifies emerging conversations from 12+ million tweets per day and selects those that are news-like. Then, it contextualizes each story by adding a summary and a topic to it, estimating its newsworthiness, veracity, novelty, and scope, and geotags it.

More: https://agency.reuters.com/en/insights/articles/articles-archive/reuters-news-tracer-filtering-through-the-noise-of-social-media.html It detected accounts of a shooting in San Bernardino, California, before any major global news organization. When an earthquake in Ecuador caused the deaths of 77 people in April 2016, Reuters News Tracer gave its journalists 18 minutes to gather more information before another news outlet broke the story. It also helped the news team get an 8-minute head start in reporting on the Brussels bombings, and a 15-minute head start on sending out a news alert on the Chelsea bombing in New York in October 2016.

And would the bombcorde really have a droop snoot at all :-)

If I ever ran a laundry files id have TSR2's after being suitably stepped on ala Staros 4

But "the user entered silly values" is almost never a good time to throw an exception (that's not exceptional, that's the norm).

...it depends...

Possibly the most successful module of code I've ever written was just after we'd persuaded the firm to spring for another advanced C++ course from a rather good trainer / ACCU member we'd used before. Reinforcement of RAII, demonstration of something close to, but better than, CPPUNIT, heavy doses of HerbSutter's "Exceptional C++".

I ended up writing several thousand lines of C++ that were to interpret the metadata of an IP library. The philosophy for the library design was simple; only valid objects may be constructed, everything else throws. One layer of catch blocks at the very top, wrapping every single API call into the library. API calls carry a result code. Test-Driven Development against the API, unit tests against the classes and all their public functions.

Suddenly, the code got simple - you don't need to do "check whether this string is empty", or any other error checking, because the object is fully and correctly formed, by definition. It must be correct, or you'd already be back at the API boundary for "OpenFile" returning an error code, with a statement saying "Could not open file 'stuff.xml'".

It works beautifully; it just needs a decent design, sensibly designed APIs, and exception-safe classes. You can't bolt it on half-way through or afterwards; and it doesn't make sense in a muddle of poorly-thought-through DLLs. Mixing "error handling by return code" and "error handling by exception" doesn't work either.

I was never able to persuade the rest of the team that it was a viable approach; there was a (healthy?) skepticism, or at least conservatism, that pushed back against it. But three years into service, IIRC ~4000 lines of code, we still had zero reported bugs and zero code changes (the static analysis tool was mumbling about stupidly high test coverage rates). We then made one minor change to match a change to the requirements; and it's still in service to this day in the toolchain, unchanged in a decade, still no bugs reported.

Here's a fantastic presentation on the subject at CppCon 2014 from Jon Kalb, that mentions the straw men (sometimes from people who should know better) that have been used to criticise exception-based error handling...

Here goes my rant, again, about terrorists and generals working from exactly the same playbook, "terrorize the populace, and break their will to resist", when 100% of the time, the effect is the reverse of that.

Terrorists and Tyrants, please.

Generals are usually very aware of the limitations of military force against political problems; e.g. from 1970 onwards, the General Officer Commanding Northern Ireland was making it very clear to UK politicians that the Army could support the Police; but it couldn't "defeat" the terrorists, it could only ever buy time until the political solution arrived.

Take, for example, General Petraeus in Iraq. Or Operation BANNER. I'm not saying that there weren't stupidities, errors, or unjustifiable killings in each case - I'm saying that there was never any policy to "terrorize".

Guardian Australia has this relevant article today talking about how the bought media mischaracterise protests in order to present the narrative that most Australians want refugees tortured to death. The writer tries to pretend that that's not the case but doesn't have any evidence.

Footage of young, justifiably upset and energised students is quickly harnessed in soundbites for current affairs programs. Police brutality against peaceful protesters is labelled as a demonstration randomly “turning violent”. The message this endeavours to spread is: the Australian public unanimously supports offshore detention, and the dissenters are merely dogmatic unemployed students complaining about whatever comes their way.

Etc. It's common knowledge, in other works.

saying "Could not open file 'stuff.xml'".

Yes, exactly. You're not reporting "missing close brace on line 1782" which is annoying but specific, you're just throwing up your hands and saying "something went wrong". Just like all the C APIs that return 0 on failure.

I was once paid ~$AU15,000 to write a GUI tool that loaded a file, and reported on errors in that file. Because the Fortran program it was input to said things like "could not open file stuff.xml" and that used to piss the users right off. They spend weeks tweaking a 30,000 line input file and the program says "no". I'm somewhat surprised that so few supercomputers are destroyed by fire if that's typical behaviour. My program would say things like "see this block? The header says 100 mines, but there are 101. Delete one or fix the count?" (and showed two buttons, plus cancel). Apparently it was quite popular with users of the Fortran program.

I prefer specific, actionable error codes. That's why meetings in my workplace normally start with "Moz's code says you're sending the wrong thing, and the customer says it's not working". When I am wrong, it's generally very specifically and fairly obviously wrong And in my defence, it's often "it does A. so we assumed it would do B". Fair enough, I try very hard to stay one step ahead of the remote hardware I'm supposed to support, but that is not always possible.

In my case reporting "a device sent a bad packet" is almost 100% useless, I should just silently drop the packet in that case. Because nonsense packets arrive all the time. But reporting "device serial 123456789 sent a packet with the following errors: type 14, version 123 (version must be 1,2,4,12-18), user 1234 (must be 0..999) and command 87 ("update firmware", user not allowed that action)" is much easier for everyone to make sense of. Even if the response is just "ooops, that device has gone insane", but more often it's "let me re-read the spec, I suspect I got the packet layout wrong". The more useful my error messages are, the easier life is for people trying to make a 10MHz processor with 48kB of RAM do everything it's supposed to do. Because there is no way our customers will pay to upgrade the hardware (unconceivabull!), it's hard enough to get them to update firmware to get new features. So... we support hardware from last century. Lucky us.

I happen to agree with your argument about useful error reporting; (my example was simplified) but you perhaps misunderstand what I delivered.

It wasn’t just a “computer says no”, it provided useful and localised error reporting at the point of failure, and was resilient in the face of recoverable errors. And it did so while using exception-based error handling, at an error density of zero, with no requests for changes. It just works. No code maintenance required at all, completely satisfied users. Given that it took me six to eight weeks to write, it cost my employer £10k or more (cost of employment, not salary, unfortunately).

My reflection about the ability to use exceptions as a coherent part of a design was to think about Tony Hoare’s famous: “It’s either complicated enough that there are no obvious errors, or simple enough that there are obviously no errors”

Did you watch Jon Kalb’s video? And did it address your concerns about exceptions?

Just to be contrarian ... There were groups that SHOULD have been police infiltrated & weren't, or not effectively ... Though I can only think of 2 examples. 1] The utter wankers who did an "anti-nuclear" (Power) protest by erecting a scaffold-tower on the Berkeley branch railway line ... the train ( *note) loco stopped, just in time, metres from the tower, & a very very scared train-driver .. back then, they had not remembered the "obstructing the running of the railway" legislation. That level of stupid, to came so close to a rail-crash was unimpressive. 2] "Animal Liberation Front" who were left alone as "harmless" until they started digging up people's graves & their personal-level intimidation got noticed by Plod, though it had been going on for years. I admit I'm prejudiced - I've met these people & they are probably, at a personal level, more unpleasant tha Hizb-ul-Tahrir, whom I've also had the dubious pleasue of encountering.

*note: It was a nuclear power-rod flask train, coming in empty, to pick up an old one - IIRC. Elsewhere, ( Aldeborough/Sizewell branch ) I've been a passenger on one of those - great fun!

Not even that I'm afraid. There are two, possibly three muslim sects that are approximate Quaker-equivalents: certainly the Sufi & the Ahmadhi are viciusly attacked by the extremists, as the Quakers were, & for exactly the same fake "reasons" - nameley "$_BigSkyFairy has told us to". Support in the community is irrelevant at that level of madness, I'm afraid, as the recent history of Da'esh has shown.

Indeed. But it was one of those utter-nutters that probably, finally, convinced the IRA that they had, actually - lost: Omagh.

And, along the same lines, right now in the USSA, where the Generals are trying to drag T Ronald Dump back from the brink of war with the DPRK. Saying: "Yes, we can easily "win" any DPRK war, but several millions will get killed in the process & it's not worth it.

Talking of which, he's managed to unite almost everyone in this country against him, just now. If he does come here, I think the best protest method would not be any sort of conventional demonstration or march or street protest at all. Someone suggested that as many people as possible, turn up for when he comes - & as he approches, everyone turn around & moon at him. A mass-mooning! ( I'm certainly ... err .. I think "down" rather than"up" for it! )

That is not how I read it - and, even with developments and hindsight, I haven't seen any flaws in my analysis. Which doesn't make it CORRECT, of course, but I am damn sure that the current public story about what the IRA did post 1998 is wrong.

Often, perhaps, but Afghanistan (including incursions into Pakistan), and much of the Iraq debacle (definitely including obscene atrocities like the second battle of Fallujah) show that far too many generals do NOT realise that.

In my case reporting "a device sent a bad packet" is almost 100% useless, I should just silently drop the packet in that case.

One of my college lecturers told a tale like this about when he worked in industry for $datacommshardwarefirm.

They had a banking client who, about once a quarter, called them and said "We've lost a bit." When this happened, he went out and cleaned their hardware.

One time, he was on leave when this happened, and another of the team went out and explained very carefully that there was nothing that could be done to recover the position in this case, but did not touch the "offending" hardware.

Before he got back to the office there was a call saying how useless he was and asking specifically for my lecturer to attend!

I wrote an SNTP client, with full error-detection, and was surprised to learn that a lot of people adopted it as a testing aid, because xntp DID simply drop packets, which made problems almost unfindable.

Mate, if you drongos had put aside your absurd differences, made peace with everyone, got everyone on side and then gone to the Brits and said 'we're at peace, everyone is safe, we'd like to go now', they'd have packed your bloody bags for you

The sheer bloodyminded ignorance of the history is making my eyes cross. Tackling absolutely nothing else, Ireland didn't gain dominion status like Australia et al until after we 'won' a shooting war with Britain, so the Statute of Westminster wouldn't have applied. Westminster couldn't even pass a Home Rule bill; you think they would have granted independence?!

I like how you used quotes around 'won' a shooting war with Britain...

Much like Vietnam, the shooting provoked a realisation that the political situation wasn't working, and that a different solution was required. Remember, the IRA got flattened every time it tried to fight a pitched battle (and a lot of civilians got killed as a result).

Home Rule was coming - the shooting was more about who was going to run things afterwards (Devlin, de Valera, or Carson). Notice that far fewer people were killed in the Irish War of Independence than in the Irish Civil War which followed it...

Actually Westminster DID pass a "Home Rule Bill", which was promptly fucked-over by Irish idiots in 1916, IIRC. [ I'm, for the purposes of this argument, ignoring the almost-equally brain-dead incident of the Curragh Mutiny ... ]

Ahem: you seem to have missed undercover Police agents within the ALF planting firebombs?

Government of Ireland Act, 1914. Easter Rising, 1916. Look up "promptly," Greg. And while you're in the P's, look up "postponed indefinitely" too.

Yep. What really fucked the home rule shtick in the 19-teens was the outbreak of the first world war, which put a lot of domestic changes (votes for women, for example) on ice.

Didn't know about that one ... Yuck.

Bugger - pressed "send" too soon. Alleged / supposed / C Lucas trolling for effect. Yes, it's possible, but it might also simply not be true.

Home Rule was coming - the shooting was more about who was going to run things afterwards

Well, yes, in that the shooting was about getting a Republic instead.

I read somewhere (can't recall where) that one factor in Germany's decision to go to war was the belief that Britain would be preoccupied by a civil war/rebellion in Ireland.

No idea whether or not it's true (the belief, not the preoccupation).

Really? By the start of the Irish Civil War, the Treaty had been signed, the Free State formed, and the Unionists hived off into Northern Ireland.

The real fighting was about whether Fianna Fail or Fine Gael would run things. Note that the IRA (both FF and FG sides) was ~15,000 strong, and lost ~550 dead in the War of Independence; the Free Staters of Fine Gael (who won) had a force of ~60,000 and lost ~800 dead, while the anti-Treaty Fianna Fail (who lost) had a force of ~15,000 and lost up to ~3000 dead.

It doesn't quite fit the "foundation myth" to admit that the bulk of the fighting and the killing was after independence had been achieved, and was over who got to run things; in the same way that it doesn't fit the USA's "foundation myth" to admit that slavery was an issue in the Traitorous Secession Against His Majesty King George in 1776; nor to admit that it was the Opportunist Land Grab of 1812 that caused the Righteous and Punitive Burning of the Invaders' Capital in 1814 ;) (Try this in your search engine: "Bladensburg Races"...)

Did you watch Jon Kalb’s video? And did it address your concerns about exceptions?

The video answers a different, almost unrelated question: can I use exceptions in code that I write? I knew the design stuff and the extra C++ complexity in the implementation I have become used to. So yes, I as a single programmer can choose a variant of C++ exceptions that does most of what I want for the exceptions that I throw. Remember that my original complaint comes from having worked with exceptions for 20-odd years and being used to them, but arriving in C++ and discovering that exceptions are fundamentally broken here.

When I started this project I naively started writing exception-based code, because that's what I normally design. But the first time I actually wrote an exception handler I spent several days scratching my head and going "surely I'm missing something obvious". I was, but sadly what I was missing was ... that C++ exceptions are broken. There's not even a bunch of hidden info that you can only get via the ".as_string" property. Attribute. Method. Whatever, "thingy that returns some text". That was the point where I went from "I just have to learn the words they use to talk about the language" to reading the FQA and going "OMG what a giant pile of weirdness".

There's a lot of cascades: no reflection means exception handling is hard, but it also makes everything from object persistence to unit testing weird and sad. Disagreement about exceptions makes error handling necessarily a mix of exceptions and error codes. Legacy strings means string handling is a mix of byte-bashing, C char-blobs and multiple different strings-as-objects implementations. And so on.

It did finally show me what Raymond Chen meant with his ongoing hatred of C++ exceptions, previously I had thought "yeah, he's old" but when I had to deal with them I realised what he meant. So I copied him, and google, and some other smart people, and stopped using exceptions in C++. It reduces the amount of time I spend going "am I just fighting the language? Am I trying to write Pascal using a C++ compiler?".

Now, if I were a terrorist, I'd buy an old-style heavy-duty remote-controlled car from a hobby story, put 2.5-5kg of some explosive on it, watch from a window with it parked under a parked car, and drive it out as the motorcade passes.

I've seen that used as a plot device in a movie or a TV show already. Can't remember which one it was, but someone made a bomb out of a RC toy car and tried to drive it under the protagonists car to blow it up.

I was very active in the later sixties and seenties in the Movement in the US. I was never involved in negotiations for things like permits, but everyone knew we were infiltrated.

I may have posted this before, but it bears repeating.

Whenever you and your like minded friends gather to plan how you're going to exercise your First Amendment right "peaceably to assemble, and to petition the Government for a redress of grievances", there's inevitably going to be some idiot who insists you have to SEND A MESSAGE by blowing up something.

This is the FBI's informant in your group.

Didn't it have Clint Eastwood in it? Or if it didn't it was the sort of thing that could have done.

Someone mentioned it, at least, and others have mentioned shiny boots.

But surely this is like Russian troll bots on twitter. The ones with the explosives-and-boot-polish fetish are like the ones with 8 random digits after their username: they're to get people to go "ha ha, spotted you" and not notice the other ones with a better disguise.

Wasn't there a case where the cartoon-joke actually happened & an "activist" group consisted entirely of guvmint plants, with no genuine members at all ( or maybe, just one) ??

Reverting to an earlier actual group, the ALF are deeply unpleasant - they aren't animal-lovers, they are people-haters ... & they - shall we say "strongly disapprove" of people lving with other animals ( pets) like cats or even dogs. I have asked them about : "Do you want animal rights for the cats or the rats?" & - needless to say, got shouted at & threatened.

...and it’s so quaint. I mean, 1970s...

Defeating that kind of threat had its beginnings in the fight against the Provisional IRA in Northern Ireland; thirty years on, for some reason there’s a thriving market.

http://www.allenvanguard.com/product/equinox/

And #248 - Yes and yes. I know I've seen and enjoyed the film, but I can't remember the title either.

Still, it absolutely is a thing (the film if not the use of the plot device IRL).

"The sheer bloodyminded ignorance of the history is making my eyes cross. Tackling absolutely nothing else, Ireland didn't gain dominion status like Australia et al until after we 'won' a shooting war with Britain, so the Statute of Westminster wouldn't have applied. Westminster couldn't even pass a Home Rule bill; you think they would have granted independence?! "

Pot, meet Kettle.

Irish independence couldn't have happened if you'd got your house in order and asked politely to be excused, because the Statute of Westminster wouldn't apply like it clearly did for the Australia....

Statute of Westminster 1931

Independence of Australia 1st January 1901, ie. 30 years before the Statute.

https://en.wikipedia.org/wiki/Federation_of_Australia

The Statute wasn't needed for the Crown to grant independence.

FFS. The reason the Brits didn't want to let the Irish go it alone was because they feared there would be a bloodbath (you can't say that was an unfounded fear...). If the drongos had been able to make real progress in assuring everyone that Northern Irish wouldn't be killed or treated as second class citizens (ie, put aside their Utterly absurd differences, and got everyone on side) then a simple vote would have been enough. Australia wasn't completely united in the idea of leaving, but the majority were. If the majority of Irish (north and south) had voted to leave and be a united Ireland then that's what would have happened. It was the violent idiots who assured that Ireland would be split in two. As Whitroth said, "...it's worse than stupid, it's counterproductive". I'll repeat my sentiment, the stated objective (united independent Ireland) had exactly zero to do with the motivations of those in charge of the useful idiots. Their goal seemed to be to fuck shit up and then claim rule of the smoking wreckage.

Oh... writing error messages.

As a lowly user, the best one I ever saw was when trying to save a form "Hidden Field is empty. Field must contain a value"

So whoever wrote that message knew that there were hidden fields the user couldn't see or fill in, and was thoughtful enough to know the user would want to know the form wouldn't save unless they found the hidden field and put a value in it.

Yes "Do you want a united Irealnd or a catholic Ireland, you can't have both?" What the IRA wanted, after 1924 was a united catholic Ireland, which was never going to happen. What MIGHT happen is a united non-sectarian Ireland, now that the S seem to be finally throwing off the Black Crows, & the Abortion vote - (?)next year(?) - will show if that has happened. But, now, it's the N that's stuck in the primitive past, as opposed to both sides.

Terrorists buying fissile material that went walkies after the USSR collapsed? And the only people who actually were buying it were various different spook organisations all stinging each other, no actual terrorists at all. I dunno, kids today don't know they're born, all that lovely 97% HEU and you can't even give it away...

sigh Around '99, I was working for a contracting company that was working for a VAR/OEM that provided the City of Chyicago its 911 computer system, and software. One prorammer left, his wife took over (two other contractors), and then she left, and the PM came to me, to tell me this: the City updated its geodatabase regularly.. and we'd been supposed to update the 911's system's geodatabase the same. Instead, it was updated (closed streets, construction, etc, etc) maybe twice a year, and it was two weeks of sheer terror (people's lives could be depending on this, for time to arrival). He wanted me to fix it.

First discovery: they'd been using Oracle sqlloader. And they'd get a ton of errors, and fix them, lather, rinse, repeat, until it went in. AT NO TIME did they ever talk to the City to tell them of data errors, so the redid the same stuff every time, and the new ones....

I wrote a large awk script that validated it, as well as I could. And then I'd email the errors that weren't acceptable back to the guy from the City, and he fixed them. Forever. Took me a month or so, and we went from two weeks of terror to a day, day and a half of boring email exchanges.

But then, the sqlloaders I've worked with are terrible - they do zero useful messages - it either succeeds, or fails, and they don't tell you all the errors, and put in what they can, they just gag at the first one.

There ought to be a required course in nothing but error handling.

As opposed to every socialist group, except the DSA that I recently joined, where somewhere between many and more (Viking counting method) want to MAKE A STATEMENT by sheer stupidity - only running for US President & VP, as opposed to any other office.

Wasn't there a case where the cartoon-joke actually happened & an "activist" group consisted entirely of guvmint plants, with no genuine members at all ( or maybe, just one) ?

Probably. The case that came up recently, and I'm frustrated that Google can't find it for me, was the case just a few weeks ago (in the US) where a group of police officers posing as drug dealers discovered a group of police officers posing as drug buyers...

Presumably any actual drug related crimes were happening elsewhere that day.

Can't be bothered hunting down sources but I'm pretty sure this was substantiated by the police spy Mark wossname (the one who fathered kids with activist women then bugged out on them). Also, remember the McLibel trial? The libelous leaflet was apparently written by a police officer acting as a member of the animal rights group in question.

Yeah & that was especially egregious, as the "animal rights" group involved (IIRC) were NOT the really nasty ALF, but some people who were quite rightly concerned about animal cruelty.

Minor quibble, which is a complete tangent to your point. Most of the original states were self-governing colonies, with their own bicameral parliaments, universal (white male) suffrage and responsible government (i.e, the executive government was responsible to the local parliament and not the imperial government or the crown) by the end of the 1850s.

It seems now as though Federation was inevitable, but looking backward at events always imbues a sense that what happened was the only pathway possible. If there had been no Federation movement, or it was defeated at the polls in some colonies, then the original states would certainly still be independent sovereign states today, much like New Zealand. In many ways they already were before Federation, hence my quibble about independence not simply arriving on January 1st, 1901.

Of course if anything this reinforces your point somewhat. And equally of course, this genteel enthusiasm for (white male) democracy is superimposed on a slow, dogged genocide over many decades with limited accountability for countless atrocities, of a kind almost unimaginable to anyone who grew up in Australia in the late 20th century.

From the description, it's probably this: Undercover Detroit cops brawled during a sting gone wrong (If you're curious, google keywords were: sting vs sting drugs undercover )

But, the sting didn't attract the kind of customers they were hoping for. Instead, fellow undercover cops from the 11th Precinct showed up, ordering the 12th Precinct officer to the ground. WJBK says the officers then raided a home on the 19300 block of Andover and the two sets of police officers began brawling.

You have to land it, at least while practicing. And the droop makes that easier, no?

OTOH it was a pity Tom Clancy put so many errors into the books the Republicans used as foreign policy manuals.

I'll settle for the "it'll be finished when it is finished" approach.

Sorry about events. One must allow time and processes to pass.

Wood, carbon fibre, ceramic, feather? Radar cross-section may not be large.

From the description, it's probably this: Undercover Detroit cops brawled during a sting gone wrong

Thank you, well spotted! That's the mess I had in mind. The recursive entrapment fiasco seems at least somewhat forgivable, the ensuing brawl less so.

Minor quibble with your minor quibble.

"If there had been no Federation movement, or it was defeated at the polls in some colonies, then the original states would certainly still be independent sovereign states today, much like New Zealand."

Well yeah, if none of them had wanted to form a Federation, they would still be independent, but that doesn't mean there would have been no Federation if some of them hadn't wanted to join up. Western Australia very very nearly didn't join and the other states were certainly intending to go ahead without them. New Zealand very very nearly did join and if they had would now be part of Australia (as a State). Fiji was invited but wasn't ever interested. So some of the colonies not wanting to be part of a Federated Commonwealth of Australia doesn't and didn't mean that there couldn't be a Commonwealth of Australia. Interestingly New Zealand was effectively an independent country when it was invited to join Australia, and had been since 1854, though it was technically part of Britain until 1947 (though Britain technically threw them out in 1907, so figure that out). Also weirdly, Australia was most certainly not part of Britain by 1922 but for some reason the UK was able to sign a treaty that required Australia to sink it's only battlecruiser. https://en.wikipedia.org/wiki/Washington_Naval_Treaty

"So some of the colonies not wanting to be part of a Federated Commonwealth of Australia doesn't and didn't mean that there couldn't be a Commonwealth of Australia."

I wasn't suggesting that either. My point was that joining in a Commonwealth wasn't the same thing as achieving independence.

I'd deleted a what-if alt-history speculative paragraph from my previous response, about how if the Federation of our timeline hadn't happened, Queensland would still have imperial honours and appeals to the Privy Council, NSW would be a republic (periodically at war with SA over water), and Vic, Tas and NZ would have formed their own Federation.

In the sense that NZ was an "independent country" in 1854, so were five of the six original states by the end of the 1850s and WA followed in 1890. Technically NSW had been a self-governing colony since 1788 (and Tasmania since 1825), but it got its constitution and current form of government in 1856 (along with the other 4 southern states, Queensland following in 1859).

1901 introduced a new federal government and a High Court, but still allowed appeals to the Privy Council. State Governors and the Governor General were still viceroys. We got rid of appeals to the Privy Council in the 1980s. We have responsible government in that the cabinet is responsible to parliament rather than the Governor General, but obviously 1975 and all that.

Yes, it certainly wasn't the clean break that it has been portrayed as. Very much a drifting away, with lots of minor disentanglements along the way.

Still not complete of course. Our head of state happens to also be the head of state of another country along with a bunch of other laws squirrelled away. I'm sure that if Scottish independence had gone ahead, they'd still be untangling themselves in 2150 (if there are still people in 2150)

I rather like your Alt-History south pacific, to think about rather than live in. I'm not convinced that Qld wouldn't still have flogging.

I think it's one of the 'Dirty Harry' films, probably "Magnum Force". (The car chase between a radio-controlled model and full sized car is fun but vuery silly if you've ever tried to control one of those RC toys.)

This 1888 incident where Captain Henry Townley Wright asked his gunners to target Queensland Parliament before being taken off the ship by police is interesting in the context of the piecemeal independence of Australian colonies:

https://en.wikipedia.org/wiki/Colonial_navies_of_Australia#Queensland https://en.wikipedia.org/wiki/HMQS_Gayundah#Operational_history

Most of the colonies built their own self-defence navies. Victoria had the largest, followed by Queensland. NSW had one but since Sydney hosted the RN's Australia Station, didn't see a need to invest much in it.

Thing was that Wright saw his line of reporting as being through the RN, and probably sincerely believed he outranked these miserable colonials who just happened to own his ship and the organisation it was built to serve. I imagine if he had sailed to Sydney, he'd have been arrested just the same. But it's also a great example of civilian authority asserting its overwhelming precedence over a corrupt military. Which is in contrast to what happened with the Rum Rebellion.

Also - I suppose it's worth pointing out that Queensland isn't and wasn't all anti-progressive. It elected the first social-democrat government in the world, for instance.

What an amazing chunk of history. I had literally no idea. Thanks heaps.

I have my doubts that an RC car could get up to the speeds of whatever muscle-car Harry was driving (my brain is saying Mustang, but it may be mixing in some other movie from the era, probably Bullit). Come to that, I'd imagine that even a sub-compact could probably outperform an RC car in speed and handling at moderately high speed.

Huh, it was the last Dirty Harry movie, The Dead Pool from 1988, so not the same era as Bullitt at all.

Bullitt was Steve McQueen (green Mustang) and Robert Vaughn (black Charger).

That's the one. Also, if you watch closely it's pretty clear how marginal the RC's handling is even before Harry notices it chasing him.

Who was the genius driver in the green VW Beetle that kept getting in front of both of them though?

Australia was most certainly not part of Britain by 1922 but for some reason the UK was able to sign a treaty that required Australia to sink it's only battlecruiser.

Assuming the crew wasn't on board at the time it probably saved a couple of thousand lives in the long run.

It was pretty old, over 10 years, and almost certainly would never have seen action again. Well 10 years old doesn't seem much these days, but there were huge advances over that period. They didn't even make the shells for her guns anymore so it couldn't shoot anything.

https://en.wikipedia.org/wiki/HMAS_Australia_(1911)

I noticed today that Mike Carlton has a book out at the moment about the successor ship, HMAS II and I suppose its WWII adventures.

Apologies, HMAS Australia II

A late comment: Until the 1931 Statute of Westminster was passed (and then assented to be each state/nation), the UK still had certain reserved powers over the various Colonies/Dominions, and could veto or refuse to assent to some legislation. There's a good long piece on the NZ context of it at the link below. The last time it was used re NZ was 1910. I'll assume there were similar situations in Australia and elsewhere. https://www.parliament.nz/en/visit-and-learn/how-parliament-works/fact-sheets/refusal/

My condolences!

Largely off topic at this point, but for those who are interested, I picked up the game we were playing further up-thread (the "children's book" lines) and ran with it. The result is now up on AO3 as:

A Laundry-verse A to Z.