Back to: Worldcon observations ... | Forward to: On hold

A fireside chat

You can listen to my Thursday evening chat with Paul Krugman here. (Warning: 13Mb of juicy mp3 download goodness; this is not a podcast, it's a raw file to stuff up the guts of your media player of hoice.) If listening's too hard on your ears, I gather a transcript can be found here. It may show up on YouTube later, too.

PS: Congratulations to Neil Gaiman on his shiny new rocketship (aka Hugo award) for best novel!

94 Comments

1:

Oh great stuff, that was a very interesting and funny chat. I was surprised that Krugman, from what i have read of him, so focused on the robot/flying car/jet pack promises of the past when the tangents tech has gone off into are and will continue to be far more fascinating.

2:

Interesting even if it sounds like it is being streamed from a Gwghthribd relay vessel near Vega, Have you been abducted by aliens, again?

3:

Even after reading every post (+discussion) you put here over the past two years the two of you were able to bring up new stuff. Great.

And yes, you should really link to the high-quality version. No point hiding it from your tech-savvy readers ;) .

4:

@2: it's a very compressed mp3 -- down to 13Mb for 75 minutes. I've got less compressed (56Mb and 113Mb) versions, but I'm in a hotel with sucky bandwidth and my attempts to upload them to the internet archive have failed so far. I'll probably get the high-def versions and/or video sorted out when I'm back in Edinburgh in about a week.

5:

@4 well half an hour ago, that link pointed to a 4.5 Megabyte file, not a 13 Megabyte one. and the 4.5 one definitely sounds like it is being played through a bucket full of water

6:

yes, it is a 4.5 MB file, not 13 MB. just fyi.

7:

Hah! Yeah, it's compressed down to 8kbps. 8!

I think I'm gonna wait for the higher quality version.

8:

I've asked $HOSTING_GUY to put back the 13Mb file. Let's see if he's reading his email!

9:

Truncating the URL linked in the post should prove helpful.

10:

Well done @9

11:

@9 thankyou. Charlie, you should change the link to point to the "hi" version.

Not a bad talk at all. I was a little surprised at how lightweight it was and almost no intermarriage between the future of tech and economics. The best case was the one you brought up on the cost of genome sequencing, yet even here the ramifications were not really discussed.

Whether oil companies will be selling algal based fuels in 50 years is an interesting question. Are the energetics sufficiently favorable for that as a replacement for fossil oil in terms of production volume or will it be a partial solution? One thing that you might think about is this. Suppose that the world does indeed manage to deliver algal fuels and that the production curve keeps on increasing as it does. At what point does fuel inventory and related manufactures reduce the CO2 in the atmosphere to levels below plant photosynthesis sustainability, and would even the oceans start to become more basic rather than acidic?

Am I putting 2+2 together to say that Halting State II will have people printing out their own flesh to eat with friends? Fascinating angle.

Many thanks for posting the talk. Well worth the time.

12:

Alex: link redirected.

13:

Wonderful discussion- I hope Krugman gives his own take on it; it sounded like you were both equally challenged by it.

On another note, how can you not consider differential tax/wage policies & their implications in the economic activities of near-future novels? Maybe it's because I'm U.S. and have a mother looking to retire to the state that will provide her the best tax policy, but it seems an obvious flag if you're looking at non-localized production capabilities.

14:

Thank you Charlie for arranging this taping and upload.

15:

Thanks for uploading this - fascinating listening for a Sunday morning. And now I'm off to look into algal fuels, because it's an idea I've never heard before at all in the context of all this talk about biofuels...

16:

Interesting stuff, I liked your bandwidth tax idea, seems like a sensible solution. I don't this the money distribution and information gathering part is insurmountable but I can't see enough people voting for somebody who says they are going to introduce a new tax to pay for the stuff we are all currently getting for free (by breaking the law). I can imagine the EU setting it up but nobody in Europe being able to fly to the US as the record company lawyers would be waiting at the airport. Is this the sort of thing the WTO is for?

17:

kombipom: if the WTO wasn't in the grip of corporate lobbyists trying to weasel the best international treaties money can buy, I'd say the answer was a resounding "yes".

18:

Mm. Charlie, hub-and-spoke is more efficient -- not less efficient -- than point-to-point because of economies of scale of transport. Mass transport of goods between two hubs is about an order of magnitude cheaper than doing it piecemeal.

Hub-and-spoke can be annoying for air travelers, who value every minute of their time. [1] But for goods, it's a very different story. Both overall costs and time costs have dropped significantly since its widespread adoption. Think Wal-Mart, think FedEx.

Anyway, to save the appearances, if you can get the effective supply chain distance in P2P smaller than the last leg of the spoke, then you'd be correct. But it's a little like Mao putting a blast furnace in every backyard. Industries are still concentrated (though services are not, as PK mentions with Atlanta).

You can tell I'm at minute 27.

[1] Hub-and-spoke doesn't help a airline passenger much (if at all), but it's great for the airline. Consider an airline with two hubs and ten spokes at each. It only has to fly (2 * 10) + 1 = 21 routes. Point-to-point, the airline has to fly (20 * 19) / 2 = 190 routes. Size matters too. It's cheaper to fly 200 people on a larger plane in one flight than it is to make ten flights of 20 people in smaller planes.

19:

Whoops, that would be (22 * 21) / 2 = 231 routes, point-to-point.

20:

Strangely, I don't have a media player of hoice.

21:

I have the basic transcription done. I should have the text posted this evening (California time), assuming my wife does not go into labor before then.

22:

Re #11: "the cost of genome sequencing, yet even here the ramifications were not really discussed." I've discussed this with David Baltimore and with the inventor of the gene sequencing machine, Lery Hood. He was the first to predict the full sequencing of the Humane Genome, specifically by analogy with Moore's Law. The issue has become not the mere sequencing (Genotype) but the dynamics of genetically regulated proteome/transcrptome/metabolome (phenotype). Bill Gates hired Leroy Hood and his entire team away from caltech, where he chaired Biology. He's been shortlisted for a Nobel Prize for many years. The last time (circa pi years ago) that I spoke with Leroy Hood about this, he agreed that the first Trillion dollars of revenue for his kind of institute or start-up will be a million people for a million dollars each to have a billion measurements of everything about their genes, RNA, proteins, an AI-built model of personal medical diagnosis/prognosis/treatment based on everything in the biomed literature plus automatically built analytical model of you. The next trillion dollars is ten million people paying a hundred thousand dollars each. The third trillion is from a hundred million people paying ten thousand dollars each. Your mileage may vary. This also relates to the previous thread on organismic versus organizational immortality.

23:

Carlos: put it down to a braino on my part (I was thinking on my feet -- few things are as intellectually terrifying as being on stage in front of a thousand people, trying to hold up your end of a conversation with $NOBEL_LAUREATE), and what I was groping for was network distribution, rather than large centralized depots remote from the consumption and production points. (The classic example in Scotland was a prawn factory who were taking Scottish-farmed prawns and air-freighting them to Thailand to be shelled by hand, then back to Scotland for packaging and local distribution/sale -- because it was short-term cheaper than fixing their broken prawn-shelling machine. I'm not making this up.)

24:

The transcript is posted at http://www.steussy.com/blog/?p=1867

My email address is in the text. Please email any corrections there.

25:

Oh, I think you did pretty well, Charlie, and I think your strengths meshed well with his. Though the bit about PETA and vat-grown meat seemed to set him back. That's why the two of you have the different awards you do.

26:

What everyone is missing, is that Krugman was a clear Fanboy when it came to Stross' work. And well he should be. Well done Charlie.

27:

Jonathon vP@22. I remember when Hood did his keynote speech at a bioinformatics conference and talked about $1000 human genome sequencing. We've definitely made a lot of ground since then. As for determining the full dynamics of the genes, proteins and metabolism of a multi-celled organism, we are a very long way from that.

However, dirt cheap sequencing (and assembly) will have deep ramifications not just for healthcare but for manufacturing a wide range of products.

Further down the tracks, Charlie makes an interesting point about 3D printing of cellular material. Initially it will be about making prosthetics, but what if you could cheaply print food using just a soup of factory produced macromolecules? Think of the impact on the global food industry, the current concerns over Malthusian limits. It turns the food industry completely upside down. I think there are some interesting economic ramifications of that.

28:

Re #23

The prawns used to make the scampi are trawled not farmed and aren't being air freighted they are caught then frozen and sent by ship. Young's had found that the transport cost to and from Findus's factory in Thailand was less than the additional cost of employing workers in Scotland.

29:

Alex@27: I hope you've got a magic wand supply well in order for those "factory produced macromolecules" in a world where energy is getting more expensive. What do you think a plant is in the agricultural context, other than a macromolecule factory?

30:

Just listened to the recording. Nice one.

I must say though, that the submarine not swimming is a silly metaphor. We understand (largely) how a fish swims or a bird flies. A submarine or plane may not use the same mechanics but they do use the same principles. We don't, so far as I know, understand how consciousness works.

31:

Chris L.

Turbulence.

One of the unresolved problems of physics.

Submarines or aircraft, sooner or later you have to tangle with that problem if you want to design them. There are ways of working with it, useful approximations, just as some computer software can briefly seem to be passing a Turing Test.

Consciousness and Turbulence: if I go to Heaven when I die, I may get enlightenment on the first.

32:

Dave: yes, but we built submarines and aircraft before we understood turbulence, and they worked. We understood lift and pressure, at least somewhat; you can approximate turbulence, that's what most naval architect do.

33:

You sir, are a scholar and a gentleman. Thank you.

34:

Charlie, never fear, you gave as good as you got. Krugman was in awe -- I think that's the right word -- if I'm any judge.

35:

If you listen to some of Paul's other lectures or talks on the various subjects that interest him (the economy, and inequality in America and around the world), then this one will seem like he spent a lot more time speaking rapidly and jumping from thought to thought in mid sentence.

I don't think it's unreasonable to suppose that Dr. Krugman was star-struck.

Also, at one point Paul admitted that the reason he became an economist is because he read and loved the "Foundation" series.

One of us!

36:

That was the second scariest thing I've ever done on stage. (The scariest thing: having a dude with a broadsword demonstrate various killing blows using me as a model. What I knew and the audience didn't was that he was registered blind -- he had total macular degeneration, leaving only about 5-10% of his peripheral visual field functional. Very good positional sense, though: he didn't land a blow once!)

37:

Charlie: Did you guys get a chance to chat afterwards outside of the context of a panel discussion? I sort of imagine having a drink with someone with his knowledge and taste for geekery would be an interesting experience.

38:

Brett @28: are you sure about that? The initial newspaper reports specified air freight.

Chris L @30: my point is, we don't build artificial fish or artificial seagulls; we build submarines and aircraft. Same physics (sort-of) but very different applications. If we knew how consciousness emerged, we may well find applications for it -- but they wouldn't resemble us any more than a seagull resembles a Boeing 737.

39:

Chris L @29 "I hope you've got a magic wand supply well in order for those "factory produced macromolecules" in a world where energy is getting more expensive."

So nothing in a factory can be biological? Let's see now... fermentation of soy beans, human insulin, starch based plastics are just a tiny subset of biological products that are either sourced or modified in factories. It wasn't that long ago that BP tried to produce food in factories by growing cells with methanol as the food source ("Pruteen"). Now why would all this be irrelevant for food printers?

40:

I did a web search and they specified shipping, secondly the price advantage of sea freight over air freight is fairly huge so unless a product is really time critical sea freight is preferred. As Scampi is sold frozen anyway it isn't time critical so it wouldn't make much sense to send it by air. Here's a BBC report that mentions that they are loaded into containers.

41:

Chris L@29:I hope you've got a magic wand supply well in order for those "factory produced macromolecules" in a world where energy is getting more expensive. What do you think a plant is in the agricultural context, other than a macromolecule factory?

Ahem. This being the 'far future', say fifty plus years out . . . space manufacturies? This is more of a cool imagery sort of thing with me, but back in the 50's and 60's, several sf stories were published about life in a large blob of water out in space. Apparently - this may just be bad sf - all you really need to stabilize the thing is a sheath of goopy polymer to prevent evaporation and damp any destructive vibrations. So you can have algae, In Space, which can be tailored to make all sorts of useful products.

I could see this has some sort of government-sponsored proof-of-principle construct, and once the machinery has been set up, have the private concerns move in doing the whole 70's thing, mining cometary ice and whatnot.

42:

Chris L@30:I must say though, that the submarine not swimming is a silly metaphor. We understand (largely) how a fish swims or a bird flies. A submarine or plane may not use the same mechanics but they do use the same principles. We don't, so far as I know, understand how consciousness works.

I don't know how far you can push that last. Certainly we have a good idea on what the 'atoms of thought' are, yes? By analogy, while we're not at the point where we're discussing how to generalize the Bohr hydrogen atom, we've at least gotten to the point of Dalton's law of Small Ratios. Imho, of course.

43:

I'm afraid I'm going to have to disagree with you about building brains that think like us, at least in the sense that we will build some brains to think like us.

You're forgetting to take into account that we do not socially interact with planes and submarines. We are going to be socially interacting with some AIs. We are going to want AIs that behave in a manner that we can mentally model and relate to. Now an AI in charge of a fusion reactor or controlling airport traffic need not think like us, but anything we personally interact with, we're going to want a Turing program, something that can be human enough for us to feel comfortable interacting with.

44:

Chris L@30: I must say though, that the submarine not swimming is a silly metaphor. We understand (largely) how a fish swims or a bird flies. A submarine or plane may not use the same mechanics but they do use the same principles. We don't, so far as I know, understand how consciousness works.

And there was I thinking Charlie was making a rather uncontroversial statement. His other point, that once a machine can do something that was thought to require intelligence, it is no longer AI, is also rather uncontroversial. We also have a enough trouble defining natural intelligence and whether animals have consciousness. Many people deny that chimps do. The problem I see when discussing consciousness, is that it is usually described like a switch - it is there or it is not. That gets us into the waters of dualism and the separation of mind and brain. Yet clearly general anesthetics turn off consciousness, even though the body is still experiencing sensations that are not recallable.

Frankly, I'm not even sure it matters. If we made a zombie Charlie, who could fool even his wife he was the conscious original, what exactly would being conscious mean? More than likely, any AI we make, will seem rather animal like at first - we will have no idea of what it is thinking or experiencing - and slowly ascend the consciousness level with possibly quite alien thoughts, or perhaps not. Given how much our thoughts are due to socialization (I don't want to get into the nature vs nurture argument here) how much would an AI reflect our kind of thinking if it grew up with humans?

45:

Nice talk Mr Stross. I'm sure you were too polite to mention that the so-called 'Nobel' for Economics is nothing of the sort - it's a bank-sponsored prize stapled on the back of the real Nobels (check the Nobel website if you don't believe it).

I first thought Krugman was making an ironic joke when he asked where you got your ideas from. I was pretty surprised when it turned out he was serious. Obviously being a world-famous entrail-reader doesn't mean you can't ask the same old questions!

46:

As I've said here before, we don't even know what consciousness is, let alone how to recreate it. Ask any ten people and you'll get ten different definitions of the word. Ask any ten researchers in biology, psychology, philisophy, etc., and you'll get substantially more than ten different definitions.

As for wanting to be able to socially interact with machines, we'll be doing that long before any AI becomes "conscious". See the work being done at MIT on "Affective Computing" and then ask who's sponsoring it and why. Having our computers understand our emotional states and track them would be a boon to the marketing of computers and to whoever sells things through them.

47:

Reading the transcript, I got the impression that Krugman had trouble keeping up with you, Charlie.

Respects, S. F. Murphy

48:

SoV@41: Cool! I wonder what temperature the blog would end up if it were parked in Earth orbit, though? I guess you could sprinkle the outside of your polymer with something shiny until you reached a happy equilibrium.

Alex@39: You mentioned Malthusian limits and my ecologist-muscle twitched. Making things in factories doesn't untie you from the sun and arable land as energy sources. At the very least you need fixed carbon, and that doesn't occur outside of living (or recently dead) tissues. Barring certain fossil deposits that we probably won't be able to keep using for long.

49:

Ugh: s\blog\blob. Although Charlie putting his blog in orbit would be cool. Could you fit a web host in a tubesat?

50:

Chris L@48: "You mentioned Malthusian limits and my ecologist-muscle twitched. Making things in factories doesn't untie you from the sun and arable land as energy sources. At the very least you need fixed carbon, and that doesn't occur outside of living (or recently dead) tissues. Barring certain fossil deposits that we probably won't be able to keep using for long."

Yes and no. Photosynthesis is very inefficient and farms tend to be in places where both sun and water are available, so if you can collect sunlight at higher efficiencies you could improve even traditional farming yields and perhaps even much better with factory based production of algae or bacteria. Then again, we can harvest far more energy than directly hits the earth and we could use other energy sources, like nuclear for inputs. But yes I agree that we may Malthusian limits in practice, although it doesn't have to be so for a while. Of course the risks of mass starvation from failed "harvests" just mount. OTOH, single cell production is not seasonal and so less likely to be catastrophic if a batch or two goes bad.

51:

Alex: the reason photosynthesis is "inefficient" is that sunlight is not generally the limiting factor in fixing carbon. A C4 plant is nature's way of doing it more efficiently: it wastes less water.

52:

Chris@51 Even C4 photosynthesis is only about 6% efficient in practice. Therefore there is scope to improve the overall efficiency of food production. The tropical rain forests intercept most of the light, leaving the understory dark. Arguably sunlight is the limiting factor in this setting. In other environments it is CO2, water or temperature. In this case, maximizing lighted surface area, e.g. using multi-levels, improves production. There is plenty of scope even using existing biologicals, before we think of other approaches. But what if we could grow cells in vats and print food on demand...really cheaply?

53:

PETA and VAT-grown meat: one of the problems which has plagued the vehetable-protein substitues is that meat is structurally complicated. It's a mix of cell-types, and it's organised

And, as far as the taste goes, we already have problems. There's such an aversion to fat, which is where a lot of the flavour of meat is, and which influences the texture of the cooked meat, that traditionally-cooked joints of meat can't come out well. The "proper" beef is a luxury product.

Cheap minced beef can taste better, because it still contains fat in large quantities.

So a PETA-approved joint of vat-grown meat is going to be rather dry and tasteless, at least until the structure can be reproduced. Some sort of curry dish? Haggis? We already have synthetic sausage skins.

54:

@DaveBell: Kangaroo meat. Very lean, tastes fine if you don't overcook it. If you're thinking of that American "marbled" beef when you talk of proper beef, well, good riddance to it. I suspect that the choicest and most expensive cuts of meat are also the ones with the least extra "stuff" in it. Pragmatically speaking though, we already have systems for transforming low-quality, naturally occurring precursor compounds into meat. They're called livestock, and they're cheaper to run than a miniature biotech lab.

@Alex: I've spent more time that I care to think about drinking with people who analyse photosynthetic efficiency for a living. It's nowhere near as simple as you're making out, and I doubt your 6% figure is anywhere near accurate for the process as a whole. Or at least, it wouldn't look too shabby compared to a feasible artificial process. Plants are good at this stuff. They've had lots and lots and lots of practice...

55:

DaveBell@53: "So a PETA-approved joint of vat-grown meat is going to be rather dry and tasteless, at least until the structure can be reproduced."

That sounds suspiciously like Quorn to me.

56:

DaveBell@53: "So a PETA-approved joint of vat-grown meat is going to be rather dry and tasteless, at least until the structure can be reproduced."

Surely the point of the printer is to reproduce the structure. Personally I can't wait for vat-grown/printed meat.

57:

In addition to 56: Isn't veal so tasty and tender because the muscles have hardly been used? If this is the case then a chunk of pure muscle which has never been in an animal and never flexed should be a gourmet's delight not the disaster you predict.

58:

Both Charlie and Krugman talk about robot drivers. We don't have independent robot drivers, but we do have anti-lock brakes, anti-skid acceleration and cruise control. This is partly why our cars are so much safer nowadays: real people are terrible at braking. We brake too hard and go into a skid, or we brake too gently and crash into the car ahead of us. Anti-lock brakes use sensors on the wheels and a sensor on the brake pedal to apply the brakes intelligently, so that you can really JAM YOUR FOOT DOWN ON THE BRAKE and your car will stop as quickly as it can. If you're not wearing a seat belt you might go through the windscreen, but your car will be safe.

Similar technology is used for acceleration and steering. These are little fixes for parts of one big problem, but we tend to ignore tham. We keep hearing about technology to stop crashes by decelerating when the car ahead gets too close but I think that will take much longer to be accepted - it's taking away your control of the car. Anti-lock brakes and anti-skid steering are the opposite: they make your car do what you want, rather than what you're asking for. The only counterexample I can think of to the drivers-want-control idea is cruise control which can now be quite sophisticated. Even so, it's mostly used on long trips, when the driver wants a steady speed - so you could argue that it falls into the same pattern.

Anyway: robot driving - it's here; you just don't recognise it.

59:

DaveBell@53 "So a PETA-approved joint of vat-grown meat is going to be rather dry and tasteless, at least until the structure can be reproduced."

If we are talking about western gourmet cooking, then yes, 3D printed meat may seem somewhat blah, although I am not fond of fatty meat at all. Vegetarians are quite fine with meat substitutes and I've been to restaurants which serve excellent TVP fake fish and meat dishes. But the wider issue is will people buy what amounts to soylent (yellow or red rather than green!). Looking at a typically grocery store, much of the food is processed, in the US much of the content is corn (maize). In many parts of the world, people are happy just to eat, and being a 'foodie' is very far from their minds. Decent protein intake in a palatable form that is very inexpensive would be highly desirable.

Even in the US, alternatives to "hamburger helper" and other starches would be useful.

60:

Chris L@54 :" I've spent more time that I care to think about drinking with people who analyze photosynthetic efficiency for a living. It's nowhere near as simple as you're making out, and I doubt your 6% figure is anywhere near accurate for the process as a whole. Or at least, it wouldn't look too shabby compared to a feasible artificial process. Plants are good at this stuff. They've had lots and lots and lots of practice..."

You may be correct about photosynthetic efficiency, in which case you might want to edit the relevant Wikipedia pages, e.g the entry "Photosyntheticefficiency" and "C4photosynthesis". As a biologist you should be very aware of the dangers of straying into the "evolution creates perfection" trap. Photosynthesis can only use a fraction of the sun's energy output. Machines however can be designed to extract energy at relatively high efficiencies from a variety of energy sources inaccessible to plants or even any other organism. This energy can be converted to energy that organisms can use. Furthermore, machines can be deployed in environments that aren't suitable for plants. We already use CO2 enriched, lighted greenhouses today. Therefore food production could be decoupled from the constraints of current terrestrial production. I would also make the argument this could be a good thing, as it could reduce the impact of farming of natural ecosystems, allow for 'rewilding' of the landscape.

The advances in synthetic biology are coming thick and fast and it really wouldn't surprise me at all that we may be using these technologies for low margin food production as the costs decline. Primitive food printers may be on the cards even in my lifetime.

61:

Chris L@48: This being an sf convention, I'm relaxing my standard Pretty Good Future scenario :-) Actually this is a quasi-economics commentary with a venerable tradition: Fred Hoyle has his Black Cloud express astonishment that life - especially intelligent life - could ever evolve on a planetary surface with it's limited resources of energy. Sterling's Swarm live in a belt of asteroidal material whose potential industrial output dwarfs that of any planetary civilization.

I mention these two because unlike a lot of standard Colonies . . . In Space! rah-rah, the focus is on organic material, and using living tissue to synthesize it. NASA putting up a string of these five- to ten-meter bubbles of water seems like a much more doable experimental pathway than developing mechanical analogs of everything we use on Earth. It would also be a (possibly) cheaper way of developing a closed ecosystem capable of supporting man-sized animals.

Again, this is all pretty much tongue-in-cheek. But as far as speculation goes, it strikes me that an orbiting chain of bubble factories whose outputs are organic molecules like, say, duodecane, is slightly more plausible than the stereotypical asteroid civilization.

62:

Alex@60:As a biologist you should be very aware of the dangers of straying into the "evolution creates perfection" trap. Photosynthesis can only use a fraction of the sun's energy output.

Hasn't this been used before as an argument for some sort of panspermia? That is, it's true that chlorophyll (or at least, the relevant parts of the molecule)isn't very efficient at extracting energy from the light of a yellow sun. But what about light from more common sources, say, red dwarfs? Isn't the efficiency much greater then, given the different spectrum?

63:

Re: #60, 62: Photosynthesis can only use a fraction of the sun's energy output. Evolution by Natural Selection has made quantum nanodevices, and tuned them very well. You breath the oxygen that they produce, and eat them.

Plant Cell. 2004 November; 16(11): 3059–3068. doi: 10.1105/tpc.104.024174.

PMCID: PMC527198 Copyright © 2004, American Society of Plant Biologists Optimization and Evolution of Light Harvesting in Photosynthesis: The Role of Antenna Chlorophyll Conserved between Photosystem II and Photosystem I Sergej Vasil'ev1 and Doug Bruce Department of Biological Sciences, Brock University, St. Catharines, Ontario, L2S 3A1, Canada 1To whom correspondence should be addressed. E-mail svassili@brocku.ca; fax 905-688-1855. Received May 11, 2004; Accepted August 26, 2004.

The efficiency of oxygenic photosynthesis depends on the presence of core antenna chlorophyll closely associated with the photochemical reaction centers of both photosystem II (PSII) and photosystem I (PSI). Although the number and overall arrangement of these chlorophylls in PSII and PSI differ, structural comparison reveals a cluster of 26 conserved chlorophylls in nearly identical positions and orientations. To explore the role of these conserved chlorophylls within PSII and PSI we studied the influence of their orientation on the efficiency of photochemistry in computer simulations. We found that the native orientations of the conserved chlorophylls were not optimal for light harvesting in either photosystem. However, PSII and PSI each contain two highly orientationally optimized antenna chlorophylls, located close to their respective reaction centers, in positions unique to each photosystem. In both photosystems the orientation of these optimized bridging chlorophylls had a much larger impact on photochemical efficiency than the orientation of any of the conserved chlorophylls. The differential optimization of antenna chlorophyll is discussed in the context of competing selection pressures for the evolution of light harvesting in photosynthesis.

64:

ScentOViolets@62: "Hasn't this been used before as an argument for some sort of panspermia? That is, it's true that chlorophyll (or at least, the relevant parts of the molecule)isn't very efficient at extracting energy from the light of a yellow sun. But what about light from more common sources, say, red dwarfs? Isn't the efficiency much greater then, given the different spectrum?"

I've no idea. Interesting thought though. Chlorophyll has been optimized for earth and various organisms, e.g. algae add various pigments to improve light capture int red light deficient water. Perhaps the experiments with artificial photosynthesis might shed some light on this.

I think your point about space based "bubbles" of food growing is more relevant. Nasa spends quite a lot of effort trying to grow normal crops in space, but one wonders if algae (micro/macro) might not be a better solution? We've all seen those little enclosed ecosystems with a shrimp and algae sealed in a glass bubble. Why shouldn't food growing be more in line with mass algae culture - no issues of gravity and everything grows in bio reactors.

In WWII, Britain did some experiments to harvest zooplankton from the sea for food - not very successful (mammals don't do well eating lots of chitin) but it proved that you could harvest zooplankton and process them for food. We could probably do a lot better now.

Maybe we don't even use existing organisms, but new synthetic ones tailored to produce the macromolecules we want.

65:

JvP@63: As Chris L has already pointed out, light is not usually the limiting factor in photosynthesis. This is the basis behind the concept of "vertical farms" - you can apportion a fraction of the total incident radiation to each level, thus multiplying the productive area.

On top of that, because photosynthesis is not highly efficient at trapping sunlight (but plenty good enough) it leaves room for other methods of trapping solar energy and using it to produce food.

ScentOViolets asked if red light might be more optimizable for photosynthesis. Whether it could or not, we know we can convert that energy to wavelengths that terrestrial green plants can use.

66:

Chris L@48: If I recall my climate texts correctly, the blackbody equilibrium temperature at Earth orbit is 255 K. So basically you want to have some sort of insulation to get to biologically sensible temperatures (liquid water and all that). It doesn't need to be much, really. The walls of American and Russian spacecraft are awfully thin, but they manage that much.

67:

@24 - well done! Your transcript has just been posted on Prof. Mark Thoma's "Economist's View" web site.

http://economistsview.typepad.com/

68:

http://www.sciencedaily.com/releases/2009/08/090810162007.htm In a paper to be published online Aug. 9 by Nature Biotechnology, a Stanford University professor reports sequencing his entire genome for less than $50,000 and with a team of just two other people.

69:

Joe in Australia @58, the US has robot car competitions and for the last four yers, I think, at least one car has finished the course.

Alex Tolley @59, the big problem with Hamburger Helper is the salt, not the rice or noodles. You can get decent protein from whole grains and skip the livestock.

70:

Alex@60: you're quite right about evolution not creating "perfect" solutions. But it does tend to create Very Good solutions to important problems like acquiring carbon.

I think we're talking at cross purposes here, because you're (reading between the lines) a techno-geek and I'm a bio-geek. The currency of life is carbon, and more specifically carbohydrates. Not electricity, nor any more mobile form of energy. That's why sunlight isn't the limiting factor in photosynthesis.

I'm prepared to believe that you could grow food more densely with things like multi-story farms, but ultimately they'll rely on inflows of material from the same finite ecosystems. They'll be limited by either water or nitrogen. Probably nitrogen. SoV's orbiting bubbles are a different matter entirely (rocket fuel, ironically, being the nearest relative to nitrogen fertilisers).

71:

Economics and sci-fi, a fertile and somewhat unexplored intersection. I've always wondered what would actually happen to our economy if "strong AI" came about, i.e., all human intellectual output could be done better/cheaper by computers. It's one thing to tell unskilled people to retrain into "knowledge" jobs to preserve their earning power. What happens if that line moves up to encompass all jobs?

I find Asimov's utopian ideal of people engaging in artistic pursuits while diligent machines attend to their every need, well, somewhat naive. There is always scarcity somewhere, and economics is all about who controls it.

72:

Marilee@67: the salt is just one of a host of flavour cues for the protein that isn't actually present in the food. Protein costs money, salt is cheap.

73:

re using algae to capture CO2 out of the air. Golly gee, I've been waiting for someone to notice that. However, the problem is it's just another form of solar cell, unless we use nuclear power to power the lights that the algae use to photosynthesize, in which case it's using gas instead of batteries.

Considering that the plant physiologists have been telling us for years that you can, to a good approximation, treat the photosynthetic rate of an area of forest (or field) as a single large leaf, I think we can calculate how the CO2 levels in the air will fluctuate once these algae-to-gas plants start coming on line. I suspect we might be disappointed with the answer, but you never know. Regardless, green algae aren't very different in photosynthetic rates from plants, so as a form of CO2 capture, this won't be any better than planting a forest. It will just be more usable.

Another interesting technology is the "trash to diesel" conversion that they've been fiddling with for a while. I've noticed that the Koreans have some really innovative ways of mining landfills that they're developing, and I keep waiting for their trash firms (or one of ours) to figure out that they can put a large trash sorting plant that backs up against a petroleum distillation plant, and make a city's trash straight into petroleum fractions to make plastics and fuel from. Trash disposal is an enormous and intractable problem for major cities, some of which are also have oil terminals in them. Image for 30 years from now LA as a net importer of trash and a net exporter of oil, as they combine their trash industry with the Long Beach oil industry. It could happen, if the environmental laws and economics work out.

Another idea: maybe someone will realize that all that plastic in the great garbage gyre in the western Pacific might be economically viable as an oil field, especially if they refine it and sell the oil on Hawaii. Just a thought.

On the photosynthesis and panspermia. OY! It's a great idea, so long as you don't believe in evolution. Chloroplasts (where photosynthesis occurs) are the mutant descendents of cyanobacteria, which are the first critters to evolve green photosynthesis on this planet. The reason RUBISCO (the enzyme that catalyzes CO2 capture) is so inefficient, and so prone to binding O2 instead, is that it first evolved around 700 million years ago, when there was a lot of CO2 in the atmosphere and very little free oxygen. The reason the bacteria haven't come up with anything more efficient is that it's easier for organisms to fiddle with CO2 and O2 sequestration inside their cells, rather than evolving a molecule more efficient than RUBISCO. I'd also strongly suggest that trying to find something better than RUBISCO for photosynthesis might be a really, really bad idea, especially if it gets out of the lab. Spikes in atmospheric oxygen would generally be a bad thing, unless you like seeing the world on fire.

74:

Chris L @70: "I think we're talking at cross purposes here, because you're (reading between the lines) a techno-geek and I'm a bio-geek."

Actually we are probably closer than you think. I was trained as a biologist and do not disregard the importance of ecosystems. I also think that there will be a far greater marriage of biology and other technologies in the C21st, especially as we learn to catalog and understand life and how it really works.

75:

"heteromeles @73:"I'd also strongly suggest that trying to find something better than RUBISCO for photosynthesis might be a really, really bad idea, especially if it gets out of the lab. Spikes in atmospheric oxygen would generally be a bad thing, unless you like seeing the world on fire."

Such spikes would be CO2 limited. You can't produce much oxygen when the amount of CO2 is a low as it is in the atmosphere. It seems a bit self-limiting to me.

76:

The snag with taxing the Internet and then monitoring downloads in order to pay artists is that you're routing all artistic earnings through a single government controlled pot.

Censorship would then be a simple passive act ("People are free to write this filth, but the government doesn't have to collect their dirty money for them").

Cultural terrorism would be easy ("All blasphemy must be stopped!")

Well meaning tinkering would become possible ("Does JKR really deserve all this money?")

There'd be this tempting pot for good causes ("No decent writer would object if we take a cut for grants for mentally handicapped writers/literary writers/gaelic writers/English Lit courses").

Really, it's a bit like the ID card. I like the idea of having a single easy way to prove who I am. But once the ID card is in place, it can be used for bad things.

77:

While these are all theoretically fine objections, try and take compare them with the status quo. 1: Censorship: Those governments which are inclined that way already use practices far more severe than just not paying, ranging from murder and torture down to such social engineering hacks as only approving or banning any given work after it has been printed (this makes publishers selfcensor because having the censor object to a book becomes fiscally catastropic) - And a world in which legal torrents are the standard form of distribution, hardware manufacturers and legislators will have infinitly less reason to build the arcitecture of control of information we are currently heading towards, and which is getting abused by thug governments world wide (most tools which are being developed to stop piracy are dualuse, and can also be used to supress speech, its inherent in the the whole control over dissemation of data thing) - 2nd objection is nothing but a restatement of the first one, so ignoring it. 3: There almost certainly would, indeed, be a (fairly high) cap on how much of the pot can go to any given work. Feature, not bug,- The point of any IP system is to encourage creativity, and if you make enough money off one single that you never have to work again, that actually discourages you from working.

4: pftft. Am I supposed to take that seriously? There are government programs with dedicated revenue streams without counting, some extremely similar in concept to whats proposed here (.. like the bbc) and generally speaking, they only get "robbed" in the kind of tinpot states where our present copy right system also goes unenforced.

78:

That said, it really does seem like the kind of thing that needs to be very carefully designed. Strong privacy controls, and the correct formula for distributing money would be very important.

79:

@75: Unfortunately, that experiment was already run in the Carboniferous, and (fortunately to a lesser extent) during some parts of the Ice Ages.

The Carboniferous experiment is where the O2 levels jumped enough to support giant insects, and quite possibly to a level that allowed amphibians to evolve onto land with grossly inefficient lungs. What happened was that lignified plants took over, plus the continental configuration ran favored CO2 release from carbonates. In any case, a lot of slowly rotting plants got buried, producing the massive coal seams that gave the era its current name. There's fossil evidence for massive fires, so the idea of "a world on fire" isn't hype. It's also a cooler world, so we'd probably be plunged into another Ice Age, along with frequent fires everywhere else. Not my ideal place to live.

Along around the Jurassic/Cretaceous, modern, wood rotting fungi evolved, so that such a build-up of unrotted wood probably can't happen again soon, absent some human fooling.

Actually, we're more likely heading into a period of high CO2, which could be worse for us, but that's a different thread (read Out of Thin Air for the details). In any case, the take-home lesson is that air chemistry has changed fairly significantly over geologic time, and different levels of O2 favor different organisms. RUBISCO evolved during a period of little free oxygen, and that's probably what determined the way it works now.

80:

heteromeles@73:On the photosynthesis and panspermia. OY! It's a great idea, so long as you don't believe in evolution. Chloroplasts (where photosynthesis occurs) are the mutant descendents of cyanobacteria, which are the first critters to evolve green photosynthesis on this planet. The reason RUBISCO (the enzyme that catalyzes CO2 capture) is so inefficient, and so prone to binding O2 instead, is that it first evolved around 700 million years ago, when there was a lot of CO2 in the atmosphere and very little free oxygen. The reason the bacteria haven't come up with anything more efficient is that it's easier for organisms to fiddle with CO2 and O2 sequestration inside their cells, rather than evolving a molecule more efficient than RUBISCO. I'd also strongly suggest that trying to find something better than RUBISCO for photosynthesis might be a really, really bad idea, especially if it gets out of the lab. Spikes in atmospheric oxygen would generally be a bad thing, unless you like seeing the world on fire.

First, let me say that I'm not the one advocating this theory; organic molecules . . . From Space! has been around for a long time. And on the face of it, it's not a totally unreasonable one; we see all sorts of stuff in interstellar clouds and cometary bodies, so why not suppose that the ancient Earth was seeded by something that had a limited ability to self-replicate? That's one of those big questions that needs to be tackled by big science. If at some future date photosynthesizing life forms on other planets are discovered, none of which use pigments that happen to have any common structural similarities with any other, then yes, evolution on Earth is the way to go. If, otoh, we find very close analogues of chlorophyll A, B, & C, the speculation of a common ancestor would not be completely unreasonable.

Imho, that's part of what makes a lot of the early science fiction so unreadable today; it's not the bad physics or astronomy(or the bad writing), it's the bad biology. And most particularly, the bad evolutionary paradigms. I think the reasonable presumption is that aliens won't be just weird-looking people with tentacles or an extra set of legs or perhaps a third eye, nor will early surveys reveal that human colonists can gainfully eat some of the native plant and animal life. My money would be on finding complex structures that one would be hard-pressed to define as life in the conventional sense, Wang's Carpets as opposed to prokaryotic analogues.

But that is, I repeat, just speculation on my part. For all I know, "The Omega Glory" is being played out not too far from here even as we speak.

81:

ScentOViolets@80: "so why not suppose that the ancient Earth was seeded by something that had a limited ability to self-replicate? That's one of those big questions that needs to be tackled by big science. If at some future date photosynthesizing life forms on other planets are discovered, none of which use pigments that happen to have any common structural similarities with any other, then yes, evolution on Earth is the way to go"

It is probably easy to test that theory on earth. There are companies like Maxygen that do rapid enzyme design by effectively using a genetic algorithm to modify enzymes via gene engineering and use the resulting activity in organisms as the fitness test. I have been told that sometimes the best enzymes for a task using this approach can be extremely different from the natural starting point.

This suggests that we could do something similar for trapping light energy. We might be able to tell if chlorophyll analogs have to be very similar or whether they can be very different. Of course energy production requires a suite of genes/proteins so the problem is somewhat harder. It would certainly be interesting to test whether the dominant light wavelengths have an impact on the type of chlorophyll that results, testing your question about the impact of red dwarf light on photosynthetic efficiency.

82:

Just to be clear Alex, I'm agnostic on the question; I'm merely presenting someone else's ideas. You can tell which way I tilt on the question by virtue of the fact that I tie it in with sf :-)

A similar thought would be whether or not life, once it had evolved on a planet, might actually find interplanetary or even interstellar space an exploitable niche. Maybe NASA could try seeding a short-period icy or chondritic body with some (extremely hardy) simple single-celled life forms and then do a sample survey two or three or seven years later. A proof-of-concept sort of thing that the public might just get fired up about.

83:

There are multiple non-standard (i.e. not green) photosynthetic pigments found in marine algae. I'm not sure how they tie into the photosynthetic machinery, though. I have a feeling a lot of them are associated with only one of photosystem II or I, but I can never remember the details of that stuff so I'm not a reliable source :)

84:

ScentOViolets@82 - NASA is doing some short term survivability of microorganisms in space tests. This is primarily to test whether life could survive a Mars->Earth trip. It would certainly be even more interesting if they could seed a synthetic "comet" and test really long term survival. Bear in mind that these tests are for surviving dormancy. There is no suggestion that life would be active in this environment - for the most part it would be far too cold. I suspect that bacteria, or even primitive replicating macromolecules drifting attached to interstellar dust grains would not survive millenia of cosmic rays. But you never know - it could just might be a numbers game.

85:

The last I heard (which was maybe a couple of months ago) the people studying photosynthesis in detail were talking about quantum-mechanical effects being involved. What that does for efficiency calculations is beyond my level of physics, but I would suspect that it would mean that photosynthesis is pretty good at getting energy out of the system.

86:

@heteromeles: You might want to link to Out of thin air. I was finding a link to someone framing a creation/evolution “debate”.

87:

Evans@85: I dunno, just about everything I can think of involving chemical reactions comes down to 'quantum effects'. The problem here is not the collection system, but what it's collecting from, which happens to have the spectrum of a black body.

For a related example, iirc, even back in the 70's solar cells with efficiencies >50% were demonstrated in the lab; their source of light was a laser tuned to the specific wavelength that corresponded to the specific band gap energy of the pn-junction in the cell. Photons that have an energy below the band gap are either reflected or absorbed but incapable of being converted to useful energy, and they can only heat the cell. Photons that have energy in excess of what is needed to bridge the gap will again have that excess converted into heat. So the window of efficiently utilized photons is rather narrow for each band gap; other strategies must be employed such as frequency doubling[1], or using other materials that have a different band gap energy to exploit photons in a different range of wavelengths.

The end result then is that different light-conversion devices will have different efficiencies depending upon the spectrum of light they are exposed to, and this is just as true for chlorophyll as it is for the latest multiple-junction thin film solar cell.

[1]People reading this thread who know more than I do will realize at once that I'm several years out of date and that I'm too lazy to do any googling to catch up.

88:

@86: Good point Tobu. I don't know how Charlie feels about me linking to Amazon, but this is a book. Here's the ref:

Ward, Peter. 2005 Out of Thin Air: Dinosaurs, Birds, and Earth's Ancient Atmosphere. ISBN-13: 978-0309100618

89:

SoV@87: Land plants intercept multiple wavelengths: all except green, in fact, which makes sense when you think about it. Their net peak is pretty close to the peak in irradiance from our little old sun.

Heteromeles@79: I've done a fair bit of reading on metabolic scaling, and I've never been convinced that modern insects are oxygen limited. Any flying insect larger than a grain of sand is already past the point of passive diffusion and does active pumping of air through its trachea, with one-way valves and lots of sinuses. I know that a group a couple of years ago claimed to have worked out the limiting factors in beetle ventilation, but at the time it was just a press release with no actual paper available.

Lots of ecological stuff has changed since the Carboniferous: the rise of gymnosperms and angiosperms, the evolution of ants (which tend to regard large insects as FUD FOR WEEKS!) and various insectivorous vertebrates. Throw in a life-cycle constrained by molting and things start to look a bit unprofitable for large insects. Shame, I'd love to have 50cm dragonflies around :)

None of which has anything to do with atmospheric oxygen...

90:

Chris L @72, Hamburger Helper expects you to add the hamburger (mince, I think it is in the UK). The "helper" is so it's fast and easy. I've never had it, but I know people who consider the hamburger and helper to be a reasonable meal. Most people eat a lot more protein than they need.

91:

@89: Feel free to disagree Chris. I think Peter Ward makes an interesting case that there's an apparent correlation of life-forms with different lung types being dominant under different O2 levels, and that major new clades (such as amphibians) seem to have evolved to breathe air in times when high oxygen made inefficient lungs less of a handicap.

With the insects, I think the best argument for size-limiting in insects comes from islands where there is little or no competition (and often, no ants--good point on that!). For example, we get the New Zealand weta, giant earwigs on other islands, and so forth. Yet even though insects can colonize islands readily and are way more common and diverse than any other animal, we don't see them getting bigger than about 4-6" (the wetas, goliath beetles, and bird-eating tarantulas). I think that argues for some sort of upper practical limit on their size. I also recall that there's an experiment where Drosophila raised and bred in high O2 got significantly bigger (in the statistical sense--it was a small amount). So I'd say that it's a reasonable theory that insect size is limited by atmospheric chemistry, but you're right, it's not a simple story about passive diffusion.

The more interesting case is where there is low 02. Ward hypothesizes that dinosaurs really were ectothermic and bird-lunged, because they showed up in the Triassic when O2 levels were abysmal, and their lungs gave them a major advantage over everything else. His suggestion is that mammals stayed tiny during the Mesozoic because their lungs were inefficient relative to those of dinosaurs, relegating them to the smaller niches until O2 levels rose. By the end of the Cretaceous, supposedly O2 levels were near those of today, and oddly enough, that mass extinction (unlike the previous two) was when mammals were able to gain an advantage over the dinosaur/birds in the large terrestrial animal cateogry.

It's a great theory, but that's most of what it is: theory. Ward is to be commended for laying his ideas out as a series of testable hypotheses, and I hope people take him seriously enough to test his ideas. His is one of the more logical stories I've run into in recent times, and it would be a shame if he was ignored.

How did we get here from Paul Krugman anyway? He's not a dinosaur.

92:

Is the Institutional Longevity thread closed? I wanted to add this:

Why do some cities die, and others live on? Jericho and Damscus have existed for roughly 10,000 years.

http://en.wikipedia.org/wiki/List_of_cities_by_time_of_continuous_habitation

Continuous habitation since the Chalcolithic (or Copper Age) is possible (but difficult) to prove archaeologically for several Levantine cities (Jericho, Byblos, Damascus, Sidon and Beirut). Cities become more common outside the Fertile Crescent with the Early Iron Age from about 1100 BC. The foundation of Rome in 753 BC is conventionally taken as (one of the dates) initiating Classical Antiquity....

93:

@77 (in case anybody is still listening) It's different from the status quo because it routes all literary earnings through a chokepoint/point of failure. Censorship would be passive, rather than active. Certain sorts of books would be ineligible to join the scheme.

And the sort of robbing I'm talking about is more subtle that what tinpot states do, and it already goes on. For example, local govenment expects museums to address specific social agenda. When last I looked, Arts Council grants take disability into account. There are other examples such as our Road Fund Tax, none of them very sinister on their own, but... this scheme would make it very easy to turn literature into a branch of government.

94:

Ah, here it is: http://www.sciencenews.org/view/access/id/43203/title/PHOTOSYNTHESIS_GOES_QUANTUM You want to click on 'enlarged version', and then click on the image that brings up, to be able to read the fine print.

Apparently what the scientists are thinking is that the electrons are going through the system as coherent waves, and taking the most efficient path through, rather than being particles bouncing back and forth in a pinball machine.

Specials

Merchandise

About this Entry

This page contains a single entry by Charlie Stross published on August 8, 2009 8:32 PM.

Worldcon observations ... was the previous entry in this blog.

On hold is the next entry in this blog.

Find recent content on the main index or look in the archives to find all content.

Search this blog

Propaganda