If someone was crazy enough to download one of those dodgy scans and deliver it to your mailbox in a proofread plaintext file, would that help enough to get it published again?

@Dirkjan It is unlikely that that would be helpful, in that Charlie already has a machine readable file of the book. Specifically, the one he typed and proofed while writing the book...

This is really frustrating. I feel your pain too, but I also feel that, perhaps, Tor could make an exception with a whole series like this and sell it as an e-book bundle, including 'The Merchant's War'. The promo might release enough resources to get it converted and proof-read. Wouldn't hurt if readers raised this issue on Tor.com (I don't read e-books myself yet).

There are probably companies that could scan, OCR, and proof their backlist. Heck, Google may have already done it partially.

At work we had a couple million pages of employee files scanned so we could upload legacy info into our file management system. Xerox was pretty fast about it, and reasonably priced. And we were giving them ratty, multi-sized, stapled & papercliped papers on a variety of different types and size of paper.

In a single word: no.

I'd like them to do that, too. And the time I should nudge them -- via my agent -- is now (as they're getting ready to prep "The Trade of Queens" for paperback in March: meaning, it'll be printed in late January, so typesetting/reflow should happen some time in November).

But it won't be productive to bug Tor via tor.com, and I'd request people to abstain from going that route.

I have been reading CJ Cherryh's blog about her backlist, along w/Jane Fancher's: how they re-acquired rights, created their own e-store; added other authors with backlists. It's been going on for about two years, and the amount of time they spend making their backlists file ready is enormous and takes away from writing time. They are harried, and although pretty tech-aware, have found bugs where they didn't expect, etc.

I think I would prefer that you keep writing but keep track of who owns your backlist. I use my Barnes & Noble reader when I'm on a trip or when the hands can't grip a book; but I prefer the feel and smell of a real book; but, I'm a lot older than most of the posters here. That being said, I do pay for ebook copies of hardbounds and paperbacks I already own because I switch out. The Merchant Princes series is how I learned of you, Mr. Stross, and I have gobbled up everything since and now have my husband telling everyone he knows about The Laundry series as I put finished volumes on his side of the bed. Thank you.

Oh, and please come to DragonCon!

Dragon*Con is (a) in Atlanta, (b) in summer.

Haven't I said often enough that I moved to Scotland because London in summer was too hot for me?

I just wanted to mention that, in fact, TOR wouldn't have to convert and re-layout and proof-read the book for ebook pulication. Pretty nice ebooks of it (proofread and all) already exist. But I guess Tor can't just go and use one of those (after removing the "ripped by foo, proofread by bar, distributed via baz" note at the end).

Hmmm.

Interesting question. What would be the legal issues surrounding something like that? I.e. the rights-holder of a book just going and getting a pirated copy of said book in order to save himself the trouble of doing an ebook himself. I guess there are very good reasons why companies don't do that but I wonder what they are (I mean explicitly .. that it would lend some amount of credence to people they don't want to lend credence to in case somebody finds out is clear ..)

Well, for starters there's the question of accuracy: how does a publisher confirm that a pirate copy of a book is in any way an accurate copy? There are any number of really bad pirate versions out there, with proofreading errors or -- worse -- editorial dabbling by amateurs who thought they were improving the text. And that's before we get into the outright weird territory of fake texts: there was quite a lot of heat and noise a couple of years back when some goon released a supposed early leak of one of the Harry Potter books, only for it to turn out to be 600 pages of not terribly good fanfic. (The author released their work with someone else's name and booktitle. It's one way of getting your words in front of an audience, I suppose ...)

No sane publisher would trust the accuracy of a pirate scan. If they want a scan, they can do it themselves in a couple of hours and at least know for sure that the right book went through the scanner ... and that what they're feeding to the proofreaders for correction hasn't been meddled with by parties unknown.

ah .. good points, didn't think of those in all my musings re legality.

Incredibly, lots of these "dodgy scans" are better edited than several books I can buy on the Kindle store (the problem with fake/edited scans in the wild isn't widespread).

The incredible part about it is that publishers expect me to pay for an unproofed book full of typos (which cannot be corrected even for personal use, thanks to their lovely DRM), while there're more professional-looking pirated editions in the p2p cloud...

Way to learn from the music mafia (I mean, industry...)

That's because (unless I am misinformed) Amazon insist on doing their own digitization -- and they don't bother proofreading at all; they just OCR a book, unless the publisher makes a typeset file available.

Amazon's quality control is execrable, because they can simply blame it on the poor bloody publisher.

(Have I said how much I hate Amazon recently? No? Okay: I hate Amazon.)

On OCR error rates

A lot of the pirated books are pirated from ebooks themselves not scanned paper books, so the quality is generally as good as the ebook. It's reasonable to think that a lot of the ocr errors are introduced by the need to process physical medium, page warp, printing problems, etc.

There are scripts out there that wrap around the reading programs on the pc (say Kindle PC or whatever) and just ocr off the screen feed ( think like ocr'ing a bunch of printscreens basically). I believe they are close to 100% accurate. That is assuming the file itself is not just cracked directly.

However if the book was never released in ebook, you are going to be seeing a higher error rate in pirated versions I imagine. One way to get around this would likely be to scan multiple editions in different fonts and merge them. Funnily enough there are a lot of really good bioinfomratics algorithms that have been developed for human genome work that would probably nail this without any human intervention.

Charlie if they want to reprint on paper on of these books, how do they do that, given that they don't have an electronic copy? Where/how is the original actually stored?

link to assembly algorithms.

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

They probably keep an Adobe Acrobat file of the flats. Caveat: I don't know for sure.

Typesetting is one of those jobs that is outsourced to third-party bureaux by large publishers these days, along with the grubby jobs of owning printing presses and warehousing books. Hopefully the typesetting agency keeps everything on file in case of demand from a publisher.

(I know that when I've asked Ace about getting my own copy of the DTP files for my books I was told it would cost $200 per book -- not to my publisher, but to the typesetting agency for providing the media.)

Doesn't sound like the obstacle to me is technical, if Tor had the desire to publish it's backlog, I can think of several bulk processes that would not be manually intensive, no reason to do proofreading or typesetting on something that has gone through that process once already. Its hard to imagine that a project like that would not net out a good ROI

I imagine renegotiating all those contracts is the real deal breaker, legal fees probably kill it.

In this case I have to agree with you. Even though I love my Kindle, Amazon sucks!

if Tor had the desire to publish it's backlog, I can think of several bulk processes that would not be manually intensive, no reason to do proofreading or typesetting on something that has gone through that process once already

Wrong.

The ebook output formats -- epub or mobipocket, mostly -- are different from the output formats for offset-litho print. You don't get page description level layout control; instead you have to give the ebook reader software hints on things like hyphenation, line and paragraph breaking, widow and orphan control, and so on. Just run an InDesign file through an automatic export to epub with all the defaults checked and the results are liable to be execrable.

You can probably trust the spelling, so a full proofread may not be needed, but you're going to have to page through the whole ebook to ensure paragraph flow is okay and there are no embarrassing typeface size change whoopsies or hard line breaks in embarrassing places.

(This is partly why Amazon's ebook publishing quality sucks. They try to automate things that don't automate well.)

Legal fees on renegotiating the contracts are, ironically, not the big obstacle: publishers pay once for boilerplate negotiation with the bigger literary agencies, hammer a one-page addendum into shape acceptable to everyone, then send it round the authors for signing (or not). If they want to republish everything in the backlist, that would be problematic -- but just the 70-80% of authors who're happy to see their work earning them royalties? That shouldn't be a contractual problem -- it's mostly a matter of bookkeeping (and tracking who's signed up and who's said "hell no!").

I guess it depends a lot on how much you care about formatting versus content. I read a lot of content on my Ipad that is pdf or text, or html or mobi or various other flavors of weirdness. It's usually not perfectly rendered but I don't really care I guess. I am sure some people do.

What bothers me is when words are actually transformed into other words, that can be hard to get past, and you do see that a fair amount in the sloppier OCR conversions.

I'm not convinced you could not solve the other problems you mention programatically either, but it is certainly a different kettle then just getting the words right.

Another option which I've always wondered why ebook vendors don't leverage is crowdsourcing it and letting fans clean up the output? I would never crowdsource editing but it seems like a decent idea for for fixing formatting issues...

Another option which I've always wondered why ebook vendors don't leverage is crowdsourcing it and letting fans clean up the output? I would never crowdsource editing but it seems like a decent idea for for fixing formatting issues...

Suppose you got 20 random enthusiasts to 'clean up' your output.

If you're lucky, you'll be supplied with 20 different 'fixed' versions, 19 of which have at least as many formatting problems as the original, and 15 of which have been edited (against instructions). So now you have to pay someone who does this for a living to go through them all and throw out the numerous erroneous variants. This will probably cost you more than paying them to set it up themselves.

If you're unlucky, you'll be supplied with 25 different 'fixed' versions, all of which have more formatting problems than the original, and some of which are wrong in complex ways that only show up when you do something slightly non-standard (ie, which the tester might not spot, but will make the text completely unusable for 10% of the population).

I would expect Charlie to be reliably lucky in this field because he has a large technically literate audience. But as you can see above, such luck is purely relative.

I just wanted to buy "The Trade of Queens", which is available on the Kindle.

If you don't live in Europe, that is. Amazon informs me this book is not available in Europe. Here, it seems the deal is dead wood or nothing.

Sort of silly, but I expect there are reasons for this which sound sensible to the parties involved. At least I hope so.

I'll buy regardless, but this geoism is sort of silly, and is certainly not what I hoped for. The publishing business should be bold enough to make its very own blunders in stead of plagiarizing the music industry.

I regret that I didn't get to buy The Clan Corporate while it was available on Webscriptions. I had to put it off for a couple of weeks due to cash flow, and Tor pulled their titles somewhere in that time.

That's because Tor only have the license to publish the book in North America.

Or rather: they have worldwide English language rights, but they only publish and sell books in North America, so they try to sell the overseas sub-rights to other publishers. And the UK publisher stopped buying Merchant Princes books after #3.

Your workaround is to set up a spare (free) gmail account, use it to create a new Amazon account, give it a false address in the USA and no credit card info, feed it with gift vouchers (purchased from Amazon.com using your other Amazon account, and buy kindle books from the US market using that account. Then crack the DRM on them. Sigh.

The regional rights to publishing didn't offend folks back when books were printed on dead tree because people didn't notice what they didn't see on the bookshelves -- and the odd grey market import wasn't a problem. (The restriction is on publishing and selling in the wrong country, not on buying. The buyers were in the clear.)

Eventually we'll be selling world rights and publishing globally ... but right now if I do that, the publishers try to pay me less (i.e. the same as for North American rights only, not NorAm plus 50% which is what I get for selling them separately).

(The restriction is on publishing and selling in the wrong country, not on buying. The buyers were in the clear.)

I've never been clear why sites like Amazon don't just do mass "grey market" imports of books. Right of first sale should permit that, no?

I've never been clear why sites like Amazon don't just do mass "grey market" imports of books. Right of first sale should permit that, no?

They do do mass "grey market" imports. If you're in the UK you can buy US titles from amazon.com with no problems. Just pay shipping. The point is, they do it retail -- the customer buys retail in the US then pays for shipping overseas. They don't do it wholesale, viz. exporting consignments to Amazon.co.uk ... unless there's no regional edition of the same work.

The Kindle/ebook position is murky because Amazon's contracts with publishers insist that Amazon is a publisher buying republication rights for the Kindle platform, not simply acting as an ebook retailer. As publication rights are territorial, it follows by this bullshit that Amazon are restricted in their territorial rights to publish on the kindle platform.

On one hand, it would seem sensible for the industry to move towards language-based rights; Tor owns the rights to the English-language edition, Gollancz the German-language edition and so on.

On the other hand, does this invite conflicts down the road? If machine translation advances enough that your Kindle can turn Mein Kampf English faster than you can scroll through it, we'd be in another audiobook-style mess and no mistake. While, yes, machine translation today is a bit of a joke, we used to say that about speech recognition and generation and now look where we are. Give it ten years...

DragonCon is an enclosed system. Once you're in one of the host hotels, you never have to go outside. All of the host hotels are connected by enclosed, elevated walkways. Anyway, I'd love to see you at DragonCon, too.

Ahem: Gollancz is a British publisher. I think you're thinking of Heyne.

And machine translation -- at least, of decent quality and for fiction -- is a pipe dream: it's an AI complete problem.

Oops - I've only found your blog about 5 months ago. I might have missed that. I thought perhaps you had been to DC because I purchased a signed copy of book 3 of this series there.

We moved to Atlanta because we could afford to live no where else in retirement. We do have air conditioning. It is a change for me, having lived for 62 years on the beach in Santa Monica. The only time I visited the UK, I was in London in November - not very warm at all; but I loved every minute of my time there.

Your workaround is to set up a spare (free) gmail account, use it to create a new Amazon account, give it a false address in the USA and no credit card info, feed it with gift vouchers (purchased from Amazon.com using your other Amazon account, and buy kindle books from the US market using that account. Then crack the DRM on them. Sigh.

On one hand, there is no need to do the credit card/gift vouchers dance.

You can just go into Account Options and change your "Kindle Address" to a US address. The account can still have it's primary delivery and billing address outside the US and you can still use a non US credit card to pay. None of that matters.

One the other hand, the problem is that after a while (not sure if it's time or number of books bought), any attempt to buy another one just gives a notice that you are out of the country for some time now and that you need to contact customer support.

As it turns out, the only thing relevant here is the country associated with the IP address used to access the amazon.com http server when placing the order.

Since I started using a VPN server in the US when ordering kindle books on amazon.com I didn't have any trouble again.

Once the books is actually ordered, it doesn't matter anymore if the connection to download it with Kindle4PC comes from outside the US.

Once downloaded it just takes 3 clicks more to unkindle them and archive them as unencrypted mobibooks.

I've speculated before that the history of copyright law (the USA was very different for most of the last century) may have been a big factor in the regional split of the English language market. I've seen claims that the standard contracts from US publishers still contain terms based on the old law.

History is a bitch.

It never ceases to amaze me how it's easier to simply bloody pirate the books than to legally purchase them... and the quips about them being proofed or edited are irrelevant if you simply get a version of the pirated book with a high enough number (something like v5.0 is often beyond the quality you get in the paper book, with even the typos that were in the "legal" edition having been fixed). Those people actually do go through the book, comparing it with the dead-tree edition and correct inconsistencies AND correct each other's work until you actually get a quality product.

To date I've managed to actually purchase only one book by OGH in paper, though I traveled extensively around EU cities during the past year (Wireless, in a tiny bookshop in Barcelona), and everywhere I go I make it a rule to check a few larger and at least a few smaller bookshops for a list of authors I want to give "conscience-money" to.

I've found it's possible to improve online book transcriptions pretty quickly by doing a diff against an independently done transcription.

So, if you take your own scan/OCR/light proofread, and then have a program automatically compare and highlight differences between that version and some not-really-legal fan-transcription, you can find errors pretty quickly; and it'll also usually be obvious if there are any places where there's been any "improvements" made in the fan-transcription (which can then be ignored).

That still leaves formatting to do, but most novel formatting is pretty straightforward (and being free of annoying typos is probably more important).

@35 not to mention that the technology used by the pirates seems to be about 10 years ahead of the publishing industry, and the workflow seems about ten times and fast and efficient. It took them about a day to rev "Surface Detail" to v5.0

Maybe the way to go is conscious money / donation sites for authors.

Apologies for going OT on this thread, but I see that "The Guardian" books blog, in an article about how Hollywood can't do SF, is suggesting Accelerando would make a good film...

http://www.guardian.co.uk/books/booksblog/2010/oct/15/hollywood-science-fiction

Hi Charlie.

I was reading Jerry Pournelle's Chaos Manor site and he mentioned that some of his colleagues from the new wave were asked about e-book versions. Most of them lived in the period where you typed. On paper. If you were lucky, you had correction tape.

(Ah, the smell of twink (TM). It takes me back to university).

They found torrented versions of their books and edited them.

I think the Baen model works... heck knows I have bought enough dead tree versions of things I have as an rtf file... but authors don't control that process. Neither does TOR. It's controlled by suits in New York and the city.

Besides, if i lose another copy of fuller, I can get another (I have four on "permanent loan". You can't do that with an McEbookReader.

On one hand, it would seem sensible for the industry to move towards language-based rights; Tor owns the rights to the English-language edition, Gollancz the German-language edition and so on.

NB: if this happens, you'd better get used to reading me in British English. Because I don't, y'know, write American English.

You get one of:

"Bob opened the bonnet of his Chelsea Tractor and unhooked the distributor cap. Then he went around and pulled the heavy suitcase out of the boot. Grunting, he manhandled it through the door of the block of flats and into the lift ..."

or:

"Bob opened the hood of his truck and unhooked the distributor cap. Then he went around and pulled the heavy suitcase out of the trunk. Grunting, he manhandled it through the door of the apartment block and into the elevator ..."

Wow, that's a really interesting criticism.

I help produce textbooks for students in schools in fourteen countries. (US, Russia, much of Asia, Africa, etc.) And I can appreciate the effort it takes to take older books and move them to digital form. We're in the process of digitizing our printed textbooks and it takes a horrendous number of hours and multiple edits, reviews and help from outside reviewers. (Your comments are good ammunition for my argument we move to InDesign.)

I had no idea Amazon was doing such a poor job with their books like that. (I haven't tried one yet.) It's inexcusable. Do you know if Apple is doing any better with their store? I'm reading The Grand Conjunction right now from the iBook store, and it sure seems like the books in the Astropolis series that came before. That would justify my sticking with more expensive iBooks rather than adding Kindle to my reading tools.

http://normanspinradatlarge.blogspot.com/2010/09/pirating-pirates.html Norman Spinrad utilized pirate e-books to get his earlier stuff on kindle.

I'm not certain, but I believe Apple, who license books on an entirely different basis from Amazon (google on "agency model") act as a software reseller -- much like the App store. In other words, they take ebooks generated by the publishers and pump 'em out, subject to basic QA checks. Amazon tried to position themselves as a publisher licensing the right to repub the text on a different platform, hence the radically different approach.

To make matters worse, I believe Amazon's approach varies depending on who the original publisher is and whether they can provide an ebook to Amazon in a format Amazon can republish, or whether Amazon generates one themselves. Even the Amazon file format is non-standard -- they've got a bastardized version of Mobipocket, and something else ("Topaz") that is entirely proprietary.

And I have pirated legally purchased copies of my own ebooks in order to get my hands on the final, corrected, copy-edited text. (Hey, what can I say? It's cheaper than paying a typesetting agency $200 a pop for the InDesign files!)

On the downside, Apple threw their file specs at the publishers weeks before iBookStore went live, and demanded a significant catalogue at launch. This works really well (NOT!) with the wondrous iTunes, which will only let you download a file you have bought once, so you can't get a later better-proofed version.

I'm afraid I've ordered a copy through Amazon of "The Trade of Queens" - because I CAN'T GET A COPY ANY OTHER WAY .......

Says something about the book-market, doesn't it?

What turns me personally of ebooks is currently the pricing policy. If you look at amazon.com list of your books all the ebooks cost more than the paperbacks. Most of them more than 30% more. Sometimes even the discounted hardcover is cheaper than the ebook.

I have read all your posts about the costs of producing books. But i cannot understand who would think it is good idea for the ebooks to cost that much more.

Add in the DRM and vendor lock in and ebooks are mostly dead to me at least until the publishing industry leaves DRM behind and i can be sure, that i can still read my books in 10 years time.

It's a pity i would like to own an ebook reader but not under this conditions.

46

Yep, publishers that don't like Amazon's dominance...but only sell at Amazon. They are the only MacMillans in the world that understand that illogic.

You can check out the Book Depository, too, though. Free shipping, all books.

Charles, being "forced" to read you (and other UK authors) in UK English instead of US English is a feature, not a bug.

Incumbent US publishers may disagree with me, but to quote from a great fictional politician, they are "mistaken about a great many things".

If I understand my wife correctly (she's put a lot of time into understanding Amazon file formats), the Topaz version is actually two things: a (badly) OCR'ed version of the original text, and a bunch of scans of the original text. When you read it, your Kindle (or local variant) displays the scanned text, so it's a 'perfect' replica, but for searching, notations, and so forth, it uses the OCR'ed electronic text.

This lets Amazon rerelease backstock very cheaply and in an automated manner. The human input that would normally be taken up in turning OCR'ed text into proofed, correctly formatted text is skipped, replace with the scans. However, it means that the Topaz formats, even ignoring DRM, are useless for translation into, say, epub format. (I suppose Amazon might consider this a feature, not a bug.)

Assuming the Kindle is presenting scanned images, I don't know how it deals with text scaling and so forth (perhaps it scans down to the word, rather than by the page, for example). But it does make Topaz an unholy mess to be avoided if at all possible.

M. Ellis@50, that's basically correct. The good news is that Topaz-format ebooks are extremely rare - in a couple of hundred books I have on my Kindle, I have exactly zero of them. Amazon doesn't push them because they provide a really crappy user experience, they're used as a file format of last resort. I did download a couple of samples of books in that format and yech.

Oh, and let me also put in a vote for one edition, and British English from authors from Over There - otherwise you run into things like when I listened to some recent Peter F. Hamilton. The audiobook was based on the UK release, and not only were things substituted as above, but there were sections that were edited fairly differently, whole paragraphs missing or added in places.

Amazon also doesn't like Topaz-format because it's larger, and they still have to pay for the bandwidth on the cellular connections.

Skip@51: I'd imagine most (all?) new titles are their mutant mobipocket format, but apparently quite a bit of backstock is Topaz. My bibliophile wife has grumbled more than once that a particular older, backstock title was only available in Topaz format.

The main problem Topaz that I see is that it lets Amazon legitimately claim they have a much bigger backstock of titles that nobody else has, but that it actively suppresses a market for 'useful' backstock in a properly done format (again, disregarding DRM for the moment). The mediocre is the enemy of the good.

As for rare, all the Pyr books are Topaz at Amazon. As mentioned above, as a format it makes good dunny roll.

Huh. "The Quiet War" by Paul MacAuley is a regular AZW (mobipocket) ebook on Kindle; its sequel, "Gardens of the Sun" (published a year or so later) is Topaz.

My (cough) problem with Topaz is ... I won't buy a DRM'd product if I can break the DRM on it, thus ensuring that (a) I'll still be able to use it in 12-18 months' time, and (b) I'll be able to read it in my preferred e-reader (I don't like the Kindle software). Mobipocket? No problem. Topaz? not so good ... and there's no way of knowing in advance which is which.

Actually, OCR does (fairly) well with PDF files. I did some work for electronic publication on a PDF of one book and while the recognition quality was a lot better than some of the paper editions I've used as source, since I could get "HERE is the text" set up as a template, it still had a fair number of glitches, including one major character's name getting repeatedly mangled.

However, typeset source files have problems of their own. For example, typesetters frequently use fonts that look like italics, but aren't marked-up that way, and when the stuff is exported from, say, inDesign, the italics disappear.

What it comes down to is that it still requires a human eye to double-check before releasing anything, treeware or electronic, to the recipients.

Topaz is why I don't buy books from Amazon in any format. They're DRM-infested to begin with, and the fact that there's no way to display the books in my Palm simply adds injury to insult.

Barnes and Noble is DRM-infested, too but:

They're real files(coughjailbreakable) and ePubs are amenable to being processed into something my Palm will deal with without upchucking.

For internal policy reasons, this did not meet with approval from Tor's parent corporate multinational. The brake pedal was well and truly stomped on, and Tor's involvement in ebook publishing ceased for a number of years.

Which is why, these days, I may buy four or five Tor books in a given year(e.g. Weber and Niven), and buy twice literally everything Baen puts out, once electronic for my Palm, and once paper to give to my local library's circulating collection as fan-bait.

Living in Australia I find the comments about ebook pricing amusing. I'm mostly using Kindle ebooks on my iPad - Every book I've bought so far (around 10) has been 20 to 50% cheaper than the paperback version let alone the HC. Living at the bottom of the world the transportation/wholesaler costs are so high that Kindle is a real bargain. Given I have a rather large library already and running out of space fast, being able to buy ebooks has been a boon. The only annoyance is the whole regional garbage with Amazon making some purchases impossible. In one case I have got a pirated book but all this did was prevent me making a bad decision on an author who had started a new series I did not like saving me the HC cost.

Much like William Moore@59 those of us living in SA are also affected by the spectacular concept of geographical restrictions. There are a few of your books available on Kindle but none of the Merchant War Series. Would love to know why something this restrictive would be put on books that are available locally in print.

My guess is "internal policy reasons... of Tor's parent corporate multinational" was an objection to e-book prices that properly reflected production costs compared to paperbacks and hardbacks.

Amazon deserves huge credit for getting the delivery model perfected after years of industry failure (including Apple) so that e-books and e-readers finally took off. Tor.com has provided us with the highest quality and most active SF site for readers I've seen. But Baen has the only really customer-friendly business model for e-books from a "big" SF publishing house.

With Webscription, lower production costs = lower retail price = more sales, and everyone wins. Hate DRM? Once you've purchased an e-book from Baen, assuming the digital rights don't expire or aren't withdrawn (looking at you Tor), you can re-download in a new format compatible with your new device. I've moved from Palm to HP to Nokia to my current iPhone and Kindle with no issues.

E-books haven't been popular long enough for everyone to really notice one more benefit - there's no economic reason for e-books to fall out of publication. One of the posters a few weeks back referred to Heinlein's "Take Back Your Government", which is out of print publication. No "photocopy" needed, go get the e-book on Baen. It only sells 10 copies this month? Who cares - the cost to maintain that database entry on the website is negligible.

How do you compete with "free" in a digital world? Content quality with value-adds (such as an instant delivery method) at prices that compare favorably with the labor required to pirate. DRM in e-book publishing is simply a technological effort by a naturally conservative industry to artificially maintain historical retail price levels in a digital system with drastically reduced real production costs. (Say that 3 times fast!) As the cost, ease, and availability of the technological tools to "rip" your own e-books proliferate, the publishing industry had better wake up to the lessons of the music industry. DRM doesn't prevent piracy, business models do.

As I sat down on the tube, yesterday, and took out my newly-arrived copy of "Surface Detail" ... I noticed the guy next to me was readin Jane Austen - on a Kindle. I didn't realise they were THAT THIN, or, quite frankly, so easy to read - I could see every word, from quite an oblique angle. So, maybe, some time, I'll get one or similar. BUT I'll want one that will accept any i/p, and without stupid "regional" boundary licansing.

Any suggestions, or is it a good idea to wait a year or so?

Greg, regional licensing for books isn't going away this year ... it's probably going to take a decade or two, and even when it's gone for all new acquisitions it won't affect backlist titles (i.e. older works). Likewise DRM. From a purely pragmatic point of view the best solution is to figure out what DRM formats you can crack, then set up accounts in both UK and US to buy the ebooks you want, and crack the DRM for your own personal use.

Charlie-Bob notes with interest that the Kindle app for iPhone runs happily on an iPod Touch, even an ancient first generation 8Gb one such as can be had for a relative pittance (£55-65 on eBay; probably cheaper elsewhere). And using an app called PhoneDisk from www.macroplant.com you can dig around the filesystem and see what files are stashed under "Books" by "Kindle.app" and copy them to your desktop computer to do nefarious things with them, ultimately transcoding them to DRM-free ePub for reading on, say, a second-hand Sony Reader (screen as good as the Kindle, but again, a damn sight cheaper on eBay).

Greg, have a look at the Sony Readers as well. They tend to be even smaller than the Kindles, and use EPub, and it uses the EPub format, for which there's now a reasonable selection of non-DRM'd books available from places like Fictionwise and Baen. For stuff available only in DRM'd format, Sony's ebookstore is convenient, cheap (pricing is about the same as Amazon), and has a surprisingly wide selection, including things not available on Amazon.

For example, typesetters frequently use fonts that look like italics, but aren't marked-up that way, and when the stuff is exported from, say, inDesign, the italics disappear.

Publishers are getting themselves into a whole new set of problems by using inDesign as the digital master version of a book. inDesign is great at producing print copy with highly-specified layouts, but it requires a specific discipline to get it to export coherent and properly-formatted reflowable text, and the exported epub always requires manual post-processing to tweak the css. Of course it doesn't help that inDesign CS5 is on its 3rd update and still has a glaring fault in its epub export filter.

There are far better options for producing the master proof of a book. In the majority of cases html/css2 is perfectly adequate - css is certainly incapable of handling the finer nuances of typography, but a lot of the time the work doesn't require them. Even a .doc file is a better choice for a proof that can then be sent to separate workflows for print and electronic output. There should be no need for further correction - if your typesetter is introducing errors into the print production process then you need to find a new typesetter - and publishers need to get out of the habit of doing textual corrections on galley proofs.

As far as the debate about quality of pirated versions goes, all I'll say is that I have to strip the DRM and tweak the formatting of the majority of ebooks I've bought, simply because the professional dolts who produce them often have no idea how to produce a reflowable work that conforms to simple typographic standards. The production house that Penguin uses is especially bad in this regard.