Fed up of paying through the nose for e-books? Ion, the company best known for its USB turntables, is readying a gadget that will help you digitise your paperbacks, hardbacks, magazines and comics. Called the Book Saver, it's a large frame into which you place an open book. Tap the Scan button and the spread is digitised and …
Still at the Mickey mouse stage.
7 seconds per scan, plus how long to lift, turn, replace, check? Probably not much faster than just reading the book. (For comparison, that's why I still haven't digitized all my singles, EPs and LPs.)
I suspect it would take the rest of my life to let me get rid of all the books I have on my shelves.
Nice idea 'though - I mean scanning and binning all one's books.
Is there any automatic machinery available to do this at a price that might become reasonable?
perhaps it would be quicker to cut the books up, put them in a duplex sheet feeder on a large copy/scanner and then OCR the results but you'd knacker the book irreparably.
The clue is in the name - called Book Saver - scan the pages without destroying the book.
Wait for the publishers to start yelling about piracy... someone needs to give a few of their chiefs a solid kick in the nuts when they do.
Now, if it turned the pages for you, that would be a worthwhile gadget.....
If that comes at a sane[*] price
then I want one. I've got an awful lot of books waiting to be scanned, and some shiny new automation to do the conversion.
Though I've tried both flatbed scanners (slow, and prone to introduce noise at the gutter as the page lifts) and camera/tripod methods (sensitive to exposure, but faster) and the flatbed has a much better image for scanning.
[*] pun unintentional
This will be ace for the good folks out there doing the magazine presertation projects (at least in the RISC OS world - Acorn User, Archimedes World, BBC Micro User, and so on). Home enthusiasts can't afford industrial scanning machinery but this looks like it might be reasonably affordable to the home user.
Nice idea, but it seems to hold down the pages flat using glass or plastic plates. Sure must give a bunch of reflections and prevent using a flash....
out of the diy then
Saw this on wired last year: http://www.wired.com/gadgetlab/2009/12/diy-book-scanner/
And it was by no means the first one I've seen.
Who would bother?
Seriously? Who would bother using this thing (beyond the initial New Gadget Euphoria)? A short book might not take too long to scan, say half an hour, but it's already small and lightweight enough to just take the book with you on the train. For a big fat book you're looking at chucking away a couple of hours of your life and probably a good dose of back and neck pain from crouching over this device for all that time.
If you just wanted to grab a couple of pages, it would be less hassle to just slap it on your multifunction printer/scanner. And this is the sort of thing that would sell for £119 - so who would bother?
I have a duplex sheet scanner but have not cut any of my books up yet. Have thought of it, but that would be quite the sacrifice.
Maybe I should buy and cut up and scan a book on book binding and book restoration first :-)
So I could scan the book I'm reading at the moment (The Distant Hours, by Kate Morton) in... let's see... factoring in 10 seconds for a page turn (probably very optimistic)... ignoring sleep and the endless hours in on the phisio's table recovering from the resulting RSI... 3 days!
Reminds me of "Three Days of the Condor (1975)" (Robert Redford)
In the movie they had a Digital Equipment (RIP) PDP8 scanning books and it had an automatic page turner!
Still, when this manual device hits the market it should do well, so everyone can have their very own :Google" digital library.
Nice try, but...
"just" lifting up the whole thing to turn a page? The thing looks a bit wobbly to do that more than a few times, nevermind the hundreds of times for serious book scanning. What about alignment? While I laud the initiative even if it isn't the first to try, this usability thing needs a bit more attention, I say.
I have been converting my personal library to a simple subset of HTML (device-independence FTW) since I started moving around internationally for work about a decade ago. But there is still about 1/4 that is just too awkward and/or delicate to lay out on a flatbed scanner.
This. is. brilliant.
First truely useful new gadget (for me at least) this century! I hate risking cracking the spine to get a decent scan of (part of) a book.
Ion readies book scanner for e-book buffs
Looks like it would take about as much time to read the book as it would to scan it in.
Not that it isn't a useful gizmo (I've got some old manuals and references it might be worth the time to scan to have a digital copies of), but hardly practical for everyday kind of stuff.
Better way to go.
Buy a second, dirt-cheap, poor-condition, used copy of the book. Doesn't matter if it's ugly looking or stinks of cigars, vomit, and cat pee. Cut the spine off, and scan it using a flatbed scanner with a sheet feeder. Then recycle it.
The main problem is that you get the front side of each page, and then the other side of each page, and you need to collate them, combine the files, OCR them, etc.
Ban the Title Requirement for Replies NOW!
Actually, there are some cheap scanners out there that will do a duplex scan; and I believe the bigger ones (if you have one of the office MFD's at work) will do the work in short time. I've scanned a 300pg book in short order, duplexed properly in a couple hours. Something you can do overnight easily for sure.
Re: Better way to go
Actually, that's not the main problem because it's easy to handle that problem with the excellent Finereader OCR package that will assemble the text into a single stream; you can specify the area to scan to eliminate potential errors (eg margins, headers and footers); it'll even automatically recognise and process multi-column pages - an all-round excellent match between problem and solution!
The primary problems are in A). quality of image - cheaply printed books (mass-market paperbacks) typically have poor quality paper - whilst good OCR packages can often handle the poor definition of characters on the page caused by ink absorbtion, poor paper opacity allows text on the reverse side to be visible making nonsense of the OCR process. B). page pickup by the document feeder - how many times have you seen the feeder grab two sheets at a time? And when scanning manually, can you be sure that you've not skipped a page? The final output always has to be laboriously checked for errors.
I have access to a scanner/photocopier that can turn the pages and do both sides of each page, but when scanning a big pile of papers it is more efficient (faster and fewer jams, particularly if the papers are a bit tatty) to scan all the front sides then put the pile in the other way round to scan all the back sides then collate the page images with a script.
A system whereby the capture device supports itself over the book stand on 3 edges would save significant parts of the stupidly demanding workflow (removing before flipping page, and then replacing the capture device).
Even then, the mind still boggles as to who would seriously consider this value for money. According to their own figure (doubtless on the optimistic side of a range), you're probably going to spend 25 minutes scanning the average novel. This is assuming their crop & skew / OCR tech is ground-breakingly advanced and doesn't require any editing on your part — in which case you could probably double the figure.
Personally, I value my time such that it will always be cheaper for me to buy a new digital edition. Which leads me to the belief that anybody who wants to make this thing worth their while will be of the philanthropic anarchist bent, and seed the fruit of their efforts across the interwebs in an aim to get a worthy return on their investment.
Regardless, I'm almost certain that by the time I'd found the spare time to get 1/100th of the way through my bookshelf, publishers will have devised a system whereby in person presentation of a bought hard copy will be exchanged for a digital version.
Solution for publishing industry
print text closer to the binding of the book.
It's not the scanning that takes the time...
Even with a slowish flatbed, a hundred page pulp magazine takes only a couple of hours to scan. It's the damn proofreading that takes the time - days to get it accurate. Long enough that I've just devoted a year to a Master's on the subject. As I said earlier, I've got an awful lot of books, and many are both fragile and irreplaceable (a 17th century work on watch making, for example, or early 20th century science fiction).
But it's a *pain* to scan and collate the images, ties up the computer for that couple of hours (yes, you can do other stuff, but the context switching every thirty seconds makes it distracting) - saving straight to a memory stick which I can digest at my leisure appeals to me.
Ah, an area I /can/ add professional comment at last! To address many of the comments:
Bulk book scanning is possible - indeed, there's many providers out there that cater for it today (both for rare books that need delicate handling and for 'normal' processing). You can buy hardware that turns pages - but it's pretty expensive and usually reserved for high-volume processing work. When the book can be sacrificed, I'd suggest the average joe will get far, far better value by cutting the spine off and using a quality sheet-fed scanner.
Non-destructive book scanning requires more intelligent processing too - without a glass plate over each page, which is slow to manipulate and consequently expensive in terms of time and hardware, the pages will curve and the resulting images require a camera that allows for depth of field variations - not something the cheap end of the market caters for that well. The image will also contain lines of text that curve; again, something the processing software needs to take account of (searching for 'straightening curved text lines' would be a good start!)
No matter what book scanner you select (I assume we're not in the 'pro' end of the market now), you'll have to light the work correctly. What usually happens is that two lights are provided, positioned outside the left and right edges and angled in at about 45 degrees. This avoids reflections and ensures the page is lit evenly and there's minimal contrast changes over the image. However, I'm not sure how you'd achieve this with the Book Saver as there's nowhere the lights could go without casting a glare into one camera or the other. This implies the whole workspace will have to be very well - and evenly - lit. This is expensive: to get a decent, high-contrast image requires a lot more light than most of us are usually working under. To keep illumination consistent, given the legs and overhead box casting shadows? An angle poise or two simply 'ain't going to cut it.
If you have very delicate material, it should go without saying that you're better off leaving the job in the hands of the experts - the hardware costs alone will put you off trying it yourself - but for personal or occasional use you could consider taking photographs with a digital camera, paying attention to strong and even lighting of course - I can't stress this enough, and submitting the images to your software of choice. At a push, use your Smartphone camera (again, there's software out there designed to handle the distortions that mobile camera images delight us with)
Of course, you might want to check and correct any errors in the OCR before you finally convert to your format of choice. If you're creating PDFs, you'll need to use OCR software that can give you an ability to review and correct the text - not all do this.
Someone has one.
I remember seeing* a scanner used by, I think, The British Library. The book rested on glass tent with all the scanning done from inside the tent. All the operator had to do was lift the book off the tent, turn the page and replace. Not quick but they were scanning valuable books so they could be viewed by the great unwashed.
They appear to have automated the process.
*It was probably on Tomorrows World. They don't do programmes like that anymore.
Yes, and a sordid story it is too.....
BL has Microsoft funding to digitise its works - see the announcement which just happened to coincide with BL's support for Microsoft's attempt to get its OOXML "standard" adopted by ISO.
Slightly off topic now but isn't this what the ALTO XML extensions were designed to address?
Maintained by The Library of Congress (www.loc.gov/standards/alto/)
Wallace & Gromit's World of Invention
Was close to [Tomorrow's World] but I don't think the BBC bosses noticed. Phew!
As for this thing... some documents aren't going to be available online. And sometimes you just want to see the pages. I might even buy myself one.
I could scan my genome faster ...
... than it would take to scan Jonathan Livingstone Seagull with that contraption.
Gutenberg still rules
Just another gadget that's meaningless outside of (maybe) a few small niches. The comments reflect the tech industry's tunnel-vision on the subject. Most focus on the time-efficiency of the device, which is a serious enough defect. But the real killer is the quality of the digitized text--which is useless for real reading unless days have been spent editing it. A couple of commenters note this much. What's even more rarely noted is that proofreading and editing isn't a DIY thing for the great majority of people, however literate they are (or think they are). And professional editing and proofreading is well on its way to becoming a lost art.
There has always been a disreputable (to professionals) segment of the publishing industry that markets the package rather than the text, to buyers who want the books for display rather than reading. These publishers allot their production money accordingly, and the texts are accordingly bad. Typically, these are "fine-binding" publishers. Occasionally, their pitch is fashionable typography. (Bodoni was the king of fashion typographers. His texts were notoriously bad.)
The appeal of DIY e-texts is, in a way, the continuation of this niche, appealing only to techno-suckers who don't anticipate actually doing much reading, of books or of any other sort of textual information, but want to say they're e-book collectors (or e-documentation users). (Professionally published e-books are, hopefully, better edited. But the quality of printed texts has degenerated too, and it's unlikely that e-books will be better.)
The real value of DIY e-texts, whether scanned for one's own use or posted on the web, is that someone has done the first step--capturing the keystrokes. That's a great deal of work in itself. But it's only the first step. Unless those texts have been proofread and edited to higher standards than all but a few can provide, using them for all but the most casual purposes is neither efficient nor reliable. (Recall Mark Twain's comment: "I never read medical books. You could die of a misprint.")
Seven seconds - figure wrong?
Well, i checked up on their website (http://www.ionaudio.com/booksaver) and they say:
"While similar devices require up to seven seconds per one page, Book Saver takes only one second per two pages! "
Perhaps you should update the article, it's quite a serious difference.
Been waiting for this...
Just set up my scanner for my two boys to scan some of the books I want to have on my Kindle. They get 1 Eurocent per page and herewith add to their pocket money. Well possible that this gadget will lower my "production costs"... http://www.reghardware.com/Design/graphics/icons/comment/happy_32.png