I was reading the article Why the commercial ebook market is broken (http://www.antipope.org/charlie/blog-static/2007/03/why_the_commercial_ebook_marke.html#comment-6103), when I came across this in the comments:
My suggestion was to standardize at 800x600 monochrome bitmapped pages, because anything higher drove up the minimum cost of the end-user device. It could handle simple images and anyone's alphabet, yet avoided the scope-creep that seems to drive us to laptops displaying PDFs. There's more about it, but that seems to be the core idea. That struck me as very interesting, as that would solve the cost issue in a rather neat way.

Let's assume that the hypothetical device had a extremely low-end CPU with 6" display (LCD for now, maybe eInk when prices fell down). All it would need to support is a very primitive OS and UI for browsing various books. The file format would be blindingly simple, a ZIP file with an "index" describing the TOC and various profiles available. Each profile would essentially be a set of page images in some format (PNG?) fitted to the ebook's screen. You could have a profile for small font, one for large font and another for landscape (it would all depend on the publisher). This would increase the size of the book from ~200-300kb to around 3-4MB, but storage is very cheap nowadays so wouldn't matter as much. The advantages would be many:


support any language with all the "standard" book formatting we expect in a paper book
low CPU/battery requirements, cheap to build
OS can be very small and no further "upgrades" required
Easy creation of books from any application ("virtual" printers)
Books can be read directly on the PC side too


It would suffice for most people and make low-cost ebook readers widely available. It could peacefully coexist with more costlier devices which supported much more features (People would upgrade if they felt the need).

I know this is probably a pie-in-the-sky idea, but what do you think?

What you are describing is similar to the lBOOK eReader V6 (http://www.mobileread.com/forums/showthread.php?t=9670). Which is estimated to cost ~$150 with a Microcup 6" SVGA Electronic Paper display by the end of 2007. This does appear to be a viable approach to a dedicated e-book device.

Well,

First, the LCD is going to kill the life of the battery as compared to e-ink.

Second, ebooks are already easy to read - once the proper ebook application is installed - on the PC.

Third, the various SBCs out there are dirt cheap already, even the high-power versions of the ARM and XScale cpus. So coming up with a NEW design using a cheap but under-powered cpu makes no sense.

Fourth, and here's the killer, pre-formatting every 'page' to an 800x600 monochrome bitmap wastes space.

Further, I submit to you that it's not the cost of the ebook reader which is causing the problems. No, it's the insistence by so many publishers that each ebook be locked up with DRM to 'protect' each copy from 'pirates' and 'thieves'. You really need to read Eric Flint's essays on DRM to more fully understand this.

And if I had my druthers, I'd druther make a $10 flexible e-ink-based ebook which holds one and only one story because the $50 ebook reader is too expensive - for the 'feature set' which could be installed at that price. A $50 ebook reader would have no more features than the $10 flexible ebook I mentioned first but would cost five times as much. On the other hand, the $300-$450 ebook readers that are out there have far more features I, as a reader, want and can be used without difficulty to read a wide range of ebook formats - as well as have their software expanded and upgraded when improvements come along. Also, I'll be far more willing to take care of a $300, full-featured device than I would a $50 one because I would perceive it to have more 'value'.

JMO

Derek Benner

Well,

First, the LCD is going to kill the life of the battery as compared to e-ink.

Second, ebooks are already easy to read - once the proper ebook application is installed - on the PC.

Third, the various SBCs out there are dirt cheap already, even the high-power versions of the ARM and XScale cpus. So coming up with a NEW design using a cheap but under-powered cpu makes no sense.
Well, with a LCD you can still get a life of around 16-18 hours easily -- we get that right now with the 1100/1150 readers. Also, I wasn't suggesting to make a new design -- just use the cheapest design that is available right now.


Fourth, and here's the killer, pre-formatting every 'page' to an 800x600 monochrome bitmap wastes space.
I agree that it would be a waste of space, but that makes the device simple in the extreme -- a bit like the digital photo frames which you can get very cheaply. Also, storage is not that much of an issue -- even with this "waste", the size of a typical book would not be more than 4-5MB.


Further, I submit to you that it's not the cost of the ebook reader which is causing the problems. No, it's the insistence by so many publishers that each ebook be locked up with DRM to 'protect' each copy from 'pirates' and 'thieves'. You really need to read Eric Flint's essays on DRM to more fully understand this.
I'd tend to disagree on that. It's a chicken and egg situation: you need the hardware out there before publishers will take it seriously, and people won't buy the hardware unless the content is out there. And no one (except for dedicated bibliophiles) will spend too much on a device which may not work or be supported in a few years. And yes, I have read Eric Flint's essays.


And if I had my druthers, I'd druther make a $10 flexible e-ink-based ebook which holds one and only one story because the $50 ebook reader is too expensive - for the 'feature set' which could be installed at that price. A $50 ebook reader would have no more features than the $10 flexible ebook I mentioned first but would cost five times as much. On the other hand, the $300-$450 ebook readers that are out there have far more features I, as a reader, want and can be used without difficulty to read a wide range of ebook formats - as well as have their software expanded and upgraded when improvements come along. Also, I'll be far more willing to take care of a $300, full-featured device than I would a $50 one because I would perceive it to have more 'value'.
I agree with you that a $10 book reader would be even more better, but it simply isn't economical at the moment. I came up with the $50 figure out of thin air, so don't take it as gospel :)

Also, this "format" neatly sidesteps all format wars: you can convert everything to a raster image, so there would be no problems of not being able to read this or that format on the device. I've been reading PDFs on my 1100 lately, and I can't distinguish between rendering done by the ebook with the page image generated on the PC. So essentially you can offload all the processing tasks to the PC, leaving the device to be dumb. It's a perfect example of the "worse is better" principle. The idea of behind the Info Pad (http://mobileopportunity.blogspot.com/2006/05/desperately-seeking-info-pad.html) is similar to that -- make everything as simple as it can be.

I agree that it would be a waste of space, but that makes the device simple in the extreme -- a bit like the digital photo frames which you can get very cheaply. Also, storage is not that much of an issue -- even with this "waste", the size of a typical book would not be more than 4-5MB.

I disagree most ebooks text only are from 500k-1.5 MB. You convert that to an image format, you are going to have ebooks from 50MB-1GB. Do the math, if one assumes that each image file is a measly 67k, and most ebooks have 1000 electronic pages, thats already 67megs. And I think I am being light on each image file.

I'd tend to disagree on that. It's a chicken and egg situation: you need the hardware out there before publishers will take it seriously, and people won't buy the hardware unless the content is out there. And no one (except for dedicated bibliophiles) will spend too much on a device which may not work or be supported in a few years. And yes, I have read Eric Flint's essays.

I think what you need is an ebook reader in the $150 range. Many people would pay that.

I agree with you that a $10 book reader would be even more better, but it simply isn't economical at the moment. I came up with the $50 figure out of thin air, so don't take it as gospel :)

Also, this "format" neatly sidesteps all format wars: you can convert everything to a raster image, so there would be no problems of not being able to read this or that format on the device. I've been reading PDFs on my 1100 lately, and I can't distinguish between rendering done by the ebook with the page image generated on the PC. So essentially you can offload all the processing tasks to the PC, leaving the device to be dumb. It's a perfect example of the "worse is better" principle. The idea of behind the Info Pad (http://mobileopportunity.blogspot.com/2006/05/desperately-seeking-info-pad.html) is similar to that -- make everything as simple as it can be.

I just dont see what converting ebooks to raster gets you except bigger files, and less compatibility. DRM is the problem here, as well as expensive hardware. My answer get the eReader II down to $150, and more widescale adoption will occur. Much like mp3 players, until the price of the hardware came down, there is not alot of widescale adoption. Now everyone has an mp3 player.

Also what happens when I try to read the 800x600 ebook on my 640x480 device? It resizes the image down? Now the text is unreadable. What if I try to read it on my 1024x768 device? Now the text is blocky.

I disagree most ebooks text only are from 500k-1.5 MB. You convert that to an image format, you are going to have ebooks from 50MB-1GB. Do the math, if one assumes that each image file is a measly 67k, and most ebooks have 1000 electronic pages, thats already 67megs. And I think I am being light on each image file. Well, if its a monochrome image, then it won't ever be more than 10kB, typically around ~4-6kB. I have actually had to deal with this while developing PDFRead, so I am pretty sure about the file sizes. Assuming 1000 pages (which seems a bit high, but whatever) the file size would be ~5-6MB. Pretty manageable, as cards of 1GB are very cheap nowadays.

I just dont see what converting ebooks to raster gets you except bigger files, and less compatibility. DRM is the problem here, as well as expensive hardware. My answer get the eReader II down to $150, and more widescale adoption will occur. Much like mp3 players, until the price of the hardware came down, there is not alot of widescale adoption. Now everyone has an mp3 player. Well, there is a good enough reader in the eBookwise 1150 at $125, but no one buys it much. That's because it is way too costly. Read the article I mentioned above, most people don't read more than a few books a year -- buying something above the $100 psychological point is just not going to happen. Also, MP3 players had it much easier getting adoption -- you could easily rip a CD to get MP3s, you can't do that easily for books as OCR, proofreading, formatting, etc all are very labor intensive.

Also, this ebook would not targeted for heavy ebook readers who see the advantage of buying a Sony Reader or 1150 or whatever is current then. That is just preaching to the choir -- this for people who don't use them that often and don't see the value of going in for a high investment in something they don't use that often.


Also what happens when I try to read the 800x600 ebook on my 640x480 device? It resizes the image down? Now the text is unreadable. What if I try to read it on my 1024x768 device? Now the text is blocky. Why would you have different screen sizes at the start anyway? In that unlikely case, it's upto the publisher to provide multi-format ebooks ie. support different screen sizes (or font sizes, or orientation, or whatever).