View Single Post
Old 05-04-2013, 09:10 AM   #54
fjtorres
Grand Sorcerer
fjtorres ought to be getting tired of karma fortunes by now.fjtorres ought to be getting tired of karma fortunes by now.fjtorres ought to be getting tired of karma fortunes by now.fjtorres ought to be getting tired of karma fortunes by now.fjtorres ought to be getting tired of karma fortunes by now.fjtorres ought to be getting tired of karma fortunes by now.fjtorres ought to be getting tired of karma fortunes by now.fjtorres ought to be getting tired of karma fortunes by now.fjtorres ought to be getting tired of karma fortunes by now.fjtorres ought to be getting tired of karma fortunes by now.fjtorres ought to be getting tired of karma fortunes by now.
 
Posts: 7,762
Karma: 59654515
Join Date: May 2009
Location: 26 kly from Sgr A*
Device: PRS-T1, KT, PB701/IQ, K2, PB360, BeBook One, Axim51v, TC1000
Quote:
Originally Posted by rkomar View Post
I doubt that there is any piece of software for reading epub/mobi/pdf that efficiently runs on embedded systems; it's just too complex (even without DRM).
It doesn't exist simply because nobody has bothered to do it.
The technology exists.

The Raspberry Pi shows the way to go because it still has a ton of (relatively) expensive things you don't need for a dirtcheap reader. Connectors, I/O ports, etc.

So does this:
http://www.ebay.com/itm/400480308539

Quote:
Specifications

Display 2.8" LCD TFT display
Storage capacity 2 GB
USB 2.0
Screen resolution 320 x 240
Expansion slot TF
Control type Touch-screen
E-book format TXT

Audio format MP3, WMA, WMV, WAV, ASF
Music format MP3, WMA
Video format AVI
Photo formats JPEG, BMP, GIF
Microphone Built-in
Work time Approx. 3 - 5 hours
Recharge time Approx. 2 - 3h
Replace the single chip MP3/TXT SOC for a microcontroller and a big PROM and you'd have a $20 reader instead of an $18 Media player. The electronics are cheap; the software is a one-time effort.

Now, epub may or not need a full OS worth of support (PDF certainly does) but Mobi doesn't.
Mobi ran (and still runs) fine on the very first PalmPilot. That is a 16MHz 8-bit cpu with 128Kb of memory.
Plenty of modern microcontrollers can do a lot better.
Arduino exist for these kinds of (hobbyist-level) things.
Arduino-capable chips run $3 or so, maybe lower by now:

http://www.instructables.com/id/The-...crocontroller/

A reading library Arduino design exists, The Humane Informatics Reader:
http://en.wikipedia.org/wiki/Humane_Informatics

Estimated cost $20.

The technology to do a very low cost reader (LCD-based for now, eink soon) is not the issue. It can be done. Hobbyists and enthusiasts have done it so a big tech company can do it, too.

The issue is the *business* case.
Simply put: everybody listens to music, not everybody reads. Most certainly, not everybody reads narrative text for recreation and that is where the money lies.

The limits on commercial ebook readers are, ultimately, human limits.

If somebody can make a business case for it, it will happen.

Now, if we were talking educational devices for literacy programs to be given out for free to children (something like the HI reader above) the thing could be built next week with a grant from the Gates Foundation or other charity. A million or so to do the software in assembly could easily cut the price of the device (and increase the durability and reliability) to MP3 player levels and, in fact, use "MP4 player" components for the case, screen, and controls. (320x200 color touch screen? Perfectly fine. I read on that for years.) It would make sense for everybody involved if the intent were to build ten or twenty million to preload literacy libraries and airlift the things into developing countries or rural populations.

Tech is easy.
What is hard is the people.

Last edited by fjtorres; 05-04-2013 at 09:13 AM.
fjtorres is offline   Reply With Quote