My Iliad experience

[CommanderROR]

Thanks Henrycat!

Don't bother about the other shots if you don't have the time...my Iliad should arive during the next 2 weeks (if they don't run out of supply before Nr. 470..^^)

[arivero]

Originally Posted by ali

Sorry, but no, precaching does not solve this problem, at least not for me.
(...)
However, if you skim through something - say, a scientific paper - it doesn't help at all. I read papers not in linear order, I'm jumping all the time.

Hmm, besides precaching, if you are reading .tex papers then a typical xdvi reader could be developed, in principle.

Has anybody got insights about how to crack into the OS? I would like to see a table of running processes, memory usage, etc.

[ath]

Originally Posted by arivero

Has anybody got insights about how to crack into the OS? I would like to see a table of running processes, memory usage, etc.

Check up on embedded Linux in general: there are several books to read. Yaghmour's book is often recommended.

The best way to get binaries in there will be by standard SDK methods, and I suspect that means generating a bootable diskimage (or part of one), and loading it to flash memory over USB.

But how are you going to start the binary? And how are you going to see the output? Will the kernel even have the syscalls you know? No shell here, I don't think. Do you have the appropriate header files, libraries and cross-compiler to generate the binary? (In short, do you have the SDK?)

Once you do, you may have to enable some debugging mode to make it work, or run a special debug-version of the OS, and perhaps even use a special development iLiad device to ensure you can't bypass DRM too easily.

But I suspect you knew all that already.

[ali]

Originally Posted by arivero

Hmm, besides precaching, if you are reading .tex papers then a typical xdvi reader could be developed, in principle.

Papers are usually tex, but distributed as postscript or pdf. Ps->pdf conversion is dead easy, so I just need the described viewer. Which means, I need the SDK.

[arivero]

Originally Posted by ali

Papers are usually tex, but distributed as postscript or pdf. Ps->pdf conversion is dead easy, so I just need the described viewer. Which means, I need the SDK.

Not for physics. The original texfile is provided in the ArXiV for more than 95% of the papers. So it is possible to reprocess almost any of them for a 1024x792x16 display without margins.

For pdf/ps it could be more sensible to move them to pnm images, then pnmcrop to get rid of the margin, then recompress with algorithms of compression designed for b/w text pages, as in some pdf formats or in the dejavu format.

Also depending of the pdf, it could be possible to play with the boundingbox limits to wipe the margins; sure it can be done with acrobat, and perhaps with some unix utility.