Devel Digest, Vol 32, Issue 113

Greg Smith gregsmitholpc at
Mon Oct 27 10:29:21 EDT 2008

Hi David,

Thanks for the detailed write up.

I created a Feature request for better eBook reader on the feature 
roadmap page at:

For requirements, I included a link to this thread and a few specific 
items which came to mind.

If you have the time and inclination, could you update that requirement 
with a bulleted set of things you think the reader should be able to 

(e.g. flip pages one at a time, zoom, etc.).

If John or anyone else wants to start filling in the specification 
section with design ideas that would be helpful too. Links to other 
pages or inline is fine.


Greg S

 > > david at wrote:
 > >
 >> >> opening the book in .html just produces a lot of errors becouse 
the book is
 >> >> spread across lots of files, and the jurnal isolation mechanisms 
copy the
 >> >> file being opened to a temporary directory, where all the links 
to the
 >> >> other files don't work.
 > >
 > > Note that you could have entered file:///media in Browse and 
located the HTML
 > > files; I believe the links to other files would then have worked OK 
but you'd
 > > have to wander around with the USB flash drive hanging out of your XO.

thanks, I'll have to give that a try. this should also work with a SD

unfortunantly, the mere fact of plugging in the USB stick makes the
journal go through the entire thing and index it. this takes a significant
amount of time and the XO is unusable during this time (I ran into a few
min where the mouse didn't even respond)

 > > In an ideal world the book would be packaged as a single collection 
 > > file, so downloading it would unpack it in ~/Library and add it to the
 > > content navigation in the OLPC Library home page.  You could try
 > > and
 > > .

I'll look into these.

 > > (I put Little-Brother.xol on USB, it showed up in Journal, I chose 
 > > and this all happened and Browse displayed the e-book.)
 > >
 > > In this ideal world the .xol container would gain traction as an 
 > > format and Bain, Project Gutenberg, and the other content 
repositories would
 > > offer books as .xol bundles.

I just tried a quick google search and found three different .xol file
formats (an biometrics data format, a database format, and a map format),
and I'm still looking for the one that you are using for collections.

if there is a trivial means to convert from .xol to a standard directory
(tar/zip) then it would have a chance.

if you were to define your bundle as a zip file with specific files in it
then you could also have your software use heristics to deal with zip
files without your metadata in it (if by no other way then to show the
list of files to the user and ask for the nessasary data)

 > > HTML in Browse integrates cleanly with the library/home page, can use
 > > advanced CSS for attractive layout, takes you from a link to a 
 > > without the download-Journal-Read steps, avoids PDF's fundamental 
 > > rendering a paper page on a screen, has JavaScript to add 
interactivity and
 > > features like annotations, etc. etc.  It's the future.  But PDF is 
 > > an important legacy format.

html in browse does have one nasty problem, it shows partial lines of text
at the top and bottom of the screen

as another smaller problem, you have to figure out where to put the
oversized mouse pointer to minimize it's annoyance when you are trying to
read. I initially tried to put it on the right edge of the screen, but I
discovered that it's very easy to flex the case enough to click the mouse
butten when in tablet mode, which scrolls you to whereever the mouse
happens to be sitting on the scrollbar.

David Lang

More information about the Devel mailing list