[OLPC library] wikimedia parser

[drew einhorn]

I'd really like to see you import dictionaries.

And mash them up into a big multilingal dictionary with links to
pictures, sound, and video.
Add word frequency statistics.
And difficulty levels dynamically assigned like freerice.com

And present it all in a wiki so we can expand dictionary in languages we speak.

A multi lingual freerice.com style activity would be amazing.  where
the student picks
a language or a pair of languages.

Many people could write their thesis analyzing the statistics that could
be gathered from using the activity.

Especially if we can capture some demographic info without compromising privacy.

I'm sure you will find amazing variations on these ideas I haven't thought of.


On Jan 2, 2008 9:56 PM, David Farning <dfarning at gmail.com> wrote:
> Hello all,
>
> I would like to thank you all for the great work you are doing with
> the laptop.
>
> First, a real life reaction to the laptop.  I bought the laptop for
> myself under the g1g1 program.  I tested it for a few days for before
> deciding that the xo was too valuable to sit in my office.
>
> I gave it to my 5th grade niece for Christmas. It was a big hit.
>
> How does this relate to the list?  When I saw her again for new years,
> she was showing all of her cousins how to use wikipedia.  The kids
> spent the rest of the day sledding and seaching wikipedia.  How
> amazing is that.
>
> This positive response has encouraged me to get involved with the
> project.  My interest is helping create the library.
>
> I have been looking at the process of bringing content into the library.
> I am interested in generalizing wiki splice to
> allow it to import item from different sources and automatically
> create .xols.
>
> My basic premises are that content curators should not have to worry
> about the details of making a bundle.  Programmers should worry about
> the content in the bundle.
>
> I have started rewritting wikisplice in more generalized manner so that
> it can take different types of input and build the bundles
> automatically.
>
> A interesting development is mwlib found at
> http://code.pediapress.com/wiki/wiki that is sponsored by the
> wikimedia foundation.  It is a library for parsing raw mediawiki text.
>
> I will hopefully have some prototype code ready for submission in a few days.
>
> Thank
> David Farning
> _______________________________________________
> Library mailing list
> Library at lists.laptop.org
> http://lists.laptop.org/listinfo/library
>



-- 
Drew Einhorn