[Server-devel] RACHEL, another developing world server project, would like to have a discussion with XS

[Martin Langhoff]

On Tue, Sep 29, 2009 at 5:13 AM, Dennis Nguyen
<denniskdnguyen at alumni.duke.edu> wrote:
> Hi Martin, thanks for the warm welcome!

My pleasure!

> We would definitely welcome re-use of the content, but since I'm relatively
> new to this particular project I don't have an expert sense of the
> particular permissions we have with the content providers as is. I will look
> into it.

Good point -- that is important! Content licensing should be under a
well known liberal license (GPL, GFDL, some of the Creative
Commons...), otherwise even RACHEL will have problems with it.

> Indexing: none! All of the content is currently listed on the index page
> with bare HTML. In our defense, this is a relatively new and shorthanded

Well, a bare HTML index *is* a perfectly valid index :-) and it beats
the performance and portability of any other solutions. How do you
maintain it? Is there any metadata around (dublin core or otherwise)?

> The wikislice we're using for the moment is the 2008/9 Wikipedia Selection
> for schools, which is handpicked front-to-back to fit the UK national
> curriculum. The website with details for that is at
> http://schools-wikipedia.org/

Wasn't aware it existed. Interesting. Still, of limited use to us,
being English only.

> what we can get away with as we go. We have English and Spanish content so
> far.

That's a good start.

> This project has a voracious appetite for content, and as such we'd rather
> stuff the project with content first regardless of compatibility with open
> projects like Gnash.

That is a reasonable approach I guess for your project. At the same
time, it limits its usability to OLPC and to low-cost netbooks.


> You can probably tell from this that this project is in a very early stage
> as far as maturity of the platform is concerned. I'm working to engage you
> folks at an early stage in case you have something grand to teach us :-)

I wish :-) -- all I can point to is the work and evolution of Linux
"distributions" (such as Debian, RH, SuSE, Ubuntu). They take software
developed by other projects, add a tiny bit of metadata, wrap it in an
installable package.

We are needing the same for content ATM, and the time is ripe for it.



m
-- 
 martin.langhoff at gmail.com
 martin at laptop.org -- School Server Architect
 - ask interesting questions
 - don't get distracted with shiny stuff  - working code first
 - http://wiki.laptop.org/go/User:Martinlanghoff