[Foundation-l] 80% of our projects are failing

[Gerard Meijssen]

Hoi,
When you are to build a system that connects Wikipedia / Wiktionary etc
articles to Commons, you are building a system that relies on the articles
to exist in the languages you want to get the data from. So it is restricted
to the data that you have in the projects. To build this data, I would use
the software developed by Daniel Kintzler for his master thesis and expand
it for the languages Daniel does not yet support. This approach will work
for Wikipedias. What you get is the type of data that can be included in a
system that is based on the OmegaWiki notions and that will need a database
that is quite similar to OmegaWiki.

With an OmegaWiki implementation, we can include information from languages
we do not support within the WMF. Consequently we can provide infromation
that is not provided by any of the projects. So, yes you can. However there
is more that you can do.

As you may know, in OmegaWiki we demonstrated how to connect to both Commons
and Wikipedias. The big advantage it provides that there is no need for
connecting to Commons from each Wikipedia article. You only connect from the
concept both to Commons and the various Wikipedias.

Yes, OmegaWiki is Open Source and its data is Open Content.
Thanks,
       GerardM

2008/12/1 Erik Moeller <erik at wikimedia.org>

> 2008/12/1 geni <geniice at gmail.com>:
> > People use the search feature on commons?
> >
> > I would assume they would click the link at the bottom of
> >
> > http://nl.wikipedia.org/wiki/Paard_(dier)<http://nl.wikipedia.org/wiki/Paard_%28dier%29>and get taken to
> >
> > http://commons.wikimedia.org/wiki/Equus_caballus?uselang=nl
>
> This is a valid point, especially when one uses it as a starting point
> to think about search. It might be feasible to build a search tool on
> the basis of this existing tagging of Wikipedia articles to Commons
> media -- and similarly, Wiktionary, Wikinews, and so on. This is an
> alternative to the notion of one giant ontology that's used for
> tagging. Instead you would treat a wiki -- any wiki -- as the
> ontology. So you could do a Wikinews/Commons search for terrorism, a
> Wiktionary/Commons search for pronunciations, etc. Because the
> approach would be wiki-agnostic, it would also be language-agnostic,
> and yield useful results as long as the underlying wiki is large
> enough and its articles are well tagged to Commons.
>
> What would be the technical requirements of this approach and what
> would be its disadvantages?
>
> --
> Erik Möller
> Deputy Director, Wikimedia Foundation
>
> Support Free Knowledge: http://wikimediafoundation.org/wiki/Donate
>
> _______________________________________________
> foundation-l mailing list
> foundation-l at lists.wikimedia.org
> Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
>