Slashdot Mirror
Wikimedia and KDE Cooperation Announced
eean writes "As reported by KDE Dot News, today Jimmy Wales, chairman of the Wikimedia Foundation, announced the beginning of a cooperation between Wikimedia and the KDE project at LinuxTag in Karlsruhe, Germany. As the first applications, like the media player amaroK,
start to integrate Wikipedia content the idea is to create a webservice
API to access the information from Wikimedia projects such as Wikipedia or Wiktionary."
they better not name it Kikipedia, or Wikide.
The plan WAS to get all these linux penguins playing tag, but they never quite got started. The rules consulted on Wikipedia.org were changed from moment to moment.
Finally! This is something I've been preaching for a long time.
While I understand it isn't (directly) the goal of open-source to compete with Company, Inc., the next generation of computing tools is going to be heavily service-oriented. That is to say: open-source has thus far concentrated on making software "products" -- applications, utilities, libraries, and so on. In a service-oriented community, though, open content is just as important as the tools that use it.
Furthermore, I like to see when open-source products doing a little more -- wait for it -- synergy. (Shoot me.) Thus far, open projects have, apart from sharing code and libraries, stayed mostly to themselves. But partnerships like this are absolutely beneficial to creating a cohesive, seamless user experience. Via services, you create an entire open "platform" that isn't just the tools, but the content that backs it up. It also creates an entirely new market for companies to support open-source software.
I use Trillian as an IM client and it does something similar to this. I get underlined links with more information from Wikimedia if words in the message match an existing wiki topic. I think it is pretty cool.
Insert Generic Sig Here:
"Earl Grey tea is a black tea blend, which gets a distinctive flavour and aroma from bergamot oil..."
I've looked at a bunch of different ways of playing and organizing my small mp3 collection (is 20GB still small?). Amarok has been one of my favorite aps when I'm using my linux desktop. When I've set up my file server to stream mp3s over the web to my office computer (running Windows), I've used Jinzora.
Now, both of these programs use a MySQL database backend to help organize and catagorize mp3s, and use id3 tag information. I think it'd be an absolute joy to have them share the same data, instead of using two redundant databases. Perhaps as this project matures, we could see this become a standardized format for multiple projects to use, so the information can be shared and edited more easily.
"What do you think?" "I think 'What, do you think?!'"
well, wiki is a big bag of web-accessible content that could fit nicely into desktop applications... personally i would like to be listening to a cd on my computer, say "here come the warm jets" by brian eno, and be able to pull up a short bio on mr. eno, his complete discography, the lyrics to the current song, a list of reviews of the album and artist and the cover art for the album... all without having to fire up my web browser and performing a bunch of search-n-clicks.
2 1337 4 u!
Wikipedia in your media player is one of the things you don't notice you were missing until you have it, after which it's indispensible. I hope other projects start to take advantage of the bindings, hopefully not just within KDE but elsewhere as well. This should benefit everyone.
I am trolling
I'm sure KDE's competitors at Gnome will rush to collaborate with Google in response, given how their name starts
I am trolling
> Wouldn't it be a better use of resources to incorporate this into a browser or word processor?
AbiWord has already done for some time.
> I already know who sang the song - they're my legal copies
Yes, but this isn't about looking up the song somehow to see who sang it (try MusicBrainz for that). It's about looking up the band to read about their history or other news. (Wikipedia does have articles on a surprising number of bands.) And nothing says that amaroK (not Amarock) would insist on popping up a Konqueror window with the results from every new song -- lots of features can be disabled as a preference, or not even compiled in in the first place. (In the same way that you'd only want your word processor to do this with words you highlight instead of *every* word you type.)
And of course there's already a list of other KDE stuff that could integrate this to benefit from information that's already been collected, instead of duplicating information on stars and chemical elements and countries (for instance).
So it's *already* been incorporated into a browser *and* word processor (there's a search plugin for Firefox); why not add it to amaroK and Kalzium and KStars and.... ?
/me compiled the Amarok beta earlier today.
:-)
It doesn't do the Wikipedia lookup unless you ask it to - normal operation is the same as ever. It works very nicely
I actually started building a wiki for a CS class with this goal in mind. Rather than a traditional HTML wiki, it was built around RDF data. So not only was it able to dump the data in a machine-readable format, it also had a built-in RDQL query engine that worked over http.
To conform to the data model, I had to impose restrictions on the input, but the final product is a lot more reliable than screen-scraping HTML, which is something I hope the kde/wikipedia people avoid in this project.
Badass Resumes
Rumour has it that Linus is going to dump Git and use Wikipedia to host the kernel code.
Who needs a moderator?
--- Dan
They were removed just to give a more clean interface. There's an icon to open the page in a web browser if you want to edit. amaroK is a media player, we wanted to avoid turning it into a full-blown web browser. :)
What if KDE's own manuals and help system were rsynced with its original data in wikipedia? What if all manuals were rsynced from wikis?
We could type 'man ls' and get the latest page with comments and all.
Sure beats submitting manpage patches to developers.
Even better, like in wikipedia, you'd click on a word in a manual page, and you'd get the man page of that manual and all related pages...
Now combine that with the google search engine.
"Give orange me give eat orange me eat orange give me eat orange give me you." -Nim Chimpsky