Registered Member
|
(Seems I can't delete posts. Hey, here's a new brainstorming idea for the KDE board: Add a delete button!)
Last edited by d.h on Sat Sep 12, 2009 1:55 pm, edited 1 time in total.
|
Registered Member
|
Some time ago I've converted Wikipedia to dict format via a wget --> bunzip --> sed --> dictfmt pipline, but I abandoned that approach since the dict format seems obsolete.
From my experience I can tell that the wiki dump contains a lot of data that is useless for an offline enzyklopedia, thus the actual size of the database can be dramatically reduced. Nowadays I'd would strip off the useless data, convert all links, tables etc. to xml and import it into mysql. I wouldn't need a full app to access it, klipper and kdialog might be enough. |
Registered Member
|
This is one of the most brilliant ideas I have ever heard!
Proudly dual-booting openSUSE 11.1 with KDE 4.3 and Windows Vista on a Toshiba A205-S4577 since July 2007.
|
Registered Member
|
There is an application which provides the german wikipedia content offline. It is called tntreader, can read *.zeno files and makes them accessible for a webbrowser. Without images the german wikipedia zeno files are 4.2 GB. Then there is a version with all images as thumbnails (+ 3.3 GB) and all images with a maximum resolution of 1000x1000 pixels (+ 22 GB).
It would be awesome if somebody converted the english wikipedia to this format. |
Registered Member
|
They missed off Flickr, Facebook, MySpace, Twitter, Kopete/Jabber integration, Geotagging in Nepomuk... oh, and I'd use OpenStreetMap/Marble integration instead of/by default/as well as Google Maps. It could pull in photos of the location you typed from Flickr, find people on Facebook/Myspace in Blackpool, do a twitter/identi.ca search for the location and showing relevant tweets in real-time, see contacts that have (I assume) been previously Geotagged in Kopete/KContactManager and show OpenDesktop.org people that are in/near that location. There's always room for more!
Madman, proud to be a member of KDE forums since 2008-Oct.
|
Registered Member
|
They probably don't compress it since even the complete articles dump is not that big. And as I said, a lot of it can be stripped of if one only wants to have the pure articles. The dict files, which are compressed, were around half the size of the articles dump if I remember correctly. Or were it 2/3? |
Registered Member
|
There is already an application hosted on kde-apps called wikipedia dump reader:
http://www.kde-apps.org/content/show.ph ... tent=65244 however, he seems quite ubuntu focused and i really want to be able to use it on opensuse. |
Registered Member
|
Is this idea somewhere here on the forum? An application which in it's simplest form would try to connect information from different sources.... But have support for plugins customizing the parsing of a source. So if I have a wikipedia plugin, a "news from google plugin" and a id3 tag plugin activated and search for some active band I would get the wikipedia page about them, news about them from google and songs in my music library from them. Ah... just brainstormin' |
Administrator
|
|
Registered users: Bing [Bot], claydoh, Google [Bot], rblackwell, Yahoo [Bot]