Summer of Code 2007/proposals

Category Page

I'd nominate working on the Category page to see if there's a way to precalculate the article and subcategory counts, and to make the navigation of long lists of subcategories and articles more like the All Pages special page. The SplitCategoryPage extension I've posted is undoubtedly not up to the code quality spec of the real developers, doesn't do all of the above, and is sloooow. But it might be a nice project for a mentee to work on a similar functionality with better supervision.

JimHu 14:22, 15 March 2007 (UTC)[reply]

Category math

See Category math. I'd very much like to be able to restrict searches to articles that satisfy category exressions, i.e. find all articles containing text "xyz" that are in categories a, b, or c but not in D. --DavidBiesack 17:01, 15 March 2007 (UTC)[reply]

WiXi

WiXi is copy of http://wiki.laptop.org/go/WiXi proposal to make twext translations work with wiki software(s).. a WiXi plugin for MediaWiki could transform stable Wikipedia content into a language learning platform providing comprehensible input .. use firefox browser to text basic twexter written in javascript/php Read.fm 20:22, 08 July 2007 (UTC)[reply]

xcroll

a "xcroll" function (one scrollbar controls both edit/input *and* preview) could be a useful function for both WiXi and regular MediaWiki editing.. a javascript xcroll working in FIREFOX is started here under GPL.. update: an improved xcroll is here Read.fm 21:20, 08 July 2007 (UTC)[reply]

File:XcrollWiki.jpg

Improving the MediaWiki API

The current version of the MediaWiki API lacks quite a lot of necessary functions (such as editing articles) so this might be a good idea for a student to work on. Extensively improve the API by adding and fixing features. Jayden54 18:32, 16 March 2007 (UTC)[reply]

Unified Login / OpenID support

Another good idea is getting MediaWiki to support OpenID, and possible Unified Login for the WikiMedia projects. See Help:Unified login for more information on Unified Login. If Unified Login is too much of a hassle then start with OpenID which would provide the same benefits and probably be a lot easier to implement. Jayden54 17:28, 17 March 2007 (UTC)[reply]

Currently single user login is being developed, see Single User Login. Perhaps after single user login is enabled we will then be able to move towards OpenID. Lcarsdata 09:50, 18 March 2007 (UTC)[reply]

Haven't they been working on the Single User Login for a long time now (months, if not years) and there's still no definite release date? It's probably easier to build in OpenID support in MediaWiki which provides a single login for every MediaWiki website, which would be even better than the Single User Login. Jayden54 19:56, 23 March 2007 (UTC)[reply]

Wikiproducts

How about a prototype of esp for Wikiproducts? --Fasten 13:39, 18 March 2007 (UTC)[reply]

Semantic extension improvement

Developing an implementation for n-ary relations (W3C) for semantic extensions (like Semantic MediaWiki) to MediaWiki. Optimale 08:33, 21 March 2007 (UTC)[reply]

Wikipedia offline

The idea is to make Wikipedia accessible offline, by downloading and converting XML to SQLite data, thus we can have access to any information we need regardless of internet connection. This can help making knowledge accessible even for those can't afford internet, or make a local wikipedia server on any network. it can also help reducing bandwidth charge on Wikipedia servers. — Preceding unsigned comment added by Ntlk (talk • contribs)

See Dumps. MaxSem 19:49, 26 March 2007 (UTC)[reply]

wysiwyg

What is the status of making WP editor wysiwyg? Badgettrg 22:22, 30 March 2007 (UTC)[reply]

Web mashup for automatic display of PubMed (National Library of Medicine) references

I propose a mashup that allows the author to insert the PubMed ID in some format similar to 'PMID ######' and a mashup will add this to the page's references and will place the correct bibliographic data into the reference section. Currently inserting 'PMID ######' nicely generated an automatic link to the citation at PubMed, but it does not display the citation on the Wiki page.Badgettrg 22:22, 30 March 2007 (UTC)[reply]

Development of a metric of impact of an author or institution

I think WP would benefit from such a metric. This could then someday be the substrate that would allow institutions to measure their faculty input into Wiki, the could lead to academic credit for authors, might even lead to institutions competing with one another to compare the impact of their faculty. This could lead to enticing the best experts to become authors. Badgettrg 22:22, 30 March 2007 (UTC)[reply]

Web 2.0 CMS and API

Wikimedia could offer a new type of content managment. The CMS API will be easy to use and learn and could provide all the tools a programmer could need to create CMS modules easily. The API also will be supporting Web 2.0 features like AJAX. That API could later be integrated with the Wiki API to create a even better tool.

The CMS part must also contain a Control Panel that will provide basic CMS functionality (The rest will be in the modules) and a module instalation system.

Everyone with a basic knowdelge of Object Oriented Programming in PHP5 and a good understanding of the API would be able to create their own CMS modules like blogs, static pages, image galleries, etc.

Search

Is search improvement too big a project for Summer of Code? Search is still the elephant in the room, and it doesn't appear on the MediaWiki roadmap.--67.186.162.70 04:10, 31 March 2007 (UTC)[reply]