Plyd wrote: No. -- brion vibber (brion @ pobox.com)

Because it's not that useful, not a part of an original plan and implemented by a third party, and not well implemented at that. *Maybe* when it matures a bit, and a definite need can be established. On May 25, 2006, at 2:34 PM, Plyd wrote:

On 5/26/06, Brion Vibber <brion(a)pobox.com> wrote: I didn't take part in this project and I'm sure they could convince much better than me but... First, it's very easy to add their relations in the already existing infoboxes (makes relations for all the articles at the same time). My point of vue : - With semantic, data can be used at different places and defined only once (a bit like openrecord.com, with their views). - It could create automatic lists (example lists of cities sorted by country, by population, etc.). - Relations makes browsing information much easier than with categories. (Example: I'm looking for all english scientists born between 1500 and 1550) - for the wiktionnary it could help classifiying data, and make it easier to present it in the view you like, external or not. - RDF export could help making Wikipedia an online database (as imdb, but for various subjects, and open), that could be used inside other softwares. For example, music showing Wikipedia neutral info, title, singer picture, its discs etc. while playing my mp3, or it could be for a book). Maybe also in a map, with cities etc. Imho semantic could help spreading wikipedia content (relations but also text) on more supports than the actual website. Plyd

On 5/26/06, Brion Vibber <brion(a)pobox.com> wrote: Is this any sort of discussion process for working out what does and doesn't make it in, or does it come down to existence of a developer willing to implement it, and belief of other developers that the change won't break stuff? Steve

On Fri, May 26, 2006 at 10:39:21AM -0800, Elliott F. Cable wrote: The idea of wiki syntax is to be simple. Easy to learn. MediaWiki already includes many features where this does not apply: * table markup * templates * advanced image syntax The additional changes to the syntax needed for Semantic Mediawiki would make the syntax even more difficult to understand, especially for people who are not aware of semantic markup. Things like San Diego is a [[is a::city]]. It has a population of [[population:=1300000]]. are not very intuitive any more. Good ideas how to hide the complexity from newbies would be very welcome. Another problem with SMW is code quality, which currently is not good enough for use at Wikipedia. Scalability has improved with 0.4, but more improvement is still needed. SMW also needs a security review, looking at the code for a few minutes already showed XSS issues that need to be fixed urgently. Regards, jens

On Sat, Jun 03, 2006 at 09:19:28AM +0200, Steve Bennett wrote: The category is of only limited use the way it's currently implemented. There is no efficient way to find the sub/supercategory relationships. And the {{city infoboxes}} aren't a usability improvement at all. Their syntax is even worse. And of course we could put all the information into city infoboxes, but then we could stop writing articles and just create tables. There is more information about a city then the infobox. [[discovered by::Columbus]], [[founded by::Caesar]], [[conquered by::William II.]], ... Regards, jens

On Sat, Jun 03, 2006 at 07:17:50PM +0200, Steve Bennett wrote: I'm not one of the developers of the Semantic Mediawiki extension, but I am a user (most notably on the Bible Wiki at http://bible.tmtm.com/) and for me the two key goals are: 1) Internal queries - being able, for example, to dynamically display a list of all the rulers of Judah in chronological order (http://bible.tmtm.com/wiki/King_of_Judah) 2) External queries - the data that is annotated in the text can be retrieved as RDF (http://bible.tmtm.com/wiki/Special:ExportRDF/Rehoboam) which can then be fed into an RDF reasoner. As well as being able to use this data locally, this will become increasingly useful as search engines start to gain the ability to work with structured data rather than just raw text. (See announcement just this week for Pingerati) Tony