Talk:Molecule
Because of pdb files:
Wikipedia is currently not meant to be a file-system. I'm missing support for vector-graphics too. Currently the TEΧ Team is working on an implementation for Molecules and Graphs in the Wikipedia. But I guess this will be a Project, witch will took very long. Currently we could even use inline SVG, MathML and ChemML, but the current problem is, that browsers would need to support this.
MovGP0 04:00, 17 November 2005 (CET)
btw - how can we represent the structure of 3D Molecules semantically, so that it will be searchable? Wee need a technique to identify and represent the common parts.
MovGP0 04:03, 17 November 2005 (CET)
- Perhaps for proteins, pdb files of functional domains could be used as attributes. Perhaps for smaller compounds each functional domain could be an attribute. A list of known interactions and reactions of a given component in ChemML might be used to find compound with similar reactivity. Fuelbottle 13:35, 17 November 2005 (CET)
- An interesting thing is that a lot of different databases with information on molecules are downloadable. For example BLAST, Rfam and Pfam. Rfam and BLAST are public domain, while pfam is licensed under GNU GPL. Even though databases like that are public domain, searching in those databases with sequence alignment probably takes a lot of computational resources and it also takes a lot of software. Perhaps wikidata could have attributes that are URLs for other databases that it could search with data in its own attributes. For example if you were looking at an article about an RNA molecule that has an RNA sequence stored as an attribute in wikidata, it could then lookup the attribue for where to use that data to do a search and then get results back from for example Rfam and then get semantic relations like which species have this kind of RNA molecule and link to articles for those or which molecules it has evolved from and link to articles to those etc. This would be more Semantic Web though and would require more interoperability and standard formats for exchange between databases. Fuelbottle 14:40, 17 November 2005 (CET)
- You think of a special kind of Database links:
- http://de.wikipedia.org/wiki/Wikipedia:Datenbanklinks (german)
- --sry, I haven't found this in the english Wikipedia
- MovGP0 15:05, 17 November 2005 (CET)
- Well that could be used to perform a query, the problem is that you don't get the answer back in a standard format that wikipedia/wikidata could parse for compare to data in its databases, that would require databases to agree on a standard format for data exchange. For example if I do a query in Rfam I don't get a result back in any XML format. Fuelbottle 15:52, 17 November 2005 (CET)
- In theory though I think a possible solution could be to write routines for a proxy server to convert results from a query to RDF so that after making a query from wikipedia/wikidata, the results could be possible to use for a semantic search. Fuelbottle 16:35, 18 November 2005 (CET)
Good Idea. I'm thinking further on the JENA Framework and Sesame witch are allowing RQL-Queries. RQL-Queries are very powerful, and I recommend to implement a possibility to do so. But for end users RQL is not the language of joice, because Queries in Wikipedia are even done by 6 year old childs and - more terribly - my mother ;-). So we need a easy-to-use Query interface. This could realized as an separate editor whitch generates RQL-Queries in Background and runs them against a special Webservice. I think further that most parts of Wikipedia should get realized as Webservices, so we can modularize Wikipedia more. From the standpoint of an efficient runtime environment, this will take additional resources. But therefore it would be more scalable, and programmers (even commercial ones) could develop own seach applications and better bots.
MovGP0 16:59, 18 November 2005 (CET)
^maybe I'm a bit away from the original theme - but my thoughts are already at the next steps. MovGP0 17:07, 18 November 2005 (CET)
<?xml version="1.0" standalone="no"?> <!DOCTYPE svg PUBLIC "-//W3C//DTD SVG 1.1//EN" "http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd"> <svg width="100%" height="100%" version="1.1" xmlns="http://www.w3.org/2000/svg"> <rect x="20" y="20" width="250" height="250" style="fill:blue;stroke:pink;stroke-width:5;opacity:0.9"/> </svg>
Above says "we could even use inline SVG". Can that really be enabled? Browsers can easily support using free add-ins for instance from Adobe. Would allow for some vector graphic capability using standard components. Though svg is primarily 2D, potentially it can also be used to show 3D inline to visualize some of the semantic data even if it's just wireframe. Also is there an inline query that would make adding User and date in edits more automatic? USoy May 20, 2006
[edit] good job
Hi Keep On!
Nice job!
<a href="http://www.amir-car.1website.net">Car insurance</a> <a href="http://www.car.1website.net">Car insurance</a> <a href="http://cheap-hotel-ny.narod.ru">New York City Hotels</a> <a href="http://cheap-insurance.awardspace.com">Cheaper Car Insurance</a> <a href=http://www.amir-car.1website.net>Car insurance</url> <a href=http://www.car.1website.net>Car insurance</url> <a href=http://cheap-hotel-ny.narod.ru>New York Hotels</url> <a href=http://cheap-insurance.awardspace.com>Cheaper Car Insurance</url> Good night