Hoi, It is not that they cannot address the issue, it is that they have not addressed the issue. Given the amount of work that is on their plate and given the ambitions for MediaWiki it is not surprising that this is the case. When you state that it is not helpful because my position could derail your proposal, that is fine. We both agree that your proposal is a hack and, given the reaction of Domas, it is plain that he was not even aware of this standard that is applicable and therefore it is equally plain that no thought has been given to implement the collation orders as available in the CLDR.
One other fine reason why it is good to oppose this hack is because it will not work for other scripts particularly some of the Indian scripts. Consequently I do not think that your proposal should be implemented at all. Thanks, GerardM
2009/5/13 Lars Aronsson lars@aronsson.se
Gerard Meijssen wrote:
The point of this thread is that the collation is not working well. A hack is proposed and in my opinion it makes better sense to apply what is effectively the standard for collation.
Gerard, I don't think you are being helpful. My point in proposing the DEFAULTSORT hack (which is totally ugly), is that the WMF technicians are unable to address the issue. If they were able to "apply a standard", bug 164 would not have been open for more than four years. Sure, there are standards and they should be implemented at the database level (in MySQL, not in MediaWiki), which I think they are in MySQL version 6. It remains an open question how many more years will we have to wait for WMF to upgrade to that version. In the mean time, perhaps we should start (today?) to use DEFAULTSORT more aggressively?
-- Lars Aronsson (lars@aronsson.se) Aronsson Datateknik - http://aronsson.se
Wikitech-l mailing list Wikitech-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikitech-l