Roan Kattouw wrote: Are they? Please see http://en.wikipedia.org/wiki/Special:Export/Main_Page (and the XML dumps) - you can usually get the user ID from his user page this way, or any other page he has edited; also note that the Stewards can use the user IDs instead of the user names. I think it's better to have the user IDs public, to have a unique identifier of each user, and not to have problems with non-latin usernames. You can't do anything dangerous with the user ID (if you can, this is a bug which should be fixed, but I know no such bug). P.S. This discussion should probably be in wikitech-l.

On Wed, Feb 20, 2008 at 10:07 AM, Roan Kattouw <roan.kattouw(a)home.nl> wrote: Personally, I am doing analysis on users' reputation and trust. In my program, these values are keyed off of users' ids, not users' names. This is because it is easier for me to work with numbers rather than unicode strings. In general, I would argue that while it is possible to do without user IDs, there is no reason to, because they are already publicly available -- just not via the api -- and using numbers avoids the problems which non-ascii characters give.