Hi,
Le Sunday 22 May 2005 17:05, Magnus Manske a écrit :
David Gerard schrieb:
I suggest that we allow ratings by anonymous
users (IP numbers), at least
in 1.5.
Reasons for this:
* we've always worked by leaving things as open as possible and only
restricting as needed;
* we're explicitly not using the data for anything important yet, so if
ten thousand rating spammers put [[Image:Autofellatio.jpg]] top marks for
everything, it won't actually affect anything;
* the raw data will be of great interest to people, and as wide as
possible is good. (I can see the academics studying Wikipedia slavering
for the ratings data tarball ;-)
Two reasons against this:
* Later, we will allow only logged-in users to rate articles, right?
Otherwise, we'll lose a great part of the perceived reliability
improvement, IMHO. But how can we really set up this system if the data
we use as a foundation for the decision is based on anon entries as well?
I think it better if only logged in users can validate articles.
But well it depend what we want to do with this feature: selecting articles
for an offline publication or studying psychology and sociology of Wikipedia
readers ?
We have a precedent: only logged in users can upload images.
* Currently, I store only user IDs with the ratings.
Since anons don't
have user IDs, I'd have to change the SQL table for that, in addition to
"just" some code, which means work for me ;-)
Magnus
Yann
--
http://www.non-violence.org/ | Site collaboratif sur la non-violence
http://www.forget-me.net/ | Alternatives sur le Net
http://fr.wikipedia.org/ | Encyclopédie libre
http://www.forget-me.net/pro/ | Formations et services Linux