geni wrote:
On 11/15/05, David Gerard <dgerard at gmail.com> wrote:
Damn we need the rating feature. What's holding it up right now? List please, referring to current version of code. (I know the servers are creaking ...)
Do you know if the softwear has been released yet? If not it could be a way to get it tested.
It's been in the MediaWiki code base for months and we've been screaming for it to be switched on (see wikien-l in all that time).
I've directly asked on wikitech-l if there's some reason this feature is *never* going in and if I should just stop bothering, and been told that's not the case and that there are things that need fixing first.
But the original developer (Magnus Manske) *still* can't get any clear list of what needs fixing for it to go in.
So. What's up with Special:Validate?
[cc to wikitech-l]
- d.
David Gerard wrote:
It's been in the MediaWiki code base for months and we've been screaming for it to be switched on (see wikien-l in all that time).
I've directly asked on wikitech-l if there's some reason this feature is *never* going in and if I should just stop bothering, and been told that's not the case and that there are things that need fixing first.
But the original developer (Magnus Manske) *still* can't get any clear list of what needs fixing for it to go in.
I already told Magnus, if there's anything terribly wrong with it it'll get fixed after it's turned on (and if necessary, back off).
There's at least a half dozen people who could turn it on this very moment, but nobody's done it in all these months, probably because nobody on the server team thinks it's particularly important, useful, or high-priority. (If someone does think this and has refrained from turning it on for some other reason, they haven't told me so.)
It's a solution in search of a problem; it doesn't solve the validated-version-display issue in any way, it's just a survey form that might, in theory, produce data that might, in theory, be interesting or useful to someone one day.
Will it be worth the trouble of turning it on and possibly having to deal with fixing it when further problems become evident? Who knows.
So. What's up with Special:Validate?
It's on my list for this week, I'll see about getting it turned on and working.
-- brion vibber (brion @ pobox.com)
Brion Vibber wrote:
So. What's up with Special:Validate?
It's on my list for this week, I'll see about getting it turned on and working.
I'm tweaking, fixing, and reworking various bits of it (display formatting fixes, better code reuse, using revision IDs instead of unreliable timestamps in places, etc).
Will try to have it ready to try out tomorrowish.
-- brion vibber (brion @ pobox.com)
Brion Vibber wrote:
I'm tweaking, fixing, and reworking various bits of it (display formatting fixes, better code reuse, using revision IDs instead of unreliable timestamps in places, etc).
Will try to have it ready to try out tomorrowish.
*THANK YOU* (imagine sound of released breath :-)
BTW, I used a combination of revision IDs and timestamps, because you told me the revision IDs can get out of order through deleting/undeleting revisions, and I wanted them in correct order no matter what. If revision IDs suffice, great.
Magnus
On 11/20/05, Magnus Manske magnus.manske@web.de wrote:
BTW, I used a combination of revision IDs and timestamps, because you told me the revision IDs can get out of order through deleting/undeleting revisions, and I wanted them in correct order no matter what. If revision IDs suffice, great.
Ee. What happens if we delete a page then restore all but one version? Do the validates become off by one?
Gregory Maxwell wrote:
On 11/20/05, Magnus Manske magnus.manske@web.de wrote:
BTW, I used a combination of revision IDs and timestamps, because you told me the revision IDs can get out of order through deleting/undeleting revisions, and I wanted them in correct order no matter what. If revision IDs suffice, great.
Ee. What happens if we delete a page then restore all but one version? Do the validates become off by one?
The validation data itself refers to the revision ID only, so no problem there.
The time plus ID thing is only for the sorting of the revisions. You can say "merge all may old validations for this article into the current version". For that, it needs to determine "older versions". That's all.
Magnus
BTW, I used a combination of revision IDs and timestamps, because you told me the revision IDs can get out of order through deleting/undeleting revisions, and I wanted them in correct order no matter what.
You should use IDs (and never timestamps) for purposes of uniquely identifying a row, not for sorting. You should use timestamps (and never IDs) for sorting purposes, and never for unique identification. You shouldn't ever need to use a combination of both for a single purpose.
Timwi
Magnus Manske wrote:
Brion Vibber wrote:
I'm tweaking, fixing, and reworking various bits of it (display formatting fixes, better code reuse, using revision IDs instead of unreliable timestamps in places, etc).
Will try to have it ready to try out tomorrowish.
*THANK YOU* (imagine sound of released breath :-)
BTW, I used a combination of revision IDs and timestamps, because you told me the revision IDs can get out of order through deleting/undeleting revisions, and I wanted them in correct order no matter what. If revision IDs suffice, great.
They can still be out of sync at times (for some old cases with delete/undelete this used to happen all the time, but not anymore due to changes; if history data is imported with Special:Import it may however still be).
However the main problem is that timestamps aren't guaranteed to be unique; you will find pages where two adjacent revisions have the same timestamp (due to some old bugs, or due to merging of history between pages that have been moved over each other for instance). There were bits of the code where timestamp was getting used as the index in an associative array, which would mean that ratings on both those revs would end up conflicting, possibly vanishing or attaching to the wrong rev.
Sort by timestamp, use id as keys, and things should... hopefully... work. :D
-- brion vibber (brion @ pobox.com)
Brion Vibber wrote:
Brion Vibber wrote:
So. What's up with Special:Validate?
It's on my list for this week, I'll see about getting it turned on and working.
I'm tweaking, fixing, and reworking various bits of it (display formatting fixes, better code reuse, using revision IDs instead of unreliable timestamps in places, etc).
Will try to have it ready to try out tomorrowish.
Between the image server and still unpacking furniture I haven't quite finished the validation stuff yet. Unfortunately I'm still finding some XSS-style security holes (which could eg be used to compromise a sysop account by sending them to a specially-crafted link which uses JavaScript to perform actions on the wiki under their behalf).
So it definitely won't be going live until I've done a thorough security review as well as fixing up the revision stuff.
Also, there's been some muttering that turning this on will warrant/require a press release or some other sort of big publicity. I currently can't give any specific date for having it done or for turning it on once its ready. Sighhhhhhh...
-- brion vibber (brion @ pobox.com)
wikitech-l@lists.wikimedia.org