[Commons-l] Getting rid of 'best before'

Thu Nov 23 01:11:01 UTC 2006

I'd appreciate some sort of a mechanism (maybe java backup) that will allow
me to better check the images.

It takes too much time to process every individual entry with the detail I'd
like to put it. And after the 1000th delete it starts to get annoying to
manually check history of the image, contributions of the tagger/uploader
and etc.

We need a bot to at the very least warn the uploaders if they weren't
warned. I wouldn't however want a bot that starts deleting things. No matter
how hard a bot is written, it is possible to trick a bot into doing
destructive stuff.

On 11/15/06, samuli at samulilintula.net <samuli at samulilintula.net> wrote:
>
> > 2006/11/15, Magnus Manske:
> >> Theoretically, one could write a bot which could do the deletions, so
> >> no admins get scolded. The idea of a deletion-bot with admin rights (a
> >> virtual equivalent of [2];-) might sound scary, but setting it for "no
> >> edits for a month since addition of the template", combined with the
> >> image resurrection capability, should calm this.
> >
> > If someone runs such bot, please don't delete more that, let's say,
> > 100 images a day, otherwise CommonsTicker gets stucked :)
>
> I'm not a bot, but I have deleted up to 1000 images per day :)
>
> But seriously, when I've been in a deleting mood, I have also spared
> tagged images. There are at least three kind of good images amongst tagged
> images:
> 1. Vandalism tagging (rare)
> 2. Ineligible (chemical formulas, PD-art)
> 3. Images where information has been provided after tagging for deletion.
> This is the most common group.
>
> These are also reasons why a bot wouldn't be an extremely good idea.
> Although considering that the time wasted on deleting could be spent
> better, I'm not totally against a "monster bot", either.
>
> --
> Ystävällisin terveisin,
> Samuli Lintula
>
> _______________________________________________
> Commons-l mailing list
> Commons-l at wikimedia.org
> http://mail.wikipedia.org/mailman/listinfo/commons-l
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.wikimedia.org/pipermail/commons-l/attachments/20061123/469752d5/attachment.htm 