[Foundation-l] Possible solution for image filter

[Tobias Oelgarte]

Am 21.09.2011 20:05, schrieb Andre Engels:
> On Wed, Sep 21, 2011 at 7:20 PM, Tobias Oelgarte<
> tobias.oelgarte at googlemail.com>  wrote:
>
>
>> I still can't the a rational difference between images included in
>> articles by the will of the community and text passages included by the
>> will of the community.
>
> It's much easier to note offensive text fragments before reading them than
> to note offensive images before seeing them. But I guess the more
> fundamental issue is: there are, I assume, people who have requested this
> feature for images. There are either no or only very few who have requested
> it for text.
>
I would doubt that. For me it seams only to be a technical issue. Images 
don't change over time (at least not often), while text is in constant 
movement. The images are also in constant movement. Some will be 
replaced by others, some will be updated, some might be moved to another 
sub-article and so on. That means filtering images is technically, in 
comparison to text, the only feasible element that could be implemented 
in a more or less direct way.

Thats why "no one asks for text". Actually i think that we have more 
potentially offending articles / text passages then images. Just count 
the biology/species articles with this enormous info boxes showing the 
development of species (an exploration by Darwin). If we could filter 
text, we would have more then enough claims to remove that. I'm sure 
about that.

The basic thought progress at the WMF must have been:
A: "We need to do something, otherwise we could lose some donors. We 
need to look fresh and attractive."
B: "But what do we do? All we can really do is something technically, 
without upsetting a huge amount of authors."
A: "Yeah Wikitext is so hard to parse and we have already a project for 
that. This will take ages..."
B: "Didn't we have some complains. There was a group that claimed 
Wikipedia has to many male authors."
A: "A you mean that gender-gap project. But just look at our pages. Who 
without studying informatics would really participate? It's way to 
complicated and we should represent some results now".
B: "Hey, yesterday i read a comment by Heroxxxx from FOX that we have to 
much porn. OK, they had nothing else to report, but this could be something"
A: "Great idea. Lets delete all pornographic images."
B: "We can't do that. Look what happened to Jimbo. As soon we delete the 
images it will cause problems."
A: "Just got an idea. Hiding is not deleting. How about hiding all this 
images by default."
B: "Would that be accepted? Some might ask: Why only porn?"
A: "OK then we need to make it more general"
B: "Wouldn't they cry this is despotism and censorship?"
A: "Let's see... How about we let someone write a report, praise him as 
neutral and to make sure that the report sees a great need for such a 
feature? We could argument, that it is important and not our idea."
B: "Thats great. Could we improve that also for text?"
A: "Text would be so hard and it would remind people on blacked out 
pages. I don't think that this would be an good idea. But how about to 
give them a new tool to decide if images are hidden or not? I see a lot 
of reasons to do so. It could please FOX and some other critics."
B: "Wouldn't this just move the problem to another project?"
A: "Who cares. Let them handle it. We will just say that the community 
will find a solution, as we always do."
B: "OK. Bye"