[Foundation-l] News from Germany: White Bags and thinking about a fork

[Tobias Oelgarte]

Am 23.10.2011 17:19, schrieb Nikola Smolenski:
> On Sun, 2011-10-23 at 10:31 +0200, Tobias Oelgarte wrote:
>> Am 23.10.2011 08:49, schrieb Nikola Smolenski:
>>> On Sat, 2011-10-22 at 23:35 +0200, Tobias Oelgarte wrote:
>>>> Why? Because it is against the basic rules of the project. It is
>>>> intended to discriminate content. To judge about it and to represent you
>>> No, it is intended to let people discriminate content themselves if they
>>> want, which is a huge difference.
>>>
>>> If I feel that this judgment is inadequate, I will turn the filter off.
>>> Either way, it is My Problem. Not Your Problem.
>> It is not the user of the filter that decides *what* is hidden or not.
>> That isn't his decision. If it is the case that the filter does not meet
>> his expectations and he does not use it, then we gained nothing, despite
>> the massive effort taken by us to flag all the images. You should know
> Who is this "we" you are talking about? No one is going to force anyone
> to categorize images. If some people want to categorize images, and if
> their effort turns out to be in vain, again that is Their Problem and
> not Your Problem.
It is wasted time for them as well as for us, since they are most likely 
editors that are part of "us". If they waste their time on 
categorization then it is lost time that could be spend on article 
improvement or invested in better alternatives that are illustrative as 
well as not offending.
>>>> easily exploited by your local provider to hide labeled content, so that
>>>> you don't have any way to view it, even if you want to.
>>> Depending on the way it is implemented, it may be somewhat difficult for
>>> a provider to do that. Such systems probably already exist on some
>>> websites, and I am not aware of my provider using them to hide labelled
>>> content. And even if my provider would start doing that, I could simply
>>> use Wikipedia over https.
>> If your provider is a bit clever he would block https and filter the
>> rest. An relatively easy job to do. Additionally most people would not
>> know the difference between https and http, using the default http version.
> If my provider ever blocks https, I am changing my provider. If in some
> country all providers block https, these people have bigger problems
> than images on Wikipedia (that would likely be forbidden anyway).
You can do that. But there are many regions inside the world that depend 
on one local provider that is even regulated by the local 
goverment/regime/... . Since the filter was proposed as a tool to help 
expanding Wikipedia inside this weak regions, it could be as well 
counterproductive. For the weak regions as also for stronger regions. 
Are you willed to implement such a feature without thinking about 
possible outcome?
>>> And if providers across the world start abusing the filter, perhaps then
>>> the filter could be turned off. I just don't see this as a reasonable
>>> possibility.
>> Well, we don't have to agree on this point. I think that this is
>> possible with very little effort. Especially since images aren't
>> provided inside the same document and are not served using https.
> Images should be served using https anyway.
It isn't done for performance reasons. It is much more expansive to 
handle encrypted content, since caching isn't possible and Wikipedia 
strongly depends on caching.  It would cost a lot of money to do so. 
(Effort vs Result)
>>>> If you want a filter so badly, then install parental software, close
>>> It is my understanding that parental software is often too overarching
>>> or otherwise inadequate.
>> Same would go for a category/preset based filter. You and I mentioned it
>> above, that it isn't necessary better from the perspective of the user,
>> leading to few users, but wasting our time over it.
> I believe a filter that is adjusted specifically to Wikimedia projects
> would work much better than parental software that has to work across
> the entire Internet. Anyway, I don't see why would anyone have to waste
> time over it.
That is a curious point. People that are so offended by Wikipedia 
content, that they don't want to read it, visit the WWW with all it's 
much darker corners without a personal filter software? Why does it 
sound so one-sided?
>>>> your eyes or don't visit the page. That is up to you. That is your
>>> If I close my eyes or don't visit the page, I won't be able to read the
>>> content of the page.
>> That is the point where a hide all/nothing filter would jump in. He
>> would let you read the page without any worries. No faulty categorized
>> image would show up and you still would have the option to show images
>> in which you are interested.
> If I would use a hide all/nothing filter, I wouldn't be able to see
> non-offensive relevant images by default. No one is going to use that.
It is meant as a tool that you activate as soon you want to read about 
controversial content. If you have arachnophobia and want to inform 
yourself about spiders, then you would activate it. If you have no 
problem with other topics (e.g. physics, landscapes,...) then you could 
gladly deactivate it. That is your choice and in that point the tool 
would support you.

Additionally you will have to consider the blurred image approach. It 
would not hide the image entirely. It's just strongly blurred enough to 
guess what it could be, but not so unsharp that you recognize nothing. 
Hovering over the image would make it visible at this moment. If the 
filter is off, all images would stay unblurred.
>>>> But feel free to read the arguments:
>>>> http://de.wikipedia.org/wiki/Wikipedia:Meinungsbilder/Einf%C3%BChrung_pers%C3%B6nlicher_Bildfilter/en#Arguments_for_the_proposal
>>> It seems to me that the arguments are mostly about a filter that would
>>> be turned on by default. Most of them seem to evaporate when applied to
>>> an opt-in filter.
>>>
>> None of the arguments is based on a filter that would be enabled as
>> default. It is particularly about any filter that uses categorization to
>> distinguish the good from evil. It's about the damage such an approach
>> would do the project and even to users that doesn't want or need the
>> feature.
> That is absolutely not true. For example, the first argument:
>
> "The Wikipedia was not founded in order to hide information but to make
> it accessible. Hiding files may reduce important information that is
> presented in a Wikipedia article. This could limit any kind of
> enlightenment and perception of context. Examples: articles about
> artists, artworks and medical issues may intentionally or without
> intention of the reader lose substantial parts of their information. The
> aim to present a topic neutral and in its entirety would be jeopardized
> by this."
>
> This is mostly true, but completely irrelevant for an opt-in filter.
>
As long the filter stays opt-in it could be acceptable. But who ensures 
that it will be so? If you only have access to the internet at an local 
institution and enforces an opt-in with enabled filter, then you have lost.
>> The German poll made clear, that not any category based filter will be
>> allowed, since category based filtering is unavoidably non-neutral and a
>> censorship tool.
> Who the hell are you to forbid me or allow me to use a piece of
> software? I want to use this category based filter, even if it is
> unavoidably non-neutral and a censorship tool. And now what?
>
We are the majority of the contributers that make up the community. We 
decided that it won't be good for the project and it's goals. We don't 
forbid you to use an *own* filter. But we don't want a filter to be 
imposed at the project, because we think, that it is not for the benefit 
of the project. Point.

nya~