That is a distinct possibility. And, I’m not referring to the importance of applying a process consistently, without fear or favour. At the start of this year, when I first had a go at contributing to Wikipedia, I was perplexed at the use of the hostname upload.wikipedia.org. In all the years I’ve been using computers, the convention has been that the name ‘upload’ is reserved for an area that has special properties with regard to uploading – normally for the temporary storage of files. It’s the directory on an FTP server that the anonymous account can write to. It can be the directory on a photo printing service that doesn’t have a quota, so you can build up a temporary print job of any size. As a subdomain, it’s normally the access point for maintenance of a website. Location names with the word upload in them are not usually intended to be a way of permanently referring to something.

The IWF couldn’t ask the Wikimedia Foundation how their websites work. Do the IWF have the time or expertise to perform their own technical analysis of each website hosting a potentially illegal image? I have no idea why they didn’t at least attempt to block both the page and the image. Some of the URLs on the blocklist with image extensions could be there because they are the exact URLs as submitted by members of the public. It has become clear to many that the IWF’s procedures need to be revised. Maybe the work needs to be brought within a UK police force, so that it will be possible to call on more resources to make better decisions and to give those decisions greater legitimacy.

http://www.chiark.greenend.org.uk/pipermail/ukcrypto/2008-December/085789.html

and

http://www.lightbluetouchpaper.org/2008/12/11/technical-aspects-of-the-censoring-of-wikipedia/

(BTW, SecureComputing is currently blocking only the URL to the image, not the full Wikipedia page).

@PA – Are saying that the IWF analysts cannot tell the difference between Wikipedia and a site devoted to delivering child porn that dynamically changes the paths to the images to avoid filtering?

My *impression* from the garbled things they’ve said, is that they’re used to dealing with images where URLs have tracking data in the file path, hence “specific URL” is variable. If you look at sophisticated spam in HTML, you can see what I *conjecture* is their reasoning. So they block the most static-looking way of accessing the image.

We have no money to spend on expensive legal assistance on issues like this. (We do have Mike Godwin, the world’s most famous Internet lawyer, as our general counsel, which is just unbelievably cool and very effective.) So all we have is people talking about these issues and giving us their spare bucks to keep the sites running. Thank you :-)