A GFDL corpus access provider or GFDLcap is a web service or other net media, or potentially even email or CD-ROM distributor, that provides access to a large portion of the GFDL corpus open content. If it also has wikitext standard editing facilities then it is often referred to generically as a wikipedia, because:
The best known of these is Wikipedia. There is often some confusion between that project and the corpus, and this is often facilitated byWikimedia which finds this confusion to be in its own interest. This leads to some GFDL issues: notably the technical barriers to retrieval of text from some IP addresses, and failure to clearly publish, or even XML export, the complete list of contributors. Also, Wikimedia makes spurious demands for links even of other GFDLcap - these are not required by the GFDL and are effectively ads for Wikimedia itself.
Wikinfo and Recyclopedia are lesser known access providers, in stricter compliance with the GFDL, to the degree that is possible given that it relies on XML export by Wikipedia for most content, and this export itself is not in compliance. Unfortunately, Wikimedia sysops choose to deal with this by engaging in vandalbot activity and denial of service attacks against Recyclopedia, and threatening legal action against Wikinfo or GetWiki. This is likely due to commercial conflict of interest that is threatened by the evolution of these more democratic projects.
Wikinfo takes a sympathetic point of view, refusing to mix critical and sympathetic views in an article. This is a direct challenge to the neutral point of view wiki ideology that is promoted by Wikimedia usurpers, since that view gives maximum power to those who claim to be 'neutral' but are in a position of technologically granted power.
Other corpus access providers focus on particular subsets of the material, and do not have facilities to automatically import via the Wikipedia interface:
Presumably, if they adopt such facilities, they too will be attacked by Wikimedia.
There are several challenges common to all GFDL corpus access providers:
- agreeing on a wikitext standard and priorities to guide wiki code changes to accomodate it; GetWiki and MediaWiki are so far the only options.
- meeting the rigid terms of the GFDL itself, especially with regard to attribution and source access, i.e. to the original text written in the wikitext standard, not the HTML form into which it is rendered for prints or presentation.
- resolving point of view differences, to which there are several approaches:
- neutral point of view tries to do this always within one article, attributing disputed claims, but of course, a systemic bias then applies
- multiple point of view tries to separate views into articles of their own, e.g. as at Metaweb or via the "faction" system proposed at some wikis
- sympathetic point of view tries to separate positive and negative views, whatever that means
- dealing with server load, especially for full text search
- inadequate PHP-based software like MediaWiki and GetWiki - a concern Metaweb's supporters are paying particular attention to
- compiling and collaborative filtering to present adapted articles - a concern Consumerium takes very seriously
- governance problems in deal with a very large group of editors, see rule of 150 for one possible limit to this
- Tracking with change to the GFDL corpus as a whole, so that the best material can be copied into those services interested in presenting it to their users.
There are proposals extant to track all changes to the GFDL corpus in any of the known access providers, and perhaps simplify signup for jabber or other services that could serve as basis for interwiki identity standard. Such proposals are simplified by an interwiki link standard and standard wiki URI, though they are not strictly required simply for tracking.
The Consumerium Governance Organization will probably need to take some strong interest in this process, in order to make free use of freely edited texts.