On Mon, Jan 13, 2014 at 2:27 PM, Brian Wolff bawolff@gmail.com wrote:
Id say that http://getwiki.net/-GetWiki:1.0 was similar to your "superset" concept (minus the merging part)
Yeah, per WikiIndex http://wikiindex.org/GetWiki, "Instead of red links, GetWiki uses green links to point to articles which do not exist locally. When the user follows such a link, GetWiki tries to dynamically fetch it from the wiki designated as an external source (in Wikinfo's case, the English Wikipedia), renders and displays the article text. A local copy is created only if the page is edited. Effectively, Wikinfo therefore provides a transparent 'wrapper' around Wikipedia pages which have not yet been copied."
That sounds pretty easy to implement, compared to what is contemplated for Inclupedia. If a page had been deleted from Wikipedia, presumably it wouldn't have been accessible on Wikinfo unless someone had created a fork of that page on Wikinfo prior to its deletion from Wikipedia. That's a major absence of Inclupedia (and Deletionpedia) functionality.
Also, a high-traffic live mirror would be contrary to live mirrorshttps://meta.wikimedia.org/wiki/Live_mirrorspolicy, as the load on WMF servers would increase at the same rate as the wiki's readership. A site that only polled WMF servers for changes, and stored local copies of those changes, would not have that problem. To the contrary, it might take load off of WMF servers, if some readers were to retrieve mirrored Wikipedia pages from Inclupedia that they would have otherwise retrieved from Wikipedia.
The approach used by GetWiki of combining a Wikipedia mirror with locally stored forks is probably more suitable for either (1) a general encyclopedia with a non-neutral viewpoint (e.g. a Sympathetic Point of View), or (2) a site that wishes to have a narrower focus than that of a general encyclopedia. E.g., if you ran a site like Conservapedia (conservative bias) or Tampa Bay Wiki (narrow focus), while you were building up the content, you might want to be able to wikilink to articles non-existent on your local wiki, such as "Florida", and dynamically pull content from Wikipedia for users who click on those links. In the case of Conservapedia, Wikipedia's "Florida" content might be considered better than nothing, pending the creation of a forked version of that content that would be biased conservatively. In the case of Tampa Bay Wiki, the content of the "Florida" article might be sufficient for their purposes, so they could just keep serving the mirrored content from Wikipedia forever.
If a site were to aspire to be a general encyclopedia with a neutral point of view, it would be better to discourage or disable, as much as possible, forking of Wikipedia's articles. Once an article is forked, it will require duplication of Wikipedians' labor in order to keep the content as well-written, comprehensive, well-researched, and in general as high-quality and up-to-date as Wikipedia's coverage of the subject. It would be better to instead mirror those articles, and have users go to Wikipedia if they want to edit them. If users edit articles that have been deleted from Wikipedia, on the other hand, there is no forking going on, and therefore no duplication of labor. Unnecessary duplication of Wikipedians' labor tends to demoralize and distract users from building up complementary content; therefore, NPOV general encyclopedia community builders should consider it anathema.
GetWiki and Wikinfo don't appear to have prospered. It would seemhttp://wikiindex.org/Wikinfothat GetWiki was created in 2004 and that Wikinfo abandoned it in March 2007 and switched to MediaWiki. According to Wikipediahttps://en.wikipedia.org/wiki/History_of_wikis#Other_wiki_websites.2C_2001.E2.80.932003, "In 2013, the content was removed without explanation". I'm glad I didn't devote too much labor to creating content on Wikinfo. GetWiki.net itself has had no activity since 2012 http://getwiki.net/-changes.
The main problem with a NPOV general encylopedias' forking Wikipedia is that, for the purposes of most people, there's not a very compelling reason to do it. They deem the quality of the product Wikipedia offers, within the purview of its coverage (viz. NPOV notable topics) to be good enough that it's not worthwhile to create a fork. Wikipedia's shortcoming, from the standpoint of inclusionists, is not so much insufficient quality of articles, as insufficient quantity of topics covered. Forking is more suitable for those who find the quality insufficient to meet their particular needs.