I guess if we're going to use the shared backend, then we may as well allow editing of Wikipedia from Inclupedia, and implement the integrated watchlist idea that has been floating around. I can't think of a better application for it. I can probably implement that; as a matter of fact, I already started on it over at mw:Extension:Interwiki integration. As for mirroring, I started some of that too over at mw:Extension:MirrorTools, but I suspect that it'll need to be redone to accommodate the fact that a shared backend, rather than API, will be used. The API approach was based on the assumption that a shared backend wouldn't be available.
Really, I think the main barriers to this idea will be the same that afflicted pure wiki deletion, namely, concern over legalities. People sometimes speedily delete or prod an article that they suspect, but can't prove, is a copyvio or otherwise illegal, when they can find other deletion criteria to get rid of it under. I think some people will worry that keeping those articles around elsewhere will create risks. But the solution is probably for Wikipedians to note, "Suspected copyvio" or whatever when deleting, and/or do greater diligence; or for Inclupedians to exercise their own due diligence. Tisane 03:00, 26 July 2010 (UTC)
- See the WP discussion here for a similar proposal, http://en.wikipedia.org/wiki/User_talk:Jimbo_Wales#Mediawiki_software_proposal Genjix 11:59, 17 September 2010 (UTC)
While the proposal is interesting, I have few concerns:
- Instant mirroring and a shared back-end is nice, but revocation of deletion seems to be a potential hole for pages properly deleted as vandalism, hoaxes (pages which are hoaxes, not pages about hoaxes), various abusive pages (like the "Eric is a fag" style abuse and libelous BLPs), and "outing" various people's real identities, from Wikipedia volunteers to 11-year-old YouTube viral video stars.
- It seems likely to me (though the sysadmins are free to clarify things, of course) that this would require the replication of the source project's database in entirety and additional on-the-fly duplication of actions from the RC tables. This is all to be done while ignoring certain types of deletion done in an intelligent manner and allowing for Inclupedia admins to access Wikipedia deletions in order to perform this work. Hence:
- Either Wikipedia admins are also (and only) admins on the corresponding Inclupedia project (which would receive considerable objection), or,
- We're widening Wikipedia deletions to a pool of probably separate admins which the Wikipedia community has not yet trusted to view such deleted material (which would also receive considerable objection).
- Some measure of support should be gained from the WMF sysadmins before further pursuit of this project: There is a question of technical feasibility and custom support which they may not be willing to take on.
- Inclusion of BLP material, including non-notable people where there is little or no source material, is not acceptable on any WMF project. This is a WMF Board resolution (see wmf:Resolution:Biographies of living people and, frankly, consensus and local policy trying to end-run around the resolution will fail. Specifically, a project up-in-arms ready to refuse board resolutions will likely be emergency-locked by stewards or staff. This position should be thoroughly reconsidered by the proposer(s).
- This appears to be a duplicate of the work of en:Deletionpedia, which frankly needs to be kept as a non-WMF project for both its security and ours. As that project is currently closed, perhaps a more reasonable approach to the desire triggering Inclupedia would be to start a similar site on a non-WMF server.
- Bugs happen, and information may be posted in Wikipedia which requires immediate removal. Even if it's immediately deleted there, there's always a chance that the change will "slip through the cracks" in replication and fail to delete critically deleterious material on Inclupedia.
- Frankly, on en.wikipedia, there are 1,737 admins as of the time of this writing, and there are still backlogs in handling deletions and reviewing potentially damaging articles. I can't help but wonder how a mirror of the site plans on reviewing all such actions within a timely window.
I can imagine other objections, but these problems are to me the ones whose answers would need to be significantly assuring before such a project is practically conceivable. Until then, I would consider the proposal premature and nonviable. Kylu 04:43, 26 July 2010 (UTC)
- "Inclusion of BLP material, including non-notable people where there is little or no source material, is not acceptable on any WMF project." I totally agree with you; everything must be verifiable, or it has to be deleted, both from Wikipedia and Inclupedia.
- It might not be such a bad idea for all Wikipedia sysops at the time of Inclupedia's inception to automatically become Inclupedia sysops. That would provide a large pool of experienced sysops who have already been vetted by a rigorous RfA process. If we are going to give Wikipedia sysops the ability to automatically delete (or exclude from deletion) from Inclupedia an article or revision at the same time that they delete it from Wikipedia, that is already giving them a degree of sysop power on Inclupedia. However, this can be counter-acted by Inclupedia sysops blacklisting certain Wikipedia sysops for extra scrutiny of their decisions in this regard; e.g., if a Wikipedia sysop keeps flagging deleted articles as acceptable for Inclupedia when they're not, or flags them as unacceptable for Inclupedia when they're fine, that sysop's decisions might be looked at more closely for potential reversal on Inclupedia. On the other hand, if we only allow the Wikipedia sysops to recommend page deletion from Inclupedia, that could create the very problem you (and probably others) are concerned about, which is that an article will be deleted from Wikipedia due to illegality, that should also be deleted from Inclupedia due to illegality, but ends up sitting around for awhile before anyone gets around to taking care of it. Inclupedia sysops who abuse their powers can be desysoped as needed.
- Deletionpedia is not an editable wiki, nor is it integrated with Wikipedia; those are two very serious drawbacks that limit its ability to be anything close to Inclupedia. No wiki can approach the level of success of Wikipedia without being closely integrated with Wikipedia; not even Wikia, which was founded and operated by some of the same people as Wikipedia, was able to achieve anything close to Wikipedia's success. The reason is that articles need to be part of that grand unified categorization, template transclusion, page existence detection, and watchlisting/recent changes/bot operation/etc. scheme; otherwise, there's just too much duplication of work, and not enough of the interconnection and economies of scale that are what made Wikipedia so useful and relatively efficient to administer.
- Perhaps, when an article is deleted from Wikipedia, the default should be to delete it from Inclupedia as well. It would then require a positive action to undelete it from Inclupedia, and a sysop would need to take accountability for doing due diligence to ensure that article is OK to be resurrected. Then, if there is any backlog, it will be on the side of resurrecting articles rather than deleting them. On the other hand, the deleting sysop at Wikipedia might be able to check a box saying, "This article is being deleted for notability reasons only; therefore, do not delete on Inclupedia." Sysops who deliberately or repeatedly cause content to remain on Inclupedia that shouldn't remain there should be desysoped. Tisane 05:29, 26 July 2010 (UTC)
Idea lab discussion
- Why is this being discussed in 2 different locations? Mr.Z-man 22:12, 27 July 2010 (UTC)
It is important that Inclupedia adhere to a verifiability standard, but given that non-notable articles, by definition, tend to not have significant coverage in reliable sources, it may be necessary to "stretch" verifiability a bit without breaking it. E.g., suppose there is a non-notable person, John Smith, about whom a few newspaper articles counting as reliable sources have been printed, pertaining to him and his dogs, but not mentioning his blogging activities. However, a blog does exist under his name, and many of his blog entries assert that he is indeed the same person that the news articles are about. It would be truthful to say, "According to Blog A, whose author claims to be John Smith, Smith was born on October 24, 1945..." That is a completely verifiable statement, in that anyone can look at the blog and see that Inclupedia is accurately describing the statements posted on the blog, even if it isn't verifiable that the blog is indeed authored by that same John Smith.
On the other hand, someone else might create another blog claiming to be John Smith and disclaiming authorship of Blog A, in which case we might leave the whole thing out of the article until the issue is resolved. Or, if one claim is significantly more believable than the other, we might leave out the unbelievable one.
Hopefully, we can find better ways of establishing which blog really belongs to which real-world person, though. E.g., if a university professor provides a link from his campus website to his blog, that will be enough to show that the blog is indeed his. Even in such a case, though, it would still be necessary to note that the information is coming from the subject of the article itself, e.g., "According to John Smith's blog, he was born on November 31, 1945." We can't simply say, "John Smith was born on November 31, 1945" without a reliable source verifying that he was indeed born on that date.
We're going to have to walk a fine line between sticking strictly to what is verifiable without discarding potentially useful information. I think it can be done; journalists do it all the time. When they cite sources of questionable veracity, they simply attribute the information to that source, and thus inform the reader of the potential unreliability. We might, however, also opt to do another thing journalists do, which is avoid even publishing the comments of sources that are deemed extremely untrustworthy. E.g., if some random person, with no reputation for reliability, calls up a reporter and says that Barack Obama was behind the 9/11 attacks, without providing corroborating information, the reporter will not publish an article stating, "According to Mr. X, Barack Obama was behind the 9/11 attacks." Likewise, we probably should not put such items in Inclupedia articles.
A perfect example of how to handle unverifiable information might be the handling of Jimbo Wales' own statements in w:Jimbo Wales: "Although Wales' 1997 marriage certificate (to Christine Rohan), the Encyclopaedia Britannica, Current Biography Yearbook and Who's Who in America state that he was born on August 7, his Florida driver's license gives his birthday as August 8, and he has stated that the August 7 date is incorrect according to The Oregonian's OregonLive.com blog. He has also stated that he was born on August 7, 1966." Tisane 06:37, 27 July 2010 (UTC)
- I think you're in danger of reinventing the wheel. w:WP:V and w:WP:RS have already gone over this ground many many times. Fences and windows 16:46, 27 July 2010 (UTC)
Kind of... weird?
I don't know if this has been put up before, but this project suffers from a serious problem.
- Wikipedia contains tons of articles that are not verifiable but still stay.
- So, if Inclupedia includes only verifiable content for non-notable stuff...
- Does that mean Inclupedia's rules are: Notable topics can be unverifiable, while non-notable topics must be verifiable?
- What is an example of a Wikipedia article that is unverifiable but still stays? Tisane 10:45, 27 July 2010 (UTC)
- Many are listed at Wikipedia:Accuracy dispute. There could be many more that escaped the eyes of NP patrollers. Kayau WP WB WN 11:51, 27 July 2010 (UTC)
- But it's not Wikipedia policy to allow unverifiable content, is it? Such articles get deleted eventually, even if it takes years to spot the more obscure wishful thinking and hoaxes. WTF is w:Wikipedia:Accuracy dispute?! That's gathering dust, I've proposed at w:WP:VPP to mark it as historical. Fences and windows 16:28, 27 July 2010 (UTC)
- Many are listed at Wikipedia:Accuracy dispute. There could be many more that escaped the eyes of NP patrollers. Kayau WP WB WN 11:51, 27 July 2010 (UTC)
I think the creator of that table has misunderstood some of Wikibooks' policies. While our page on original research is still a draft, OR is prohibited by WIW. Also, we do not have a notability requirement. Kayau WP WB WN 08:53, 27 July 2010 (UTC)
Annex & similar wikis
- See also existing wikis for deleted from the Wikipedia and other non-notable articles - Marjorie-wiki.de (german) and PrePedia (polish) -- Alan ffm (talk) 17:12, 30 November 2012 (UTC)
Would primary sources meet Includipedia's standard for verifiability? What about information that isn't freely or easily accessible (background checks from public records, facebook profile info shared only with friends, high school yearbooks, phone books, etc.)? If so, this would open the project up to a whole host of BLP issues. Mr.Z-man 22:23, 27 July 2010 (UTC)
- Don't the same issues also arise when someone cites those types of sources in an article about a notable person (especially a barely notable person)? Tisane 20:26, 28 July 2010 (UTC)
- Yes, but it will be amplified for non-notable people where primary sources are the only sources. A ban on original research would prevent flagrant abuses of primary source material, but there's still the privacy issue of letting people write articles about completely non-notable, private individuals based on semi-public information. For a non-public individual, even truthful information can violate a law if its presented in a biased manner. Mr.Z-man 22:25, 28 July 2010 (UTC)
The legal issues associated with founding a new project, it seems to me, are something we should leave to experts to figure out. For us to debate them is kind of like when Wikipedians used to raise concerns about proposed new bots, asking "What if the bot causes this?" and "What if the bot causes that?" A lot of times, their concerns were unfounded, and the result of a faulty understanding of bots, the MediaWiki API, the servers, etc. That's why we created a Bot Approvals Group, made up of bot experts, to make such decisions.
Likewise, it is important that legal issues be sorted out by those with an understanding of what laws will apply to our particular situation. Operating a large wiki creates certain risks, but we may also have certain protections as a nonprofit that, rather than being indifferent to illegal activity, attempts to police itself. And case law that applied in one situation might not apply in another; indeed, some old case law that we may read about might not apply at all anymore, since the law is always changing, and most of us probably don't subscribe to Shepard's Citations Service. A lot of what people have written about the legal liabilities of wikis is theoretical, since there hasn't been a lot of wiki-related cases litigated yet.
Since we're not proposing that Inclupedia deliberately break any laws, but rather we're just proposing a wiki that, like any other, will have some exposure to lawsuits, it's probably better to just formulate the proposal based on the functionality that we as a community would like to have, and let the WMF legal staff vet it later, as it gets closer to becoming a reality. If they veto certain aspects of it, or demand that additional controls be instituted to reduce the presence of content that could expose us to liability, then we can revise it accordingly. Otherwise, it seems like these discussions can't really have a satisfactory end, because the participants (including me) are not qualified to give a definitive answer as to these legal issues. Tisane 23:37, 31 July 2010 (UTC)
Awesome idea, nitpicky objection
I love this idea! It'll be interesting, fun, useful, and might even lower the cruft levels on WP, since they'll have an alternative outlet (I dunno, just a thought). Forgive my soapboxing though, for I have one tiny point I'd like to make. One of my pet peeves is this trend in calling new wikis something ending in "pedia". Wookieepedia did it, it was clever, after that, it stopped being clever. Maybe something more like "Wikiclusion"? Just some food for thought. S8333631 20:19, 16 August 2010 (UTC)
- Another good reason to create Inclupedia is to provide an alternative project for those banned from enwiki due to its arguably dysfunctional governance processes and structures. Inclupedia will adhere to wikihierarchism and dispense with the mobocracy and wikirepublicanism. Leucosticte (talk) 08:26, 24 September 2012 (UTC)
I think a simpler approach can be done by doing everything at Wikipedia (only one database) by adding a new namespace Non-notable (or a similar name) and a few rules:
- The namespace Non-notable cannot be seen by non-admins in the Wikipedia domain.
- If an Wikipedia admin determines that an article is non-notable he/she just moves the article to that namespace.
- If someone (including admins) tries to create an article with the same name used in the Non-notable namespace the attempt is denied.
- On the domain en.inclupedia.org, all articles from the main namespace and the Non-notable namespace will be present in the main namespace and can be edited by anyone which is not blocked, of course, so non-notable articles can be changed here.
- If an article is created in en.inclupedia.org it goes to the Non-notable namespace automatically so it cannot be seen on Wikipedia.
Best regards, Alpertron 19:12, 7 February 2011 (UTC)