Wikipedia has a number of processes that allow for deletion of articles, but there seem to be far fewer processes that directly encourage editors to get involved in article improvement, so I would like to propose one. The purpose of Articles for Sourcing, for an example of a name, would be to try to find sources for articles in need of better sourcing, one article at a time.
I imagine the process working somewhat similar in structure to an Articles for Deletion nomination, once an editor identifies an article in need of additional sources. The nominating editor would start a page for review of an article, and would be encouraged to suggest and what terms or search strings would be most helpful in such a search and to suggest ideas they might have on where other editors can search for sources (searches in specific databases, trade magazines, etc.).
Other volunteer editors would then perform searches both online and offline, and report their findings. Editors should indicate where and how they searched for sources (Google Books, HighBeam, other databases, search at library, personal hardcopy collections) and indicate what search strings they used, etc. If an editor finds sources that are freely available online that editor may just include a link on the search page, although if the editor finds an online source not freely available or a source that is not available online, that editor is strongly encouraged to edit the article using citations to the new source.
I imagine an Articles for Sourcing review page consisting of multiple separate sections, to keep the page better organized. The nomination would go in the first section, indicating the article to search on, as well as any other information the nominator thinks would aid someone in finding sources. The next section would be for other editors to post the results of their searches. There may be reasons to have additional sections.
Editors are also encouraged to perform whatever other needed cleanup they are willing and able to work on while the nomination is active, as part of general improvement. This may include copyediting, trimming in-universe text or original research, and building up new or short sections. The review page should last for a week, I would say, to get some attention on the article and give editors time to work on it, and after a week's time the review page should be automatically removed from the transclusion list. The review page should remain open indefinitely so that editors searching through the archives could continue to perform searches for the article, although an article that has been successfully properly sourced can have its page closed and archived by an administrator.
While I think this process would work best for single articles, an argument could be made for a provision to allow multiple closely related articles (such as different parts of a series based on the same topic) to be reviewed at the same time. Part of the reason for this would be that sometimes a search for sources will turn up a source that could be used on multiple related articles, so we might as well "share the love". An argument against allowing this would be that it would require more resources from volunteers to check for multiple topics instead of just one, but at the same time if we are talking about cases where a number of articles might later be merged into a list or another article, it would help to find out all at once which ones should stay or go. This could be accomplished in a manner similar to how multiple articles are nominated at AFD.
These articles should be kept in a log and/or transcluded to a larger list, similar to how AFD's are tracked. Likewise, these articles should be made to show up on the article alerts pages of associated WikiProjects, so that project members can be alerted to an article in need of work. Any article that received insufficient attention after a week of review can always be renominated in the future for another one week period, and have the review page added back to the transclusion list. The ultimate goal of Articles for Sourcing is to build an article up to the point where there is enough material sourced to multiple independent reliable sources to make an article larger than a stub.
- Why do we need a new process like this?
I originally conceived of this idea because I have seen far too many AFDs which resulted in "Keep" because people found sources during the discussion when the nominator either did not try to find sources or did not know where to look. There have also been plenty of instances where an article was deleted, but sources were discovered later which allowed the article to be restored. So I figured, why not have an environment where such a search can be performed without the stress of possible deletion hanging overhead? There are quite a lot of editors who like to "save" articles in danger by finding appropriate sources, so maybe it would help to have a central organized place for people like that to go, without the stigma of people trying to "rescue" an article by voting to keep it without improving it – since there would be no voting, and the focus would be all about improving the article.
This process could serve as either a prelude or an alternative to an AFD, depending on the results. Anything which is improved enough to retain could be weeded out, thus making a deletion discussion unnecessary. Anything subjected to a thorough search which turns up nothing would make for a stronger case if the article is taken to AFD, especially if it has been nominated multiple times with little result. Alternatively, if the article is still unsuitable for retention, any discussion at the review page could lead to alternate ways of dealing with the article, thus again eliminating a need for AFD in that case.
This process would also be useful for drawing a focus to an article in no danger of deletion, but still in need of cleanup and sourcing. This would be equally good for old content in need for an update or upgrade, as well as newer content which is simply not as developed as it could be. This would especially be useful for articles which have existed for several years but have received little attention while policies have changed and could stand to be improved.
This process would not be intended to replace any other existing process, although it could lighten the load on some of them. Speedy deletion, PROD, and AFD would still be available as options for any editor who can't see any reason an article could be retained (although editors who routinely bring salvageable articles to be deleted should be encouraged to check out Articles for Sourcing).
- What articles are suitable for this process?
Just about any article in need of more and better sources, I would think. It doesn't even need to be limited to currently existing articles.
This would be a good way to get help in finding sources with which to build an article that hasn't been created yet; if sources are found, a stub should be started during the Article for Sourcing process so that other editors can help you build it right away.
Former articles or not-yet articles, such as failed Articles for Creation submissions, user space drafts, and article incubator pages could be submitted for review, and then moved to article space if improved sufficiently.
Articles that have been merged or just redirected due to notability concerns could also be nominated, using the most recent good diff as a starting point, and restored if sufficient sourcing is found. If the article was merged as a result of a consensus discussion, a new discussion may be required to determine if the new sourcing has been sufficient before restoring.
Even potentially salvageable articles which have been deleted could be nominated, and restored if they can be improved enough; if anyone is concerned that the improvement is insufficient, it can go back to AFD for a new discussion. If no new sources are found, or if the article cannot be improved, it should not be restored.
- What articles are unsuitable for this process?
Well, hoax articles and other nonsense, of course – we never want to write about things that are not true, except for things like actually notable hoaxes. Attack pages and copyright violations which cannot be written over from scratch also come to mind. Likewise, any other nominations for blatant policy violations and any article which has been previously determined to be wholly unsuitable for encyclopedic coverage should be closed immediately. Exceptions can always be made if there are good enough reasons, but for the most part this sort of content should be avoided.
Quality content such as Good Articles and Featured articles already have their own processes for improvement (such as peer review), or reassessment (GAR, FAR), and should be considered outside of the scope for this process. An article that already has enough material sourced to multiple independent reliable sources to make it larger than a stub should also be excluded from Articles for Sourcing. Articles currently undergoing another evaluation process (AFD, DRV, etc) should be excluded to avoid overload and confusion, and anything currently active at Articles for Sourcing should likewise be excluded from those other processes. Patience, please!
In theory, an article could be submitted to Articles for Sourcing as many times as is needed to get the attention it needs, but there should be a mandatory interval to wait so as not to monopolize the community's time with the same topic – 3 months, 6 months? In theory, there is no need to close an Articles for Sourcing page on an article that still needs more sources, so it can remain open indefinitely even while not transcluded to the main list. There are also other places to discuss an article, such as the article's talk page or a WikiProject talk page.
Other than that, pretty much anything goes!
- Who can participate?
Anyone and everyone who is interested in improving the encyclopedia and its articles, hopefully. Users who have violated the spirit of this process after being warned can be topic banned, and other editors under certain restrictions could be disallowed, but otherwise the more good-faith participants the better.
To prevent editors from making too many requests at one time, there should be a limit on how many nominations they may have actively going at once – maybe 1, 2, or 3? Alternatively, we could say that an editor may only make one nomination once every 2 days or every 3 days. Unfortunately, this would have to be tracked to prevent abuse.
- Who would not benefit using from this process?
As I see it, that would be limited to editors who want to delete an article without scrutiny from anyone who would try to save it, as well as editors trying in vain to save articles beyond hope – both are likely to find disappointment here.
- What do you think?