Project:Schools of thought on deletion
There are differing schools of thought on what should be included or deleted from MediaWiki.org.
On many wikis, including this one, there is a tendency for users to create pages that then become seemingly abandoned, either because the topic has become less relevant due to changing circumstances or there are no longer any users interested in making edits to the page. For example, a software idea may be proposed but not be implemented because the discussion made evident a number of unanticipated drawbacks to the proposal, or the supporters became too busy with other priorities to implement it at that time.
The question arises, what to do with those pages? As on other wikis, editors divide into inclusionist and deletionist schools of thoughts, with many opinions in between the two extremes. As a practical matter, on this wiki such issues are often settled by unilateral sysop actions.
Inclusionists favor keeping such pages, because one never knows when the information might become useful in the future. Inclusionists point to Wikimedia's Bugzilla installation, or listserv archives, as example of sites with many open proposals that have gone years without being implemented, yet still appear in search results. Part of the point of keeping that content around on those sites is so that if another user comes up with a similar idea, he can search to see whether it has already been addressed before raising it again.
If a duplicate proposal is raised, people can point to the old discussions and thereby help avoid rehashing the same issues, unless someone has some new information or insight to add. MediaWiki.org is merely one other type of venue in which that sort of activity can take place; why treat it any differently? The wiki style of communication has some advantages over the Bugzilla and listserv styles, such as allowing the form of the proposal to be improved through collaboration and darwikian evolution. More communication, especially if it consists of specific ideas on how to implement an idea or fix a bug and the drawbacks of those ideas, rather than vague complaining about software shortcomings, can help to focus thought on how to effectively solve problems, rather than distracting.
Even ideas that prove unworkable can still provide food for thought. Aspects of those ideas might be adapted or combined into other ideas to produce an idea that is workable. Inclusionists therefore view the appearance of these pages in search results, categories, etc. as beneficial rather than harmful. If categories get too full with too wide a range of content to be very useful, that is what sub-categories are for.
In many cases, the only reason an idea does not get implemented is lack of programmer labor. The idea could be quite promising, but lacks attention at that moment from those with the necessary skills. Given the existence of a volunteer developer community; a wikisphere of many wikis operating independently of one another; and a software architecture allowing customization through extensions, skins, and so on, there is not necessarily any way of knowing that there will never be developer interest in any given idea. Even frivolous ideas sometimes get implemented, and even some ideas with high-level WMF developer support can be shelved for long periods before implementation.
No harm comes from lengthy incubation periods; keeping the idea on the wiki may promote eventual implementation even if it takes awhile. Software development requires long-term planning, and keeping in mind what types of features will be desired in the future can help ensure the code is designed in a way that makes it easier to implement those features later. Otherwise, one can paint oneself into a corner and have trouble getting out of it without extensive revisions.
Opening up more types of pages (e.g. "abandoned" or "useless" pages) to deletion inevitably leads to some useful pages occasionally also getting deleted. Deletion of pages close to the margin of what is allowable can chill participation in the project by making users hesitant to create new pages, lest they be deleted. It is better to err on the side of keeping useless content than deleting useful content. As long as the content is within project scope, it should be kept.
Deletionists favor deleting such pages. They clutter up search results and AllPages and category listings, thereby hindering readers from finding content on actually implemented software. The continued existence of such pages also may, by example, encourage users to create pages on ideas they do not intend to implement. It has been pointed out that "good execution is hard to find, but good ideas are cheap". There appears to be a high ratio of people with technical ideas to people with the technical skills needed to implement them, and it is typically much less labor-intensive to come up with an idea than to implement it. So, the vast majority of ideas will never be implemented. A search of open proposals on Bugzilla and the upward trend of these numbers makes this obvious.
The most common type of spam that has been encountered on MediaWiki.org is that in which a (possibly automated) new user signs up, makes a few edits outside the scope of the wiki mission in order to promote a business, product, person, etc. and does not make any useful edits to the wiki. On rare occasions, a new user with good intentions may engage in an editing pattern that appears superficially similar; i.e. he will edit his user page to introduce himself and link to his personal or business website, as a means of informing the community of his interests and activities, the better to find and collaborate with others who share those things in common.
There is not necessarily any way of knowing in advance whether that user will ever make useful edits or not; therefore, one has to decide whether it is better to err on the side of deletion (in an effort to deter spammers) or inclusion (in an effort to avoid accidentally biting newcomers who would have become productive editors). Newcomers can perhaps reduce the likelihood of such deletions by establishing an edit history of useful edits to the wiki before posting personal links, keeping such content to a minimum, adopting an informational rather than promotional tone, and avoiding linking to the types of blatantly disreputable sites (e.g. quack cures, pyramid schemes, etc.) commonly linked to by spammers.