Imagine for a minute that you are a new user. You have just created an account on Wikipedia, "the encyclopedia that anyone can edit." The idea of sharing your knowledge interests and excites you, and you are eager to jump in. After making a few edits here and there, you decide that you're ready to start your first article, perhaps about an organization or person that you know of. You write a sentence or two and save your changes, feeling proud of contributing your very first article. You probably intend to continue work on your new article, so you go off and look for more information to add.

Then, several minutes later, you come back to your page. You can't even see what you wrote, because a huge red box has been slapped on the top of your article. It complains of " no context" and " no indication of significance," and threatens the immediate deletion of your page. It has all kinds of strange letter-number abbreviations, like "A1" and "A7," and links to a long page filled with rules. Meanwhile, a red notifications icon has lit up, and it tells you that you have a message on something called your " talk page." You click on it, and on your "talk page," you find another long message, accompanied by another red sign (this time with an exclamation point), once again filled with abbreviations, jargon, warnings, and links to rules pages. Totally confused and discouraged, you go back to your article one more time, only to find that it has already vanished, having been replaced with yet another red sign announcing that the page has been deleted.

Furious at the total lack of appreciation for your contributions, you simply log out and never come back. "The encyclopedia that anyone can edit" — what a lie!

Now, coming out of this thought experiment, it is important to realize that these events are not merely hypothetical. Much to the contrary, they occur innumerable times every day on Wikipedia. Good-faith new users try to contribute by creating new articles, but some self-appointed "new page patrols" see these articles as pathetic jokes and as threats to the quality of Wikipedia. So, they slap them with all kinds of jargon-laced deletion templates, and, with admins happily going along, they wipe out the hard work of countless new users who were just trying to help.

Meanwhile, we wonder why Wikipedia has an editor retention problem. It is a known fact that Wikipedia has difficulty gaining and retaining editors. Simple statistics can easily demonstrate this. For example, the official numbers show that of all users who have edited Wikipedia at least once, not even 20% stayed to make 10 edits, less than 3% made at least 100 edits, and less than 1% made 500 to 1,000 edits. The active editor count reveals that Wikipedia is largely dependent on a small group of about 3,000 editors, and this number has remained stagnant for over half a decade, following a period of noticeable decline between 2009 and 2011.

Wikipedia only becomes more popular by the year, and a Wikipedia article is invariably one of the top results on any internet search. It is now the fifth most popular website in the world, surpassing even Yahoo!, Reddit, Amazon and Twitter. But instead of Wikipedia's editor retention rate increasing, it has actually been stagnating or even decreasing, despite Wikipedia's soaring popularity. This is clearly unsustainable. There might not be much concern about this now, because everything seems to be working smoothly. But what some people do not realize is that many active editors are still from the early days of Wikipedia, and it is inevitable that these old-timers will gradually retire as the years pass by. At the current rate, many retiring editors will not be replaced. If Wikipedia is to survive and not degenerate with time, it must gain and retain more editors.

But why does Wikipedia have such difficulty with editor retention? The answer is that many editors perceive Wikipedia as a hostile place. This survey does a good job of illustrating that point. With particular regard to deletion, 60% of respondents were discouraged from editing because their article was deleted.

It is always best to prevent rather than cure a problem, so if the long-term stability of Wikipedia is to be assured, it is imperative that the community address the issue of editor retention.

Deletion is a good place to start. As shown by the survey mentioned above, Wikipedia's deletion system is one of the top factors that discourages editors. In addition, we are currently in the middle of the autoconfirmed article creation trial. New page reviewing and deletion are inextricably connected, and in light of the ACTRIAL, further discussion of those topics would be especially relevant right now.

To reduce hostility and increase editor retention, we could start by making more use of the draft namespace. The entire point of drafts is to incubate articles that are not yet of sufficient quality for the mainspace. So instead of taking the extreme and discouraging step of deletion by default, articles that are created in good faith, that are not vandalism, and that comply with copyright requirements can be simply be moved to the draft space for a time.

G13 would still apply to these drafts, and to reduce the workload on admins, a bot could be created to auto-delete all stale drafts. On top of making the environment more friendly, this solution could also have the added benefit of resolving the long-standing conflict between inclusionists and deletionists. Low-quality articles would be removed from the public eye (as deletionists want) but they would still be preserved elsewhere (satisfying inclusionists). Of course, AFC reform of some sort would also be a necessity.

An alternative, but related, possibility is to give more force to the refund process and to the procedure for contesting speedy deletions. Again, this would only apply to articles that meet the same basic criteria (not vandalism and copyright-compliant). We could make more categories of deleted article eligible for restoration and draftification, and we could require that all requests for articles in those categories be granted. Additionally, we could require that contested nominees under certain CSD criteria be automatically moved to draft space. Finally, article creators could be more prominently informed of their ability to contest a speedy deletion, and of their ability to request restoration and draftification for certain types of articles.