User:Tyrol5/Op-ed

RfA reform discussions: An alternative analysis by an observer and participant

---

Tyrol5 has been an editor on Wikipedia since November 2008 and an administrator since July 2011. Below, he records some observations pertaining to his experience at and around the "Requests for Adminship" process and discussions pertaining to it and offers an analysis of the effects of discussions surrounding RfA reform.

---

The road of RfA reform discussion is winding, but seems to be leading somewhere nonetheless. Where it leads depends on the participants.

Requests for Adminship (RfA): it's a touchy subject, one that's the topic of countless discussions each year, few of which reach any semblance of consensus. But they're discussions nonetheless. In 2010, having been an active participant at RfA for a number of months since becoming fully involved in the project in July of that year, I took an interest in these discussions. I participated actively in the discussions at the RfA talk page and elsewhere. Late that year, I made a decision to withdraw from participation in RfA discussions in favor of holistically observing the process without worrying about formulating an opinion on every candidacy that came up. Thus began a fascinating exercise in observation that culminated in my participation at the 2011 RfA reform project—perhaps the most advanced and comprehensive analysis of RfA in the history of the process—and enlightened me to the observations that follow.

While I've since returned to full participation in individual RfA discussions and the 2011 reform project resulted in very little change in policy, participation in that facet of discussion alongside my long-term observation of the process changed the way I participate there very much for the better. While resulting in very little policy change, the 2011 RfA discussions resulted in an influx of successful requests from some excellent candidates in early 2011. Additionally, WereSpielChequers' August 2010 article indicating a downward trend in candidacies galvanized a burst of successful candidacies that month and later (including my own in July 2011). Further, the 2007 discussions coincided with a sharp increase in candidacies in May of that year. This suggests, to me, that the discussions surrounding RfA reform during the past couple of years has had an unexpected affect. Increased publicity of the process and talk of it being "broken" appears to have led, perhaps ironically, to short-term increases in qualified candidates coming forward. My contention isn't that RfA isn't or is no longer broken, but that discussion alone seems to help in unexpected ways.

Effect of discussion and publicity

January and February 2011 saw a spike in discussion on the RfA talk page regarding proposals for reform and the process in general, culminating in the creation of the RfA reform 2011 project in March. This coincides with nine successful candidacies at RfA in both February and March 2011, an increase from three in January 2011, one in December 2010, and four in November 2010. Spring fever, perhaps? With a statistical analysis of another instance in August 2010 in mind, I think not.

In August 2010, WereSpielChequers published his excellent article documenting a statistical analysis of the worsening RfA drought, which generated interest in RfA and efforts to reform it (including the massive project launched in early 2011 mentioned above). That same month, thirteen candidates passed RfA. This was the most successful month in terms of RfA candidacies in 2010 and was almost exactly twice the average number of successful RfA's per month in 2010 up to that point (6.29). But this isn't the earliest example.

In April 2007, the "golden age" of RfA, a rather large reform effort was launched. A significant influx of discussion and proposals followed, to little avail. A statistical analysis, however, suggests that the effort wasn't all for naught. In May 2007, during which the discussions initiated the previous month continued to balloon, fifty-four editors passed RfA. This was the second highest month in 2007 (following only November with fifty-six successful candidacies) and was a sharp increase from the previous month's thirty successful RfA's.

This all, I think, is illustrative of something. What, exactly? It illustrates that there are editors out there somewhere that are perhaps not privy to the RfA process and its intricacies or have not given much thought, if any at all, to requesting the administrative toolkit. When made aware of the apparent drought at RfA, which WereSpielChequers astutely analyzed in August 2010, these editors offer their services to serve Wikipedia as administrators. We saw it in May 2007, again in August 2010, and yet again in February–March 2011. This leads me to believe that if more qualified editors were encouraged to run at RfA, we'd see more successful RfA's. I don't mean the bursts we saw when RfA was being discussed and publicized by the community, but a sustained increase in successful RfA candidacies.

Has RfA "bottomed-out"?

The possibility that RfA has "bottomed-out" and that the process could be following a "boom and bust" cycle has been the subject of recent discussion at the RfA talk page. This discussion in particular highlighted that RfA appears to have started out in 2013 on a good note, with four successful and zero unsuccessful RfAs in January. February saw five successful candidacies, tying the second highest month in 2012 (November) and falling one successful candidacy short of the highest month last year (July). It was suggested by several discussion participants that RfA could be coming off of a "bust" and entering a second "boom" (following the original in 2006–2007). I'm skeptical of this hypothesis, but the 2013 numbers so far are encouraging. Four and five successful candidacies in January and February, respectively, of this year is a marked increase from one and three candidacies during the corresponding months in 2012.

Not surprisingly, the increased numbers thus far in 2013 roughly coincide with yet another influx of discussion surrounding RfA. January 2013 saw numerous discussions and RfC's (see here and here, along with the 2013 RfC's on RfA). This provides an alternative explanation to the success at RfA so far this year, the same explanation for the influx in Spring 2007, August 2010, and February–March 2011: the positive ramifications of discussion in the amount of qualified candidates that run at RfA.

My contention here is this: that the number of successful RfA's seems to be proportional to the amount of discussion about the RfA process and that this seems to be the result of public interest in RfA generated by these discussions rather than a natural "boom and bust" cycle. But, while RfA might not have a naturally cyclical nature, I think there is a terminal limit to how low RfA can go. There were several months in 2012 in which there was only one successful RfA (January, March, May, June, and October) and one in which there was none (September). As I've said, the numbers thus far in 2013 are a bit more encouraging. I don't want to raise hopes prematurely, but I'm cautiously optimistic in terms of RfA in 2013 think the suggestion that RfA has "bottomed out" is one worth considering.

Discussions in 2013

In early 2013, following a slew of discussions on the RfA talk page, a multi-stage RfC was launched. Unlike the free-styled discussion of the 2011 effort, the RfC took on the structured "!voting on proposals" structure of the 2007 effort. Now, a substantial portion of the community seems to be coalescing around the conclusion that RfA is in need of some sort of change. As a result, a few proposals, including unbundling some U1 and G7 deletion functions, automatic candidate prospecting, and a project to support and coordinate nominators seem promising in their potential to gain consensus with the community-at-large.

This is an indication that, contrary to the light-hearted jokes cracked about efforts to reform RfA, progress is indeed being made (albeit very slowly). While the community may not be moving very quickly towards any major policy changes, movement is occurring nonetheless. In addition to the positive immediate effects discussion has had on the number of successful candidacies, some discussion of late has shown promise for those advocating reform at RfA. Although a mere glance might lead one to believe that discussions about RfA appear to have sunk into an endless perennial cycle, an in-depth analysis reveals that these discussions seem to make just a little bit more progress each time and result in small bursts of successful RfA's to boot.

Conclusions

There has, no doubt, been a marked decline in successful RfA candidacies (and candidacies in general) over the past few years. A few instances here and there have proven to be an exception to the rule (although a "high" number of candidacies is much different in 2013 than it was in 2007). These instances have, for the most part, coincided with efforts to reform RfA and find ways to convince more qualified editors to seek the bit. While these discussions and efforts are oftentimes written off as failures for their lack of success in changing policy, they have had oftentimes unnoticed positive effects, as a statistical analysis reveals. As an active participant in these discussions surrounding RfA, I too was often pessimistic in terms of the feasibility of RfA reform and found myself questioning how worthwhile the discussions were and are. However, while these discussions very rarely resulted in direct policy changes, they seem to play a part in encouraging good candidates to run at RfA. What I mean to say here is that RfA reform discussions have had positive effects at RfA. They're not long-term, but this is a good sign nonetheless.

A pessimistic conclusion derived from RfA and the discussions surrounding it would suggest that talk of the process being "broken" would actually result in a decline in candidacies, and I must confess that I once held this opinion. But the apparent effect has been the opposite: an increase in public interest in RfA (and thereby positive interest, resulting in a short-term increase in successful candidacies). With this in mind, I'm not so pessimistic.

In conclusion, there is a solid foundation in place for progress at RfA. A significant amount of research has been gathered (see here, here, and here) and a lot of proposals have been discussed. While no major proposal has come to fruition (and even this seems to be changing slowly with the latest discussions), I'd submit that these discussions are still worthwhile and are cause for very cautious optimism. The increases in successful RfA candidacies coinciding with these discussions indicate that there are concerned editors out there who would consider running at RfA if they knew the full extent to which their assistance with administrative tasks would be useful. I'm not saying RfA is or isn't broken but that, contrary to the belief of a significant portion of editors, RfA reform discussions haven't been a quagmire of endless discussion without any avail, but have had largely unnoticed (and positive) effects. So, in this light, I wouldn't be so quick to write-off the value of the seemingly perennial discussions surrounding RfA and the reform thereof.