Wikipedia talk:RFA by month

Comments
I share the frustration at the relatively low admin promotion rate in recent years, and although I do believe that standards for admins have gotten higher (itd be pretty hard to argue not), I think that a large part of the problem is that there are fewer new users who have the desire and the ability to become admins. To keep pace with 2005-2007, we would have to triple the rate at which we're promoting admins. Do you really think that for every candidate who passes RfA, there are two more that "should have" but either werent interested or were denied for spurious reasons? I'm sure there are some ... I can think of some candidates who I'd think would be good admins but ran and just barely missed the critical percentage ... but I really dont think there are hundreds of them. Perhaps it is simply that most of the people who would make good Wikipedia admins heard about Wikipedia and signed up early on, whereas those who are signing up now are mostly people who just aren't that interested. (I stress mostly because I know of plenty of counter-examples — people who would have been great in the early years too but were just too busy with other things to register an account.)  —  Soap  —  16:13, 28 February 2010 (UTC)
 * Sorry if the above comes off sounding hostile, particularly the first few sentences ... I didnt mean it that way because I wasnt really addressing you specifically, but rather the people who have been posting on WT:RFA blaming the decline only on higher standards. Then I decided to post it here right before I hit submit.   —  Soap  —  16:53, 28 February 2010 (UTC)
 * I think there's a combination of things.
 * I suspect that most editors who have both Rollback and Autoreviewer would meet my RFA standards, and would easily meet the standards of four years ago. But persuading them to run is difficult and even if they do, many good candidates fail first time.
 * Adminship is no longer seen as the norm for all longterm users, and that itself is self reinforcing.
 * Stricter standards at RFA are certainly delaying RFAs from serious candidates, as they tend to look at recent RFAs and judge themselves against the defacto standard, and if you delay long enough it doesn't happen during someone's wiki career. Standards have also risen on three separate axis, tenure; edits and type of activity.
 * Tenure expectations have also increased sharply, once one could become an admin within three months; Now it is very rare for a new admin to be appointed with under twelve months tenure.
 * The rampant editcountitis at RFA means that many for whom this is just a small part of their life know they can never meet the expectations that currently apply at RFA.
 * The unbundling of Rollback in Jan 2008 seems to have caused a step change with "good vandalfighter" no longer being sufficient to get through RFA, I have some sympathy with that change and agree with the idea that all admin candidates need to have done something to build the pedia rather than just protect it. But it has shifted the focus of RFA away from blocking users.
 * We are still getting lots of new editors, and I believe there are hundreds if not thousands out there who would meet the standards of 2004/5. But I don't know how many today could meet current RFA standards.  Ϣere Spiel  Chequers  18:29, 1 March 2010 (UTC)
 * Youre probably right. Higher standards in itself isnt a bad thing, but I think we've gone too far and will probably start to notice problems in a few years.  We can wait and fix them then, or we can fix them now.   —  Soap  —  18:24, 25 March 2010 (UTC)

change colour scheme
I think that whilst the current scheme illustrates the drought, a more fine grained scheme with more than three shades would be useful. I'm thinking a hot/cold type of scale, 0-5 6-10 11-15 etc strikes me as one possibility if we can find a 12 shade scale. Do others agree and if so what bands do people suggest and what range of colours?  Ϣere Spiel  Chequers  19:18, 25 March 2010 (UTC)


 * I'm not sure this answers your question, but if you want the color-scale to better reflect the meaning of the chart, then I have three suggestions:
 * #1. Showing total incoming admins is not useful by itself, in terms of colorizing it to give a quick overview.  If you want a quick overview of incoming-only counts, just make a line-chart.
 * #2. Colorizing *does* make sense for a slightly different kind of chart -- one that shows whether the active number of admins is growing, or declining, year over year.
 * #3. So, I suggest you make a line-chart showing incoming-admin-counts, another line-chart showing retiring-or-banned-or-inactive-or-minimally-active-admin-count, and then a colorized chart showing percentage-change-trends.  Dark green for 2006 when relative-admin-growth was explosive, dark red for some hypothetical year when all current admins were struck down by bubonic plague, fade to a neutral white in the middle (very pale red for slow decline and very pale green for slow growth).


 * And hey, if you want to really make my day, then showing the number of articles per admin, or the number of bytes-of-non-boilerplate-content per admin, or the number of admins per unique-visitor-per-month, or some other variation on admins relative to the scale of wikipedia, would also be useful.  There were not very many admins in 2002... but there were not very many articles, or unique vistors, either.  English wikipedia is effectively 'complete' at this point... there are 150k articles in class-C-or-higher plus simultaneously class-medium-or-higher, which cover the major encyclopedic topics.  Perfection is not the same as effectively complete, of course... still plenty to accomplish, and always will be.  (The rest of the 4M articles are mostly stubs, bands, videogames, B-movies, local politicians, esoteric mathematics, esoteric biological species, census-sourced demographic data, and the like... but I'd love to be wrong about this assertion.)
 * &npsp; Point being: is there a correlation between the active-ness of the admins, and the articles multiplied by the uniques?  Or something a bit more subtle?  I'd be less shocked about a low growth in admin-count, if it was highly correlated with a levelling-off of the growth-slash-churn amongst the top-100k-most-visited articles wikipedia offers.  Those are the biggest targets for spam, and vandalism.  Maybe the slowing growth of admin-types is simply due to lack of real-world demand, in other words.  There is pressure from spammers, but wikipedia broke into the top-fifty-websites list a long time ago, so that is arguably not *way* worse now than it was a while ago.  Vandals are becoming easier to handle, as the admin-toolset increases.  There are always editor disputes, bans for behavior, and so on... but except in inherently controversial cases (infoboxen being an exception apparently!) have these been getting worse, or better, since 2002?  I'm happy to try and help create some of these graphs, btw, if you think it will help any.  74.192.84.101 (talk) 01:41, 15 October 2013 (UTC)

Additional useful totals
Do you think you could add totals for each row in the bottom area in a final column? That would be useful, though one can obviously do the sums by hand. Thanks. 64.142.103.143 (talk) 20:37, 25 March 2010 (UTC)
 * Done. Though I'm afraid this is pretty much all done by hand.  Ϣere Spiel  Chequers  21:16, 25 March 2010 (UTC)

No successful RfA's in the first 27 days of December
It looks like an all-time record, if you exclude 2001 and 2002. Perhaps this could be the worst month for RfA's in general too? Probably not many people are gonna want to run around Christmas, but then maybe someone will prove me wrong.  — Soap  —  00:22, 21 December 2010 (UTC)
 * Sorry for not replying earlier - my watchlist has got so long that I sometimes miss threads even on pages that matter to me, and I try not to go round saying RFA is broken when I'm about to nominate a candidate myself. Yes, in terms of successful RFAs this was the worst month since 2002 and the last quarter of 2010 was the worst quarter since the first quarter of 2003. Poor Panyd had to make the coffees for longer than any new admin in about eight years. I think one reason for the shortage is a lack of active nominators, I nominated two of the last four successful candidates, and while I may well nominate more candidates, I can see a backlash coming if I continue to nominate as frequently as I do.  Ϣere Spiel  Chequers  13:55, 11 January 2011 (UTC)

An error
I doubt this affects the results very much, but there is one mistake in this. I'm guessing you used the same data for promotions per month as Successful requests for adminship, as you've made the same mistake which I just fixed in Successful requests for adminship/2004. December 2004 shows 24 promotions, but there were actually 25. Alzarian16 (talk) 22:57, 24 March 2011 (UTC)
 * Thanks, I've fixed that now.  Ϣere Spiel  Chequers  22:36, 27 March 2011 (UTC)
 * Looks good. I've been working on something not unrelated at User:Alzarian16/RfA participation by year, so you might see me making all sorts of strange edits relating to old RfAs in the near future... Alzarian16 (talk) 22:44, 27 March 2011 (UTC)

Templates
Hi, it's me again. I was thinking, we could probably get this page to work with just one template, the way the temperature templates work ... using a switch function that determines the color of the cell by what value is in it. I'm not much good with MediaWiki syntax, but I could probably figure it out if you're interested in doing it that way.  ☮ Soap  ☮  01:57, 24 September 2012 (UTC)
 * Proof of concept at user talk:soap/T, though I wouldnt use that much red. I coded it wrong at first, I like it the way it is now (though would still be OK to change it).  ☮  Soap  ☮  02:19, 24 September 2012 (UTC)

I wonder what the whole story is behind this graph
It's a beautiful graph. Thanks for maintaining it. Do we have statistics on the admin backlogs over time? It would be great to see an overlay between the two over time. Does anyone know? Also with the # of active admins plotted. Thanks. Biosthmors (talk) pls notify me (i.e. ) while signing a reply, thx 13:02, 23 September 2013 (UTC)

Ping to  Ϣere Spiel  Chequers . Thanks. Biosthmors (talk) pls notify me (i.e. ) while signing a reply, thx 12:53, 27 September 2013 (UTC)
 * The intersection of those data points with this would indeed be interesting. I don't know of any tool/bot/page that has recorded backlog status over time. There is a bot that notes how many active admins there are, and updates this every day. User:Rick Bot has been doing this since 2007. At the time, there were 912 active admins. There are 640 now. See . --Hammersoft (talk) 16:10, 27 September 2013 (UTC)

Admin Trends
This chart is so depressing. People must have noticed this trend 3 or 4 years ago. What solutions were proposed? People at the top are aware of this, right? I mean, considering that less than 50% of Admins on en.wiki are actually still active, this is not a good trend. Liz Read! Talk! 01:32, 3 October 2013 (UTC)


 * An enormous amount of work  has been done in  attempts to  address these issues. WP:RFA2011 was the big  initiative. In early  2013 several  RfC were launched in  further attempts to  find solutions. Please see my  response to  your messages on  my  talk  page. Kudpung กุดผึ้ง (talk) 03:14, 3 October 2013 (UTC)


 * I don't find it depressing until I can see all the data as described in the section above. Biosthmors (talk) pls notify me (i.e. ) while signing a reply, thx 07:14, 3 October 2013 (UTC)
 * I find it fascinating, otherwise I wouldn't have been working on this off and on for four years now. There are times that I've found it depressing, but I am now much less worried about the change to RFA than I once was. The community has changed, and adminship has changed. You can see one change quite starkly on the graph, after Rollback was unbundled in early 2008 "good vandalfighter" ceased to be sufficient for someone to become an admin. Now I see an anomaly there as we really need our good vandalfighters to be able to block vandals as well as quickly revert them. However my preferred solution to that is to unbundle a limited block right, and to allow good vandalfighters to block the IPs and new accounts that do almost all of the vandalism, but not to give them the ability to block the regulars.
 * Since Feb 2008 adminship has changed from something that all established editors could have within months of joining us, to something we are willing to give to well rounded sensible regulars after a year or so, and provided they meet certain criteria. I would still prefer to have a larger proportion of the community as admins, and I worry that our focus on certain easily measured criteria is making RFA less effective as a way of differentiating between good and not so good candidates. But with this year producing at least as many new admins as last year I consider that RFA is no longer in a death spiral.  Ϣere  Spiel  Chequers  13:34, 3 October 2013 (UTC)
 * Even a dead cat bounces. After many years of decline, it's nice to see some positive figures for once. But, it's meaningless. July 2012 saw 6 successful noms, more than any other month within a year of that month. Some people were gushing over how great it was (Blast of nostalgia, RFA is not broken?). Two months later, we had a month with zero successful noms...the first time that happened in ten years, with a rather unsurprising thread about the decline of RFA (The Year Without a Santa Claus). This year started off fast in the first quarter, outpacing the same time period in 2012 14-5, almost a 3:1 improvement over the prior year. In the two quarters since then, 14 this year vs. last year's 15, or a slight decline vs. last year. A minor bump in the first quarter is no cause for celebration.
 * The problem here is this data is in extreme isolation and fairly meaningless. The number of active administrators is another data point; we're down ~33% since 2007. In this year, 47 net admins have gone inactive (includes the 28 promotions). If the patterns for the last six months at WP:RFA and at List of administrators (active admins) continue, we will promote 8 more admins before the end of the year, and lose 47 active administrators. This would result in the second worst year in six at a -11.5% decline vs. beginning admin count (2010 had -11.9%) in replacement pace.
 * But guess what? Even this data is meaningless. If it were tied to the administrative burden of the project in some meaningful way, we'd have the beginning of understanding of our real situation. Sadly, we don't have that. It would take an enormous amount of work to come up with that data, but I think it is achievable with some bot work.
 * WereSpielChequers (love the name by the way!), your analysis of the changing nature of adminship is absolutely fantastic! I wholeheartedly agree with your approach here. I also would like to take the opportunity to praise you for the work you've done on this page. I know it took a lot of effort to create it, and further to continue to maintain it for four years. I huge THANK YOU for doing it! I certainly hope you continue to do it for a very long time to come! The data, while meaningless in isolation, is something we must have. --Hammersoft (talk) 14:25, 4 October 2013 (UTC)
 * Thanks Hammersoft, I know I can get a spike for a month by doing a signpost article, and I aim to get such a spike each year, and yes this year it was in q1. I appreciate that 28 admins so far this year means we are still facing a sharp decline in active admins, but the pattern has changed, if the annual drop that has applied since 2008 had happened then this year we would be getting fewer than twenty admins, instead we will get more than thirty and yes this enough of a difference to be significant, so we are still heading in the wrong direction, but we have managed to get our foot off the accelerator.
 * Ultimately an 11% attrition rate would mean that if we maintain the RFA level of the last two years we can maintain an admin community of around 250 notionally active admins. Of course "active" in our definition doesn't mean active as an administrator and it would take a ton of work to calculate what our minimum is for admin resources and how close to that we are coming. But I'm not volunteering for that, not because it would be hard and I don't have the time or the tools, all of which are true, but because I don't want that sort of a community. I want a self governing community where all sane, civil sensible long term members are admins or could be if they wanted to be. So I'm not really interested in an analysis that works out how close we are to having too few admins to keep the site going, I'll let someone who wants admins to be a scarce "powerful" minority figure out how close we are to running out of admins. My interest is in whether we can go back to the days when adminship wasn't a big deal because most of the regulars were admins, or in other words how close are we to the maximum not the minimum.
 * There are other phenomena emerging that I find interesting, in particular the issue of wikigenerations, in my 2010 article I worried that the admin cadre still underrepresented my wikigeneration, the editors who joined us in 2007. Three years on and we probably no longer under-represent the class of 2007, but we don't have many admins whose first edit was in 2008 and we really badly under-represent those who joined in 2009, 2010 and 2011. All our current admins first edited over two years ago and only 16 of our non bot admins have accounts that are less than four years old. So the wikigeneration gap has widened sharply since I first got alarmed about it. We still rely on the admins who started editing and became admins more than five years ago as the bulk of the admin cadre, and that means that many admins are people whose RFAs would have snow failed if current de facto criteria had applied when they ran. I suspect that much of the tension about adminship is from that unvoiced wikigeneration contrast and the inevitable perceived lack of legitimacy; contrasted by those pointing out that the solution to RFA being broken through standards inflation is to ditch the artificial inflation of standards and go back to a sustainable set of standards at RFA.
 * I'm also interested in the equally touchy subject of what RFA criteria accurately predict who will make a good admin. My suspicion is that the standards inflation has largely been misguided, and that while it has become harder for good candidates to get through RFA, it has perversely become relatively easy for bad eggs to slip through unspotted. But it is tricky to explain why on wiki without writing a trolls guide to gaming RFA.
 * PS Always loved your name too, it makes me think of the sort of toy one bought to throw at the TV, back in the day when I watched TV.  Ϣere Spiel  Chequers  21:24, 4 October 2013 (UTC)
 * I am one of those 16 with the account less than four years old (my first edit was in May 2011), but in fact I previously used another account an started editing in 2007. (Admittedly, I was not very active in the English Wikipedia before 2011). So the actual number of really recent admin accounts may be even lower than that.--Ymblanter (talk) 06:48, 5 October 2013 (UTC)
 * @WSC: Sane, civil and sensible are not the prime criteria for adminship as you of course know. If they were, I'd be an administrator. It's part popularity contest and part number crunching. I've kept a log of late of my actions that I could be doing if I were an administrator. Out of 183 actions, I was wrong once, an error rate of about 0.54%. But, that doesn't count. I'm controversial and unpopular, in part because I have no interest in having wiki friends and am quite willing to call a spade a spade. Becoming an effective administrator is more a matter of accident than RfA actually producing what we need.
 * I like your idea of 'maximum' over 'minimum'. It's an ideal I wish would could achieve, but alas. We'll never go back to the days when it was no big deal. Now it's a huge deal, and fewer and fewer people are administrators. The fewer there are, the bigger the deal it becomes.
 * There are pressures that are increasing the generation gap. I don't follow RfA closely, but I do know that the number of edits one needs before they are considered acceptable is ever higher and higher and higher. That creates upward pressure on the age of accounts that pass RfA. There's no way to overcome this by any 'reform' of RfA that depends upon people voting for candidates. It's how people perceive legitimate candidacy. The only way to overcome that is to get rid of the voting system.
 * Hmmm. A toy to be thrown...like a stress brick :) Fun! I hadn't thought of it that way. I've always thought of it in an oxymoron sort of way. --Hammersoft (talk) 01:22, 6 October 2013 (UTC)
 * Civil and sensible are my criteria, not necessarily the community's, I think to be honest I should replace sane with clueful. I'm aware that others have different criteria, or different indicators that they use to predict who would be civil and sensible and I hope that our endless discussions at wt:RFA and elsewhere do get some to rethink some of the more unusual criteria we've seen people use.
 * I'm aware that we are far closer to the minimum than the maximum, but I still have my dream and I reserve the right to challenge proposals that take us further from it. More importantly I'm not going to spend my time working out how close we could go to the minimum as I think we should be aiming to steer well away from it.
 * Edit count is one of the arbitrary criteria at RFA, tenure is another. a couple of years ago I nominated a candidate who was close to both minima with circa 3,500 edits and just over a year of tenure. He got through but I'm not sure a similar candidate would now. However I suspect that the very sharp fall in RFAs may help us here by slowing down the process of standards inflation, if I'm right the process is a side affect of consensus - people on the losing "side" of an RFA feel that they want to bring their criteria in line with the winning side, and if the threshold was 50% then that would tend to a stable set of criteria. But because it only takes about 30% to veto a candidate there is a tendency for criteria to drift upwards. I think the best solution to that is not to lower the threshold to 50%, but to set a criteria, and require consensus to change that criteria.
 * But all hammers are soft, or soft as iron gets, if not they'd shatter!  Ϣere Spiel  Chequers  15:39, 6 October 2013 (UTC)

-1
It looks like the switch to the single template did not properly handle months with no admins and showed them as the default yellow, as there was no "-1" option in the new template. I have now added one and it should work properly  Snowolf How can I help? 18:41, 18 December 2014 (UTC)
 * Thanks Snowolf, much appreciated. Of course a very different heatmap would be to show change by month, but I'm not volunteering to do that.  Ϣere Spiel  Chequers  10:52, 20 December 2014 (UTC)

Bot automation
Just as with Desysoppings by month, I feel like updating this can't be enormously fun. Are you interested in a bot maintaining it? I'd love to help :) &mdash; MusikAnimal  talk  23:58, 30 January 2017 (UTC)
 * Thanks MusikAnimal whilst I'm very fond of this and it doesn't take long to update I'd be delighted to have its future ensured in that way. I have moved it out of Userspace to Wikipedia:RFA by month to indicate that it is now published under an open license. Incidentally I was thinking long term we can't keep making this wider all the time, but we could make it longer. What do you think it would look like with the months across the page? Unless screens get bigger we might need to rotate it in the next few years.  Ϣere Spiel  Chequers  14:47, 31 January 2017 (UTC)

Rotation
Nice work rotating this so that the months are on the horizontal axis, I was wondering if it would be possible to have the vertical axis ascending? That would means people could see the recent info without paging down.  Ϣere Spiel  Chequers  18:48, 19 September 2017 (UTC)
 * Like this? --Hammersoft (talk) 21:12, 19 September 2017 (UTC)
 * Thanks, yes much better.  Ϣere Spiel  Chequers  09:01, 21 September 2017 (UTC)

Colour Scheme
Now that it is extremely rare to get more than 5 successful RFAs in a month, does the colour scheme still make sense? With only two exceptions, everything since 2012 has been either the lightest shade of orange, or white, so having colour coding is arguably no longer especially helpful. It might at least be more useful to have another colour for 1, and then one for 2-5, or something like that to provide more granularity? -- K orr u ski Talk 14:14, 13 September 2018 (UTC)
 * We currently have 12 colours, 24 would be nice. Provided that is they still were a complex blend that gives differentiation even to people with colour blindness, and they visually work as a heat map. But I'm not convinced we should make a change just for the very low bands, If we were only displaying the last three years then I'd agree that everything in the last eight years is in one of three bands, the vast majority 1-5. But the whole heat map also contains the other very different era when things could get as busy as 68 RFAs. 24 colours would give us a range with one colour for zero and 23 bands of 3.  Ϣere Spiel  Chequers  21:33, 13 September 2018 (UTC)
 * That makes sense - although adding another 12 shades might be a challenge. Perhaps the current shades of black to orange could be used for the higher 12, and the lower 12 could be increasingly 'cold' shades of blue and/or green? I'm happy to have a pass at this however I'm unsure of the best approach. I don't want to just start editing Template:Hotcold and break other pages that use this. Can you help at all? -- K orr u ski Talk 08:09, 14 September 2018 (UTC)

I suggest that first we consider the colours and bands then sort out how to do it. This is my suggestion for the bands, but simply repeating the colours: If you are OK with the bands, lets try and see how some colour ranges would work (I think cool blues sound like something that might have problems with one form of colour blindness).  Ϣere Spiel  Chequers  17:17, 14 September 2018 (UTC)


 * Considering how rare the larger numbers were, and even more unlikely that they will be exceeded again, you could probably cut off at a lower number. For example, making the last band "over 52" would only add two months to the current highest band. That would reduce the number of additional shades needed by six. --RL0919 (talk) 17:55, 14 September 2018 (UTC)
 * I would be uncomfortable doing something like that. It would be a form of massaging the statistics, it also misses the point that to some people looking at these stats the interesting thing is the contrast between peaks and troughs, and working out the link between them and wiki history.  Ϣere Spiel  Chequers  22:13, 14 September 2018 (UTC)
 * How about this as a first pass. I have used color bands from http://colorbrewer2.org/ which are supposedly colorblind-friendly, although slightly adapted to get the range we require.


 * {| class="wikitable" border="1"

! Key!!
 * {{legend|White|0 successful RFAs}}
 * {{legend|#FFF382|37–39 successful RFAs}}
 * {{legend|#deebf7|1–3 successful RFAs}}
 * {{legend|#FBEC5D|40–42 successful RFAs}}
 * {{legend|#c6dbef|4–6 successful RFAs}}
 * {{legend|#FFDF00|43–45 successful RFAs}}
 * {{legend|#9ecae1|7–9 successful RFAs}}
 * {{legend|#FCC200|46–48 successful RFAs}}
 * {{legend|#6baed6|12–15 successful RFAs}}
 * {{legend|Orange|49–52 successful RFAs}}
 * {{legend|#4292c6|16–18 successful RFAs}}
 * {{legend|DarkOrange|53-54 successful RFAs}}
 * {{legend|#2171b5|19-21 successful RFAs}}
 * {{legend|#E9692C|55–57 successful RFAs}}
 * {{legend|#08519c|22–24 successful RFAs}}
 * {{legend|#CC5500|58–60 successful RFAs}}
 * {{legend|#0F3D80|25–27 successful RFAs}}
 * {{legend|#b06030|61–63 successful RFAs}}
 * {{legend|#082F69|28–30 successful RFAs}}
 * {{legend|#804020|64–66 successful RFAs}}
 * {{legend|#03234E|31–33 successful RFAs}}
 * {{legend|#602000|67–69 successful RFAs}}
 * {{legend|#001C42|34–36 successful RFAs}}
 * {{legend|Black|More than 70 successful RFAs}}
 * }
 * K orr u ski Talk 08:15, 18 September 2018 (UTC)
 * It would be nice if the color saturation monotonically increased, although I like the idea of using two different colors. Maybe fade out to white again at the top of the scale? Enterprisey (talk!) 06:34, 29 September 2018 (UTC)
 * I don't see sufficient difference between the middle of the scale and the top of the scale - compare 34-36 with 70 plus or 31-33 with 67-69. It looks like two scales, one increasingly blue, the other more autumnal. It must be possible to have two dozen colours in one range, even if that meant that subtly different results were subtly different from each other.  Ϣere Spiel  Chequers  10:40, 29 September 2018 (UTC)
 * {{legend|#082F69|28–30 successful RFAs}}
 * {{legend|#804020|64–66 successful RFAs}}
 * {{legend|#03234E|31–33 successful RFAs}}
 * {{legend|#602000|67–69 successful RFAs}}
 * {{legend|#001C42|34–36 successful RFAs}}
 * {{legend|Black|More than 70 successful RFAs}}
 * }
 * K orr u ski Talk 08:15, 18 September 2018 (UTC)
 * It would be nice if the color saturation monotonically increased, although I like the idea of using two different colors. Maybe fade out to white again at the top of the scale? Enterprisey (talk!) 06:34, 29 September 2018 (UTC)
 * I don't see sufficient difference between the middle of the scale and the top of the scale - compare 34-36 with 70 plus or 31-33 with 67-69. It looks like two scales, one increasingly blue, the other more autumnal. It must be possible to have two dozen colours in one range, even if that meant that subtly different results were subtly different from each other.  Ϣere Spiel  Chequers  10:40, 29 September 2018 (UTC)
 * It would be nice if the color saturation monotonically increased, although I like the idea of using two different colors. Maybe fade out to white again at the top of the scale? Enterprisey (talk!) 06:34, 29 September 2018 (UTC)
 * I don't see sufficient difference between the middle of the scale and the top of the scale - compare 34-36 with 70 plus or 31-33 with 67-69. It looks like two scales, one increasingly blue, the other more autumnal. It must be possible to have two dozen colours in one range, even if that meant that subtly different results were subtly different from each other.  Ϣere Spiel  Chequers  10:40, 29 September 2018 (UTC)

Why did the number of requests drop dramatically after 2007?
I was looking at this chart, and the number of requests seem to have been significantly reduced after 2007. What happened to requests that made there not be so many requests in the modern era? Matthew Cenance (talk) 15:43, 25 June 2019 (UTC)
 * the biggest drop came in early 2008 with the unbundling of rollback. At the time you needed Rollback to use Huggle, the tool that all self respecting vandalfighters wanted to use. Nowadays there are other tools and rollback itself is not so important. But it made a big deal at the time, and with rollback unbundled the defacto criteria came to include contributing content as well as needing the tools. So from early 2008 "good vandalfighter" was no longer enough on its own for a candidate to pass RFA.  Ϣere Spiel  Chequers  11:07, 2 August 2019 (UTC)

End of voting or end of Crat chat?
I don't remember when we last had a crat chat that went into the next month and how we handled that. But in such circumstances should we date the end of the RFA from the end of the crat chat or the end of RFA !voting?  Ϣere Spiel  Chequers  11:15, 2 August 2019 (UTC)
 * I'd say go by when the crat closed it. --RL0919 (talk) 21:56, 2 August 2019 (UTC)

Memory of ancient days
Looking at the very bottom of the chart, I'm surprised to see that there were as many as 44 Admins in 2002. From my memory of the time, there were only a very few people with Admin rights on Wikipedia: Jimmy Wales, Ed Poor, maybe one or two others (who kept a lower profile), & maybe the developers. (I remember one Admin privilege that no longer exists was permission to run queries on the database that managed the wiki; I don't know if this right privilege still existed when I got the bit, but I know I never used it because I couldn't think of a good -- or any -- use for it.) And at the time one became an Admin thru direct communication with Wales, & having to convince him to give it you. I don't know how hard it was to persuade him, but I imagine one had to do more than ask politely for it.Then in 2003, about the time of the "Jimbo doesn't scale" declaration, Wales sent out an email saying he'd appoint anyone an Admin simply if they contacted him. Then that Summer, people who wanted to be an Admin had to submit to an approval process; the earliest Admin election is documented here in 14 June 2003. I put my name a few months later in 20 August 2003; I delayed because (to be frank) I didn't have the confidence that I could do a good job.One must remember that having Admin rights in the first few years truly wasn't a big thing: it was used more as an administrative function, deleting pages or renaming them, or running queries on the wiki database. At the time it was felt that peer pressure was sufficient to make people conform to community norms -- which worked 90%+ of the time -- & that having one of the senior members like Jimmy Wales or Ed Poor talk to them would work the rest of the time. (Of course it didn't.) The idea of an Admin being akin to a police officer only came later. (It'd be an interesting sociological study to compare the nomination statements over time, to see how the role of Admin changed. I bet "vandal fighting" peaked around 2006 or 2007, then slowly declined afterwards. I also expect that when the role of Admin became so identified with being a wiki LEO, the attraction of being an Admin faded: who wants to volunteer to be a wiki LEO, with all of the drawbacks of the job & few if any benefits?) -- llywrch (talk) 16:07, 8 October 2021 (UTC)
 * Being an administrator these days isn't really a very attractive prospect. I stayed away from it for a long time. Why would I want to do that? Why would I subject myself to the crap admins go through? There was plenty for me to do without having the admin bit. What motivated me was finding it frustrating to deal with bad actors on the project in a timely way. It's a complete guess, but I think a significant number of people have no interest in being an administrator because there's nothing appealing about it. There's sort of an inverse that's happened I think. Back in the day, being an admin wasn't a big deal, but since anyone with experience could be one it was kind of like being recognized as a senior editor, more part of the community or something. Now that it's become a very big deal, there's a crap ton of layers on it that burden it with potential for being attacked. So, now it's not a big deal on a personal level to be an admin if you've been in the community for a while. There's other ways to "belong" without putting up with the crap. I hope I'm making sense :) --Hammersoft (talk) 18:04, 8 October 2021 (UTC)
 * Another reason RfAs have become a nasty gauntlet is the tall poppy syndrome. And HS, your comment does make sense. -- llywrch (talk) 23:03, 8 October 2021 (UTC)
 * and anyone else interested: for more info about admins in 2002, see User:NoSeptember/Early admins (20 Sept 2002) and follow the links from there. Graham 87 07:17, 16 August 2022 (UTC)
 * Also, per the page history at Database queries, that feature seems to have been disabled some time in 2004, but I don't know exactly when. There's more interesting contrast at this September 2003 version of the blocking policy. Graham 87 07:33, 16 August 2022 (UTC)