1051:, which has completely independent data gathering. This broken trend line has been discussed for years. I don't know of any good explanation for why the change in 2007 was so sudden. More generally speaking I do think the steep growth up to 2007 was partly because more and more people got the news about this Next Big Thing Called Knowledge. At some point, presumably around 2007, nearly all internet users in many countries with high internet reach were aware of Knowledge's existence, saturation near 100%, S-curve flattened out. Then as novelty effect wore off, some people who were marginally interested, but wanted to experience the newest hype, left for the Next Big Thing. Therefore comparing our current metrics with the peak in 2007 feels somewhat artificial to me. It would have been different if we had had a longer history, say with a steady state for 10 years, and only recently numbers started to drop. Of course the often mentioned aspects like less welcoming attitude towards newbies, and progressively difficult syntax, emerging social sites, etc all played their part. But my personal hunch is that Knowledge's fast rise to fame, and the resulting the hype factor, is often underrepresented in explaining our modest decline in editors since 2007.
985:
asymptote like a logistic growth curve, or level off and gradually decline, but instead the trend suddenly screeches to a halt and reverses. If this were a scientific graph I'd immediately begin looking for a change in data collection methods. Then, if we could assure ourselves that the data series is consistent, we would look for a new process or other change that was introduced at this point. So, what happened in March-April 2007? Second, the lack of any relationship between the "active editors" and "retention rate" seems slightly odd. The retention rate began decreasing around the start of 2005, and has been quite smoothly decreasing at a decreasing rate up to the present. It's looks just a little strange for this curve to be totally decoupled from the number of active editors. (Yes, I'm aware that one is a number and the other is a ratio, but it's still a little puzzling that they're totally independent). To summarize my slightly long-winded musings, I think it's worth asking (1) if we're certain that the curves are showing us what we think they're showing us and (2) if so, what happened in March 2007?
1138:"perished" from articles, and when I found myself enjoying it I started to branch out, eventually finding NPP. I did it quite voluntarily, but the vast majority of new users don't, and that's when they run into trouble. People who want to immediately do everything, be it trying to gain every possible userright, immediately try to set speed records as vandal fighters/NPPers, or write a brand new article, tend to get smacked down because there's no possible way they can use those tools/complete those tasks as effectively as is necessary, and as much as we want to be nice we can't mess around in those areas. As Sue says in the interview, we can't expect people to immediately become great editors; it'd be helpful to promote, for instance,
1492:
frontier mentality prevailed and we didn't have to compete quite so keenly for our reputation on the internet. Things started getting more serious from about 2005 onwards, and in the process, some articles that university lecturers might have once scoffed at gradually became better written than they themselves are capable of producing. We became more rule-bound (like all quality publications) and more demanding of all editors. In effect, we pulled away from "anyone can edit", and we should accept as inevitable that quality enforcement has made WP less "welcoming" to newcomers, many of whom lack the skills and patience to learn the patterns demanded of this cultural product. Openness and quality
127:
1501:
happen a bit), but that a culture of actively communicating and empathising with the volunteers needs to run more deeply. One disappointing event last year was the WMF's vetoing of our community's decision to stop anon drive-by article creation; this would have required that newbies show that they're prepared to edit existing articles for just a little while before creating their first article. The WMF appears to see increases in article numbers as a highly significant metric of success, whereas a groundswell of en.WP editors know that committed long-term members come from newbies who are keen to
326:
editors get discouraged. They feel like theyâre making mistakes, that theyâre getting in trouble, people donât want their help. And so they leave, and who can blame them? We can mitigate some of that by toning down the intimidation factor of the warnings: making them simpler and friendlier. We can also help by adding some praise and thanks into the mix. When the
Foundation surveys current editors, they tell us one of the things they enjoy most about editing Knowledge is when someone they respect tells them theyâre doing a good job. Praise and thanks are powerful.
880:
135:
314:
464:
83:
793:
way to retain good new people and turn them into long-term editors. It certainly is a lot of work, and takes up a lot of time. Sometimes it prevents me from working as much as I would like on other aspects of
Knowledge, but when you take the long-term view, it is definitely extremely worthwhile. I would suggest that encouraging and even organizing this kind of activity from the most active projects may be one good solution to help reverse the curve on editor loss.
73:
177:
high quality and attracting new editors are both critical goals, her view is that quality has not been the problem, although she didn't define exactly what article quality is. What we didnât know in 2007, she said, was that âquality was doing fine, whereas participation was in serious trouble. The
English Knowledge was at the tail end of a significant drop in the retention of new editors: people were giving up the editing process more quickly than ever before.
173:, the boss of one of the world's most powerful websites is all charm and professionalism. Much of the interview concerned the issues she raised in a landmark address in November to the board of Wikimedia UK, in which she said the slide showing a graph of declining editor retention (below) is what the Foundation calls âthe holy-shit slide". This is a huge, "really really bad" problem, she told Wikimedia UK, and is worst on the English and German Wikipedias.
251:
and institutional partnerships?" For
Gardner, "The editor retention problem is our shared problem. ... it's easiest for the Foundation to help when there's a technical or tools aspect to the problem. But when the issue is purely editorial or cultural, itâs harder for us to play a role." She singled out two areas: the first is behavioral problems, and the second the sheer quantity of policy and instructional text ("simplifying it would help everyone").
1733:
358:
to optimize for one thing (no bad edit should survive for very long) than for many things (good edits should be preserved and built upon, new editors should be welcomed and coached, etc.). So I donât think itâs an attitudinal problem, but more an issue of focusing energy now on re-balancing to ensure our processes for patrolling edits, deleting content, etc. are also designed to be encouraging and supportive of new people.
93:
53:
335:
collectively make? How can we better distinguish in the patrolling process between good faith new-user mistakes, and bad faith edits? What are the three most essential pieces of advice that every new editor should be given, and how do we make them front and centre, early in their experience with us? In general, how can we best equip new editors to edit well, as quickly and enjoyably as possible?
103:
56:
63:
1113:
documented (see Ed Chi's paper). I actually spoke with him a while back and he was telling me that when his team started analyzing the data, they were also surprised at the sudden plateau and thought it might have been due to a glitch in the data. They conducted their own investigation and concluded that the plateau was in fact a real effect.
1160:, would indicate they're convinced they know what we need better than we do. New Page Triage will be helpful (at least to those of us who know what we're doing, which is a very small percentage of NPPers), but (the great majority of) the community rather clearly indicated what it wanted and thought was best for itself; alas, we're
1718:, the total of all-language active Wikipedians has been growing, slightly, for the past 6 months (October: 80,630 active editors, all-languages). Anyway, the so-called "mass exit" of editors since 2007 has clearly ended. Tell the Foundation not to turn off the lights yet: those 34,000 editors intend to stay all year at the party! -
1509:
We have reached a stage in our evolution in which we need a multi-pronged strategy to keep the labour-force growing, upskilling, and increasingly diverse. It's one thing for the
Foundation to develop technical functionalities (although the visual editor does sound excellent); but we need from the WMF
1464:
There is still a drop in WP editors for May/June each year, but the number returning in
September has steadied over the past 2 years. The feared "free-fall decline" has ended, and globally, there are more active editors now, but slight drops in enwiki or dewiki, while Swedish WP has been unchanged as
1137:
I'm someone who started here in March 2010 and am still going quite strong here. It is my firm belief that the people who stay here are usually the people who take the time to make minor edits and read around the project pages to find something they enjoy. I started off by trying to remove the word
792:
In my experience I have found that (within WikiProject
Gastropods), identifying new contributors using NewArtBot listings, then welcoming them personally, inviting them to join the project, and then (assuming they express some interest) mentoring them carefully and enthusiastically, has been the best
341:
Jimmy went to
Wikimedia and said "quality ⌠we need to do better", there was this moral panic created around quality ⌠what Jimmy said gave a whole lot of people the license to be jerks. ... Folks are playing Knowledge like it's a video game and their job is to kill vandals ... every now and again a
753:
The fundraiser for money has been working exceedingly well with our number of donors increasing 10 fold since 2008. What we need now is a fundraiser for editors. I meet well educated professionals who use
Knowledge but have no ideas that they can edit it. We need to run a banner with the same energy
424:
Regaining our sense of openness will be hard work: it flies in the face of some of our strongest and least healthy instincts as human beings. People find it difficult to assume good faith and to devolve power. We naturally put up walls and our brains fall into us-versus-them patterns. Thatâs normal.
387:
arenât intended to conduct comprehensive, authoritative studies, but to produce actionable insights that guide and support our work. Thatâs why we tend to focus on small, fast, tightly defined research projects that will answer simple questions, such as "Which warning templates deter new editors the
254:
We queried her take on this second area, pointing out that all publishers that aim to present high-quality information find they need complex rules, whether explicit or via accepted standards of writing and scholarship. Could she give specific examples of areas where we could simplify policy without
250:
Does the
Foundation have any solutions to enable the editing community to address the cultural issues that might be driving editors away â beyond the WMF's technical initiatives such as an easier editing interface and means of empowering kitten distribution, and external initiatives such as outreach
198:
wanted to know more about where article quality fits into this model â specifically whether the three factors are sometimes at odds with each other and whether a purely one-way causality is involved. Deletions and reversions might be distasteful to new editors, but how can we, for instance, maintain
156:
In 2007, she took up consulting for the Foundation on operations and governance and within a year was in the top job. Her tenure has seen a precipitous increase in the staffing at the San Francisco office, which now employs some 100 people, up from 65 just six months ago, and a budget well in excess
1571:
10 edits per month), since June 2011 (October: 35,028 editors). We had discussed this likely outcome, several months ago, that the "free fall" or "hemorrhaging" of editors was obviously ending, at a bottom-out count of 34,000 people who will always edit English Knowledge each month. The usage data,
1086:
What happened was the start of deletionism as a major force and crackdown on the very open culture. The first larges cale example was the Userbox wars in 2006, and crackdowns on wheel warring in 2007. Those crackdowns changed the culture from the laid back culture wikipedia had when it started.
357:
I wouldnât characterize it as bunker mentality at all. Itâs just a system thatâs currently optimized for combating bad edits, while being insufficiently concerned with the well-being of new editors who are, in good faith, trying to help the projects. Thatâs understandable, because itâs a lot easier
271:
I don't have specific examples of where policy should be simplified, but I do think it would be helpful for us to visibly embrace "be bold" again, as well as "break all rules." People get embarrassed when they make mistakes, and some of our policies seem almost impossibly intricate. So, I think one
1460:
Because U.S. schools voluntarily adopted the WP-bans, the decline would not be as drastic as if a bizarre U.S. Federal law (would be from 2006) had censored WP use after 2007, and the U.S. Govt rarely passes drastic legislation. Instead, however, when the growing support for school-bans is coupled
367:
As I said, we do want to work with community members to figure out how to reduce policy and instructional cruft: probably the first step in doing that, though, would be a convening of interested editors. The community department has been talking about doing that. I think it would be very difficult
1711:
Well, along with active users who were edit-banned, about 3,000 "average editors" left in June 2010, and do not seem to have returned. I am suspecting that they were some groups of students who left in June 2010, but now the remaining 34,000 editors do not take "summer wikibreak" as others did in
984:
As someone whose day job involves time series analysis there are several things that come to mind. First, the "active editors" curve is smooth and well behaved up until about (eyeballing) March 2007. Up to this time we see a rising trend that is beginning to decrease a bit. We might expect it to
811:
I think the electronics / engineering topic could really use a few people doing that. Somewhere between 95% and 99% of new contributors really want to work collaboratively to make Knowledge better. Alas, a small percentage of the existing editors are arrogant and somewhat abusive with an attitude
726:
Could Gardner explain what was meant by "quality was doing fine"? As Doc James says, the quality of Knowledge articles unsatisfactory. The only meaning of "doing fine" that I'd be comfortable with was one showing the chart going up. We need to be careful that efforts to attract new editors do not
362:
If the presumably less important issue of controversial content merits outside study by consultants, why isn't the Foundation putting resources into having social scientists diagnose the problems of editor retention and offer suggestions to the community to reform its internal culture? Why aren't
325:
If you look at new editorsâ talk pages, they can be pretty depressing â theyâre often an uninterrupted stream of warnings and criticisms. Experienced editors put those warnings there because they want to make Knowledge better: their intent is good. But the overall effect, we know, is that the new
207:
I donât believe that quality and openness are inherently opposed to each other. Openness is what enables and motivates people to show up in the first place. It also means weâll get some bad faith contributors and some who donât have the basic competence to contribute well. But thatâs a reasonable
176:
A prominent issue on the English Knowledge is whether attempts to achieve high quality in articles â and perceptions that this is entangled with unfriendly treatment of newbies by the community â are associated with low rates of attracting and retaining new editors. Although Gardner believes that
1525:
Also riding on point 2, collaborating with the community to develop effective infrastructures for the human mentoring of those identified by a combination of statistical probability and human savvy. Our mentoring resources are limited and need to be carefully allocated; if we can learn how to do
1500:
I find Ms Gardner's professionalism and dedication impressive. But the Foundation's great challenge is to engage more deeply with the editors of the Wikipedias; and by that I don't mean that their now massively increased numbers of employees should sit around passively browsing en.WP (which does
1452:
The closest factor seems to be a trendy movement among colleges and entire local schoolboards to ban use of Knowledge in schools near the end of the 2007 summer break, and fewer returned for the 2007-2008 school year. To research this school-ban concept, I wrote an essay listing 18 major sources
416:
will be available within a few weeks. My hope is all this will create a virtuous circle: support for openness will begin to increase openness, which will begin to increase new editor retention, which will begin to relieve the workload of experienced editors, which will enable everyone to relax a
259:
Yes, the premise of this question is absolutely correct. The analogy I often use is the newsroom. Anybody whoâs curious and reasonably intelligent can be a good journalist, but you do need some orientation and guidance. Just like a newsroom couldnât invite in 100 random people off the street and
334:
How do we counter systemic bias when it comes to defining reliable sources and notability â that is, in a context where decisions are made by consensus, and in which many types of people are underrepresented, how do we ensure systemic bias doesnât weaken and harm the quality of the decisions we
1491:
I think there's some truth in Erik Zachte's observation that there was always going to be flattening of the curve after the hype and after the project reached a certain level of maturity. Back in the early days, article quality and comprehensiveness were queasy, but we didn't care that much: a
1112:
To my knowledge, the sudden plateau of active editors in March 2007 is not the result of a change in data collection. I haven't seen any evidence that suggesting a change in either the way we capture the data, or the way "active editors" is defined. This plateau has actually been pretty well
267:
used by Knowledge Ambassadors arenât ideal in every respect, but theyâre increasingly road-tested and optimized for the real-world instruction of new contributors. They're pretty good. In general, to the extent that weâre showing instructions as part of the user interface, we need to make them
217:
If you believe thereâs an inherent tension between quality and openness, then yes, you might believe that when I advocate for openness, Iâm speaking against quality; but I donât believe that. Quality improvement work â like page patrolling, the FAC, developing content partnerships, and staging
1518:
Identifying which early editing patterns are associated with people who are more likely to join the community of article improvers. Ms Gardner talked of these patterns in broad terms in the Wikimedia UK video, but can we achieve the statistical depth that is necessary? Is the WMF trying to do
595:
My long observations from the past years, tell me the experience, that the most likely people to have the energy to speak an-mass to the newbies, and at the same time the people who are the strictest, are actually former newbies just recently grown in to self-confident Wikipedians - about the
376:
We put it to Gardner that much of the supporting research she cited to the Wikimedia UK board was informal, small-scale or not entirely rigorous (such as the study based on a dubious assumption that any editor who was not a sockpuppet, vandal or spammer was editing in "good faith"Â â ignoring
1496:
thus pitted against each other in key respects, and although Ms Gardner points to causal connections between these phenomena, in other parts of the interview she acknowledges them as competing forces ("if we say that becoming an editor should be easy, really, that's a little delusional").
1447:
I agree the drastic decline in editors by mid-2007 does not follow patterns of normal population growth. Such drastic changes often indicate a key factor, such as a change in data-collection methods, or a massive societal change, such as a war, natural disaster, or government regulation.
189:
Participation matters because it drives quality. People come and go naturally, and that means we need to continually bring in and successfully orient new people. If we donât, the community will shrink over time and quality will suffer. Thatâs why participation is our top priority right
1712:
past years. The final core of 34,000 editors seem to work each month, regardless of the northern hemisphere summer-break period beginning in June. However, it could be that more students (or others) use home computers to continue editing Knowledge when school ends (or on vacations).
958:
I don't know if I should be replying to you Tony, but the indentation indicates you are replying to me. I assume you are not referring to the talk section I linked to. Because it is not as you described: "rather diffuse thread-fest: everyone's wishing each other happy new year."
519:
1514:
Targeting demographics out there that have the time, skills, and knowledge. Professional retirees, especially women, are just one obvious quarry, and here's a strong basis for leadership by San Francisco among the chapters, starting with the anglophones and proceeding to other
1505:
articles, and are painfully aware that every stub-article created imposes large overheads on existing editors to reach a minimum level of utility for readers. Ms Gardner is aware of this "scut-work" problem, but I felt she didn't place it in this practical context.
352:
Many people have complained that Knowledge patrollers and administrators have become insular and taken on a bunker mentality, driving new contributors away. Do you agree, and if so, how can this attitude be combated without alienating the current core contributors?
212:
What do you say, we asked, to editors whose focus has been on improving quality and who may have taken your comments and the recent focus of Foundation initiatives as an indication that their contributions aren't valued, or even that they are part of the problem?
182:
There aren't enough people to do the work ... people are stressed and they're burned out ... you still have lots of doing scut-work ... So where are the new generations of people, relieving them of the need to do all this scut-work? Â â Sue Gardner, UK address
1526:
this, there could be a big bang for our bucks. San Francisco is best placed to coordinate the analysis and interpretation of a mentoring program, given that it probably needs to be intertwined with the technical side the WMF sees itself as occupying.
666:. It's been up for several weeks, but gets far less new feedback than English so far. Anything you can do to improve the interface for Dutch speakers (try logging in under a new account to see the tool for leaving feedback) would be most welcome.
425:
But we need to resist it. The Wikimedia projects are a triumph of human achievement, and theyâre built on a belief that human beings are generally well-intentioned and want to help. We need to remember that and to behave consistently with it.
1116:
I actually agree with Erik's Zachte's comment on the "hype factor." I do think part (though not all) of the rise and subsequent plateau may be explained by more and more people finding out about Knowledge and becoming interested in the
1182:
This graph is more than two years out of date. Are current data available? (I just noticed this because the labels are far too small to be legible unless the graph is expanded to its full size. Somebody needs to read Tufte...)
1164:
where we were before that started. Not to say the WMF doesn't do a great deal of good work (indeed, without them I'm not typing this), but saying and doing are two different things; I'd like to see some more of the latter.
372:
Having said that, Gardner pointed out that the Foundation tries to persuade social scientists who do understand Knowledge to study it, and has done "lots of informal consultation on participation-related issues" with them.
76:
1461:
with the yearly pattern of school vacations in May/June, then the rapid decline is worsened by students leaving for vacation, and fewer students returning to use Knowledge in September for the next school year (2007-2008).
1727:
1702:
On average, people have stopped leaving: the count never drops below 34,000 active editors. That stability gives the WMF resource planners the controlled usage pattern they need to expect 34,000 active editors each
743:
First of all quality is not "fine". Yes it is okay but our combined number of FAs andd GAs is still less that half a percent of total articles. We do not need new articles we need to improve the quality of what we
619:
They are very important for the community, for there is non, with so much energy to work hard, nevertheless they should be schooled a bit more - on regular basis about the importance of new editors to the project.
645:"The new-editor feedback dashboard is live on the English and Dutch Wikipedias" - are you sure you didn't misquote here? I am a Dutch editor and unaware of the feedback dashboard being live there :) Check it out:
368:
for most social scientists to help us with a highly specific problem like new editor retention, because it's a deep community issue, and not many social scientists understand very much about how Knowledge works.
288:
At this point, Gardner stepped back to take a big-picture view of how the community and the WMF should interface, saying that the Foundation isn't the expert in either the behavioural or the cultural aspects:
1777:
1192:
222:â makes Knowledge better and more valuable for readers. Where we run into problems is when we do things that repel or frustrate good-faith new editors. But Iâm not sure thereâs a fixed relationship between
830:
The presentation Sue gave to WMUK included an analysis of the "new editor story" - discouragement on opening the edit tab, discouragement on feedback being the first two hurdles. Things we can do
96:
66:
303:
to make new editorsâ experiences visible to experienced editors, and to give experienced editors an easy mechanism for coaching the new people. Weâve started working with editors to create
106:
1436:
1757:. The table has monthly page views for Knowledge over years. It looks like the total monthly page views for all Wikimedia projects in all languages has almost doubled in around 4 years.
1060:
994:
377:
self-promoters and POV-pushers). Now that we know editor decline is a very serious problem, is a comprehensive and rigorous research initiative being planned to analyze the phenomenon?
1766:
With more articles we need more editors to bring up the quality of the articles. Many articles need to be filled in, and need a higher level of quality info, charts, images, etc.. --
1222:
1129:
1014:
1570:
The October and November usage stats have confirmed, along with 3rd quarter editor counts, that the count of active editors has stabilized at nearly 34,000 active editors (: -->
297:, and I think the editing community should lead on others. There are other initiatives where weâre partnering with the editing community â for example, the Foundation built the
293:
The community understands them better than we do, and will probably have better ideas about how to solve them. ... The Foundation will lead on technical initiatives such as the
39:
1120:
I'm not sure what you mean by the lack of any relationship between "Active editors" and "retention rate". If anything, the two seem to be inversely related. Can you clarify?
412:
is live on the English and Dutch Wikipedias. New warning templates are being tested on the English and Portuguese Wikipedias. And the first opt-in user-facing prototype of the
1722:
678:
1176:
278:
We need to be able to experiment, to do stuff. Weâre going to consult when when we think itâs helpful and necessary, ⌠but we need to do tiny bits of experimentation ...
1544:
1142:
so editors can get a start doing something which won't run them into really severe trouble while simultaneously giving them a glimpse of just how diverse our topics are.
1474:
1376:
1346:
1270:
1402:
508:
498:
1048:
The chart is from a study by Howie Fung and Diederik van Liere. I can't comment on the retention plot. The active editor plot corresponds with the one from wikistats
858:
321:
While staking the Foundation's claim to the more technical side of the equation, Gardner doesn't shrink from providing advice on how we can fix the cultural problem.
657:
363:
there Foundation usability experts recommending overhauls of dense and daunting policy and procedural pages and not just of technical aspects such as the interface?
1572:
for the past 6 months (June-Nov.) has confirmed this same pattern of editors staying: each month in 2011 is nearly 99% of the 2010 active-editor counts. See table:
1096:
1081:
821:
503:
717:
263:
So if we say that becoming an editor should be easy, really, that's a little delusional. And itâs exactly why people need easy instructional text and videos. The
1549:
1316:
1290:
970:
953:
309:, a tool that will make page patrolling easier, and will offer options that support and encourage new editors as well as repelling vandalism and other bad edits.
1479:
802:
812:
that they are always the smartest person in the room, and I really think that this drives away newbies. It's hard to figure out how to address this problem. --
534:
493:
696:
86:
1747:
1456:
1239:
1035:
913:
785:
1484:
As one of the two journalists involved in this story, I probably shouldn't comment; but please indulge me a little. What I say is just my personal takeânot
734:
938:
To whoever indented Gardner's direct quotes in the interview above: I don't like it, I was happy with it before; but I can't be bothered to revert it all.
208:
price to pay for the overall effectiveness of an open system, and it doesnât invalidate the basic premise of Knowledge: that openness will lead to quality.
481:
1442:
1421:
1255:
927:
727:
make the quality chart go down. That's not just speculation: one recent large-scale student recruitment had precisely that effect on our articles.
475:
897:
887:
636:
1088:
382:
1845:
1741:
902:
1170:
1391:
The two articles were meant to run in the same edition, but unfortunately, they were accidentally split into two separate editions. --
230:, so I donât think editors who focus on quality improvement should feel attacked or unappreciated when openness is being emphasized.
149:, the non-profit, non-commercial organization that operates nearly 300 online projects. Originally from Canada, she worked with the
396:
My hope is that the community will become less risk-averse as the Foundation makes successful, useful interventions. I believe the
1432:
1188:
990:
539:
21:
1303:"Spiked" doesn't make sense in this context either. Why not just use plain English? "Use jargon and acronyms judiciously..." --
194:
At the core of Gardner's philosophy, then, is an intrinsic connection between editor retention and what she calls openness. But
1820:
523:
200:
392:
How can a culture that has a heavy status quo bias be changed? How can the community be persuaded to become less risk-averse?
1815:
1810:
1166:
546:
298:
1522:
Riding on point 2, developing both bot-based and human-supervised identification and messaging of such likely candidates.
663:
1805:
1428:
1184:
986:
1427:
Ah, so you meant "split" and not "splice" (the latter means to join together, not to separate). Everybody happy now?
150:
1233:
443:
138:
The âholy-shitâ graph. Active editors (blue) and the one-year retention rate (red) on the English Knowledge, 2004â09
1557:
236:... weâre not falling off a cliff; but we are having serious difficulty retaining good faith new editors, and that
330:
What, then, does Sue Gardner believe are our significant social challenges? She puts these questions in response:
892:
600:
1 year or 2 years of experience in Knowledge. They tend to extrapolate the behavior they witnessed on themselves
264:
33:
1789:
400:
usability improvements are generally seen as successful, although they of course haven't gone far enough yet.
203:(BLP) without reverting problematic edits and deleting inappropriate articles? Gardner rejected the premise:
1800:
777:
463:
17:
653:
1415:
1340:
1249:
1152:
on that premise. A look at the talkpage of New Page Triage (linked in the interview) and, dare I say,
935:
Timeshifter: Looks like a rather diffuse thread-fest: everyone's wishing each other happy new year. OK.
687:"means of empowering kitten distribution" ... Wha...? Is that a placeholder that was never filled in?--
616:
Its like, if they are testing themselves, whether they do already know all the related rules, actually.
163:
as one of ten âmedia game changers of the yearâ for the impact on new media of her work for Wikimedia.
852:
126:
1488:
views or those of my co-author, Skomorokh, or anything Ms Gardner said aside from what we printed.
840:
Don't delete good stuff. We speedy and prod rather wilfully, without proper checks for notability.
781:
754:
we use to raise money to raise editor numbers. This idea has been trialed to a limited extent here
575:
I believe both the points are at the heart of the matter. I do not think there are easy remedies.
1408:
I've been using the word the wrong way all these years...thanks for the grammar lesson I guess.
408:
is on the English Knowledge and is currently being used in seven other projects. The new-editor
1761:
672:
613:
They are the strictest and the mutest (mute, except placing the template and sign to talkpage).
404:
is a small feature, but itâs been adopted by 13 Knowledge language-versions, plus Commons. The
1238:
As I noted last week, we accidentally spliced the publication of this issue - see last week's
260:
expect them to make an immediate high-quality contribution, neither can Knowledge expect that.
170:
1773:
1056:
966:
923:
1826:
1398:
849:
712:
146:
1209:
That is the same data file I analyzed to conclude the editor-decline has ended. See below:
8:
1714:
Meanwhile, because some other-language Wikipedias are growing in active editors, such as
1410:
1372:
1335:
1312:
1266:
1244:
817:
798:
758:
but the effort did not have sufficient data crunching behind it to determine if it works.
649:
1754:
304:
1750:. It also has a timeline chart of active editors over the years. See it on the right.
1153:
561:
So if we say that becoming an editor should be easy, really, that's a little delusional
879:
268:
concise, and emphasize the must-read items instead of trying to cover every edge case.
1719:
1715:
1470:
1219:
1157:
1092:
1078:
1032:
692:
668:
602:(and themselves they survived it, so why not take it as role-model of interaction...)
530:
429:
272:
helpful thing we could do is to tell people that making mistakes is normal and okay.
1767:
1367:
Yes. One day and four comments after asking, I still have no idea what you mean. --
1304:
1125:
1052:
960:
917:
773:
633:
134:
1740:
Editors are leaving for various reasons. Some editors are being driven away. See:
1148:
understand our own dynamics better than the WMF does, and it'd be nice if the WMF
1538:
1008:
947:
732:
706:
159:
1689:
1204:
313:
1368:
1308:
1262:
1139:
837:
Make page text simpler, especially minimising, unifying and clarifying mark-up.
813:
794:
120:
1839:
1393:
1728:
More articles and page views, but less editors. Low quality of many articles
1466:
1280:
688:
914:
User talk:Jimbo Wales/Archive_93#Declining number of editors and donations
1121:
1049:
765:
622:
446:
shown in Gardner's address to the board of Wikimedia UK, 17 October 2011.
142:
381:
No. The Foundation isnât a research organization, and programs like the
1531:
1001:
940:
728:
116:
755:
342:
nun or a tourist wanders in front of the AK-47 and gets murdered ...
169:
interviewed Gardner on her fourth anniversary as executive director.
1732:
1218:
The related graph could be updated from the new data of 2009-2011. -
1212:
1764:. Money can not buy editors, nor quality info on Knowledge pages.
584:
One area to focus on, in my opinion, is to address the most likely
1755:
Page Views for Wikimedia, All Projects, All Platforms, Normalizeds
870:
The lack of enough arbitrators was not discussed in the interview:
1074:
1028:
388:
least?" and "Which help avenues are most used by new editors?"
1510:
a commitment to partner with us in the following strategies:
240:
cause our community to dwindle if we donât fix the problem.
1690:
http://stats.wikimedia.org/EN/TablesWikipediansEditsGt5.htm
1205:
http://stats.wikimedia.org/EN/TablesWikipediansEditsGt5.htm
157:
of $ 20M a year. In October 2009, Gardner was named by the
417:
little and allow for more experimentation and playfulness.
1333:
Are you guys seriously arguing over my word usage -_-.
1785:
Get the latest headlines on your user page â just add
1201:
The current month-by-month data is maintained on page:
702:
592:
who do actually do interact with the newbies the most.
1748:
Knowledge:Knowledge Signpost/2011-12-26/Opinion essay
756:
http://en.wikipedia.org/Wikipedia:Invitation_to_edit
1279:"Spiked" rather than "spliced", perhaps? â Cheers,
544:If your comment has not appeared here, you can try
1837:
1213:#Editor base stabilized at 34,000 active editors
1550:Editor base stabilized at 34,000 active editors
1457:WP:Schools and colleges banned WP in 2007-2008
568:People get embarrassed when they make mistakes
153:in production, journalism and documentaries.
114:
1762:Fundraiser statistics - Wikimedia Foundation
1742:User:Timeshifter/Unresolved content disputes
1480:Tossing the ball into the Foundation's court
646:
1760:Here are fundraiser stats over the years:
1075:#Schools/colleges banned Knowledge in 2007
1029:#Schools/colleges banned Knowledge in 2007
1443:Schools/colleges banned Knowledge in 2007
228:activities that hurt new editor retention
1731:
312:
133:
125:
547:
14:
1838:
224:activities designed to improve quality
874:
1846:Knowledge Signpost archives 2012-01
1453:about the growing WP-ban from 2007:
604:while they were newbies themselves.
27:
843:Be kind. Be personal. Be friendly.
662:The Dutch FeedbackDashboard is at
462:
130:WMF executive director Sue Gardner
28:
1857:
1746:This article has been mentioned:
999:I'll ask Erik Zachte about this.
912:For more info and discussion see
529:These comments are automatically
151:Canadian Broadcasting Corporation
145:is the executive director of the
1558:User_talk:Jimbo_Wales/Archive_89
1167:The Blade of the Northern Lights
1144:I also think we, the community,
878:
664:Speciaal:DashboardTerugkoppeling
101:
91:
81:
71:
61:
51:
1708:Who were those people who left?
1567:The drop in editors has ended:
1242:for "the other take." Cheers,
540:add the page to your watchlist
436:
13:
1:
1429:Short Brigade Harvester Boris
1185:Short Brigade Harvester Boris
987:Short Brigade Harvester Boris
637:16:57, 28 December 2011 (UTC)
317:Sue Gardner at Wikimania 2011
1778:05:49, 10 January 2012 (UTC)
1261:What does "spliced" mean? --
515:
201:biographies of living people
18:Knowledge:Knowledge Signpost
7:
1723:11:30, 7 January 2012 (UTC)
1545:14:38, 6 January 2012 (UTC)
1475:09:53, 6 January 2012 (UTC)
1437:14:30, 6 January 2012 (UTC)
1422:03:35, 4 January 2012 (UTC)
1403:01:09, 4 January 2012 (UTC)
1377:00:34, 4 January 2012 (UTC)
1347:00:17, 4 January 2012 (UTC)
1317:21:04, 3 January 2012 (UTC)
1291:06:51, 3 January 2012 (UTC)
1271:04:50, 3 January 2012 (UTC)
1256:21:57, 2 January 2012 (UTC)
1223:11:30, 7 January 2012 (UTC)
1193:14:38, 6 January 2012 (UTC)
1177:04:21, 6 January 2012 (UTC)
1130:16:55, 4 January 2012 (UTC)
1082:09:53, 6 January 2012 (UTC)
1061:07:13, 4 January 2012 (UTC)
1036:09:53, 6 January 2012 (UTC)
1015:05:52, 4 January 2012 (UTC)
995:03:54, 4 January 2012 (UTC)
971:04:47, 3 January 2012 (UTC)
954:02:02, 3 January 2012 (UTC)
928:01:45, 3 January 2012 (UTC)
859:20:32, 2 January 2012 (UTC)
822:18:42, 2 January 2012 (UTC)
803:18:23, 2 January 2012 (UTC)
786:17:09, 2 January 2012 (UTC)
735:17:58, 2 January 2012 (UTC)
718:19:59, 4 January 2012 (UTC)
697:16:39, 2 January 2012 (UTC)
679:17:54, 2 January 2012 (UTC)
658:15:45, 2 January 2012 (UTC)
10:
1862:
1465:a steady number in 2011. -
982:These graphs look strange.
1736:Active editors over time.
1686:
347:â Sue Gardner, UK address
283:â Sue Gardner, UK address
1556:(Updated from thread in
1097:22:15, 30 May 2012 (UTC)
647:http://nl.wikipedia.org
255:sacrificing standards?
199:strict standards about
1737:
1450:What happened in 2007?
537:. To follow comments,
467:
427:
390:
370:
360:
344:
337:
328:
318:
311:
280:
274:
242:
232:
210:
192:
185:
139:
131:
1790:Signpost-subscription
1735:
898:editors and donations
834:to improve this are:
466:
406:article feedback tool
394:
379:
365:
355:
339:
332:
323:
316:
291:
276:
257:
234:
215:
205:
187:
180:
137:
129:
34:The Gardner interview
669:Steven Walling (WMF)
533:from this article's
220:Wiki loves monuments
147:Wikimedia Foundation
885:The lack of enough
1738:
524:Discuss this story
509:Arbitration report
499:WikiProject report
468:
410:feedback dashboard
384:Summer of research
319:
300:feedback dashboard
218:competitions like
140:
132:
1716:Spanish Knowledge
1696:
1695:
909:
908:
862:
548:purging the cache
1853:
1829:
1794:
1788:
1688:Monthly counts:
1579:
1578:
1543:
1541:
1536:
1420:
1418:
1345:
1343:
1288:
1283:
1254:
1252:
1173:
1013:
1011:
1006:
952:
950:
945:
882:
875:
857:
770:
715:
709:
677:
675:
628:
551:
549:
543:
522:
504:Featured content
486:
478:
471:
447:
440:
348:
284:
246:
123:
105:
104:
95:
94:
85:
84:
75:
74:
65:
64:
55:
54:
1861:
1860:
1856:
1855:
1854:
1852:
1851:
1850:
1836:
1835:
1834:
1833:
1832:
1831:
1830:
1825:
1823:
1818:
1813:
1808:
1803:
1796:
1792:
1786:
1782:
1781:
1730:
1552:
1539:
1532:
1530:
1482:
1445:
1416:
1409:
1341:
1334:
1284:
1281:
1250:
1243:
1236:
1234:Note to readers
1171:
1009:
1002:
1000:
948:
941:
939:
910:
766:
713:
707:
673:
667:
626:
553:
545:
538:
527:
526:
520:+ Add a comment
518:
514:
513:
512:
479:
474:
472:
469:
451:
450:
441:
437:
432:
350:
346:
306:new page triage
286:
282:
248:
244:
186:
160:Huffington Post
124:
113:
112:
111:
102:
92:
82:
72:
62:
52:
46:
43:
32:
26:
25:
24:
12:
11:
5:
1859:
1849:
1848:
1824:
1819:
1814:
1809:
1804:
1799:
1798:
1797:
1784:
1783:
1780:
1729:
1726:
1713:
1705:
1704:
1700:
1699:
1698:
1697:
1694:
1693:
1684:
1683:
1680:
1677:
1674:
1671:
1668:
1665:
1662:
1659:
1655:
1654:
1651:
1648:
1645:
1642:
1639:
1636:
1633:
1630:
1626:
1625:
1620:
1615:
1610:
1605:
1600:
1595:
1590:
1585:
1563:
1562:
1551:
1548:
1528:
1527:
1523:
1520:
1516:
1486:The Signpost's
1481:
1478:
1463:
1462:
1459:
1454:
1444:
1441:
1440:
1439:
1406:
1405:
1388:
1387:
1386:
1385:
1384:
1383:
1382:
1381:
1380:
1379:
1356:
1355:
1354:
1353:
1352:
1351:
1350:
1349:
1324:
1323:
1322:
1321:
1320:
1319:
1296:
1295:
1294:
1293:
1274:
1273:
1235:
1232:
1230:
1228:
1227:
1226:
1225:
1216:
1210:
1207:
1202:
1196:
1195:
1143:
1135:
1134:
1133:
1132:
1118:
1114:
1106:
1104:
1103:
1102:
1101:
1100:
1099:
1084:
1066:
1065:
1064:
1063:
1043:
1042:
1041:
1040:
1039:
1038:
1020:
1019:
1018:
1017:
978:
977:
976:
975:
974:
973:
936:
907:
906:
883:
873:
872:
871:
867:
866:
865:
864:
863:
845:
844:
841:
838:
827:
826:
825:
824:
806:
805:
789:
788:
762:
761:
760:
759:
748:
747:
746:
745:
738:
737:
723:
722:
721:
720:
701:That would be
684:
683:
682:
681:
642:
641:
640:
639:
617:
614:
608:
607:
606:
605:
598:(half year or)
593:
579:
578:
577:
576:
572:
571:
564:
528:
525:
517:
516:
511:
506:
501:
496:
494:News and notes
491:
485:
476:2 January 2012
473:
461:
460:
459:
458:
456:
453:
449:
448:
434:
433:
431:
428:
421:
418:
338:
275:
270:
269:
262:
261:
233:
179:
110:
109:
99:
89:
79:
69:
59:
48:
47:
44:
38:
37:
36:
35:
30:
29:
15:
9:
6:
4:
3:
2:
1858:
1847:
1844:
1843:
1841:
1828:
1822:
1817:
1812:
1807:
1802:
1791:
1779:
1775:
1771:
1770:
1765:
1763:
1758:
1756:
1751:
1749:
1744:
1743:
1734:
1725:
1724:
1721:
1717:
1710:
1709:
1692:
1691:
1685:
1681:
1678:
1675:
1672:
1669:
1666:
1663:
1660:
1657:
1656:
1652:
1649:
1646:
1643:
1640:
1637:
1634:
1631:
1628:
1627:
1624:
1621:
1619:
1616:
1614:
1611:
1609:
1606:
1604:
1601:
1599:
1596:
1594:
1591:
1589:
1586:
1584:
1581:
1580:
1577:
1576:
1575:
1574:
1573:
1569:
1568:
1561:
1559:
1554:
1553:
1547:
1546:
1542:
1537:
1535:
1524:
1521:
1517:
1513:
1512:
1511:
1507:
1504:
1498:
1495:
1489:
1487:
1477:
1476:
1472:
1468:
1458:
1451:
1438:
1434:
1430:
1426:
1425:
1424:
1423:
1419:
1414:
1413:
1404:
1400:
1396:
1395:
1390:
1389:
1378:
1374:
1370:
1366:
1365:
1364:
1363:
1362:
1361:
1360:
1359:
1358:
1357:
1348:
1344:
1339:
1338:
1332:
1331:
1330:
1329:
1328:
1327:
1326:
1325:
1318:
1314:
1310:
1306:
1302:
1301:
1300:
1299:
1298:
1297:
1292:
1289:
1287:
1278:
1277:
1276:
1275:
1272:
1268:
1264:
1260:
1259:
1258:
1257:
1253:
1248:
1247:
1241:
1240:Opinion essay
1231:
1224:
1221:
1217:
1214:
1208:
1206:
1200:
1199:
1198:
1197:
1194:
1190:
1186:
1181:
1180:
1179:
1178:
1174:
1168:
1163:
1159:
1155:
1151:
1147:
1141:
1131:
1127:
1123:
1119:
1115:
1111:
1110:
1109:
1108:
1107:
1098:
1094:
1090:
1085:
1083:
1080:
1076:
1072:
1071:
1070:
1069:
1068:
1067:
1062:
1058:
1054:
1050:
1047:
1046:
1045:
1044:
1037:
1034:
1030:
1026:
1025:
1024:
1023:
1022:
1021:
1016:
1012:
1007:
1005:
998:
997:
996:
992:
988:
983:
980:
979:
972:
968:
964:
963:
957:
956:
955:
951:
946:
944:
937:
934:
933:
932:
931:
930:
929:
925:
921:
920:
915:
904:
901:
899:
894:
890:
889:
884:
881:
877:
876:
869:
868:
860:
855:
854:
851:
847:
846:
842:
839:
836:
835:
833:
829:
828:
823:
819:
815:
810:
809:
808:
807:
804:
800:
796:
791:
790:
787:
783:
779:
775:
771:
769:
764:
763:
757:
752:
751:
750:
749:
742:
741:
740:
739:
736:
733:
730:
725:
724:
719:
716:
710:
704:
700:
699:
698:
694:
690:
686:
685:
680:
676:
670:
665:
661:
660:
659:
656:
655:
651:
648:
644:
643:
638:
635:
634:
631:
630:
625:
618:
615:
612:
611:
610:
609:
603:
599:
594:
591:
587:
583:
582:
581:
580:
574:
573:
569:
565:
562:
558:
557:
555:
554:
550:
541:
536:
532:
521:
510:
507:
505:
502:
500:
497:
495:
492:
490:
487:
483:
477:
470:In this issue
465:
457:
454:
445:
439:
435:
426:
422:
419:
415:
414:visual editor
411:
407:
403:
399:
393:
389:
386:
385:
378:
374:
369:
364:
359:
354:
349:
343:
336:
331:
327:
322:
315:
310:
308:
307:
302:
301:
296:
295:visual editor
290:
285:
279:
273:
266:
256:
252:
247:
245:â Sue Gardner
241:
239:
231:
229:
225:
221:
214:
209:
204:
202:
197:
191:
184:
178:
174:
172:
168:
164:
162:
161:
154:
152:
148:
144:
136:
128:
122:
118:
108:
100:
98:
90:
88:
80:
78:
70:
68:
60:
58:
50:
49:
41:
23:
19:
1768:
1759:
1752:
1745:
1739:
1707:
1706:
1701:
1687:
1622:
1617:
1612:
1607:
1602:
1597:
1592:
1587:
1582:
1566:
1565:
1564:
1555:
1533:
1529:
1508:
1502:
1499:
1493:
1490:
1485:
1483:
1449:
1446:
1411:
1407:
1392:
1336:
1285:
1245:
1237:
1229:
1161:
1149:
1145:
1136:
1105:
1003:
981:
961:
942:
918:
911:
896:
895:drives away
886:
848:
831:
767:
652:
632:
623:
621:
601:
597:
589:
585:
567:
560:
556:Exactly so:
488:
482:all comments
455:
452:
438:
423:
420:
413:
409:
405:
401:
397:
395:
391:
383:
380:
375:
371:
366:
361:
356:
351:
345:
340:
333:
329:
324:
320:
305:
299:
294:
292:
287:
281:
277:
258:
253:
249:
243:
237:
235:
227:
223:
219:
216:
211:
206:
196:The Signpost
195:
193:
188:
181:
175:
167:The Signpost
166:
165:
158:
155:
141:
1827:Suggestions
1769:Timeshifter
1073:See below:
1053:Erik Zachte
1027:See below:
962:Timeshifter
919:Timeshifter
893:arbitrators
708:Orange Mike
590:talkbackers
531:transcluded
143:Sue Gardner
1515:languages.
1158:WP:ACTRIAL
888:moderators
853:Farmbrough
430:References
45:Share this
40:Contribute
22:2012-01-02
1821:Subscribe
1369:Guy Macon
1309:Guy Macon
1305:WP:JARGON
1263:Guy Macon
903:More info
814:Guy Macon
795:Invertzoo
768:Doc James
586:notifiers
535:talk page
489:Interview
265:resources
171:In person
121:Skomorokh
31:Interview
1840:Category
1816:Newsroom
1811:Archives
1394:SMasters
1117:project.
778:contribs
703:WikiLove
650:effeiets
402:Wikilove
87:LinkedIn
67:Facebook
20: |
1720:Wikid77
1682:34,764
1653:34,516
1503:improve
1467:Wikid77
1220:Wikid77
1162:exactly
1140:WP:GOCE
1089:CD-Host
1079:Wikid77
1033:Wikid77
689:greenrd
77:Twitter
1703:month.
1679:35,443
1676:34,874
1673:36,429
1670:35,856
1667:36,270
1664:39,286
1661:38,991
1650:35,028
1647:34,767
1644:35,651
1641:35,501
1638:35,747
1635:36,930
1632:37,294
1540:(talk)
1172:芹ăăŚä¸ăă
1154:WT:IEP
1122:Howief
1010:(talk)
949:(talk)
654:anders
444:slides
398:Vector
97:Reddit
57:E-mail
1806:About
1588:April
1519:this?
1150:acted
782:email
744:have.
729:Colin
117:Tony1
16:<
1801:Home
1774:talk
1753:See
1658:2010
1629:2011
1603:July
1598:June
1583:Year
1534:Tony
1471:talk
1433:talk
1399:talk
1373:talk
1313:talk
1282:Jack
1267:talk
1189:talk
1156:and
1126:talk
1093:talk
1057:talk
1004:Tony
991:talk
967:talk
943:Tony
924:talk
916:. --
891:and
850:Rich
818:talk
799:talk
774:talk
714:Talk
693:talk
674:talk
588:and
442:The
238:will
226:and
190:now.
119:and
107:Digg
1623:Nov
1618:Oct
1613:Sep
1608:Aug
1593:May
1494:are
1417:Mar
1412:Res
1342:Mar
1337:Res
1286:Lee
1251:Mar
1246:Res
1211:⢠"
1077:. -
1031:. -
832:now
115:By
42:â
1842::
1793:}}
1787:{{
1776:)
1560:.)
1473:)
1455:â˘
1435:)
1401:)
1375:)
1315:)
1307:--
1269:)
1215:".
1203:â˘
1191:)
1175:)
1146:do
1128:)
1095:)
1059:)
993:)
969:)
959:--
926:)
856:,
820:)
801:)
784:)
780:¡
776:¡
711:|
705:.
695:)
671:â˘
1795:.
1772:(
1469:(
1431:(
1397:(
1371:(
1311:(
1265:(
1187:(
1169:(
1124:(
1091:(
1055:(
989:(
965:(
922:(
905:.
900:.
861:.
816:(
797:(
772:(
731:°
691:(
629:o
627:e
624:R
570:"
566:"
563:"
559:"
552:.
542:.
484:)
480:(
Text is available under the Creative Commons Attribution-ShareAlike License. Additional terms may apply.
