Jump to content

Wikipedia:WikiProject Guild of Copy Editors/Membership/News/2019 Annual Report

From Wikipedia, the free encyclopedia
The Guild of Copy Editors


Report by Reidgreg, Baffle gab1978, Jonesey95, Miniapolis, Tdslk and Twofingered Typist. | February 5, 2020.

Introduction

Welcome to the Guild of Copy Editors' (GOCE) annual report for 2019, a summary of membership activity, progress on the copy-editing backlog, and throughput of our Requests page.

Over 100 editors joined the Guild during the year, and more editors were active throughout the year than we have seen during the last three years. One member stepped-up as a first-time coordinator to help run the Guild behind the scenes, and another was named a coordinator emeritus.

Line graph showing a diminishing trend
The backlog hit new record lows!

The copy editing backlog was reduced by 43% overall, with a series of record lows reached on the four last month-long Drives of the year (see graph to right). Our Requests page was busy: copy editors there continued to attend to submissions in a timely manner, with an average wait time of under 19 days. Blitzes focused on articles from the backlog selected by theme, and on Requests when that queue was large.

With the diminishing backlog and 11% fewer submissions to the Requests page in 2019, there have been some explorations of other possible activities for the Guild. One pilot project began screening the list of upcoming Today's Featured Articles (TFA) to see which ones might benefit from copy editing, and met with mixed success. There have also been discussions of joint editing drives with other WikiProjects.

Thanks to everyone who has helped to make the Guild a friendly and collegial editing environment, and for insisting on clear, comprehensible, and grammatically correct writing.

Reidgreg, second lead coordinator for 2019.

Membership

[edit]

GOCE membership is open to all editors in good standing who are interested in copy editing. Please see our membership page for information about signing up and remember to add your username to our mailing list to receive Guild news. Guild membership is voluntary and informal, and editors do not have to be members of the Guild to participate in its activities. Many editors perform copy editing outside of our organized events but we have no means of tracking such efforts.

At the end of 2019 there were 1,805 pages populating the Guild's membership category, 1,312 names on the Guild's membership list, and 613 names on the mailing list. This is a growth of 108, 97, and 38, respectively, from the end of 2018.

Of the approximately 104 editors who are listed as having joined the Guild in 2019,

  • 21 completed a copy edit with a Guild-organized activity during 2019, including:
    • 14 who participated in one or more Drives
    • 13 who completed Requests
    • 8 who participated in one or more Blitzes
    • 4 who participated in all three activities (A big welcome to UnnamedUser, Bobbychan193, Robotxlabs, and Willbb234)

If you were one of these new editors, thank you, and welcome!

Coordinators

[edit]

The GOCE coordinators are responsible for coordinating Guild activities, maintaining our pages, organizing Drives and Blitzes, looking after the Requests page, and performing other maintenance tasks that keep the Guild running smoothly. They are chosen twice a year through consensus voting.

  • Miniapolis served as lead coordinator for the first half of 2019, assisted by returning coordinators Baffle gab1978, Jonesey95, Reidgreg and Tdslk.
  • For the second half of 2019, Reidgreg served as lead coordinator, assisted by Baffle gab1978, Miniapolis, Tdslk, and first-time coordinator Twofingered Typist. Jonesey95, who had served as a coordinator since July 2013, took a WikiSabbatical for this term.
  • In the December 2019 election, Jonesey95 was chosen as lead coordinator for the first half of 2020, with Baffle gab1978, Reidgreg, Tdslk, and Twofingered Typist assisting as coordinators. Having served as a coordinator for twelve terms, Miniapolis was awarded coordinator emeritus status (and encouraged to stick around!).
Term 2019 2020
January 1 – June 30 July 1 – December 31 January 1 – June 30
Election December 2018 June 2019 December 2019
Lead coordinator Miniapolis Reidgreg Jonesey95
Coordinators

Thanks to everyone who took the time to participate in the elections and thanks to our coordinators for their diligent, drama-free volunteer labor. Elections are scheduled for June and December each year; links to the election pages are posted in the Guild's message box (add it to your Watchlist). All Wikipedians in good standing are welcome to nominate themselves or others. Please consider helping out if you can.

Requests page

[edit]

Requests for copy editing are one of the Guild's main focal points. Our Requests page remained busy in 2019 but saw decreased use compared to the previous year, with 9% fewer editors submitting 11% fewer copy-editing requests.1 Consequently, there was a reduction in the number of requests that were processed and completed. Of these requests, 53% were for good article or featured article nominations or reviews. Any editor can accept requests but because articles posted to the Requests page are often being held to a higher standard of compliance with Wikipedia's guidelines, we prefer editors who are new to or inexperienced with copy editing to gain experience with articles in the backlog before handling requests.

Among the 616 requests received in 2019 were:2

  • 242 for GAN
  • 3 for GAR
  • 76 for FAC
  • 3 for FAR
  • 8 for FLC
  • 9 for PR
  • 10 for ACN
  • 55 for DYK
  • 1 for OTD
  • 16 for TFA
  • 1 each for tone, cleanup, and grammar
  • 20 declined
  • 7 withdrawn
  • 188 unspecified
Monthly breakdown
2019
Month
Requests Mean days to complete
Received[a] Processed[b] Completed
January 63 51 48 17
February 40 71 72 20
March 57 39 39 14
April 49 61 59 15
May 50 42 40 17
June 51 36 36 26
July 49 62 55 23
August 55 55 50 20
September 59 46 42 20
October 40 63 63 20
November 52 44 41 16
December 51 64 62 14
Average 51.3 52.8 50.6 18.46
  1. ^ "Requests received" between 1 January and 31 December 2019, inclusive.
  2. ^ "Requests processed" includes 36 requests made in 2018 and completed in 2019. Includes withdrawn and declined requests.

Note that because some requests are declined or withdrawn, over time there will be more requests received than completed.

Some analysis and comparison to previous years:

  • 607 requests were completed3 in 2019, compared to 668 in 2018, 571 in 2017, 665 in 2016, 537 in 2015, and 486 in 2014.
  • 616 requests were received in 2019, compared to 690 in 2018, 588 in 2017, 690 in 2016, 543 in 2015, and 489 in 2014.
  • The mean (and median and mode) completion time was 19 days. Previous averages were 15 days in 2018, 26 days in 2017, 17 days in 2016, 30 days in 2015, and 41 days in 2013.
  • Fewer than 19% of requests waited more than 26 days, which was the average wait time in 2017. Thirty-five requests waited longer than 30 days, and only one waited longer than 40 days.

Distribution of time-to-completion of requests (in days) in 2017 (blue), 2018 (orange), and 2019 (green).

At the beginning of 2019, there were 36 pending requests, a queue which grew steadily and peaked in the mid-50s in early February. The shortest month was the most productive for requests: by the end of February, the number of outstanding requests fell into the mid-teens. It stayed in the mid-20s to mid-30s for most of the year, with a week in the high 40s around the mid-year elections and again in late September, and a week in the teens during the October Blitz. There were just 18 pending requests at year's end.

Of the 616 requests submitted in 2019, 27 were declined or withdrawn. 39 copy editors completed the 589 remaining requests, which were submitted by 215 editors.

Notes
^1 Figures were arrived at by exporting tables from the 2018, 2017, 2016, 2015, 2014, and 2013 request archives into a spreadsheet for analysis, particularly for completion periods.
^2 Some requests were made for multiple reasons; these were listed in each applicable category.
^3 Figures for completed requests discount any that were declined or withdrawn. Figures for requests submitted, received, or addressed include requests that were ultimately declined or withdrawn.

A full list of 2019 requests is available at our 2019 archive page.

Archiving of requests

[edit]

GOCE coordinators and helpers had always performed manual archiving of processed entries on the Requests (REQ) page; it's not a difficult task, but it consumed volunteers' time and resources, and introduced occasional human errors. In August on REQ Talk (permalink), Bobbychan193 suggested automating this task. After a lengthy discussion (permalink), Zhuyifei1999 began coding a bot to meet our specifications. Zhuyifei1999 submitted a request for approval to the Bot Approvals Group (BAG) and in September, a successful trial run led to a further discussion and tweaks to improve the bot's edit summaries. YiFeiBot 2 began full-time archiving of the Requests page in November. Guild coordinator Baffle gab1978 asked both editors about the process of setting up the bot:

  • Baffle gab1978: What prompted you to think about automating the Requests page?
    • Bobbychan193: I joined the GOCE around July 2019 and found the GOCE Requests page soon after. I copy edited quite a few requests, but I only archived like one or two myself before I got too lazy and gave up. I noticed that Baffle gab1978 and Twofingered Typist were doing most of the archiving. I felt bad that they were archiving pretty much all of the completed requests (not just the ones I completed, but requests other copy editors completed as well). The existing talk page archiving bots were what really got me thinking... "Wouldn't it be nice if the GOCE had their own lowercase sigmabot III?" I decided to reach out to Zhuyifei1999, an admin on Wikimedia Commons who also happens to be a good offline friend of mine. I knew he had a lot of coding experience, so I asked him to help create a bot to archive requests.
  • Baffle gab1978: How do you write a Wikipedia bot script for a particular purpose?
    • Zhuyifei1999: To write a bot, well first you have a problem, then you evaluate the potential methods to solve it; should it be a bot, a tool, or a gadget, or what? I personally prefer bot (if the task is doable by bot) because the other two involves user interface design, while bot is pure programming logic and least amount of work for me.
    Then you think about the main pieces of code (shall I say component?) that are necessary for the bot. For most bots, it's often two: one "page generator" to get a list of pages to execute on, which could be recentchanges, a category, template transclusions, database scan, XML dump scan, or something else like that. Another component "treats" each page based on some logic, which could be doing a regex replacement, finding somewhere to add or remove text, or even invoking some other logic on another page.
    As for my GOCE archiving bot specifically, the "page generator" is [not needed]; it is only Wikipedia:WikiProject Guild of Copy Editors/Requests. The "treatment" is more complicated. I decided to do it in two parts, a parsing part and the archiving part, where the first part finds where all the sections are and their statuses, and the second part adds to the archive page and removes text from the requests page.
    Then you go through the process of discussing the plan, writing the bot, debugging with simulate / sandbox edits (run the code, either show diffs but don't actually save page, or run on some sandbox page), BRFA.
    Bingo!
  • Baffle gab1978: Were there any difficulties in setting up the archiving bot?
    • Zhuyifei1999: No, I've been writing MediaWiki bots since 2013. Getting close to 7 years ;). Now that I think about it, yeah I avoided things that could be complex, for example, parsing status templates in a more human way rather than simple regex matches, like <s>{{done}}</s> being a non-match rather than a match, but the benefit is probably too marginal to be worth it.
    • Bobbychan193: I think the overall process was pretty smooth. During the initial brainstorming process with GOCE coordinators and other community members, there were a few minor disagreements, but they were quickly resolved through constructive discussion. The bot approval process took a while, but was relatively straightforward. I wasn't really involved in the programming or technical side of the bot, so I can't really speak to how that aspect went. However, given that Zhuyifei1999 is a prolific coder, I imagine he probably did not encounter any major difficulties.
  • Baffle gab1978: Are you happy with YiFeiBot 2? Any further thoughts?
    • Bobbychan193: Yes, I am very happy with how the bot turned out. It is very predictable and methodical. Depending on the situation, it can correctly detect multiple acronyms and credit multiple copy editors. The bot has an intentional 24-hour delay to allow GOCE members to double-check requests before the bot archives them. It can also handle a variety of edge cases and special situations; it can even delete the month header after all of a given month's requests are done. Overall, I think Zhuyifei1999 did an outstanding job, and I'm really glad that the bot can handle mundane archiving while GOCE members can focus on fun things, like copy editing! :)
Thank you to both Bobbychan193 and Zhuyifei1999 for taking time to answer my questions, and of course for setting up the bot. CC-BY-SA: replies were copied with minimal changes from the respondents' talk pages (User talk:Bobbychan193 and User talk:Zhuyifei1999).

Backlog Elimination Drives

[edit]
Outstanding tags at the
end of each month
2018 2019
January 1,141 842
February 1,325 1,039
March 1,239 811
April 1,449 1,011
May 716 598
June 1,024 897
July 937 585
August 1,031 814
September 969 519
October 1,244 684
November 825 479
December 1,112 638

The Guild's main efforts toward reducing the number of articles tagged with {{copy edit}} were six backlog-elimination Drives held in alternating months, starting in January. At the beginning of the year, there were 1,112 articles in the backlog. Record lows were achieved in each of the final four Drives, reaching a backlog of 479 articles in November before rebounding to 638 articles at year's end.1 The backlog ended the year at seven months, the same as at the end of 2018, though it dropped as low as five months during the year.2

  • January Drive: In January, we removed copyedit tags from all of the articles tagged in our original target (oldest) months of June, July, and August 2018, and by 24 January, we ran out of articles. After adding September, we finished the month with 8 "old" articles remaining and 842 left in the backlog. GOCE copyeditors also completed 48 requests for copyedit in January. Of the 31 people who signed up for the Drive, 24 copyedited at least one article. Final results, including barnstars awarded, are available here.
  • March Drive: We removed copyedit tags from 182 of the articles tagged in our original target months October and November 2018, and the month finished with 64 target articles remaining from November and 811 in the backlog. GOCE copyeditors also completed 22 requests for copyedit in March; the month ended with 34 requests pending. Of the 32 people who signed up for this Drive, 24 copyedited at least one article. Final results, including barnstars awarded, are available here.
  • May Drive: Guild copy editors removed copyedit tags from 191 of the 192 articles tagged in our original target months of November and December 2018, and January 2019 was added on 22 May. We finished the month with 81 target articles remaining and a record low of 598 articles in the backlog. GOCE copyeditors also completed at least 24 requests for copyedit during the May Drive, and the month ended with 35 requests pending. Of the 26 people who signed up for this Drive, 21 copyedited at least one article. Final results, including barnstars awarded, are available here.
  • July Drive: The year's fourth backlog-elimination Drive was a great success, clearing all articles tagged in January and February 2019, and bringing the copy-editing backlog to a low of five months and a record low of 585 articles while also completing 48 requests. Of the 30 people who signed up, 29 copyedited at least one article, a participation level last matched in May 2015. Final results and awards are listed here.
  • September Drive: Of the 32 editors who signed up, 23 editors copy edited at least one article; they completed 39 requests and removed 138 articles from the backlog, clearing March 2019 and bringing the backlog to a record low of 519 articles. Final results, including barnstars awarded, are available here.
  • November Drive: For the final drive of 2019, of the 28 editors who signed up for this event, 20 editors completed at least one copy edit; they completed 29 requests and removed 133 articles from the backlog, clearing April 2019 and bringing the backlog to its fourth consecutive record low of 479 articles. Final results and awards are listed here.

Drive totals for the year: 65 editors removed 1,171 articles from the backlog and edited 210 articles from the Requests page for a total of 1,381 recorded copy edits of 2,742,745 words.

21 editors participated in two or more Drives, 17 editors participated in three or more Drives, and 11 editors participated in all six: thanks to UnnamedUser, Baffle gab1978, BlueMoonset, Dhtwiki, FiveFaintFootprints, Gog the Mild, Jonesey95, Miniapolis, Reidgreg, Tdslk, and Twofingered Typist! 127 barnstars were awarded for contributions to Drives.

Notes: About 300–400 articles are tagged with {{copyedit}} every month, so we need to process somewhere around 4,000 articles from the backlog each year just to break even. The number of articles formally listed as copy-edited during each Drive includes only articles for which editors took credit on the Drive page. Additional articles are removed from the backlog by editors who do a quick or minor copy edit and don't record it on the Drive page, edit during non-Drive months, remove a copyedit tag because it is no longer needed or not applicable, or nominate an article for deletion.

Blitzes

[edit]

The Guild organized six one-week Blitzes in 2019, in alternating months with the backlog drives.

  • February Blitz: This Blitz focused on outstanding requests. Of the 15 people who signed up, 13 copy edited at least one article. Participants claimed 32 copyedits, including 22 requests. Final results, including barnstars awarded, are available here.
  • April Blitz: The Blitz ran from 14 to 20 April, and the themes were sports and entertainment. Of the 15 people who signed up, 13 copy edited at least one article. Participants claimed 60 copyedits. Final results, including barnstars awarded, are available here.
  • June Blitz: From 16 to 22 June, we copy edited articles on the themes of nature and the environment along with requests. 12 participating editors completed 35 copy edits. Final results, including barnstars awarded, are available here.
  • August Blitz: From 18 to 24 August, we copy edited articles tagged in March 2019 and requests. 12 participating editors completed 26 copy edits during the Blitz, including 17 requests. Final results, including barnstars awarded, are available here.
  • October Blitz: This event ran from 13 to 19 October, with themes of science, technology, and transport articles tagged for copy edit, and requests. Sixteen editors helped remove 29 articles from the backlog and completed 25 requests.
  • December Blitz: We focused on articles tagged for copyedit in May and June 2019, and requests. Of the 23 editors who signed up, 17 copyedited at least one article—including 19 requests—and removed 42 articles from the backlog.

Blitz totals for the year: 30 editors completed 85 requests and 172 articles from the backlog for a total of 257 copy edits (603,451 words).

15 editors participated in two or more Blitzes, 11 editors participated in three or more, and seven editors participated in all six: thanks to Dhtwiki, FiveFaintFootprints, Gog the Mild, Jonesey95, Miniapolis, Reidgreg, and Twofingered Typist! These seven editors also participated in the Guild's six Drives, making them a perfect 12 for 12!

The Guild awarded 74 barnstars for contributions to Blitzes.

Annual Leaderboard

[edit]

The following table records the top contributions for combined activity in the Guild's Drives and Blitzes and on the Requests page throughout 2019. We appreciate the many hours of volunteer work represented here. Will you be on this table in 2020?

1 2 3 4 5
Total
Articles
Twofingered Typist (404.5) Miniapolis (191.5) Baffle gab1978 (145.5) Reidgreg (145) Bobbychan193 (109)
Total
Requests1
Twofingered Typist (242.5) Baffle gab1978 (122) Miniapolis (52.5) Gog the Mild (32) Bobbychan193 (29)
Blitz and
Drive Totals2
Twofingered Typist (270) Reidgreg (144) Miniapolis (143) Bobbychan193 (105) Tdslk (97)
Oldest
Articles2,3
Twofingered Typist (135) Reidgreg (129) Miniapolis (124) Ben79487 (70) Logophile59 (62)
Total
Words2
Twofingered Typist (791,603) Miniapolis (339,514) Reidgreg (332,963) Baffle gab1978 (195,221) Logophile59 (172,484)
5k+
Articles2,4
Twofingered Typist (54) Reidgreg (33) Gog the Mild (16) Dhtwiki (11) Logophile59 (10)
Longest
Article2
Reidgreg (22,687) Twofingered Typist (13,607) Gog the Mild (12,970) FiveFaintFootprints (12,292) Dhtwiki (11,802)
Leaderboard
Awards5
Twofingered Typist (30)
Reidgreg (30)
Miniapolis (24) Logophile59 (13) Ben79487 (11) Gog the Mild (10)
Gold
Awards5
Twofingered Typist (19) Reidgreg (5) Logophile59 (3) Bobbychan193 (2)
Csgir (2)
Gog the Mild (1)
Miniapolis (1)
Notes
^1 Requests completed in 2019, counting a half-point when credit was split.
^2 Includes Blitzes and Drives only.
^3 Articles from the oldest or target backlog month(s) of each Drive and of the February, August and December Blitzes.
^4 Number of articles of 5,000 words or more. An article of 10k+ words counts as 2, of 15k+ as 3, etc.
^5 Awards from Drive leaderboards only.

Plans for 2020

[edit]

By Jonesey95, lead coordinator for the first half of 2020.

As I write this in early February 2020, we have just completed the January 2020 Drive. At the end of that Drive, we had reached a new record low backlog count of just 323 articles. At the beginning of our first Drive, in May 2010, we had exactly 8,000 more articles in our backlog (8,323). When I joined the Guild in January 2013, just seven short years and a few gray hairs ago, the backlog was at 2,922 articles. Since that time, I have had the pleasure of participating in every Drive and Blitz, along with serving you as a Guild coordinator for over six years. To be clear, I don't take any credit for the massive numerical reduction – you won't find my name in the impressive leaderboard above – as I prefer to serve by doing the sweeping and tidying. The editors named above, along with some contributors who prefer to keep a lower profile, get the lion's share of the credit.

As shown in the graph at the top of this page, we have made slow and steady progress over the last ten years, to the point where our backlog total is about the same as the number of new articles that are tagged for copy editing every month. What does this mean? Are we done? Can we sit out on the front porch with a cool glass of lemonade, glowing with the satisfaction of having completed the monumental task that we set for ourselves a decade ago? I have a few thoughts.

One is about how we got to this place. We got here by establishing and maintaining a culture of hard work, collegiality, and relatively drama-free interaction, something that is not true of every place on the internet. I am proud of our Guild, not for the numbers, because numbers can be gamed, but for our dedication to quality and teamwork. We are all volunteers here, and we are mostly self-policing, so I am sure that some questionable copy edits slip through, but it has been my experience in checking other copy editors' work that our edits almost always improve article prose substantially, even when they leave behind – or, inevitably, introduce – a few errors.

Obviously, we are not done. See all those red links? There are a lot more articles to be written, which means a lot more copy editing is coming our way. Because this is the encyclopedia anyone can edit, editors have a wide variety of competence, skill, and experience. Our work takes the contributions of dedicated contributors whose language skills may not be top-notch and improves those contributions so that knowledge can be delivered to the world. It is no small thing that we do here.

The reduction of our backlog to essentially one month of articles means that we may want to discuss ways to offer our skills to other groups of editors who want to improve subsets of articles, or actively seek out valuable articles that need copy editing but have not been tagged. I encourage all interested editors to continue these thoughts in a conversation at Wikipedia talk:WikiProject Guild of Copy Editors. Happy editing to you all, and may 2020 be a great year for you.