Jump to content

Wikipedia talk:WikiProject Good articles/Newsletter

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia

Leave comments on the WikiProject Good Articles newsletters here.

Comments on the first (November 2007) issue

[edit]

Comments

[edit]

Good job! I think perhaps a lighter shade of green might be in order. The current issue was a bit hard to read. The List of colors might be a good place to find other options. Also, it's now defunct, but the old WikiProject Wine newsletter used to be really great. One feature I remember liking was the interview with a member of the project. Maybe not quite as broad and applicable for our topic, but it might be a good way for people to hear different approaches to reviewing. Again, good work - VanTucky Talk 02:01, 1 November 2007 (UTC)[reply]

Looks like LaraLove has fixed this issue. Thanks. It is easier to read. I was originally basing the color scheme on the formatting used on the WP:WGA page. Maybe we should switch those colors, too? Dr. Cash 02:36, 1 November 2007 (UTC)[reply]
I never much cared for the WGA color scheme. I changed the newsletter to the colors I used for the Sweeps page. I'll update the others tomorrow. LaraLove 05:19, 1 November 2007 (UTC)[reply]
Yes, please do! The lighter colours are much better and more accessible. Geometry guy 22:01, 2 November 2007 (UTC)[reply]
I've now updated the project pages to use lighter colours. I used a different, greener colour, because although LaraLove's choice is very nice, it is rather similar to the FA light blue. The colour can be changed at Wikipedia:WikiProject Good articles/Shell, but if others like the greener choice ;-) then it could be used for the newsletter too. Geometry guy 19:58, 11 November 2007 (UTC)[reply]
I've now tweaked the colours on this page too. Geometry guy 16:54, 17 November 2007 (UTC)[reply]
I'm reading the same newsletter for an uncharacteristic (for WP) second time and am enjoying myself. The presentation is great and there are intelligently placed links to many key aspects and current projects available in the first issue itself. Must have taken some work to write something like this and still make it interesting. I'm hooked !
Havelock the Dane Talk 12:14, 1 November 2007 (UTC)[reply]
I agree. It reads well. I like the little section that showcases the reviewer of the month, giving them a short bio. Nice touch. A section to list some important discussions going on throughout the project, linking to them, would be helpful. Particularly those discussions that require participation to reach consensus. Perhaps, also, some mention of the Collaboration of the Week, as it seems to be dying. Lastly, something for little announcements or whatever, like the current, on-going issue with errors in the AH template caused by GA participants incorrectly building/updating the template. LaraLove 16:58, 1 November 2007 (UTC)[reply]
I think highlighting the correct use of the {{ArticleHistory}} template is a good idea. Any volunteers to write a short guide for the next newsletter? Dr. Cash 06:20, 2 November 2007 (UTC)[reply]
I like the first issue, and I'd be happy to write a short guide on {{ArticleHistory}}, but really I'm not sure if we should be encouraging reviewers to use {{ArticleHistory}}: there are so many possible mistakes, and I really think it would be better if this template were updated automatically. As pointed out by SandyGeorgia, reviewers even make mistakes with the {{GA}}, {{DelistedGA}} and {{FailedGA}} templates. Updating ArticleHistory is mundane, time-consuming and error-prone. Humans are not good at any of these three things: computers have no problem with any of them. Geometry guy 21:45, 3 November 2007 (UTC)[reply]
Ah, yea, but if a guide was written, as Dr. Cash suggests, that could remedy these issues that are occurring with the AH template. If everyone would check for errors once the complete their changes, SandyGeorgia and Gimmetrow would have less worries. LaraLove 05:18, 5 November 2007 (UTC)[reply]
I admire your optimism, but it falls down because no one reads the instructions. For example, someone has just now closed a GAR discussion without archiving it, despite the fact that there are clear instructions. Most article history errors would not occur if everyone read the template documentation, but they don't. A guide in the newsletter would only be received by the small proportion of GA reviewers who subscribe, and these reviewers are probably the ones who make the fewest errors anyway. Geometry guy 19:19, 6 November 2007 (UTC)[reply]
You may be surprised. It's actually often the editors that use it the most. It's just a matter of volume. They use it more, thus a greater chance for mistakes. Everyone really just needs to check for errors before saving. LaraLove 19:29, 6 November 2007 (UTC)[reply]

Idea, list for GA participants

[edit]

See User:Anonymous Dissident/List of Wikipedians by number of DYKs. Anyone think it would be an interesting idea to create something like this for amount of successful GA noms or to keep track of number of GA reviews someone has done? It could then be put in the newsletter in a small subsection. Curt Wilhelm VonSavage 03:27, 1 November 2007 (UTC).[reply]

That list has made DYK a less happy place to be, I wouldn't suggest using that system here, it promotes competition and an attitude that "I am better than you, look at all my GAs." IvoShandor 07:32, 1 November 2007 (UTC) (This comment was mine, I didn't realize I wasn't logged in IvoShandor 15:29, 2 November 2007 (UTC)).[reply]
I think it would do some good but people may use the "I am better than you, look at all my GAs" tone at some point. That said, different people nominate different articles for GA review, and GA contributors are usually more well behaved than POV pushers. I can't see how that situation would easily rise in the first place.
The project can always do with recognizing the top contributors and I, for one, feel that it might even encourage some editors to make more GA level contribs instead of scattered contribs here and there. Havelock the Dane Talk 07:57, 1 November 2007 (UTC)[reply]
I think this is a good idea. Not sure the easiest way to make it work, but showcases such as these promote participation, which we desperately need. LaraLove 16:48, 1 November 2007 (UTC)[reply]
  • If you folks are intent on doing this, you can just set an arbitrary cut off, like five noms, this would eliminate all of the users who have contributed one or two noms and never come back. I suggest making the list and notifying the editors that are on it (notification could probably be done with an existing bot). That way any users, like myself, who wish to opt out of such a hierarchy type list have an opportunity to do so. IvoShandor 15:29, 2 November 2007 (UTC)[reply]

Automating Good Article contribution recognition

[edit]

I wrote, and run, user:Rick Bot which (among other things) maintains WP:WBFAN. The bot essentially recreates WP:WBFAN every time I run it based on the content of the by-year lists, like Wikipedia:Featured articles nominated in 2007. These are in turn maintained by the bot, but only semi-automatically, by parsing the nominated article name and nominator from the monthly FAC archive files (e.g. Wikipedia:Featured article candidates/Featured log/October 2007). The bot gets the article name right 100% of the time, but the nominator is identified correctly only 80-90% so I run the bot manually assisted (it suggests a nominator and I confirm its choice or enter a correction). To make the counts roughly right, I've created by-year lists back to when WP:FA first started by parsing the historical FAC archives and manually augmenting these so that every article in WP:FA (and WP:FFA) has an entry (the bot checks every time it runs whether all FA and FFA articles are accounted for in WBFAN).

I'm working on a tool to automate WP:DYK recognition. At this point, it can create lists like Wikipedia:Recent additions 145/History based on the DYK archives but it's not 100% accurate (so takes manual proofing) and isn't hooked into any automatic update mechanism. The DYK tool has been significantly more difficult than the FA tool since it's based on parsing historical versions of the DYK nomination page (T:TDYK) which are nowhere near as regular as the FAC archives.

To do something similar for GA would require deciding what to recognize and figuring out from where and how to mine this information. The what to recognize is not as clear to me as for FAC since it seems nominating something for GA does not carry the same sort of responsibility as a FAC nomination. Should we recognize all nominations or only successful noms? All reviews so a quick-fail, fail, and pass all count the same, or only "pass" reviews (recognizing only passes would seem to encourage reviewers to pass anything while recognizing any review might encourage quick-fails as a low effort way to get a review credit). Looking at the current GA process (by the way, was it ever different?), I think nominations could be parsed from historical versions of WP:GAN. Since these are at least currently template-based, rather than free-format, the nominator and nominated articles could be identified with pretty close to 100% accuracy. Reviews seem to be trickier. Is removing a nom request sufficient indication that a user performed a review? Do reviewers always update the page's talk page? I think the bottom line is that there are significant issues regarding any GA recognition mechanism (bot-assisted or not) that would need to get worked out.

BTW - there are also Wikipedia:List of Wikipedians by featured list nominations and Wikipedia:List of Wikipedians by featured portal nominations (as far as I know both manually maintained), but nothing similar for Wikipedia:Featured picture candidates. -- Rick Block (talk) 17:43, 4 November 2007 (UTC)[reply]

Discussion, Responses to above

My feeling:

  1. Recognize those that nominate and then successfully have GAs that pass.
  2. Start with those with at least 5 successfully nominated articles passed as GAs.
Problem I predicted to encounter: Some GA are promoted prior to the guidelines established in mid-2006. They are not removed yet as Sweeps is still under progress. How can you address this issue when all it needs is an editor pasting a template and that article becomes a GA? OhanaUnitedTalk page 18:31, 6 November 2007 (UTC)[reply]
Well, we could start by adding those on the list with five or more successfully nominated and listed GAs under the newer criterion... Curt Wilhelm VonSavage 22:34, 6 November 2007 (UTC)[reply]
So, parse the history of WP:GAN looking for noms (addition of a line with {{la|...}} or {{article|...}}? I'm not sure when this started, but it isn't consistently like this back to the beginning of GAN), check if the article is a GA now (based on WP:GA or Category:Wikipedia good articles, or what?), count these noms, sort by decreasing number of noms with a cutoff at 5. This would miss articles that were GA at one point but later demoted and would be expensive to re-run (requires re-parsing the whole history of GAN). An approach more similar to the FA approach would be to parse the history of GAN producing one or more tables of nominations (similar to Wikipedia:Featured articles nominated in 2007), then write a second tool to produce the "credit table" (like WBFAN) from the pre-parsed nomination history tables. The code I have that does WBFAN could be easily modified to do the second step. The first step of parsing GAN's history is similar in some ways to what I've been doing with the DYK history and allows history to be rewritten (or corrected) if the tool messes up. To avoid re-parsing the entire history of GAN every time this is re-rerun requires a way to identify successful noms (I assume there really isn't a maximum "linger" time in GAN). Perhaps perversely, the success/failure of a nom is clear only when it's removed from GAN which in general isn't done by the same person who nominated the article (FAs don't have this problem, since the FA director adds all successful noms to an archive file when the articles are promoted). I think this may mean all noms have to be recorded in the nom history tables (when the article is nominated) and then a pass/fail indicator needs to show up in these tables (later, when it's clear) as well. Another option would be to keep the pending nominations in the tables and delete failed noms (but keep, and mark, passes).
I won't have a significant amount of time to spend on this for a while. You might get better response at WP:BOTREQ. -- Rick Block (talk) 01:18, 7 November 2007 (UTC)[reply]
I like your idea that starts with: "An approach more similar to the FA approach would be to parse the history of GAN ..." - Modeling it after Wikipedia:List of Wikipedians by featured article nominations with the addition of the cutoff for 5 GAs, seems like the most straightforward way to go. Curt Wilhelm VonSavage 01:23, 7 November 2007 (UTC)[reply]

Idea: Common Manual Style of Issues

[edit]

Maybe we can have a section for reviews about common MoS that are often overlooked in GA Reviews? Some things off the top of my mind:

  1. No birthplaces in the lifespan per WP:DATE
  2. No redlinks in infoboxes
  3. All one-two sentences should be either expanded or merged with surrounding paragraphs, as they cannot stand alone.

Etc. etc. Useful for new reviewers and nominators alike. Would be an ice collaborative effort.

Also maybe a section settling the "references in the lead" issues once and for all. Just my thoughts. Cheers, CP 22:13, 6 November 2007 (UTC)[reply]


It seems to me that it would be more useful to have one talk page at Wikipedia:WikiProject Good articles/Newsletter rather than one talk page per issue. If there are no objections I will move this talk page there and tweak it for such a purpose. (Thanks to me, this would involve fixing a lot of double redirects from User talk pages, but I'm willing to do it.) Geometry guy 01:05, 12 November 2007 (UTC)[reply]

Done. Of course the redirect was not an issue, as it was the middle redirect that needed to be fixed. Geometry guy 17:39, 12 November 2007 (UTC)[reply]

GA Sweeps

[edit]

If interested, the current completion rate of the GA sweeps can be included. I added a progress table to the top of the running total of swept articles. It may not be extremely accurate, but still gives an acceptable estimate of the current progress. If anyone wants to tinker with the table more, please go ahead. At our current rate, the sweeps is going to take more than a year to complete. --Nehrams2020 08:42, 12 November 2007 (UTC)[reply]

December Newsletter

[edit]

Looks great, guys. Very good work! LaraLove 01:36, 4 December 2007 (UTC)[reply]

FA sweeps

[edit]

Hi, It's written in the newsletter that there is a "GA sweeps work" to review GA articles. I believe similar issue should be done for FA articles. I proposed it in July but they didn't accept it and I didn't have enough time to follow the issue.(here). Now you have a good experience and you can persuade them to do similar work.--Seyyed(t-c) 03:48, 4 December 2007 (UTC)[reply]

I was reading SandyGeorgia's comment and couldn't stop myself laughing. She said "Soon, there will be 1500 FAs; reviewing each of them routinely isn't feasible". Let's see, how many GAs are we sweeping? 2808 GAs. OhanaUnitedTalk page 14:15, 4 December 2007 (UTC)[reply]
The GA and FA systems are different: a single reviewer vs multiple reviewer model. The FA model can't easily handle sweeping so many article systematically. It is even a challenge at GA. This probably means that experience of GA sweeps will not transfer, and in any case GA experience is unlikely to carry much weight at FA. With FA, the main issue is pre-2006 articles, which are often rather weak, but I think there has been some informal sweeping of these articles by bringing them to FAR. Rather than a sweeps process, FAR might benefit from a list of articles promoted by date, where editors can sign and date the article each time they check it and think it is still okay. That might make it easier to find the articles which are not up to standard. Geometry guy 19:43, 4 December 2007 (UTC)[reply]

Ideas for January 2008 issue

[edit]

We're looking for new ideas and contributors for January 2008 issue. Please post your suggestions here (until a newsroom page is created). OhanaUnitedTalk page 23:27, 4 December 2007 (UTC)[reply]

I mentioned at WT:GA that I'm not completely convinced that "Recently listed good articles" is a particularly sensible entry in the newsletter, as it (somewhat arbitrarily) showcases the articles which happen to get listed towards the end of the month. Does anyone have a better idea? Geometry guy 20:11, 11 December 2007 (UTC)[reply]

highlight GAs going for FAC?

[edit]

I believe the GA statistics section should highlight GAs that are currently also WP:FACs to encourage people to try to make the leap from GA to FA. I'm not sure if it should be in a numerical form (ie. 7 GAs are currently at FAC) or if they should explicitly name the articles. -Malkinann (talk) 23:27, 30 October 2008 (UTC)[reply]

Delivery?

[edit]

I just saw that this has become active again. Is there any chance of someone sorting out talk page delivery? —Tom Morris (talk) 23:58, 29 December 2012 (UTC)[reply]