User talk:WereSpielChequers
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
|
Home | Bling | Content | Userboxen | Editcount | Talk | Guestbook |
- Welcome to my talk page. If you just want to make a short comment why not put it in my guestbook. If you want to add something to one of the existing topics go ahead, Or click here to start a new topic.
Poop patrol[edit]
Thanks for the message about poop patrol. I had a look at it and been thinking, but I've not had a chance to do any work on it yet. Sorry for not contacting you sooner. I'll try and find some time to have another look. Edward (talk)
- I've had a go at implementing my own version of poop patrol. My prototype is available here: https://edwardbetts.com/page_patrol
- a few issues with the prototype:
- Some pages don't become safe articles despite multiple clicks, it is almost as if the mark as safe article button is disabled for pages such as shoegazing, or possibly if it is a safe page for stared it can't also be so for staring.
- The fifty exceptions limit seems to be on fifty examples of the word being found rather than fifty examples after screening out safe pages and safe phrases. This makes it a bit of a faff going through a report such as staring which has over 20 screens. not subdividing it might be the best solution here.
- Sometimes the same exception keeps recurring, almost as if the next batch of fifty is calculated on a sometimes differing index.
- @Edward: An odd one. When I clicked on Who's That Boy? I wound up at the article that Who's That Boy redirects to, as if the questionmark had been displayed but ignored.
- Would it be possible to import the existing lists of safe pages into the new system? I noticed with causalities that many of the examples are on the existing safe page list. This is particularly helpful for ones like pubic where it can take a bit of research to add each example to the safe page list. ϢereSpielChequers 08:16, 6 July 2014 (UTC)
- Currently there are two numbers, one on the list before you select a report and one in the report. For example Strecher appears as Strecher (8) with a result that says 7 results. I think the 8 comes from the first time it was run with your software and 7 is the number of examples screened out by safe words and safe phrases. Both are interesting, though it would help if the 7 was subdivided into safe pages and safe phrases. But more important is the number of articles still to check, if as in the case of strecher this is zero then there is no need for me to check that report.
- Smelly and Poop can come off the list, I started the thing with poop and cleared up lots of old vandalism, but I think one of the edit filters is stopping the new stuff. However there are a number of others worth adding. Idealy the current list would be as per User:DeltaQuadBot/Job requests, though I suppose I should move that. ϢereSpielChequers 06:01, 5 July 2014 (UTC)
I've dome some work on Page patrol. The latest prototype is in the same place: https://edwardbetts.com/page_patrol
There are now pages that list the safe phrases and safe articles for each term. It is possible to add new safe phrases.
The fifty examples after screening problem is due to the way that the mediawiki search works. I can only get up to 50 results at a time. I get 50 results, then filter them and display them. The solution is to use more Javascript. I should show the first set of filtered results, then in the background request more results from the search engine, filter and add them to the current patrol page.
I have a look at adding the existing lists of safe pages and at adding the terms from User:DeltaQuadBot/Job requests.
The numbers are all off, maybe of the hit counts are returned from the search engine before filtering. To fix this I need to download all the search results, filter and safe to the database. This would take more time, I didn't make it a priority.
I'll take a look at the question mark problem. I need to URL encode the article titles when generating links to Wikipedia.
Edward (talk) 13:43, 7 February 2016 (UTC)
- I've been working on fixing some of these problems. The new plan is to cache search queries in the database instead of trying to use the live search engine.
- The article titles from the search results are saved to the database with safe articles filtered. The search engine is used to find articles that contain safe phrases and filtered them as well.
- The API will be used to download and save the full text of each article. Snippets showing patrol terms within article text will be generated and stored so the patrol pages can be rendered quickly. The patrol pages will include a pager that works correctly, showing a fixed number of articles per page.
- A user can either identify an article as safe by clicking the 'safe article' button, or follow the link to the article and fix the vandalism. When the user clicks 'safe article' it will trigger Javascript add to the list of safe articles in the database. The user will see the article removed from the list without needing to reload the page.
- There will also be a 'vandalism fixed' button. This will trigger the article text to be downloaded again and checked to ensure the problem has been fixed. The article will be removed from the patrol page using Javascript without a reload.
- Using this technique means the search result numbers on the index page and patrol pages will be correct. Edward (talk) 08:52, 9 February 2016 (UTC)
- Hi Edward, I think what is happening is that the search includes the safe phrases, but the fifty are then filtered against the safe pages. So searches with a lot of safe phrases work well but ones with lots of safe pages can involve quite a few tedious trawls through empty batches before you find anything. Also there seems to only be one safe page list for all queries, and the system errors if you need to put the same page as safe in more than one query. ϢereSpielChequers 09:17, 9 February 2016 (UTC)
- Hi Edward, I noticed you were back editing today. Hope you can switch your server back on soon. BTW any chance of tweaking this so that it looks at draftspace as well as mainspace? ϢereSpielChequers 07:41, 4 July 2020 (UTC)
- @WereSpielChequers: Server is back. Let me know if there are any problems. I'll investigate adding draftspace. Edward (talk) 22:06, 4 July 2020 (UTC)
And in this corner...[edit]
I hope you don't think I'm beating up your idea. I actually find it quite interesting with great potential, it just raises a lot of questions of how to manage since it is so different than anything we have done before. I'm peppering you with a lot of questions, but others will be asking the same in time anyway. I'm curious as to if the Foundation would interfere and say no to such an idea for some reason, ie: "it is against the open nature". Anyway, it is very thought provoking and I just wanted to make sure you knew the questions were because it intrigued me, not because I was against it. Dennis Brown | 2¢ | WER 15:19, 7 July 2014 (UTC)
not curly[edit]
Saw that on my watchlist and wondered what the tarnation does that mean? Looked at the diff and laughed. Well done. It is nice to laugh for a change amongst all the vandalism entries on the watchlist. Bgwhite (talk) 07:24, 29 July 2014 (UTC)
- Thanks, it took a while to get that one under control, and there are still a few music lyrics where I'm really not sure whether to put them on my safe page list or not. But I reckon to make two or three typo fixing not curly edits per week. Curiously I'm seeing less vandalism nowadays, especially with that particular search, I suspect someone has set an edit filter to stop edits that include removing the first l from "public schools" as I used to get one or two of that sort of vandalism every week. I also stopped patrolling for "poop", lots of vandalism when I first went through all the articles that contained that word, but nowadays not enough to be worth manually checking for. ϢereSpielChequers 11:45, 29 July 2014 (UTC)
- And in the case of Empire Air Day, does show I wasn't copy+pasting form the sources
!--Shirt58 (talk) 10:48, 6 August 2014 (UTC)
- And in the case of Empire Air Day, does show I wasn't copy+pasting form the sources
Editor retention[edit]
Any watchers interested in looking at User:WereSpielChequers/2010-2014 Editor retention test? ϢereSpielChequers 17:03, 25 August 2014 (UTC)
List[edit]
https://meta.wikimedia.org/wiki/User:Rich_Farmbrough/British_places_with_coord_and_no_pic
Note: that this may include non-place entities, e.g. people in the category tree that have burial coordinates. But then there may well be a nice photo of their tomb/grave/crypt.
If you let me know any items without coords or with images, improvement may be possible.
All the best: Rich Farmbrough, 19:47, 5 February 2015 (UTC).
- Thanks, that's brilliant. I spotted that where the infobox and image are being delivered by a template your program doesn't realise there is an image see Lochaber Narrow Gauge Railway. Also Zeta Island, Bermuda a Dutch volcano Zuidwal volcano a pass in South Georgia Zigzag Pass and a museum in Abu Dhabi Zayed National Museum are not quite in the UK, though the volcano is close. ϢereSpielChequers 22:39, 5 February 2015 (UTC)
- There's not a lot I can easily do about abuses like Lochaber Narrow Gauge Railway - content is not supposed to be abstracted away like that, though I understand the reasoning if there are many uses of the template - in this case there is only one. Conversely, of course, if the template was used on many station pages, then a generic image would not suffice.
- Zayed National Museum was in the category "British Museum"
- Zeta Island was included because Category:Islands of British Overseas Territories was a sub-cat of "Islands of the United Kingdom"
- Zuidwal volcano was included because Category:North Sea is a sub-cat of "Bodies of water of the United Kingdom" (among others) (not changed yet...)
- Zig-Zag Pass was probably included because of a similar British Overseas Territories miscat, possibly Category:British Overseas Territories and Crown Dependencies being a sub-cat of "Government of the United Kingdom"
- All the best: Rich Farmbrough, 18:02, 8 March 2015 (UTC).
- There's not a lot I can easily do about abuses like Lochaber Narrow Gauge Railway - content is not supposed to be abstracted away like that, though I understand the reasoning if there are many uses of the template - in this case there is only one. Conversely, of course, if the template was used on many station pages, then a generic image would not suffice.
Going off the boil?[edit]
Hey. I took up your invitation. Some comments:
- Theories are great; and I have another one. But, without verification, they are pure speculation and nothing more.
- You mentioned about typos being less common, and are often fixed quickly. Of note, there's an abuse filter that picks up on canned edit summaries including "fixed typo" (see this). Thus, in recent changes log a well meaning user fixing a typo will come under increased scrutiny, with a higher chance of having their edits reversed.
- Along with typos, you mentioned vandal fighting as a gateway task. It would be interesting to conduct a comprehensive analysis to find out what productive new editors of this generation are doing. If we knew that, we could focus our resources on maximizing opportunities for new editors to become connected with now-relevant gateway tasks.
- There's a saying I learned a long time ago. The factory of the future will have two employees; A human and a dog. The dog's job is to keep the human away from the machines. The human's job is to feed the dog. Wikipedia is heading towards this paradigm. Increased automation, decreased editorship, fewer things that need to be worked on.
- You commented on increased mobile surfing. Very true. Desktops are dying. It will take time, but desktops will become artifacts as obsolete as typewriters. [1] Portable devices make horrible editing devices and always will. Nothing yet can replace the efficacy of a keyboard. Until that happens, mobile platforms will be permanently handicapped. The world record for texting is 25 words in 18.19 seconds [2]. This works out to 82 words per minute, if it could be sustained. I type faster than that on a keyboard, and I'm just an average guy on a computer. The world record top speed for sustained typing is 150 words per minute (and the record holder sustained that for better part of an hour). Whoever comes up with a better interface for mobile devices is going to be a billionaire. I digress. The point is the mobile platforms mean editing will permanently decline.
- All the good stuff is already done. We have nearly five million articles now. People like to feel good about their contributions. If all you're doing is tweaking a few things here and there, it's not very rewarding. If you're starting the articles on the Nile, God, Superbowl X, etc. you feel like you're contributing. If instead any article you can think to create has already been created, you feel like you can't contribute. I had a personal case of this; there's a place I've visited a couple of times that is on the national historic register. It's an amazing place! It is frequently visited and toured. Yet, it did not have an article on it here. Rather bizarre. So, I had a not-so-devious plan. I was going to create an article from absolute nothing to featured status with just one substantive edit. I even bought a book on the place in support of this. I took my own photographs, took notes on the tour to verify in other sources, found a multitude of solid sources, and began writing. Just as I began writing, someone created the article. I was sorely disappointed. I haven't even touched the article. Pout :) But, I hope you get my feeling on this.
- It isn't just that we need a wysiwyg editor. Trying to get people to learn new systems is very difficult. We have to create something that mimics how things are done in MS-Word and other significant desktop editing platforms. The markup language we use, while intuitive for me after years of use, is horribly archaic and difficult to learn. It is as archaic as TeX. Case point for me; at a job I had I installed a wiki on our Intranet, to allow people to easily write things up and share them with everyone. There were a couple of geeks who took to it like ducks to water, but the organization as a whole never did anything of consequence with it. We later moved to SharePoint, which allows people to edit things in MS-Word, and it took off.
- On snark; I've long held that the WP:NPA policy is null and void. I've been directly attacked by a whole host of editors, admins and even a bureaucrat. Nothing every happened to them. Personally, I think the policy should be deleted. It creates more controversy and disappointment than it solves. Witness that we did away with Wikipedia:Requests for comment/User conduct. Doing away with WP:NPA isn't much of a stretch of the imagination. It is very rare that people who are deliberately insulting ever have more than a soft warning tossed their way. People don't like confrontation. Case example for me; I know of an administrator who is extremely rude, and frequently bullies people who disagree with him. He doesn't directly insult anyone, which is how he gets away with it. Plenty of people have complained about him, but nothing is done. I've committed to avoiding him at all costs, knowing that any complaint against him will fall on deaf ears and it will only lead to more grief. As a result, there's a section of the project he frequents that I have largely avoided. Result; less things for me to work on that I want to work on.
- Simplifying our policies and guidelines isn't going to happen. They are ever expanding. I concur with the reality that this generates difficulty for new users. The learning curve is enormous and fraught with a huge number of potholes where you get negative feedback telling you how badly you messed things up, and how you would have done better if only you'd read War & Peace first.
- As you know, there's been quite a bit of change in the stratification of editors. Unbundling has contributed, but so has 'upbundling'. Excuse my neologism; I mean rights normally available to brand new editors are no longer available. Given that we have bots that quickly deal with vandalism, given that much of our content comes from IP editors, partially shutting them out of the process is antithetical to our purposes. I refer in large part to the pending changes and autoconfirmed rights. They sound good on paper, but I would speculate they have significantly impacted gateway tasks for new editors. I also find the template editor right insulting. I should not have to plead my case before someone to get the right. The bureaucracy for it is already large (Wikipedia:Template_editor#Guidelines_for_granting,Wikipedia:Requests for permissions/Template editor) and this will increase. We want to reduce the monstrosity that is WP:RFA yet create more crap like Wikipedia:Requests for permissions. Unbundling achieves only a stratification of bureaucracy; it does not eliminate it.
So, there are my comments. Enjoy. --Hammersoft (talk) 14:11, 14 July 2015 (UTC)
- PS: I took so long to write this I got "Sorry! We could not process your edit due to a loss of session data. Please try saving your changes again. If it still does not work, try logging out and logging back in. "
- Thanks, I don't agree that all the good stuff is started as I know of subjects where that is far from true. But I'd concede that it is true for many of our existing and former editors, so I've added User:WereSpielChequers/Going_off_the_boil?#All_the_Good_stuff_has_already_been_done. I'm familiar with the term upbundling, though I use it more for various proposals to shift certain admin rights just to crats. I agree that there have been changes that are disconcerting for newbies, the classic being that if you cite your first edits you are going to have to complete a capcha, if you don't you will get reverted and maybe bitten by others. I became autoconfirmed in 2007, and in those days you could add uncited edits without getting bitten. I was autoconfirmed long before my first cited edit so never had the capcha hassle, but we need a better less bitey way to handle that. ϢereSpielChequers 15:12, 14 July 2015 (UTC)
- On the 'good stuff'; I saw a study a while back that showed disparity in areas of Wikipedia. The types of people attracted to Wikipedia will of course have subject areas where they are plentiful and subject areas where they are not. The areas where we do not have a lot of contributors certainly have lots of 'good stuff' to start. But, we don't attract the types of editors that fill those voids. The types we do attract find little in the way of voids where they can contribute. As to the captcha problem...holy crap! I had no idea that was part of the editing process for newbies now. That's sick! On rights... I eschew having any special rights. I shouldn't have to jump through a bureaucratic nightmare to edit a template after I've been here for many years with tens of thousands of edits and one inappropriate block that was placed by an admin who has (for an unrelated event) since been relieved for cause. The idea that I could be here so long, have done so much work here and still not be trusted without jumping through a bunch of hoops is absurd to me. I am an editor. Nothing else. If you can't trust me, you might as well shut down the project. --Hammersoft (talk) 15:31, 14 July 2015 (UTC)
- Admin, Crat Oversight and checkuser are the only rights that involve jumping through hoops. If you want Reviewer and Rollbacker you have but to ask. As for admin, RFA may be crazy but a 2008 block should be ignored, I'd be stunned if even one maverick opposed over that. ϢereSpielChequers 20:30, 14 July 2015 (UTC)
- On Capatchas... holy crap. I got a new account for my WMF internship and had to ask to be manually confirmed. It was way more of a hassle than I ever thought it would be. Ed [talk] [majestic titan] 15:37, 14 July 2015 (UTC)
- Capchas are a complete pain, I can sort of understand IP editors needing to do them and also them being needed for new account creation - apparently a test of dropping the requirement saw us spam botted. But I don't see any gain in requiring capcha for links added by new accounts. PS are you going to Mexico? ϢereSpielChequers 20:33, 14 July 2015 (UTC)
- I am not. I have a wedding to attend this weekend. :-) Ed [talk] [majestic titan] 22:32, 14 July 2015 (UTC)
- Thanks, I don't agree that all the good stuff is started as I know of subjects where that is far from true. But I'd concede that it is true for many of our existing and former editors, so I've added User:WereSpielChequers/Going_off_the_boil?#All_the_Good_stuff_has_already_been_done. I'm familiar with the term upbundling, though I use it more for various proposals to shift certain admin rights just to crats. I agree that there have been changes that are disconcerting for newbies, the classic being that if you cite your first edits you are going to have to complete a capcha, if you don't you will get reverted and maybe bitten by others. I became autoconfirmed in 2007, and in those days you could add uncited edits without getting bitten. I was autoconfirmed long before my first cited edit so never had the capcha hassle, but we need a better less bitey way to handle that. ϢereSpielChequers 15:12, 14 July 2015 (UTC)
Sorry for the late response. I think my brain went off the boil. Anyway, to your 20:30, 14 July 2015 comments above; people oppose RfAs for all sorts of crazy reasons. I would be surprised if someone didn't oppose an RfA for a years old block. As to additional rights, I don't want to ask for additional rights. If my time here isn't enough for me to have rights to perform editing functions without having to ask, then I don't deserve them. Five years ago, an administrator gave me additional rights without me having to ask. I asked him to undo his changes (which he did without any resistance). Wikipedia has long walked away from the idea of empowering the simple editor to being able to build something magical. This is wrong, and is a foundational issue to the problems we have with editor recruitment and retention. But, to date, our response has been to continue to stratify users into ever more rights groups. By keeping my simple status as an editor I stand against this stratification. It also has the added benefit of understanding the viewpoint of editors who are relatively new to the project. My anecdotal conclusion; we treat them like crap. I think we treat any editor that does not have additional rights like crap. --Hammersoft (talk) 18:49, 23 July 2015 (UTC)
- I think we mistreat lots of our volunteers in various different ways, but I don't agree that it is as simple as mistreatment of editors who lack additional rights. I occasionally edit as an IP when I'm on an insecure connection, and my typo fixes as an IP don't get reverted. I've also been involved in several outreach editathons training new editors. Sometimes the participants ignore my advice and create an article that they can't find independent reliable sources for and then have it deleted, but newbies who add referenced content rarely have problems here. If anything lack of userrights can be used as a shield by some editors. ϢereSpielChequers 09:42, 5 August 2015 (UTC)
Nominating for Autopatrolled user rights[edit]
Hello WereSpielChequers! At our earlier discussion at WT:Autopatrolled you said "Many people get nominated or are nominated for this user right, and occasionally when we have a list I and others trawl through the list of prolific article creators and appoint suitable ones as autopatrollers."
Well, I'm starting to parse through the data to try to figure out how many editors there are with 20-50 (non-redirect) articles created (which I'm having to do manually via Wikipedia:List of Wikipedians by article count!) But what I'm finding alarming is the number of (still active) editors that have 50-100 (non-redirect) articles created and who don't have "Autopatrolled" rights! So, I'm thinking that I'd like to nominate some of these for Autopatrolled rights – How do I do that? Thanks in advance! --IJBall (contribs • talk) 20:14, 4 August 2015 (UTC)
- Hi IJBall, the simple answer is that you can nominate them at Wikipedia:Requests_for_permissions/Autopatrolled - though you might want to check a bit more than just the raw number of articles created. However I'm hoping to get one of my contacts to start producing the list of prospects again, 14 months after it last ran there should be a good crop of editors ready for this userright. Extracting it manually from the Wikipedia:List of Wikipedians by article count sounds like a much more time consuming thing as a lot of those editors will already have the right directly or as admins, and those that don't need to be checked out, some will have already lost the userright due to copyvio or creating unreferenced BLPs, and some will still be creating articles that get deleted for notability reasons, doing some of those checks is a lot quicker if you have admin rights and can look at deleted revisions. So I don't want to sound discouraging, but I wouldn't want you to waste hours of time doing something that a computer may be about to resume doing. ϢereSpielChequers 21:01, 4 August 2015 (UTC)
- Yep, that's exactly what I'm having to do: weed out Admins (and the occasional Bureaucrat), and exclude anyone who hasn't edited in a year or more. I'm also checking to see if they're already "Autopatrolled" at less than 50 articles created (I'd say about 20% are...). I'm not sure I have the stomach to do this for the Top 10,000 Article Creators, but I will definitely finish this out for the Top 5,000 (I'm already up to about #2,200 right now...) So, progress!! I figure some numbers here for my proposal are better than none! (I'll hope the people that lost Autopatrolled due to Copyvios, etc. is a very small number here – hopefully ~0.) P.S. Thanks for the answer on nominating – I won't be nominating any of those editors today (with maybe just a few exceptions! – I've already seen 2–3 editors that I know should be Autopatrolled!), but I may come back later, on a day I'm bored, and check the 50–100 article editors, and think about nominating some of them for Autopatrolled... --IJBall (contribs • talk) 22:02, 4 August 2015 (UTC)
Data[edit]
WereSpielChequers, as you're one of the people who likes data, I thought I'd report what I found out... I manually looked through Top 5,000 in the List of Wikipedians by article count looking for editors who had created 40-49, 30-39, 25-29, and 20-24 (non-redirect) articles (note: the data wouldn't let me filter out Diambiguation articles easily so, unfortunately, they'll be included in the count - oh well...), respectively, excluding from the count any:
- Admins or Bureaucrats
- editors blocked for cause, and
- editors who hadn't actively edited within approximately the last year.
It was a "hand count" so, unlike a bot, it's probably not "100%" accurate (I'm sure I missed an editor from the count here or there...). And it's certainly not the "complete data set" of everyone in the Wiki. But I think it's decent enough data to draw some conclusions from. Here are the results:
# of articles created | Total # of editors | # of editors with Autopatrolled rights |
% editors with Autopatrolled rights |
---|---|---|---|
40–49 | 72 | 8 | 11.1% |
30–39 | 61 | 8 | 13.1% |
25–29 | 47 | 4 | 8.5% |
20–24 | 38 | 6 | 15.8% |
A couple of follow-up thoughts: While I didn't keep a "hard count", I would say about half of the editors in the table above already had some kind of "extra permissions" (e.g. Rollbacker, Reviewer, File Mover, or combinations of any of these) even if they didn't have Autopatrolled. Also I recognized a significant percentage of the names from this list, and nearly all of them appeared to be long-time editors. IOW, I'd say nearly all of these can be put in the category of "trusted editors". So considering all of that, and the percentage of editors with less than 50 created articles who have already been granted Autopatrolled status, I can't think of any good reason not to go ahead with my proposal to drop the Autopatrolled permissions requirement to somewhere between 20–30 (non-redirect, non-disambig.) articles created (with 25 articles created being my likely suggested level). There's clearly a significant number of medium- and long-term "trusted" editors who would benefit from a lower requirement for Autopatrolled, and lowering the requirement will also hopefully help take some of the burden off the Page Curation crew.
I may try and put together the proposal for VPP tonight, but if I don't I may not get to it until 24 hours from now (as I'll have to run errands tomorrow during the day...). As always, any thoughts on your end are welcome. Thanks again! --IJBall (contribs • talk) 03:32, 6 August 2015 (UTC)
- The tricky thing is not establishing that we have lots of editors who have created that many articles, but establishing the point where such a large proportion will be ready for autopatroller that it is worth lowering the bar. I realise that my own experience is likely to be somewhat skewed, in going through the list of people who have created lots of articles but who don't yet have Autopatroller I am looking at the list after many people have pulled out good candidates and given them autopatroller. So maybe if someone went through a batch of people who have created forty articles they will conclude that it is worth doing, but there are two prices to remember. Firstly assessing candidates takes much longer than patrolling a single article, and can only be done by an admin, we may be short of new page patrollers but we are even shorter of admins. Secondly we don't want to set people up to fail; If we lower the bar to forty but more people then get refused because their work isn't yet good enough, then we don't just waste admin time, we give a group of goodfaith editors a negative experience. ϢereSpielChequers 21:38, 12 August 2015 (UTC)
WMF / Auto sign on talkpages[edit]
The WMF put out some extremely misleading statements. Some of your comments in the Village Pump discussion were incorrect (not your fault).
- Now that Flow is deprioritised - Nope.
- The WMF is asking for community input into things they could do instead of FLOW - Nope.
I spoke with the Flow project manager[3]. They have not diminished work on Flow, they are working full speed ahead on specific features for Flow. When the WMF says that their new work is going to be driven by the needs of the Community, they mean they decided what they wanted to give us, then they did research interviews with a couple of editors, then they shoehorned those responses to fit what they wanted to build. I spoke to one of the people they interviewed - the WMF staff interviewing him didn't even know the Community had already built the functionality they are working on. Specifically they are building a replacement for scripts like Twinkle, except their version won't work on existing pages. The Flow team is going full speed ahead, building a project that largely duplicates functionality we already built, and they are deliberately designing it so it won't work unless we convert every goddamn page on Wikipedia into Flow chatboards. Oh.... and it doesn't work unless you switch to Visual Editor too.
In the last election for WMF board of directors, all three elected candidates ran on a platform that Flow could not be deployed if the community didn't want it. So.... the WMF is restricting new development and support to Flow. If we want any new features, if we want any continuing support, we have to take Flow first. There's not much chance of the WMF willingly picking up the autosign project. They want our editor *gone*. Alsee (talk) 19:40, 8 September 2015 (UTC)
- Thanks, interesting. Wikipedia:Wikipedia_Signpost/2015-09-02/News_and_notes "Flow on Ice" did seem to me a pretty clear sign that Flow was no longer in fashion at the WMF. The dinosaur takes time to change, but in my view that sort of comment is usually code for "this project is dead", due to the obfuscation the message may get out more slowly, and due to its sauropod like nature the WMF's hind legs might still be moving for a while after the brain has squawked end. But I had stopped worrying about Flow. Though this implies that you may be right, staffers may still try to get it deployed wherever they can get consensus. ϢereSpielChequers 19:55, 8 September 2015 (UTC)
Hello again[edit]
It's possible you may vaguely remember me. We had a fair bit of contact several years ago, but I became pretty disillusioned by certain aspects of the whole project (specifically with regard to fair use images) and I have been on pretty substantial wiki-break. Now I'm willingly getting pulled back in to things and I'm once again finding everything for the most part interesting, challenging and enjoyable.
Anyway, I was wondering if you might have any observations you might share about how editing culture has changed since 2011. There are a few things I have noticed that are at least as bad, and in some ways worse, than they were before:
- I still see speedy deletion being horribly misapplied, in ways which can only discourage new editors. I'm looking at an article which was tagged as A1 and G2, neither of which was valid, eight minutes after it was created. I doubt that user will be back.
- The new (to me!) Articles for Creation process doesn't seem to have significantly changed things. I regularly see drafts that are really in pretty good state being declined by editors who want to hold them to punitively high standards of referencing. I'm looking at a draft that has three quality references to reliable sources which specifically discuss the topic in depth, which has been declined now by the same editor three times in a row, with the same reason given - "needs more references". The same editor declines drafts for hours on end at a rate of roughly two per minute. Unsurprisingly, the same editor is also an offender for inappropriate speedy deletion tagging.
- Free-content idealists are still chipping away at fair use images, unnecessarily making the encyclopedia progressively less useful as they work in pursuit of a goal that I do not share (an encyclopedia entirely built from free content, even if it means removing screenshots from articles about historically-significant GUIs). I'm staying away from this topic area because it's what led to my quitting in frustration last time. I do feel it's unfortunate that those policy and deletion debates are so dominated by editors with a stridently anti-fair-use point of view, and I'm not sure that they reflect the true consensus of the community.
- I keep seeing what appear to be deletion campaigns based on racial or religious grounds. For example there is an editor who is a significant contributor to articles on Sunni Islam, who also seems very keen on getting articles about Shia Islam-related topics, and Shia-believing individuals, deleted. I recently rescued an article about a significant topic in Sikhism (to be fair, the article was a pig's breakfast, and the "rescued" version is basically a stub). We have well-written, well-sourced, multiply-linked articles about "XXX in Catholicism", "XXX in Islam", "XXX in Buddhism" etc, but there was a serious argument from the nominator and an AfD participant that "XXX in Sikhism" was "not notable", a position that is so removed from reality (even reality as revealed by ten seconds of Googling) that I struggle to assume good faith.
- There seems to be an increase in the number of new articles from India, the Philippines, etc, and other places where English is not the primary or only language. I think this is a good thing - it reflects the global reach of the project, and increasing global participation is something to be proud of, not a problem to be solved. But new page patrollers in particular seem to disregard non-English sources. They don't seem to know how to Google non-latin-alphabet names. I don't think this is deliberate racism, but the behavior is racist in effect.
I suppose a lot of these things could have been written in 2011. Maybe not much has changed. Certainly the ongoing trend away from the "optimistic content creators" of 2006 towards the "pessimistic gatekeepers" who will eventually control everything is very noticeable when you take a few years away.
Cheers,
Thparkth (talk) 15:09, 1 December 2015 (UTC)
- Hi and welcome back! Yes CSD is an ongoing problem, but as we found with WP:NEWT resolving it one way or the other risks tearing the community apart. I do occasionally decline incorrect speedies and try to educate patrollers, but its a thankless task. There is a new speedy deletion process but it largely missed the point and for example changes such as disabling A1 and A3 as options for the first few minutes were rejected out of hand. I'm more enthusiastic about our chance of protecting more of our established writers from deletionist mistakes, once I get a bot writer to refresh the prospect list we can appoint a load of overlooked ones, the sort of people who create an articles a fortnight and have done so for years.
- AFC is a failure and I'm sure I'm not alone in thinking so. Part of the problem is that it keeps articles out of mainspace so they don't get collaborative editing, and anyone can decline an AFC submission whilst only an admin can delete a new page in mainspace. I would advise against any goodfaith newbie using it.
- Disregarding non-English sources is a problem, people who !vote delete at AFD with rationales such as "no English sources" should soon get the message when admins close as no valid reason given for deletion. A bigger issue is when people who rescue articles don't know how to use non English articles. Is there an essay explaining how to do so? If not would you consider writing one?
- Yes there has been greater growth in areas such as India, I assume this reflects growing internet use in such places. My view is that editing Wikipedia is not usually an entry level Internet task, I don't know if the rule of thumb still holds that it takes two years from getting internet access to using it for shopping. But I'm expecting there to be a lag between growth in internet use in an area and growth of the Wikipedia community in that area. This is greatly complicated/hindered by the problem that the smartphone is not as good an editing devise as a pc, and in many areas the increase in Internet use is basically a smartphone phenomena. I'm hoping that we have little overt racism here, but we've yet to come to terms with issues such as people in poorer societies having less access to free reference sources, or even the reliable sources in some countries being much more extensive than others. On the plus side there has been a huge growth in Commons, not least because it is much more inclusionist than Wikipedia, and part of that growth has been European museums digitally releasing images from their former colonies.
- I can't comment as to whether partisan deletion tagging has become more or less common. The biggest such battle I've been involved in was before you left and I don't know whether my experience is indicative of anything.
- I'd agree we have a problem with gatekeeping, but I'd diagnose it slightly differently. The community is roughly the same size today as it was four years ago, but we have significantly more articles; So we probably have less of a problem with people owning articles and rejecting any change to "their" work. But we have more of a problem with people who treat unsourced changes in the same way we used to treat vandalism. Personally I'm quite moderate on this issue, but I draw the line at an unsourced change to sourced information, and with a growing proportion of the pedia being sourced even with my stance there is much less room for editors who contribute uncited content. ϢereSpielChequers 17:00, 1 December 2015 (UTC)
- Lots to digest there. For now, is there any chance you could share the kind of criteria you're thinking of with regards to protecting established editors? Is there any existing "prospect list"? Thparkth (talk) 14:06, 2 December 2015 (UTC)
- Wikipedia talk:Database reports/Editors eligible for Autopatrol privilege is worth looking at and explains why Wikipedia:Database reports/Editors eligible for Autopatrol privilege isn't currently fit for purpose. ϢereSpielChequers 11:24, 3 December 2015 (UTC)
- So from what I understand, you're looking for editors who have created articles in the last 30 days, who don't currently have autopatrol, and who meet the general autopatrol eligibility criteria of having created at least 25 articles. At that point presumably you and other administrators would want to manually review the persons article creation history with a view to proactively offering them the autopatrol right. I've written a little bit of MediaWiki::API code that might be helpful in identifying those users. It isn't a bot - it doesn't make any edits, and it's run manually for now.
- I have put the output from a limited test run in my userspace here (based on editors who created pages in the last day, rather than the last 30 days). Note that I don't bother counting pages created if it goes over fifty, because it takes longer and is unnecessary since the guideline threshold is twenty-five. Even from this limited amount of data it is clear that there are many users who should really have autopatrol.
- I would be grateful if you would let me know if this going in the right direction. If so I will work further on it.
- Cheers, Thparkth (talk) 05:00, 6 December 2015 (UTC)
- Further to this, I updated the page with my output data with the complete data set, and also upped the maximum creation count to 200. This should pretty much be the complete list of users who meet the criteria of having created 25+ articles, not currently having autopatrol, and having created at least one article in the last month. Thparkth (talk) 12:16, 6 December 2015 (UTC)
- Wikipedia talk:Database reports/Editors eligible for Autopatrol privilege is worth looking at and explains why Wikipedia:Database reports/Editors eligible for Autopatrol privilege isn't currently fit for purpose. ϢereSpielChequers 11:24, 3 December 2015 (UTC)
- Lots to digest there. For now, is there any chance you could share the kind of criteria you're thinking of with regards to protecting established editors? Is there any existing "prospect list"? Thparkth (talk) 14:06, 2 December 2015 (UTC)
- Hi, that's good, almost there, and you've got rid of the bots. Running it daily means we might miss the person who creates an article every week or two, and they are a key group I'm trying to find an appoint as autopatrollers here; so if it could pick up anyone who has created mainspace article in the last 30 days that would be better. Also I'm not sure whether you are excluding redirects and creations of pages outside mainspace. This editor has over a thousand according to your list, but I could find only twenty mainspace articles they had started. ϢereSpielChequers 23:20, 6 December 2015 (UTC)
- Hello, apologies for the confusion from the contradictory information as I worked on the code. The current version of the report does indeed consider all of the users who have created articles in the last 30 days, not just those from the last few days. It also does exclude redirects. However it appears that the "count how many articles the user created" method I am using inadvertently includes the File: namespace, which is why the numbers are anomalous for the user you referenced. I'll go work on that now :) Thparkth (talk) 01:03, 7 December 2015 (UTC)
- Update - I have fixed the issue with the inclusion of non-mainspace contributions in the count. Updated data (for all users who have created articles in the last thirty days) is in the same place, User:Thparkth/autopatroltest. Thparkth (talk) 02:45, 7 December 2015 (UTC)
- Nicely done, I've started appoint autopatrollers from that list. Do you want to start refreshing Wikipedia:Database reports/Editors eligible for Autopatrol privilege, that's where people look for it, and we can mark the request as fixed once you refresh it there. ϢereSpielChequers 13:13, 16 December 2015 (UTC)
- I agree with much of what Thparkth says but he's perhaps getting too black a view by looking at speedy deletions and AFC which, by their nature, tend to be negative in nature. The number of articles is still going up steadily – see WP:5MILLION – and the rate seems to have increased in the last year or so. So, overall, the content creators are still gaining ground. Where they have got the hang of this, this seems to happen without much fuss and so is perhaps just hard to discern. Anyway, welcome back and feel free to ping me if you need help with something. Andrew D. (talk) 09:49, 23 December 2015 (UTC)
- Hello :) What is perhaps not obvious from what I wrote is that speedy deletion was something I was heavily involved with in 2011 as well. I really do notice a less patient, less forgiving, less encouraging mood in the project today versus then. It was already pretty WP:BITEy back then, but it is worse now.
- That is the continuation of a long-term trend that I have observed from my earliest days as a Wikipedian when as a brand new editor I created a number of shitty unreferenced articles (in mainspace directly, no less!) and was immediately welcomed by a real person and constructively engaged in cooperative editing. Today I wonder how many templated warnings I would have earned instead.
- In my mind, the early days of the project were dominated by optimistic, imaginative, open-minded content-creators who loved the idea that they could significantly contribute to an encyclopedia article on a major topic. "Geology", "James Watt", "Scotland", "Christianity" - they were all missing or stubs. How exciting is that? But over time the trailblazers got burned out, or bored, or just weren't interested in writing about minor topics. On the other hand, there was a need for a new kind of editor - one who would make incremental improvements, and create new articles on minor topics, but who would also defend the encyclopedia from the growing menaces of vandalism, POV-pushing, and commercial interests. This time also corresponded with the explosion of policies, guidelines and the zoo of acronyms that we cheerfully throw at each other nowadays, and it was also the time when newcomers began to be treated with suspicion and distrust.
- A few years later and we have a situation where new article creation is becoming an eccentric minority interest (and we more or less have a default assumption that any article a new user wants to create is probably a bad idea), where WP:NPP is played like it's a MMPORG where score is kept by counting the number of pages deleted and new editors banned (the goal being to level-up to admin, at which point you have completed Wikipedia and presumably will stop playing it), and where a huge amount of human effort is spent discussing, criticizing, documenting, and ultimately supporting an ever-more-self-referential quasi-judicial bureaucracy that is so many steps removed from the stated goal of building an encyclopedia that it is sometimes alarming. (Why are we arguing about the fairness of the process for establishing a consensus as to how to run an election for membership of an arbitration panel? Don't these people realize that Semiconductor is still a WP:SHITTYARTICLE?
- Now you may take issue with the factual accuracy of some or all of the above, and you may be right - it is largely bullshit. I am looking at the past through rose-tinted eyeglasses, and I am being more cynical then necessary about the present. But I do genuinely believe that somewhere along the line we have stopped assuming good faith about new editors, and the foundational principle of being the encyclopedia that "anyone can edit" (and which anyone else may fix afterwards if necessary) may no longer be the consensus position of the community.
- I guess this is where I see myself fitting in, in a small way. I am motivated to look out for new users, particularly those who have created new articles. There are some new users whose contributions we would not miss, it's true, but it breaks my heart to see someone's genuine, well-intended best work be slapped with an A7 and their talk page pasted with warning banners all because some wikipedian with a hundred thousand twinkle edits and a complete lack of empathy is ignorant of its importance. I know this is a concern WSC and others share. I'm not proposing any major changes to the processes - they are mostly quite well designed. I'm just advocating for a kinder, more empathetic, and perhaps less-rushed approach in dealing with newbies. Much as we need new pages patrollers, we can't ever get into the situation where we are significantly more welcoming to new patrollers than to new good-faith content contributors.
- Thparkth (talk) 17:02, 23 December 2015 (UTC)
- 2011 wasn't perfect either- have a look at some of the problems we battled over in WP:NEWT. My theory is that the drift from the wp:soFixIt society of the pre 2007 era to the subsequent WP:SoTagitForSomeHypotheticalOthertoFix era is an ongoing process. the article rescue squadron is one of the responses. My hope is that we can get the WMF to do some technical changes, in particular reducing the edit conflicts caused by people categorising and templating new articles. The difficulty is in getting WMF support for change. ϢereSpielChequers 22:22, 23 December 2015 (UTC)
I know and have been following Wikipedia articles for many years though I never knew how to write or edit one until a few months ago. I am from India and yes, people are getting more involved about from where and how all that information is coming. I am new writer and I have made a few edits; making edits is easier now due to visual formatting. It's very helpful.
Dishita Bhowmik (talk) 17:49, 22 November 2018 (UTC)
Users eligible for Autopatrolled[edit]
Thought I would move this discussion into it's own section, hope that's OK.
I could definitely update Wikipedia:Database_reports/Editors_eligible_for_Autopatrol_privilege but that page "belongs" to the Database Reports project and Community Tech Bot is updating it with its slightly-less-useful information twice each month. I suspect it is theoretically possible to have my script run through that mechanism, but I haven't worked in that environment before and there would likely be a learning curve involved. Going that way may end up being dependent on the same overworked volunteers who haven't yet been able to action your request to modify the current report.
It would be much, much easier to put my data somewhere else. Any thoughts?
Finally I have updated User:Thparkth/autopatroltest with data from December 18th.
Cheers,
Thparkth (talk) 22:22, 23 December 2015 (UTC)
- @Thparkth: Your list does look useful. I hope you get some help on this, as your list will be much more helpful than the current Database Report version... You might want to contact NKohli (WMF) about your version. --IJBall (contribs • talk) 21:08, 27 December 2015 (UTC)
New list[edit]
I have forgotten how I created the old list, but I found a note on Meta from you asking for a list split by {{Infobox UK place}}
vs everything else.
As I remarked I am still severely restricted in what I am permitted to do, so the list(s) will have to be placed in my user space, from where they can be copied, moved or transcluded.
I have been tussling with a huge (somewhat related) issue of producing accurate region codes for some 100k+ articles, success rate is now about 87-88% (compared with much less than 50% before I started) but of course the last 20% is going to be 80% of the work.
Nonetheless the first list should be uploaded soon. The rest will have to go on my todo list for now.
All the best: Rich Farmbrough, 03:46, 12 January 2016 (UTC).
- Wikipedia:Bots/Requests for approval/Femto Bot 7. All the best: Rich Farmbrough, 21:11, 17 January 2016 (UTC).
Precious anniversary[edit]
![]() | |
native tongue as tool | |
---|---|
... you were recipient no. 432 of Precious, a prize of QAI! |
--Gerda Arendt (talk) 06:46, 22 March 2016 (UTC)
Four years now! --Gerda Arendt (talk) 06:51, 22 March 2017 (UTC)
... and five! --Gerda Arendt (talk) 07:47, 22 March 2018 (UTC)
... and six! --Gerda Arendt (talk) 07:47, 22 March 2019 (UTC)
... and seven! --Gerda Arendt (talk) 07:38, 22 March 2020 (UTC)
Users eligible for autopatrol... finally![edit]
It's been a while but the all-singing, all-dancing, new version of the report code is now working and its first results are on the report page. I would be grateful if you would let me know about any problems you identify in the output.
Thanks for setting me along this road by the way - this kind of work is a lot of fun for me, and I now have all the access required to fix, modify, and create new reports without undue delay. Thparkth (talk) 23:37, 15 April 2016 (UTC)
- WSC – I can confirm that this list is now useful again! I'll start looking through it on my end for editors I recognize to let them know about applying for autopatrolled... --IJBall (contribs • talk) 17:51, 18 April 2016 (UTC)
The the[edit]
Thank you![edit]
You've fixed a lot of my shoddy reduplicative errors ("the the") etc recently and I appreciate it a lot. › Mortee talk 21:54, 25 August 2018 (UTC)
the ''[[The → ''[[The in AWB[edit]
Hi - I noticed that you've removed the second the in "the German adaption of the The Best Singers series.", and that you've performed a large number of similar edits with AWB. I don't think this is correct, however, as the final the is part of a proper noun which forms a separate syntactic unit, so the preceding article is still necessary, in my understanding - although a rewording of the sentence would most likely be preferable, as the repetition of the is obviously awkward. If there is an obscure grammatical rule permitting omission of the article, I would be pleased to read it, but otherwise I'd request that you be more careful or stop using AWB for this particular change, and go through the previous such edits again. Thanks. Kranix (talk | contribs) 18:09, 15 January 2019 (UTC)
- Hi Kranix, I have edited a number of "the the" instances, but not the one you link to, did you mean to link to something else?. Glad you agree that the repetition of the is obviously awkward. Do you want to give an example where you would have resolved this differently? ϢereSpielChequers 18:20, 15 January 2019 (UTC)
- Sorry, I forgot the link. The sentence I was referring to appears in Silbermond. Kranix (talk | contribs) 18:23, 15 January 2019 (UTC)
- OK, so you prefer: of the The Best Singers series over of The Best Singers series. I think that of the The is awkward and my solution an improvement. What alternative do you propose? ϢereSpielChequers 18:38, 15 January 2019 (UTC)
- I don't think it should be either way, but I also don't think we should change an awkward construction to an incorrect one. I went ahead and changed it to the "the TV series The Best Singers". The best course of action when going through this with AWB (or otherwise) is to perform a minor rewording like this; if that's not obvious, I would just skip the page. Kranix (talk | contribs) 18:46, 15 January 2019 (UTC)
- OK, so you prefer: of the The Best Singers series over of The Best Singers series. I think that of the The is awkward and my solution an improvement. What alternative do you propose? ϢereSpielChequers 18:38, 15 January 2019 (UTC)
- Sorry, I forgot the link. The sentence I was referring to appears in Silbermond. Kranix (talk | contribs) 18:23, 15 January 2019 (UTC)
Also would it be possible to make a box including his name dob and place of birth top right ? Thanks Eddie98706 (talk) 00:18, 3 May 2020 (UTC)
Thanks for the help x Eddie98706 (talk) 00:23, 3 May 2020 (UTC)
Wikipedia:Correct typos in one click[edit]
I saw the edit you made beenath->beneath. This kind of edits can be easily done using the tool in the title. Let me know if you need any help setting it up. Uziel302 (talk) 15:19, 28 September 2019 (UTC)
- Thanks for helping correcting typos from this list, I would love it if you can use the script and give me some feedback. If you prefer to correct one by one, please remove the ones you checked after you finish. Thanks again, Uziel302 (talk) 11:27, 29 September 2019 (UTC)
- Hi, I already use both AWB and another tool that Edward wrote for me - top thread on this list. I'm finding your list useful, but I'm just using the generic search, that way I get to see the same typo in other projects like Wikivoyage and Wikibooks as well. Plus I have something in Firefox that helps me find other typos around that one. If you are working from the top of your list I could start working from the bottom? What would make it even more interesting is if you could list words by number of occurrences, as using AWB I would start with the most frequent ones. ϢereSpielChequers 20:58, 29 September 2019 (UTC)
- I understand you already use tools, I am also using AWB. As of Edward's tool, I would love to try it out. The point of my script is to make the obvious corrections as easy and fast, without looking for similar typos and typos in the area, in order to manage to fix the 6000 typos my software have found. I would appreciate it a lot if you could just try it out, by adding
importScript('User:Uziel302/typo.js');
to your User:WereSpielChequers/common.js. This tool adds simple buttons to the lis: replace, remove, no much to learn. If after 5 minutes you don't like it, you can revert the edit on you common.js. No need to divide work as every checked typo should be removed from the list. The typos are sorted alphabetically so you can see those who happen couple of time, these are the top reccurring ones:
- I understand you already use tools, I am also using AWB. As of Edward's tool, I would love to try it out. The point of my script is to make the obvious corrections as easy and fast, without looking for similar typos and typos in the area, in order to manage to fix the 6000 typos my software have found. I would appreciate it a lot if you could just try it out, by adding
- Hi, I already use both AWB and another tool that Edward wrote for me - top thread on this list. I'm finding your list useful, but I'm just using the generic search, that way I get to see the same typo in other projects like Wikivoyage and Wikibooks as well. Plus I have something in Firefox that helps me find other typos around that one. If you are working from the top of your list I could start working from the bottom? What would make it even more interesting is if you could list words by number of occurrences, as using AWB I would start with the most frequent ones. ϢereSpielChequers 20:58, 29 September 2019 (UTC)
unveilled - 9 receving - 9 awll - 8 teching - 7 coveres - 7 appearanced - 7 adiation - 7 sceni - 6 unnotated - 6 predomnantly - 6 featureed - 6
Thanks again, Uziel302 (talk) 05:56, 30 September 2019 (UTC)
I made: Wikipedia:Correct typos in one click/list by occurrence. Uziel302 (talk) 18:03, 30 September 2019 (UTC)
- Thanks Uziel, that's interesting, I have fixed a few and also fed some of those to AWB. Taking aritst as an example. You had it down as 4 examples, I have just fixed 35. Part of that difference is that I think your search is case sensitive, and also misses plurals. Perhaps another involves non-alphanumeric characters, for example ". Another thing I would point out is that sometimes all your examples are very recent, like from the last month or so. That probably means that AWB already picks up that typo, its just that AWB usually has a backlog of typos to fix. That's why it is useful to have people like me do AWB fixes whilst we are using AWB to do other things. If the people who just use AWB in typo fixing mode were on top of the flow of new typos it would be rare for me to find AWB typo fixes when I'm searching for something else. Would it be possible for you to exclude either existing AWB typo fixes, or articles updated in the last three months? If so that would make your tool a really useful way to find additional typos to go into AWB, or if they have lots of false positives or multiple things they could be, to go into Edward's tool. ϢereSpielChequers 10:56, 3 October 2019 (UTC)
- Thanks for the elaborated feedback. I only got to see it now, haven't gotten notification on it. I know my scan is selective and doesn't find all occurrences, I only look for words with space on both sides so that the replacement won't replace part of other word. There are other ways to protect replacements, right now the bottleneck is not finding more typos but actually correcting the current typos. I may run a fuller scan in order to find the most frequent typos to add to AWB. I can also exclude current AWB replacements. I'm less of a fan of AWB since it requires a windows app (even if possible to run on other desktop), my script is easy to use right from browser, both desktop and mobile. Current bottleneck is finding more editors to help with this mission, and after we finish the current list of 6k typos, I will improve the scanning script.
- Another issue is that the edits aren't recorded anywhere except for the project history page. In Hebrew Wikipedia I created tag for edits of this script and it makes it easier to keep track of the edits (current count over 16k typos fixed). Is there a chance you can create a tag here for "fix-typo-tool" or something like that? Uziel302 (talk) 09:55, 8 October 2019 (UTC)
- Hi Uziel302, I'm sorry, but I have no idea how to create a tag. Perhaps Rich_Farmbrough can help us? I agree that AWB isn't ideal, and I resent having to run a Windows machine in order to use it, but it has an established userbase such that I assume any typo loaded into it is fixed long term whether I'm around or not. Plus I'm really interested in the sort of typos that have so many false positives that they need a human like me to look at them, or they could be typos of more than one word and you need a human to decide which. I'm not really interested in going through thousands of typos in the way that AWB does, especially not if AWB is going to fix most of them I the next few weeks. But if you can tweak your scanning script to screen out current AWB replacements then I would be very interested in going through some of what you then found. As for other occurrences, if you can treat commas, semicolons, the equals sign and brackets in the same way as spaces then that would be very helpful. If you could also treat "fullstop and space" in the same way that would be good (fullstop without a space would give us a deluge of urls). Once again, thanks for this. ϢereSpielChequers 11:31, 8 October 2019 (UTC)
- There should be a simple interface to create new tag on Special:Tags, but it only available for sysadmins. Uziel302 (talk) 13:32, 8 October 2019 (UTC)
- I made a list of odd words that don't appear on scowl biggest wordlist, English Wikipedia titles and wordnet. I looked only for lower case words, capitalized words have so many false positives of names etc. I posted here only words that appeared on latest dump more than 5 times. I then used AWB to see what it does, and excluded all the words it changed. The result is mainly legit variations of words, some foreign language words and some real typos like duringthe. Let me know if this list helps you and if you have any idea how to exclude more correct words without excluding common typos (like, if you exclude all wiktionary words, you get over 5000 common misspelling recorded there, and many many typos that are words in foreign languages. One of the reasons I try to help Lexeme namespace on Wikidata, where words can be querried and filtered by langauge and many other parameters) Uziel302 (talk) 15:14, 9 October 2019 (UTC)
- Thanks Uziel, I've skimmed through part of that, and used it to fix more than a hundred typos and suggest several new ones for AWB. I'm sure there's more to find as well, though most of those words are either jargon or quoted foreign words. ϢereSpielChequers 10:12, 10 October 2019 (UTC)
- Dear Uziel302, I'm still working through that useful list of yours. I have found a number that I have suggested incorporating into AWB and which have started going into it. I have also found examples such as "startet" where I corrected 36 articles where it was a typo, but hundreds more are a correct use of a non English word. So far too many false positives to load into AWB, but I might put into Edward's tool as I'm sure there will be more such typos in the future. There are also some words where the meaning is clear and which may in time be recognised as new words in English. I suspect there are also some that are either picked up by the Regex function in AWB or because someone else occasionally patrols for them, so if you have the time to rerun it some time it would be interesting to see the difference. ϢereSpielChequers 10:10, 19 October 2019 (UTC)
- Dear WereSpielChequers, I ran a new scan, this time focusing on capitalized words. Many of them are names, but I expect every name appearing frequently on Wikipedia to have an article containing it, or at list a disambiguation page. Here is the list I found the most, please note that all of it are words similar to known words, so it won't surprise me if many are typos: User:Uziel302#Missing names that appear frequently on Wikipedia. Thanks a lot, Uziel302 (talk) 04:33, 11 November 2019 (UTC)
- Thanks Uziel, I have fixed a couple of them, and gone through a couple more that had some to fix. But a lot will be like "Saachi", occasionally a typo for Saatchi but mostly a rare Indian name. So rare that there aren't enough articles on people with that name to justify a disambiguation page - just passing references such as a minor character in the plot of a Bollywood film. Unfortunately I am a bit pressed for time and or editing away from my home setup for the next few weeks, so I'm not sure when I will have time to go through more of this. ϢereSpielChequers 12:56, 12 November 2019 (UTC)
- There is a new list you might want to take a look.This time all the words are variations of actual words so it may contain higher rate of real typos. Uziel302 (talk) 12:04, 21 March 2020 (UTC)
- Hi Uziel302, good to hear from you. Hope all is well with you and yours. "15$$$Deptt" is a duplicate, and gets no hits for me, but Deptt does. Should i ignore the dollar signs? ϢereSpielChequers 13:00, 23 March 2020 (UTC)
- Indeed, the dollars just helped me separating numbers from words. I started grouping words by 15 and went from there down to 3 so words with over 15 occurrences may appear twice. It was easier than doing a full count. Usually I count on excel but this list was too long for excel to process in a timely manner, so I stayed on the text editor and ran regex replacements. Uziel302 (talk) 14:47, 23 March 2020 (UTC)
- User:Uziel302/sandbox - new list of lower case words that don't appear on aspell list of words but do appear multiple times here, and are very similar to common words. Some are foreign words and some are names, but you might find some that can be real typos, e.g. appointement that appears in non-french text. Uziel302 (talk) 06:29, 25 May 2020 (UTC)
- Indeed, the dollars just helped me separating numbers from words. I started grouping words by 15 and went from there down to 3 so words with over 15 occurrences may appear twice. It was easier than doing a full count. Usually I count on excel but this list was too long for excel to process in a timely manner, so I stayed on the text editor and ran regex replacements. Uziel302 (talk) 14:47, 23 March 2020 (UTC)
- Hi Uziel302, good to hear from you. Hope all is well with you and yours. "15$$$Deptt" is a duplicate, and gets no hits for me, but Deptt does. Should i ignore the dollar signs? ϢereSpielChequers 13:00, 23 March 2020 (UTC)
- There is a new list you might want to take a look.This time all the words are variations of actual words so it may contain higher rate of real typos. Uziel302 (talk) 12:04, 21 March 2020 (UTC)
- Thanks Uziel, I have fixed a couple of them, and gone through a couple more that had some to fix. But a lot will be like "Saachi", occasionally a typo for Saatchi but mostly a rare Indian name. So rare that there aren't enough articles on people with that name to justify a disambiguation page - just passing references such as a minor character in the plot of a Bollywood film. Unfortunately I am a bit pressed for time and or editing away from my home setup for the next few weeks, so I'm not sure when I will have time to go through more of this. ϢereSpielChequers 12:56, 12 November 2019 (UTC)
- Dear WereSpielChequers, I ran a new scan, this time focusing on capitalized words. Many of them are names, but I expect every name appearing frequently on Wikipedia to have an article containing it, or at list a disambiguation page. Here is the list I found the most, please note that all of it are words similar to known words, so it won't surprise me if many are typos: User:Uziel302#Missing names that appear frequently on Wikipedia. Thanks a lot, Uziel302 (talk) 04:33, 11 November 2019 (UTC)
- Dear Uziel302, I'm still working through that useful list of yours. I have found a number that I have suggested incorporating into AWB and which have started going into it. I have also found examples such as "startet" where I corrected 36 articles where it was a typo, but hundreds more are a correct use of a non English word. So far too many false positives to load into AWB, but I might put into Edward's tool as I'm sure there will be more such typos in the future. There are also some words where the meaning is clear and which may in time be recognised as new words in English. I suspect there are also some that are either picked up by the Regex function in AWB or because someone else occasionally patrols for them, so if you have the time to rerun it some time it would be interesting to see the difference. ϢereSpielChequers 10:10, 19 October 2019 (UTC)
- Thanks Uziel, I've skimmed through part of that, and used it to fix more than a hundred typos and suggest several new ones for AWB. I'm sure there's more to find as well, though most of those words are either jargon or quoted foreign words. ϢereSpielChequers 10:12, 10 October 2019 (UTC)
- I made a list of odd words that don't appear on scowl biggest wordlist, English Wikipedia titles and wordnet. I looked only for lower case words, capitalized words have so many false positives of names etc. I posted here only words that appeared on latest dump more than 5 times. I then used AWB to see what it does, and excluded all the words it changed. The result is mainly legit variations of words, some foreign language words and some real typos like duringthe. Let me know if this list helps you and if you have any idea how to exclude more correct words without excluding common typos (like, if you exclude all wiktionary words, you get over 5000 common misspelling recorded there, and many many typos that are words in foreign languages. One of the reasons I try to help Lexeme namespace on Wikidata, where words can be querried and filtered by langauge and many other parameters) Uziel302 (talk) 15:14, 9 October 2019 (UTC)
- There should be a simple interface to create new tag on Special:Tags, but it only available for sysadmins. Uziel302 (talk) 13:32, 8 October 2019 (UTC)
- Hi Uziel302, I'm sorry, but I have no idea how to create a tag. Perhaps Rich_Farmbrough can help us? I agree that AWB isn't ideal, and I resent having to run a Windows machine in order to use it, but it has an established userbase such that I assume any typo loaded into it is fixed long term whether I'm around or not. Plus I'm really interested in the sort of typos that have so many false positives that they need a human like me to look at them, or they could be typos of more than one word and you need a human to decide which. I'm not really interested in going through thousands of typos in the way that AWB does, especially not if AWB is going to fix most of them I the next few weeks. But if you can tweak your scanning script to screen out current AWB replacements then I would be very interested in going through some of what you then found. As for other occurrences, if you can treat commas, semicolons, the equals sign and brackets in the same way as spaces then that would be very helpful. If you could also treat "fullstop and space" in the same way that would be good (fullstop without a space would give us a deluge of urls). Once again, thanks for this. ϢereSpielChequers 11:31, 8 October 2019 (UTC)
I would appreciate if you could try using the script for fixing and dismissing typos of Wikipedia:Correct typos in one click and give some feedback. Thanks. Uziel302 (talk) 17:19, 26 June 2020 (UTC)
Just updated the list on User:Uziel302/sandbox to current dump. Uziel302 (talk) 18:13, 6 September 2020 (UTC)
- Added some 10K capitalize words that appear at least 5 times and very similar to common words. User:Uziel302/sandbox. Some are typos, most are names. Would appreciate any help locating and fixing the typos. Uziel302 (talk) 18:10, 7 November 2020 (UTC)
- You may want to check insource:/ whats /. Uziel302 (talk) 07:08, 20 March 2021 (UTC)
Strange sequence of edits.[edit]
It is not immediately clear whether it's vandalism or good faith editing, but I am required by policy to assume the latter in the absence of evidence to the contrary. What makes it so difficult to assume good faith is that you have edited over 200 articles contributed to by well in excess of that number of editors and yet you assume that you are right and therefore every one else must be wrong. You have been industriously changing "dependant" in all those articles to "dependent". "dependant" is the correct spelling of the word (see [here]). "dependent" seems to be creeping in as a alternate spelling (historically: it is an olde English spelling often still used by the legal profession). There was no reason to change the spelling per WP:NOTBROKEN.
On the subject of plurals, an apostrophe is not used when pluralising a word. However: English, as ever, has its exceptions and decades is one of those exceptions. Thus referring to the 1970's as a decade is correctly apostrophised. Another good (topical) example would be, "The United Kingdom will be leaving Europe on the 31st of October, no if's or but's". -86.130.28.61 (talk) 12:31, 14 October 2019 (UTC)
- Thanks for your note. Over the years I have corrected tens of thousands of easily confused words and have significantly reduced the occurrence of some words in Wikipedia, especially posses which when I started was mostly a typo of possess. It is because of me that very few actors are now staring in films as opposed to starring in them, and neither pubic schools nor pubic libraries last long without me adding the missing l. So I'm well used to finding that hundreds of editors have made minor typos for me to fix, and as encouraged by Wikipedia I have been bold and corrected many contributions that were made by people who knew that their contributions were liable to be "ruthlessly edited". It is possible that wiktionary or my reading of it are wrong re dependant, and I have paused my dependent corrections to give you time to convince the good people at wiktionary to change their records accordingly. As for your suggestion that "in the 1970s" could be "in the 1970's". I find that an odd use of the grocers' apostrophe, but since that one is embedded in AWB it would probably be better to discuss there, as you presumably want AWB changed. ϢereSpielChequers 13:57, 14 October 2019 (UTC)
- I would not regard "dependant" as an obsolete spelling. The link that I provided lists "dependent" as the US English spelling so this may be an WP:ENGVAR issue. My own (paper) dictionary gives the spelling as "dependant" and notes no alternative. The fact that over 200 articles (and probably a lot more) used the -ant spelling must be a good indication that it is still very much in use.
- I often wonder if the exceptions to non-apostrophised plurals is the source of the misused grocer's apostrophe (to give it its correct name). I know not what this Autowiki Browser is and I doubt that I can use it anyway. -86.130.28.61 (talk) 15:14, 14 October 2019 (UTC)
- Dependant isn't entirely an obsolete spelling, and it is partly an EngVar issue. Dependant is still valid for (Britain) A person who depends on another for support, particularly financial support (= US dependent). So I was careful not to change all the examples that I encountered, only the ones that according to my reading of Wiktionary were wrong. If Wiktionary needs updating, then, well it is a wiki. As for AWB, you don't need to be a user of it to take part in the discussions there, so I have started a thread at Wikipedia_talk:AutoWikiBrowser/Typos#In_the_1970's. ϢereSpielChequers 15:49, 14 October 2019 (UTC)
- OK, now that we have one issue resolved, I have started a thread on Wiktionary. Given the relative frequencies of Dependent and Dependant I would be surprised if the Wiktionary article was wrong, but I'm OK to continue pausing my patrol of dependant for a few days more. ϢereSpielChequers 12:47, 17 October 2019 (UTC)
- BTW Do you have any opinion re "publicistic"? ϢereSpielChequers 13:51, 17 October 2019 (UTC)
- 86.130.28.61 Discussion at wikt:Talk:dependant doesn't seem to be taking off, are you OK if I resume correcting those examples of dependant that don't fit the wiktionary definition? ϢereSpielChequers 17:21, 27 October 2019 (UTC)
- BTW Do you have any opinion re "publicistic"? ϢereSpielChequers 13:51, 17 October 2019 (UTC)
- OK, now that we have one issue resolved, I have started a thread on Wiktionary. Given the relative frequencies of Dependent and Dependant I would be surprised if the Wiktionary article was wrong, but I'm OK to continue pausing my patrol of dependant for a few days more. ϢereSpielChequers 12:47, 17 October 2019 (UTC)
- Dependant isn't entirely an obsolete spelling, and it is partly an EngVar issue. Dependant is still valid for (Britain) A person who depends on another for support, particularly financial support (= US dependent). So I was careful not to change all the examples that I encountered, only the ones that according to my reading of Wiktionary were wrong. If Wiktionary needs updating, then, well it is a wiki. As for AWB, you don't need to be a user of it to take part in the discussions there, so I have started a thread at Wikipedia_talk:AutoWikiBrowser/Typos#In_the_1970's. ϢereSpielChequers 15:49, 14 October 2019 (UTC)
- I often wonder if the exceptions to non-apostrophised plurals is the source of the misused grocer's apostrophe (to give it its correct name). I know not what this Autowiki Browser is and I doubt that I can use it anyway. -86.130.28.61 (talk) 15:14, 14 October 2019 (UTC)
- Here's what Merriam-Webster has to say on dependent vs. dependant. They note British usage versus American, as well as adjective form vs noun. isaacl (talk) 18:34, 28 June 2020 (UTC)
if you haven't already done so, please fix your script[edit]
Yeah, I know, an old edit, but your script broke a citation template by corrupting its |url=
value. If you have not already fixed it, please do so.
—Trappist the monk (talk) 14:14, 25 April 2021 (UTC)
- Whoops, sorry about that. I don't currently have a working machine that supports AWB. But thanks for spotting and fixing that. ϢereSpielChequers 14:20, 25 April 2021 (UTC)
False Information under article titled"Samaale"[edit]
You have put false information under the article titled "Samaale". For starters, Darod is not a Samaale tribe. They married into Samaale through the Samaale tribe Dirs daughter Dombira. Also Gardhere is not a pre- hawiye but a separate lineage that includes - Garre, Saransoor (Degodia, Gaaljecel, Massare, Isa), Owrmale, and Garjante.
The Ajuran are also a separate somali lineage. What are you sources since this information is readily available everywhere. 45.221.74.222 (talk) 10:34, 13 November 2021 (UTC)
- I replaced the word decent with descent. Fixing typos does not require a source. ϢereSpielChequers 19:33, 13 November 2021 (UTC)
You've got mail (2)[edit]
![Mail-message-new.svg](http://webcf.waybackmachine.org/web/20220405133241im_/https://upload.wikimedia.org/wikipedia/commons/thumb/4/49/Mail-message-new.svg/40px-Mail-message-new.svg.png)
It may take a few minutes from the time the email is sent for it to show up in your inbox. You can You've got mail}} or {{ygm}} template. at any time by removing the {{TheSandDoctor Talk 06:47, 25 December 2021 (UTC)
RFA 2021 Completed[edit]
The 2021 re-examination of RFA has been completed. 23 (plus 2 variants) ideas were proposed. Over 200 editors participated in this final phase. Three changes gained consensus and two proposals were identified by the closers as having the potential to gain consensus with some further discussion and iteration. Thanks to all who helped to close the discussion, and in particular Primefac, Lee Vilenski, and Ymblanter for closing the most difficult conversations and for TonyBallioni for closing the review of one of the closes.
The following proposals gained consensus and have all been implemented:
- Revision of standard question 1 to
Why are you interested in becoming an administrator?
Special thanks to xaosflux for help with implementation. - A new process, Administrative Action Review (XRV) designed to review if an editor's specific use of an advanced permission, including the admin tools, is consistent with policy in a process similar to that of deletion review and move review. Thanks to all the editors who contributed (and are continuing to contribute) to the discussion of how to implement this proposal.
- Removal of autopatrol from the administrator's toolkit. Special thanks to Wugapodes and Seddon for their help with implementation.
The following proposals were identified by the closers as having the potential to gain consensus with some further discussion and iteration:
- An option for people to run for temporary adminship (proposal, discussion, & close)
- An optional election process (proposal & discussion and close review & re-close)
Editors who wish to discuss these ideas or other ideas on how to try to address any of the six issues identified during phase 1 for which no proposal gained are encouraged to do so at RFA's talk page or an appropriate village pump.
A final and huge thanks all those who participated in this effort to improve our RFA process over the last 4 months.
This is the final update with no further talk page messages planned.
01:47, 30 December 2021 (UTC)
Administrators' newsletter – March 2022[edit]
News and updates for administrators from the past month (February 2022).
|
|
- A RfC is open to change the wording of revision deletion criterion 1 to remove the sentence relating to non-infringing contributions.
- A RfC is open to discuss prohibiting draftification of articles over 90 days old.
- The deployment of the reply tool as an opt-out feature, as announced in last month's newsletter, has been delayed to 7 March. Feedback and comments are being welcomed at Wikipedia talk:Talk pages project. (T296645)
- Special:Nuke will now allow the selection of standard deletion reasons to be used when mass-deleting pages. This was a Community Wishlist Survey request from 2022. (T25020)
- The ability to undelete the talk page when undeleting a page using Special:Undelete or the API will be added soon. This change was requested in the 2021 Community Wishlist Survey. (T295389)
- Several unused discretionary sanctions and article probation remedies have been rescinded. This follows the community feedback from the 2021 Discretionary Sanctions review.
- The 2022 appointees for the Ombuds commission are Érico, Faendalimas, Galahad, Infinite0694, Mykola7, Olugold, Udehb and Zabe as regular members and Ameisenigel and JJMC89 as advisory members.
- Following the 2022 Steward Elections, the following editors have been appointed as stewards: AntiCompositeNumber, BRPever, Hasley, TheresNoTime, and Vermont.
- The 2022 Community Wishlist Survey results have been published alongside the ranking of prioritized proposals.
Happy Bureaucratship Anniversary![edit]
![]() | Happy bureaucratship anniversary! Hi WereSpielChequers! On behalf of the Birthday Committee, I'd like to wish you a very happy anniversary of your successful request for bureaucratship. Enjoy this special day! CAPTAIN RAJU(T) 10:38, 3 March 2022 (UTC) | ![]() |
Precious anniversary[edit]
![]() | |
Nine years! |
---|
--Gerda Arendt (talk) 06:45, 22 March 2022 (UTC)
Deleted key content[edit]
Hi, WereSpielChequers and bbb23. I had added a large amount of content to this article. Why did you delete it? You can re-write what I had written. You can use your vocabulary, but please keep the information, which is extremely important and helps enrich Wikipedia. Thanks in advance. — Preceding unsigned comment added by Brightstarrrr (talk • contribs) 21:24, 5 March 2022 (UTC)