Jump to content

Wikipedia talk:WikiProject Red Link Recovery

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

Before posting questions here, remember to check the FAQ in case they are answered there already.

Database dump

[edit]

When will be the next dump?Smallman12q (talk) 19:49, 8 January 2010 (UTC)[reply]

I've worked up a newer method of generating links directly from the live database, rather than relying on the periodic database dumps. See: Red Link Recovery Live tool. - TB (talk) 14:34, 20 April 2010 (UTC)[reply]

Extra test for punctuation errors

[edit]

Is it possible to have the punctuation test check for replacement of single quotes (’) with apostrophes (')? I have done this roughly myself (example here) and discovered it is quite common. If the red link check can be expanded to deal with double quotes etc. that might also be useful. Tassedethe (talk) 14:31, 21 January 2010 (UTC)[reply]

Well spotted. When we find a mess of apostrophes and quotes, what should we change it to? Wikipedia:MOS#Apostrophes suggests that we replace single quotes by apostrophes everywhere, and even (that's a quote not an apostrophe) redirects to Apostrophe. Do we simply get rid of all quote characters other than ' and " as found on the keyboard? Certes (talk) 14:52, 28 February 2010 (UTC)[reply]
Red links containing unusual quotes (left and right single, left and right double, apostrophes, acutes and graves) can now be see in the Unlikely Links tool. Some quite interesting hits in there. - TB (talk) 22:14, 30 November 2010 (UTC)[reply]

"add a note saying why"

[edit]

1) I don't see the point of this. If the project is to determine which suggestions are correct, and remove the incorrect ones for the next database dump, the "add a note saying why" step just seems to create unnecessary extra work both during and when 'restarting' the project.

2) Understood too that you want to avoid random blanking of the pages, but reverting to an old edit just because you yourself did not work on the entries kind of defeats the purpose of other users' work on the project.

Please do not include "that's the way it always has been done" in your response. Thank you. Sct72 (talk) 00:51, 26 March 2010 (UTC)[reply]

What is this a reply to? —Paul A (talk) 01:16, 26 March 2010 (UTC)[reply]
  • I have received a notice on my Talk page, regarding:

Wikipedia:WikiProject_Red_Link_Recovery/Punctuation/7 To keep the list tidy, remove suggestions that were correct and have been dealt with. Remove sections that no longer contain any lines. If you couldn't repair the link and the suggestion is incorrect, strike it out and add a note saying why. I do not agree with the procedure, therefore I am opening it to discussion. If there is a reason for it past what I have in #2 above, I would like to know. Thank you. Sct72 (talk) 01:35, 26 March 2010 (UTC)[reply]

The "add a note saying why" step is to prevent needless duplication of effort. The person who processes incorrect suggestions so that they don't appear in future lists needs to know why the suggestion is incorrect. If you just strike through the suggestion without explanation, that person has to go and look at the link and figure out why it was struck out. If you leave a note explaining why it's incorrect, that saves them having to do what you've already done.
I don't understand your question #2. Where did somebody "revert to an old edit just because they themself did not work on the entries"?
Paul A (talk) 04:58, 26 March 2010 (UTC)[reply]
I have had to do that, see diff. I had previously asked the OP about removal of items from these pages but never received a reply (diff). I found that the OP had also removed items from the list without fixing them e.g. M14 rifle entry removed from list on Feb 28 (diff) but not fixed until after I had restored the entries and rechecked them all on Mar 7 (diff). This is the exact sort of reduplication of effort we are trying to avoid. Tassedethe (talk) 06:45, 26 March 2010 (UTC)[reply]
In regard to reverted edits, I believe I have undone some of the edits in question which had no summaries. I did this simply because there was no way to tell if the work had actually been done. There is nothing to stop someone from going on any page and crossing out a few lines without even reading them, the summary prevents work from having to be repeated. EmanWilm (talk) 11:16, 26 March 2010 (UTC)[reply]
  • Many things to address from "the OP" (crimony).
Trust. You want people to review suggested entries, but you don't trust their review enough without a stated reason. Instead of the person actually running the database dump, it sounds like an intermediary between the database person and the project volunteer.
Vandalism. With all of the types of vandalism done to Wikipedia - blanking, insults, inserting information, advertisement - the chances of a 'vandal' having the knowledge to go to a specific WikiProject to strike out random entries is almost nil.
Removal/"Fixed". When I originally started working on the project over a year ago, I did incorrectly remove struck-out entries. After the notification (in Feb '09), I learned from my mistake and did not think a reply was necessary.
The entry for M14 rifle on the project page was: M14_rifle links to Fleet_Anti-terrorism_Security_Team, try Fleet_Antiterrorism_Security_Team.
The only "Fleet" string visible on the M14 article page is Fleet Marine Force, so it was removed. Upon later investigation, the acronym FAST (for Fleet_Antiterrorism_Security_Team) on the article page is the red link in question. This could only be found by clicking Edit Page, or hovering over every red link on the page. Talk about needless effort!
  • I do not have a quarrel with eman, open2, tass, or the project, my concern is also for time and effort. If there's a quicker way to work through a project, I'll suggest it, if it's better or not. However, in the time going back over a year into my edit hiliststory, must have missed all of the other work I have done on the project, including the percentages I added to the main page to better track progress on the project. Whatever, may be time to find something different. Sct72 (talk) 04:22, 28 March 2010 (UTC)[reply]
Hi guys, TB here, please excuse my tardiness in joining this discussion. The original list handling instructions and associated templates were very much a thrown-together effort on my part four or five years ago. The vast majority of the overhead incurred in the current system is in the form of project volunteer time, and I'd be quite amenable to someone optimising the process to minimise this burden.  ::List entries currently exists in one of four 'states': new, fixed, falsepositive and needsanexpert. The most common 'transformation' is (and always should) new->fixed, hence the simplest operation (removing the entry) is used to indicate this. As the most common being made to the lists, it's likely that the majority of list handling errors will therefore also fall into this category - ie things will be deleted rather than left in or scored out. Entries that are deleted without being handled have a very low cost in that they'll simply re-appear on the next generation of lists. The next most common transformation is new->falsepositive, currently indicated by striking out and comnmenting. While the striking out is a faily quick and easy edit, I'd agree that adding a comment makes it a lot more onerous. My thinking it requesting it was twofold; firstly, the cost of marking a 'good' list entry as a false positive is quite large - the relevant red link(s) will not only not get fixed, they'll never be suggested again in future lists. Requesting a brief explanation of why it's a false positive acts as a stumbling block for an information saturated brain - possibly jarring it awake enough to notice something's wrong. Secondly, I had expcted to use the explanations to fine-tune future generations of lists. In practice, I've not made too much use of them as hit-rates have so far been well within acceptable bounds.
- TB (talk) 14:59, 3 April 2010 (UTC)[reply]

New data added to RLRL

[edit]

I've generated a fresh set of a few thousand suggested fixes for red links. As is my intention from now on, these are available from through the Red Link Recovery Live system on the toolserver. - TB (talk) 20:37, 26 October 2010 (UTC)[reply]

Marking false positives

[edit]

I've fixed all the incorrect "double quotation marks" links that I felt were wrong. Is there any way to mark the others as valid, so that people fixing future errors don't waste their time rechecking them? (Sorry if this is a perennial FAQ; I've been away for a while.) I'm not sure how to escape quotes in Wikipedia:WikiProject Red Link Recovery/Exceptions, or whether the Unlikely Links tool would look there. Certes (talk) 01:24, 30 October 2010 (UTC)[reply]

The unlikely tool doesn't use Wikipedia:WikiProject Red Link Recovery/Exceptions, largely because I too struggle to escape all those mis-matched quotes. I look into adding a mechnism for white-listing unlikely links as time permits - something akin to a button beside each listing would I think work best. All suggestions welcome. - TB (talk) 06:59, 30 October 2010 (UTC)[reply]
Thank you. I think a whitelist button would be useful if it's not too much trouble. Alternatively, the tool could make its own list if the technique[1] worked well for RLRL and is easy to copy. Certes (talk) 09:34, 30 October 2010 (UTC)[reply]
Sadly, while the RLRL tool is able to 'detect' a red link being fixed by examining the live Wikipedia database, the same is not possible for unlikely links. A button it'll have to be. - TB (talk) 13:26, 30 October 2010 (UTC)[reply]
Okay - try that now please. It's a little clunky but seems to work. - TB (talk) 17:10, 30 October 2010 (UTC)[reply]
The whitelist button works well for me. Thank you. Certes (talk) 21:19, 1 November 2010 (UTC)[reply]

Thanks to all the project members who have used the above tool and more thanks still to those what have provided feedback regarding its operation. Pretty much all of the rough edges have I now beleive been dealt with, and the majority of features I'd hoped to put into the tool are complete. Re-reading the "add a note saying why" thread above, I'm particularly pleased that almost all of the overhead in the old system of wiki-based lists has vanished - the setting up of lists, 'scoring out' of fixed entries and marking up of entries as 'wrong' is all entirely automated, and the 'AUTOFIX'' links remove much pain of hunting down the position of red links in articles and filling in a decent edit summary.

There's now a full set of 70,000 or so suggestions posted there to work on, so have at them please. I've some interesting ideas (of my own and others) for new facilities and improvements, and will continue to warmly welcome feedback from project members.

- TB (talk) 22:17, 13 November 2010 (UTC)[reply]

The Great Backlog Drive

[edit]

PanydThe muffin is not subtle 22:59, 23 November 2010 (UTC)[reply]

I've read through the project's page and, while its goals are sensible and worthwhile, wouldn't myself consider our red-link lists to be a 'backlog' in the sense used there. Our corrections to red links are useful in that they improve Wikipedia for readers (making it more likely they'll find material related to their area of study rather than blank pages) and editors (making it less likely they'll spend time and effort creating duplicate articles), the overall result is more of a slight imrpovement to the encyclopedia rather than bringing an espect of it 'up to standard' as it were. Far better that those with the time and inclination work through the other tasks being promoted by the TGBL2010 than our lists here, IMHO. - TB (talk) 23:16, 23 November 2010 (UTC)[reply]

Joining The Project

[edit]

I have been removing red links on pages, unaware that this project existed. Would it be possable to join this project and if so, how?Oddbodz (talk) 21:22, 22 March 2011 (UTC)[reply]

Hello, welcome and thanks for your improvements. I see you've found the list of participants and userbox. There's no formal membership process, so just keep on linking and do use the project talk pages if you want help with anything. Certes (talk) 20:38, 23 March 2011 (UTC)[reply]

Why no Special page for red links?

[edit]

I'm disappointed there's no Special page for red links as there is for blue links (Special:Allpages) - unless I'm looking in the wrong place. What would it take for such a page to be created? -- Jack of Oz [your turn] 21:58, 23 April 2011 (UTC)[reply]

Howdy there, JackofOz. I'm pretty sure that what you are imagining doesn't exist. Special:Allpages lists each and every of the several million articles in Wikipedia. An equivalent list for red links (all pages that don't exist) would be incredibly long and hard to generate automatically. There is however a manually generated list of pages that don't exist but are linked to from quite a few places at WP:MWA. - TB (talk) 10:10, 24 April 2011 (UTC)[reply]
Thanks, Topbanana. Actually, it's not all pages that don't exist, period, which is way longer than "incredibly long" - infinite, in fact. It's all pages that don't exist but which someone thinks exist or should exist, as evidenced by their linking the supposed title, which is a far smaller list.
I'm not an IT person's bootlace, so could you explain why it would be so hard to generate? -- Jack of Oz [your turn] 11:41, 24 April 2011 (UTC)[reply]
Ah, okay, I see where you're coming from now. It is perfectly feasible to generate the list of 'all red links on Wikipedia' that you're after. Indeed such a list is periodically generated to drive the Unlikely Links tool used by this project, which is then searched for various pre-defined patterns such as mismatched brackets or repeated letters. My original intention was to expand this tool into a more general 'red link explorer', allowing people to search for whatever they wanted in the red link list. This I beleive would have met your needs admirably. I'll need to check why I abandoned the project - from memory it was not technically trivial, largely due to the sheer number of red links we have.
In the meantime, I've started the process that rebuilds the list of red links on the toolserver now - it takes around 3 days to run. Once it's done, if there was something in particular you wanted to search for, I'll happily run a manual search. - TB (talk) 13:24, 24 April 2011 (UTC)[reply]
Oh no, nothing in particular, but thanks for the offer. I'm an inveterate browser, and I never know what my next article is going to be until the heavens combine and say "Jack, this is the one, start researching and writing now". I'm up to 795 articles now, but that's just a drop in the bucket compared to what I will have achieved by the time I'm done here ("here" being planet Earth). To be able to browse red links in exactly the same way I can browse blue links, categories, my too-enormous watchlist, and lots of other stuff, would assist me greatly by indulging my ceaseless appetite for new topics to write about.
I cannot believe I'm the only person who'd be interested in this. Cheers. -- Jack of Oz [your turn] 19:27, 24 April 2011 (UTC)[reply]
Have you had a look at WP:MISSING? - TB (talk) 21:30, 24 April 2011 (UTC)[reply]
It's far from a complete solution, but the (for now rather basic) Red Link explorer tool may be of interest. - TB (talk) 20:31, 24 August 2011 (UTC)[reply]
[edit]

Well, the toolserver based portion of this project has now been running for a year. Time for a quick summary of what's been achieved:

  1. Participation: We've had around 300 editors use the tools of whom 20-mumble spent considerable time working through the lists of red links there. This is a historical low for the project (we had around 90 active red-link-fixers a few years back). I suspect a combination of attrition from the burgeoning number of alternative fixup projects, a general dislike of using sites 'outwith' Wikipedia. All suggestions for increasing our numbers welcome.
  2. Results: 41813 red links fixed. 14462 (35%) of these were generated using the live 'check this page for red links and suggest alternate targets for them' tool with the remainder coming from the more carefully pre-prepared lists. This is down a bit on last years figure (46800), but in the same ballpark we're still doing good.
  3. Lists: Of the 70290 pre-prepared suggestions, 49128 (70%) were looked at. As the new system makes it much easier to try out new methods of suggesting alternate targets for red links, I've probably gotten a little carried away and put too many up this year. I'll adjust the quantity/quality parameters a bit to give fewer but hopefully more accurate suggestions for the upcoming year.
  4. Innovation: The new system has made it a lot easier for me to generate and manage lists of suggested alternate targets for red links, including a few new and interesting methods. I'll be producing full lists of red links containing 'fancy' quote characters and selected genitive-forms. regularly from now on.

- TB (talk) 22:27, 1 July 2011 (UTC)[reply]

Where are the red links?

[edit]

Posted for interest only - have been doing a little analysis of where the red link are in the namespaces. Italicised figures are expressed as a percentage of all red links in the english-language Wikipedia. By way of an explanation, the figure 793 (middle column, 4 rows down) indicates that there are 793 links from pages in the Template namespace to User pages that do not exist.

  From namespace
Main (0) Template (10) Portal (100) Book (108)
To namespace
Main (0) 17446072     (89.28%) 498968     (2.55%) 37755     (0.19%) 89
Talk (1) 16287 635 3847 0
User (2) 17 793 1754 2
User talk (3) 11 372 208 0
Wikipedia (4) 1023 5006 397 820
Wikipedia talk (5) 3 262 56 0
File (6) 45062     (0.23%) 4665 1906 1
File talk (7) 0 41 14 0
Mediawiki (8) 0 77 0 0
Mediawiki talk (9) 0 18 0 0
Template (10) 141915     (0.73%) 257437     (1.32%) 296 0
Template talk (11) 875040     (4.48%) 46873     (0.24%) 213 10
Help (12) 0 7 0 0
Help talk (13) 0 2 0 0
Category (14) 47746     (0.24%) 30891     (0.16%) 2114 0
Category talk (15) 3 34 111 0
Portal (100) 3828 1147 62097     (0.32%) 0
Portal talk (101) 0 8 5002 0
Book (108) 139 10 0 25
Book talk (109) 0 3 0 1

- TB (talk) 21:52, 26 July 2011 (UTC)[reply]

[edit]

A list of the most frequent words (for a given value of word) in all red links in the main namespace of the English Wikipedia.

  1. of (344948) - At least *some* of these are possessive forms of expressions - see Wikipedia:WikiProject_Red_Link_Recovery/RLRL#New suggestions - genitives
  2. in (110877)
  3. the (98125)
  4. The (95588) - New set of suggestions posted, see Wikipedia:WikiProject_Red_Link_Recovery/RLRL#New suggestions - leading the's
  5. and (89204)
  6. de (86243)
  7. Cite (60393) - Some of these arise from errors in using the {{cite}} template. A report listing these can be found at http://toolserver.org/~tb/cite/
  8. School (51454)
  9. ISO (46874)
  10. 3166 (44000)
  11. John (38541)
  12. New (36378)
  13. Act (33454)
  14. List (29819)
  15. v. (29800)
  16. House (26797)
  17. & (26301) - New set of suggestions posted, see Wikipedia:WikiProject_Red_Link_Recovery/RLRL#New suggestions - ampersands
  18. at (26061)
  19. Coat (25455)
  20. arms (25281)
  21. code (25091)
  22. for (24990)
  23. team (24908)
  24. season (24470)
  25. South (24216)
  26. - (23468)
  27. National (22012)
  28. District (21803)
  29. State (21524)
  30. William (21076)
  31. County (20710)
  32. United (20029)
  33. name (19492)
  34. football (17664)
  35. James (17583)
  36. World (17111)
  37. Park (17072)
  38. Robert (16960)
  39. David (16813)
  40. River (16807)
  41. County, (16558) - New set of suggestions generated. See Wikipedia:WikiProject_Red_Link_Recovery/RLRL#New suggestions - unspecified counties
  42. to (16507)
  43. Cup (15746)
  44. North (15637)
  45. J. (15524)
  46. High (15493)
  47. Club (15268)
  48. Championships (15246)
  49. Company (15006)
  50. A (14991)
  51. Thomas (14931)
  52. George (14665)
  53. International (14297)
  54. Museum (14207)
  55. Road (14111)
  56. La (13876)
  57. Association (13742)
  58. City (13721)
  59. – (13669)
  60. Lake (13609)
  61. A. (13330)
  62. Charles (13183)
  63. St. (13001)
  64. F.C. (12789)
  65. Michael (12730)
  66. States (12658)
  67. University (12603)
  68. railway (12546)
  69. film) (12406)
  70. Men's (12393)
  71. (film) (12087)
  72. Peter (12062)
  73. Island (12029)
  74. Paul (11954)
  75. Flag (11893)
  76. Department (11694)
  77. College (11609)
  78. Route (11463)
  79. San (11443)
  80. Center (11433)
  81. West (11431)
  82. American (11113)
  83. Women's (10892)
  84. Richard (10867)
  85. Summer (10859)
  86. station (10818)
  87. USS (10791)
  88. M. (10791)
  89. Creek (10742)
  90. R. (10700)
  91. Open (10628)
  92. a (10543)
  93. HMS (10352)
  94. la (10298)
  95. Society (10251)
  96. York (10190)
  97. Church (10184)
  98. Station (10146)
  99. Joseph (10029)
  100. I (9987)
  101. SS (9910)
  102. League (9691)
  103. Henry (9664)
  104. C. (9645)
  105. Railway (9624)
  106. di (9620)
  107. Police (9519)
  108. W. (9476)
  109. H. (9394)
  110. Records (9384)
  111. E. (9364)
  112. on (9331)
  113. Institute (9242)
  114. Pennsylvania (9168)
  115. (disambiguation) (9078)
  116. Games (8991)
  117. (band) (8903)
  118. José (8838)
  119. Township, (8818)
  120. Union (8612)
  121. S. (8604)
  122. User (8531)
  123. Group (8421)
  124. Council (8378)
  125. von (8327)
  126. del (8298)
[edit]

Is there a tool that is leading to making these sorts of edits? MOS:DASH is clear that the en-dashed form ("foo–bar") is correct, so the "problem" is that this form of the title doesn't exist (hence changing to a hyphenated form ("foo-bar") that does). In many cases, it appears that the hyphenated form that does exist is just a redirect to somewhere else anyway ("foo-bar"→"foo") , so a better solution is to use a piped link—resolves the redlink problem but also doesn't make the visible content typographically incorrect. Since there appear to be tools used to suggest the replacement link to use, should be easy for them to look if the proposed replacement (for example, naive and often MOS-incorrect replacement of dash with hyphen) is a redirect, and if so, propose piped syntax instead of simple link-change. DMacks (talk) 14:43, 5 September 2011 (UTC)[reply]

Howdy DMacks. The tool generating the suggested repairs to red links is Red Link Recovery Live (http://toolserver.org/~tb/RLRL/). It's current understanding of dash-like characters is (like my own) rather basic, yup. Looking over MOS:DASH, I see that:
Other dash-like characters are more complex. If you can confirm the above is correct, I'll look into improving things. - TB (talk) 15:44, 5 September 2011 (UTC)[reply]
I cannot fathom how you read such things in MOS:DASH. There's nothing wrong with hyphens in titles. En dashes sometimes have spaces around them, and sometimes digits, and sometimes other things. Read it again. It does not make anything illegal. It provides guidance on choosing; it's up to thoughtful editors to make the right choices after examining the situation. Your automated tools is doing way too much harm. Please stop. Dicklyon (talk) 06:14, 8 September 2011 (UTC)[reply]
Hyphens: "Two forms of dash are used on Wikipedia: en dash (–) and em dash (—)" ... "When naming an article, do not use a hyphen as a substitute for an en dash that properly belongs in the title"
Em/en dashes and spaces: "There are two options. Use one or the other consistently in an article." ... "1. Unspaced em dash" ... "2. Spaced en dash"
I'm not suggesting that there aren't exceptions to these; just that page titles meeting these criteria should be considered by the RLRL tool as 'suspicious' enough that they be avoided as potentially good alternative targets for red links. Or more restrictively, the inverse - that red links metting these criteria should be considered dubious enough that a search should be made for a better-formed page title as a target - TB (talk) 12:15, 8 September 2011 (UTC)[reply]

Happy to offer my services

[edit]

I am very happy to join this Wikiproject. I had actually written a plea on the Villiage pump for this very idea - to sift through the pages upon pages of endless red-links for notable subjects, and try our best to clear Wikipedia of as many of these as possible, via stubbing if necessary. I think we should aim to just get the articles done in any shape or form (quality may very well be reminiscent of a first draft, as per WP:IMPERFECT and WP:DEV. i.e. building laterally before building vertically. At the moment, I am in finishing off my Year 12 course (I'm in Australia so I finish my WACE in a week), but in my spare time I sift through Wikipedia:WikiProject Film/Requests and try to make at least decent articles for them. Anyone who would like to give me a hand with this, please do. I am sure that revisiting that page after completing an article and seeing 10 or so newly-blued links instead of one would make the monstrous project seem a lot easier.

Any thoughts?--Coin945 (talk) 12:50, 3 November 2011 (UTC)[reply]

Welcome to the project, Coin945. As a starter on the Wikipedia:WikiProject Film/Requests list, running it through the Red Link Recovery Live tool suggests that at least some missing films exist, just using slightly differing titles. - TB (talk) 13:31, 3 November 2011 (UTC)[reply]
Ahh cool. I didn't even know there was a tool for that. That's very helpful. Once I chose 3 redlinks that already had article written about the slightly differently named film before I chose one that didn't. So what happens to them? Is there a quick and easy way to redirect each of the redlinks to their respective film articles?--Coin945 (talk) 15:15, 3 November 2011 (UTC)[reply]
Oh just be careful btw. One of them is a film called "Flush" and the autocorrect thingy has "Flesh" and "Flash" - obviously different films. I had a quick search on google and it seems the film is not notable anyways.... so, yeah, we need to be careful with that.--Coin945 (talk) 15:17, 3 November 2011 (UTC)[reply]

Getting started

[edit]

How do I actually get started with the project? I added importScript('User:Topbanana/RLRL_SR_Utility.js'); to my .js page, and I pressed Ctrl-Alt-Delete to bypass my browser's cache. However, I don't see the "AUTOFIX" link next to any red links.

Am I doing anything wrong? Or, is there anything else that I haven't done? Or, could the edits be required to be done from some Toolserver page that I haven't seen? Thank you for your help. Allen (Morriswa) (talk) 01:22, 4 January 2013 (UTC)[reply]

Welcome! The AUTOFIX button appears in the RLRL tool at http://toolserver.org/~tb/RLRL/, once you have selected a type of redlink to fix and generated some suggestions. welsh (talk) 07:34, 4 January 2013 (UTC)[reply]
So, there is no way to correct red links directly from Wikipedia? Allen (Morriswa) (talk) 09:52, 4 January 2013 (UTC)[reply]
The RLRL tool is just one way to find suggestions of possible alternatives to redlinks. There are many other ways, ranging from manual searching, guessing, and other tools, as explained in the project page. At its simplest you don't need any tools to edit articles with redlinks - just go ahead and make a change if you know a bluelink alternative or synonym. welsh (talk) 17:17, 4 January 2013 (UTC)[reply]
I would love to pop a small icon beside each red link for which a suggested target exists. Clicking on this would pop up a small list of possible remedial actions that might be carried out. Selecting an action would cause it to be applied to the article text, and a preview shown, ready for approval or manual correction. Alas, the facilities needed to make this work aren't available currently on my own servers, the Toolserver or Wikimedia Labs. One day, perhaps. Until then we've the RLRL tool as described by Welsh above. - TB (talk) 00:25, 5 January 2013 (UTC)[reply]
TB, that's a great idea. In fact, that's where I thought the "AUTOFIX" links were to show up, not on the Toolserver page. The only problem I have is if there are multiple red links on a page. If there is a way to implement it, I'm all for it. Allen (Morriswa) (talk) 01:08, 5 January 2013 (UTC)[reply]
[edit]

Is there a way to find redlinks on a page (and then strip them from the page) automatically?

I'd like to be able to build city outlines like these...

...using a template (as a starting point) that lists every possible topic (like "Government of ___", "Neighborhoods in ___", "Taxicabs of ___", "Cycling in ___", "Rats in ___", "Port of ___", "Fireboats of ___", "Climate of ___", "___ City Council", "___ Board of Trade", etc. There are hundreds of such potential links.

The problem is, for many cities, most of the links would show up red. Yet this would be feasible if there was a redlink stripper. Strip out the redlinked items, and leave the blue.

Of the outlines listed above, Outline of San Marino and Outline of Monaco still have a redlink problem. The redlinks were stripped out of the other outlines by hand. The necessity to remove template-generated redlinks by hand limits the rate at which city outlines can be created using the subst:template approach.

So the question is: Is there a way to find redlinks on a page (and then strip them from the page) automatically? The Transhumanist 05:28, 19 May 2013 (UTC)[reply]

It's a rather manual answer, but you can use AWB to list out redlinks and bluelinks from an article and paste the answer back in. Any ordering or formatting would be lost, so this is only a sensible idea when you are first creating a page, or if the blue links are sparse. welsh (talk) 07:01, 19 May 2013 (UTC)[reply]

Comment on the WikiProject X proposal

[edit]

Hello there! As you may already know, most WikiProjects here on Wikipedia struggle to stay active after they've been founded. I believe there is a lot of potential for WikiProjects to facilitate collaboration across subject areas, so I have submitted a grant proposal with the Wikimedia Foundation for the "WikiProject X" project. WikiProject X will study what makes WikiProjects succeed in retaining editors and then design a prototype WikiProject system that will recruit contributors to WikiProjects and help them run effectively. Please review the proposal here and leave feedback. If you have any questions, you can ask on the proposal page or leave a message on my talk page. Thank you for your time! (Also, sorry about the posting mistake earlier. If someone already moved my message to the talk page, feel free to remove this posting.) Harej (talk) 22:47, 1 October 2014 (UTC)[reply]

this is my first time taking classes online..

[edit]

can anyone help me by telling me what exactoy the assignment is that is due tonight ? or at least point me in the right dircti0n? there so many links!!! thank you, Mark — Preceding unsigned comment added by 2602:306:BD5A:740:B11C:E1AC:E4B7:ACBF (talk) 22:09, 11 January 2015 (UTC)[reply]

WikiProject X is live!

[edit]

Hello everyone!

You may have received a message from me earlier asking you to comment on my WikiProject X proposal. The good news is that WikiProject X is now live! In our first phase, we are focusing on research. At this time, we are looking for people to share their experiences with WikiProjects: good, bad, or neutral. We are also looking for WikiProjects that may be interested in trying out new tools and layouts that will make participating easier and projects easier to maintain. If you or your WikiProject are interested, check us out! Note that this is an opt-in program; no WikiProject will be required to change anything against its wishes. Please let me know if you have any questions. Thank you!

Note: To receive additional notifications about WikiProject X on this talk page, please add this page to Wikipedia:WikiProject X/Newsletter. Otherwise, this will be the last notification sent about WikiProject X.

Harej (talk) 16:56, 14 January 2015 (UTC)[reply]

Ten Years

[edit]

The Red Link Recovery WP:WikiProject has now been formally active for 10 years - yay!

Over that time, over 350 volunteers have help us to turn approximately a half-million red links blue. In the spirit of WP:WikiGnomeery, these incremental improvements to the encyclopaedia have been made with a minimum of fuss and upheaval. Good job all.

- TB (talk) 13:50, 17 March 2015 (UTC)[reply]

And to you!

- EmanWilm (talk) 03:50, 18 March 2015 (UTC)[reply]

[edit]

As User:Jarble and User:Fayenatic london have pointed out on the project page, another source of red links are deleted pages. Pages can be deleted for a number of reasons, at least one of which indicate that Wikipedia should not contain an article on the given subject. In this circumstance it would almost always be appropriate to de-link the title, but I'm having trouble both tracking down relevant policy and clearly evaluating the logic involved in identifying 'titles than have unambiguously been deemed inappropriate'.

By way of a start, I've added a report at Wikipedia:WikiProject_Red_Link_Recovery/DPWIR listing the 250 deleted pages with the largest number of incoming links. Being very-linked-but-deleted articles, these probably represent the most difficult cases. All ideas welcome. - TB (talk) 08:07, 20 May 2015 (UTC)[reply]

The admin instructions include removing backlinks, see WP:AFDAI. WP:CFDAI and WP:PROD#Deletion. I occasionally remind admins about these requirements when I find red links that have been left after a recent deletion.
Occasionally, if incoming links indicate notability that deserves an article, but the article was deleted for other reasons e.g. COPYVIO or entirely promotional tone, then it's better to leave the links. – Fayenatic London 08:36, 20 May 2015 (UTC)[reply]
[edit]

In an effort to pick out interesting features of red links that may point towards useful RLR work to be done, I've been carrying out some mechanised clustering of the current crop of red links into related groups. So far nothing too interesting has shown up, but I'm posting the results here in case it inspires anyone else. - TB (talk) 08:07, 6 October 2015 (UTC)[reply]

Metacode Set size Salient Notes
KNTKRT 3171 Kentucky_Route_[0-9]{4} One outlier missing the space between 'Route' and number has been fixed
FRMTMRKTRT 2685 Farm_to_Market_Road_[0-9]{1,4} Variants use "Farm-to-Market_Road" and "Farm_To_Market_Road". One outlier missing the space between 'Road' and number has been fixed.
N0SNLKNKTM 1805 {1,4}_in_the_Sinhala_Kingdom Number is a calendar year
TSMPKXN 1735 [0-9]{1,4}_\(disambiguation\)
FFTXPRTPLNT 824 Voivodeship_road_[0-9]{3}_\(Poland\)
ASLST 804 USS_LST\(H\)-[0-9]{1,3}
ASLSL 755 USS_LCI\(L\)-[0-9]{1,4}
NPRTKL 728 [0-9]{1,4}_in_Portugal Number is a calendar year
AKSKTFRTR 719 Executive_Order_[0-9]{4}
PPRSKSRNXS 698 Papyrus_Oxyrhynchus_[0-9]{1,4}
NNR0RNRLNT 622 [0-9]{1,4}_in_Northern_Ireland[0-9] Number is a calendar year
NFLSF 607 [0-9]{1,4}_in_philosophy Number is a calendar year
ARKNPLTMSR 579 Oregon_Ballot_Measure_10_\([0-9]{4}\)
NSKTLNT 574 [0-9]{1,4}_in_Scotland Number is a calendar year

The table below shows large sets of links arising from {{SLBY}}.

Metacode Set size Salient
LSTFNTRNXNLRKNSXNLTRSN 2087 List_of_international_organization_leaders_in_[0-9]{1,4}
LSTFFRNMNSTRSN 1987 List_of_foreign_ministers_in_[0-9]{1,4}
LSTFRLJSLTRSN 1855 List_of_religious_leaders_in_[0-9]{1,4}
LSTFSFRNSTTSN 1833 List_of_sovereign_states_in_[0-9]{1,4}
LSTFKLNLKFRNRSN 1663 List_of_colonial_governors_in_[0-9]{1,4}
LSTFSFRNSTTSNPK 677 List_of_sovereign_states_in_[0-9{1,3}_BC
[edit]

I have compiled a list of 'lists of red links'; these can often benefit from being run through the 'checkpage' module of the Red Link Recovery Live tool. - TB (talk) 16:56, 20 February 2016 (UTC)[reply]

New here

[edit]

I just discovered this wikiproject, but don’t have the time to study it closely. I have always wondered if there is an organized process at wikipedia for discovering, categorizing, and building redllinks. I am guilty of introducing many redliniks as I just did with Real Estate Visionary. Is this something of interest to others here? Ottawahitech (talk) 15:22, 23 September 2016 (UTC)please ping me[reply]

I have also redirected the above redlink to real estate visionary which is also a redlink. Just wondering if there are any rules here prohibiting this type of redirect? Ottawahitech (talk) 23:15, 24 September 2016 (UTC)please ping me[reply]
Wikipedia:WikiProject Missing encyclopedic articles seeks to identify useful articles not present in Wikipedia - in essence red links we ought to have. Wikipedia:Most-wanted articles compiles lists of common red links and tries to classify them. And this project (Wikipedia:WikiProject Red Link Recovery) tries to remove red links by identifying ones we shouldn't have - spelling errors, typos, titles that should redirect to an existing article. The field of completing redlinks by adding new articles is largely addressed by the many subject-oriented WikiProjects such as Wikipedia:WikiProject Sports, Wikipedia:WikiProject History and Wikipedia:WikiProject Spain. - TB (talk) 09:04, 29 September 2016 (UTC)[reply]
@TB: Thanks for explaining. So I understand that this wikiproject is only involved in the removal of redlinks which the projects members(?) deem un-encyclopedic? But I read in the link you supplied that articles with more than 20 red-links are considered worthwhile topics, so if you remove wp:redlinks doesn't it defeat the purpose? Ottawahitech (talk) 15:46, 29 September 2016 (UTC)please ping me[reply]
[edit]

The wonderful https://tools.wmflabs.org/tb-dev/RLRL/ is no longer suggesting any links for me to mend, and I miss it! Further details at Wikipedia:WikiProject Red Link Recovery/RLRL#Not seeing suggestions. Please put any replies there to avoid two parallel threads. Thanks, Certes (talk) 16:49, 2 January 2017 (UTC)[reply]

Apologies, I've been inactive for quite some time. Will try and resurrect RLRL in the next few days. - TB (talk) 21:19, 4 July 2018 (UTC)[reply]
[edit]

I have a regex for Greek Letter Organizations (so it gets Lambda Lambda Alpha and Beta Eta , etc) and I'd like to find the redlinks on wikipedia that match that. Ideas?Naraht (talk) 15:26, 27 August 2018 (UTC)[reply]

You could either convert your regex to a MySQL pattern and run a Quarry query on pagelinks (joining to page, to eliminate blue links), or download a dump of those tables and run a script in your favourite language locally. There may be easier ways. Certes (talk) 15:47, 27 August 2018 (UTC)[reply]
MariaDB (our DB) supports PCRE. However, I had to segment dumping with bit-manipulation to build my lookup table without the query killer noticing it. — Dispenser 00:54, 28 August 2018 (UTC)[reply]
...or put together 24 sandboxes, like this one for Lambda, and see what appears other than the sandbox itself. Certes (talk) 17:39, 27 August 2018 (UTC)[reply]
-- ; mysql -h tools.labsdb
SELECT COUNT(*), rl_title_ft
FROM u2815__p.redlinks_enwiki_p
WHERE rl_title_ft REGEXP
'(?i)^(\\b(alpha|beta|gamma|...|psi|omega)\\b[.]? *){2,10}$';
+----------+-------------------+
| COUNT(*) | Sample            |
+----------+-------------------+
|      351 | Alpha Alpha Gamma |
+----------+-------------------+
The table updates weekly and has a (high speed) Full-Text index for Dabfix. — Dispenser 00:44, 28 August 2018 (UTC)[reply]
My apologies for responding late,
Dispenser, I don't understand how to access the database that you are querying, and given that searching for insource:/\[Alpha Alpha Gamma/ only gives one hit, I'd be surprised if there were 351 redlinks. I'd love to understand more.
Certes, so that would ultimately require me to create 24*24*24 lines?
Certes, I've used Quarry before but not for much complex, so any help there would be useful.
The other question is whether these techniques would include entries where the redlink is in a template.Naraht (talk) 21:28, 29 August 2018 (UTC)[reply]
Dispenser's suggestion sounds better than either of mine, if we can find a way to access the u2815__p.redlinks_enwiki_p table on host tools.labsdb. I think Alpha Alpha Gamma is just an example; most of the 351 will be for other pages. Quarry would need something like this query. Infuriatingly, although all the data we need is in there, the useful indices have been redacted (and we don't have permission to create our own) so there's no realistic prospect of any useful query ever finishing. And yes, the last resort of a sandbox would require 24*24*24 lines, probably best done by creating a page with 24*24 entries ΑΑΑ-ΑΩΩ then globally replacing the initial letter and previewing 23 times. Certes (talk) 22:11, 29 August 2018 (UTC)[reply]
Special:Permalink/857180897Dispenser 02:18, 30 August 2018 (UTC)[reply]
Gorgeous. My ultimate goal would be to find even the ones that aren't linked at all. :)

A new newsletter directory is out!

[edit]

A new Newsletter directory has been created to replace the old, out-of-date one. If your WikiProject and its taskforces have newsletters (even inactive ones), or if you know of a missing newsletter (including from sister projects like WikiSpecies), please include it in the directory! The template can be a bit tricky, so if you need help, just post the newsletter on the template's talk page and someone will add it for you.

– Sent on behalf of Headbomb. 03:11, 11 April 2019 (UTC)[reply]
[edit]

While the admin instructions include removing backlinks, see WP:AFDAI. WP:CFDAI and WP:PROD#Deletion that does not always happen.

Can a group/category/bot list/etc. Be created that identifies articles that have an AfD AND have incoming (red) links. At the very least a bot could probably scrape it from the what links here page of deleted articles. Before I consider posting something at Wikipedia:Bot requests I was wondering if there was any other way to get the info. I searched the talk archives and did not find a previous discussion about this.

I am assuming that it is obvious that in most (always there are exceptions) red links to articles deleted by AfD should be unlinked. Jeepday (talk) 18:08, 11 June 2021 (UTC)[reply]

  • I hope someone will answer your question, but I would only like to point out the caveats. An article X may be judged non-notable, but if the topic is treated within article Y, then X can be turned into a redirect to it and almost all incoming links preserved (that's relevant here because AfD participants, and closers, aren't always careful in distinguishing these cases, so even after an AfD deletion there might sometimes be a need to re-create as a redirect). Another sort of complication arises when some of the incoming links are not intended for the deleted article, but for another topic with the same name. It's for reasons like these that a bot task is unlikely to be workable. – Uanfala (talk) 18:19, 11 June 2021 (UTC)[reply]
  • It would certainly be useful to identify links to deleted articles, but we still need a human to decide what to do with the identified links. It could make a good AWB task for an interested gnome, with the suggested replacement being [[Cruftsville]]Cruftsville to be overridden at the editor's discretion. Certes (talk) 22:13, 11 June 2021 (UTC)[reply]
    Certes Yes that was pretty much what I was thinking but had not considered AWB (I might be a little old school) I was thinking put them in a list, and work/research from the 'what links here' page of the deleted article to figure out what to do. Jeepday (talk) 10:16, 13 June 2021 (UTC)[reply]
    Experience may suggest systematic refinements, such as removing the entire line from dabs and See also sections rather than unlinking. Certes (talk) 10:29, 13 June 2021 (UTC)[reply]
    I posted at Wikipedia:Bot_requests#Red_links_to_AfD_deleted_subjects I guess we see what happens now. Jeepday (talk) 15:43, 22 June 2021 (UTC)[reply]
Is there a reason that this is only for AFDs? Because plenty of articles are deleted via PROD & CSD as well. Liz Read! Talk! 23:09, 24 June 2021 (UTC)[reply]
@Liz: Since CSD isn't necessarily passing judgement on notability, it doesn't make sense to delete links to what could be a viable article. Articles deleted using PROD are usually non-notable, so they probably could be included in this effort if the bot operator wanted. Vahurzpu (talk) 01:03, 25 June 2021 (UTC)[reply]
We also need to beware of articles AfD'd per WP:TNT, where the topic is notable but this attempt at an article is worthless. Certes (talk) 01:23, 25 June 2021 (UTC)[reply]
Hi, I'm just about finished delevoping this bot. Where would be a good location to post the output? At the moment there's a work in progress at User:Pi/Sandbox/Bot Pi (Talk to me!) 23:17, 13 July 2021 (UTC)[reply]
In the mean time I've started outputting the page here: Wikipedia:WikiProject Red Link Recovery/Red Links to AfD Deleted Pages Pi (Talk to me!) 22:48, 15 July 2021 (UTC)[reply]
Pi sorry I was out for a couple days and missed this post. At first glance Wikipedia:WikiProject Red Link Recovery/Red Links to AfD Deleted Pages looks good. I hope to get a chance to work through some of them in the next couple of days. Jeepday (talk) 11:02, 3 August 2021 (UTC)[reply]
Pi I worked the 2 at 2020-07-07 that looks good. Of note Hannah Bacon shows two main page articles W-League transfers for 2017–18 season ‎& W-League transfers for 2016–17 season not listed by the bot as red links This is GOOD as best I can tell we can't change those red links, as they are attribute of the table, not specifically added. I will continue to work some as time allows, and let you know what I find. Jeepday (talk) 16:25, 3 August 2021 (UTC)[reply]
Pi I worked 2021-07-20 and three from a year earlier 2020-07-17, 2020-07-18 and 2020-07-19. I like the format and the data returned. I was thinking about asking the bot to return a link to the AfD, but I don't think we want it, there may be multiple and I need to give the "what links here" a review before I do anything anyhow.
Can you make the bot update the page? Maybe weekly (day of the week, your choice)? Return the last 362 or 365 dates with articles that meet the criteria. Just drop the existing content and completely replace with updated. As we work through them, the most current will be at the top, the oldest at the bottom, and at places in-between date gaps where those that have been addressed used to be. I don't think we need any history beyond what is the pages history. Jeepday (talk) 12:28, 5 August 2021 (UTC)[reply]
Sure, I'm just rushing for a deadline at work, but I'll get it done tomorrow evening or Saturday Pi (Talk to me!) 23:47, 5 August 2021 (UTC)[reply]
Pi I am looking at 2021-07-22 Peter Hand is showing and as best I can tell the only main space link is for the redirect to Peter Hand Brewing Company this is not unusual, and is often the out come of AfD, can filter the report to exclude deleted articles if the only link is a redirect? Jeepday (talk) 16:25, 10 August 2021 (UTC)[reply]
[edit]

This is largely directed at @TB, but does anyone know if the Red Link Recovery Live and Unlikely Titles are still maintained? The current links listed in this project give this 410 Error saying:

Tool disabled
The tool responsible for the URL you have requested, https://tb-dev.toolforge.org/unlikely/, has been disabled as a part of the Toolforge Grid Engine shutdown.
If you have reached this page from somewhere else
This URI is managed by the tb-dev tool, maintained by Tb.
If you maintain this tool
You need to migrate your tool to the newer Kubernetes runtime. For more details, see the instructions on Wikitech.


If not, the links should probably be removed from the project overview. Thanks! TB5ivVaO1y55FkAogw1X (talk) 17:41, 7 January 2024 (UTC)[reply]