It would be interesting to know, but probably difficult to determine, how many redirects were deleted on November 5, the day before this list was created. 100? 1,000? More? There are already 50,000+ here but it seems like a lot had been deleted over the previous 24 hours. LizRead!Talk!00:22, 7 November 2015 (UTC)[reply]
I've been removing some red links to get this page under control but I realized that I don't know what the purpose of this page is. Is the goal to keep all the contributions for reference? Else, should we first remove all the red links? Then can we remove the pages that were decided as keep per RFD? -- Ricky81682 (talk) 01:05, 23 November 2015 (UTC)[reply]
I did that mainly to preserve numbering, since people were referencing the list by number at RFD. The second reason was to preserve redlinks, as I like to see what's been deleted. Breaking this up into multiple pages does seem to be the best solution though. --Tavix(talk)01:22, 22 January 2016 (UTC)[reply]
Now I see what you are talking about at ANi. I see value in retaining the redlinks as they help us see patterns. It is also a lot of work to remove the redlinks because you can't see what is red in edit. I'd love to chunk this into roughly 0-9999, 10000s, 20000s, 30000s, 40000-end or even smaller chunks. Another good idea would be for editors to adopt a chunk and put their name on that chunk to screen check everything - reduce the overlapping effort. Legacypac (talk) 22:20, 21 January 2016 (UTC)[reply]
Again, that can be done by going back to the first version here if people want. I'd prefer no redlinks and clearing out sections already reviewed. -- Ricky81682 (talk) 01:43, 22 January 2016 (UTC)[reply]
I mostly edit using a tablet, so loading a list like that is pretty much impossible for me, and probably a lot of other users as well, which is at least partly why I have just been dipping in and out at RFD instead of coming at it directly. If you guys can find some way to break it up into reasonable chunks, that'd be great. I'm also thinking that adding a temporary note at WP:CSD#G6 about this might help prevent further misunderstandings. I imagine someone could have done this a while ago but probably nobody thought it would take this long to deal with. Its such a ridiculously huge number of pages the mind just kind of recoils from the reality of it. Beeblebrox (talk) 00:57, 22 January 2016 (UTC)[reply]
Awesome! Now I can edit the sublists. Anyone object if we remove deleted or confirmed ok redirects from these sublists as we go? If an editor thinks its unquestionably reasonable, it survives RfD or gets retargeted/built into a page etc we delete from the sublist. That way we can eventually declare the job finished. Legacypac (talk)
What were you section breaks based on? Was there a set number? Like I said, I did this before for 45k pages so I know how to do this in Excel. Want breaks at each 500 or so pages (with some logical adjustments of course)? 10,000 a page is still a burden. -- Ricky81682 (talk) 06:15, 22 January 2016 (UTC)[reply]
I was still having trouble editing the lists (loading, finding specific lines to delete) so I tried throwing in section breaks which helped a lot. No set number, but I try to put them between different targets at least. Each section can be worked out of existence so set numbers in each is somewhat unnecessary? Legacypac (talk) 06:19, 22 January 2016 (UTC)[reply]
For sure. I throw a section break in when I want to tag an area to work on deleting now too. Then I can count the first 5 or whatever after the break. Legacypac (talk) 06:46, 22 January 2016 (UTC)[reply]
Talk page redirect
Are the talk page redirects relevant? Aren't they just paralleling the main redirect? Do you think they could just be removed? Or at least removed if the mainspace page one is there too? -- 06:10, 22 January 2016 (UTC)
I guess they are redirects from moves. Many have no content. I was sending for deletion but a different admins saw them differently so I stopped. Legacypac (talk) 06:15, 22 January 2016 (UTC)[reply]
I don't see the need. They are G8s and let the admins reviewing them deal with them. I think we should remove them as irrelevant, they'll just follow the main versions. Please it's a cheap way to move the progress bars fast. -- Ricky81682 (talk) 06:27, 22 January 2016 (UTC)[reply]
Nice work. User:Anomie/Neelix_list/1 to User:Anomie/Neelix_list/5 are working great to process down the redirects as checked/deleted etc. Generally he built the redirects to a given target all in a row so they are already grouped by target on these lists. The by target info sure shows the breadth of stupidity and some areas we should check carefully.
7. String trimmer (125 redirects, 123 remaining) could use a review.
The next time anyone gets a hard time for working on this, you should show them this list. 400 redirects to a page about a type of window is a perfect demonstration of the utter lunacy of this situation. Beeblebrox (talk) 23:31, 24 January 2016 (UTC)[reply]
There is another crusade to crucify me for working on these redirects lead by User:Hullaballoo Wolfowitz who is posting in ANi, RfD, AfD against my efforts very pointy style. He's been removing CSD tags too on redirects that have zero real world use. He needs a topic ban on Neelix redirects User:Beeblebrox.
Unfortunately we are not done with breasts related redirects yet. I just got over 20 100% fake phrase/word redirects to a breast cancer screening process deleted, and there are many more related ones to check. Legacypac (talk) 01:40, 25 January 2016 (UTC)[reply]
It should be possible, but all I can manage is a crude copy/filter/paste-back, filtering out any whose URLS have "redlink" in them. NebY (talk) 14:27, 24 January 2016 (UTC)[reply]
There are bots who can do that. I asked Anomiebot but it may be worth making a request at WP:BOTREQUESTS. There are tricks to do with AWB (have it pull only blue links and copy it over section by section as well. -- Ricky81682 (talk) 00:56, 29 January 2016 (UTC)[reply]
@Oiyarbepsy:, I was the one who started using that workflow because it is far easier for me as a non-admin to move the redirects into a section where it is understood that they are all tagged than tag all of the redirects individually. Tavix was simply preserving the section for the use of non-admins. Tazerdadog (talk) 06:31, 28 August 2016 (UTC)[reply]
(edit conflict)There was an AN discussion which resulted in this solution. Instead of tagging the redirects for CSD, non-admins are to list them in the WP:X1 section for an admin to review/take care of. A CSD-regular admin complained that he's tired of having to deal with Neelix redirects and this was decided to be the optimal solution to bypass them listed at CAT:CSD. I'm on mobile right now or I'd link the discussion. --Tavix(talk)06:35, 28 August 2016 (UTC)[reply]
and to answer the follow-up question, it's easier for all parties involved to do it this way. Non-admins don't have to individually tag the redirects. Also, as an admin, I can just do a spot-check and batch-delete them instead of having to individually delete individually tagged redirects. --Tavix(talk)06:38, 28 August 2016 (UTC)[reply]
It's faster, basically. I can put redirects in the section much faster than I can tag them individually by maybe 5-10 seconds per redirect. I assume that it is more convenient (or at least just as convenient) for Tavix, or another admin to check my work by going down the page and using the delete link in the template rather than opening individually tagged pages. It saves maybe 15-20 seconds per redirect between me and the deleting admin, which when multiplied over hundreds or thousands of deletions is real amounts of time. I'm not sure why you object to this workflow? Tazerdadog (talk) 06:39, 28 August 2016 (UTC)[reply]
Twinkle's batch delete function is a Godsend for this task. Just a couple days ago, I took care of 175 redirects in one swing and it only took about 10 minutes of checking the redirects that weren't obvious to me. All I have to do is cut the X1 section after I've done my double-check, paste the redirects into my deletebox, hit the d-batch button, and enter the rationale. I could probably handle several hundred per hour if someone's already checked them. --Tavix(talk)06:53, 28 August 2016 (UTC)[reply]
Got it. I just added a short sentence below each X1 section explaining this that could have avoided this entire time-wasting conversation. I'm rather amazed that people think I object to something that I'm only learning about right now. Documentation is a good thing, people. Oiyarbepsy (talk) 06:46, 28 August 2016 (UTC)[reply]
User:Tazerdadog, Okay, next question. We broke these into sections to improve load times. But now the speedy noms require moving to a different section. If I hit the edit section button, this different section won't even be on my screen. If I don't, I have to wait for the entire page to load. How do you handle this? Oiyarbepsy (talk) 06:48, 28 August 2016 (UTC)[reply]
Three ways> Firstly, If I'm editing without other people around creating edit conflicts, I'll just load the whole page. They load fine now. Secondly, I'll use copy and paste over two edits. Finally (the preferred method), I can edit the top non-X1 section, and by moving the redirects to be deleted above the section heading, it will plop them into the section heading. Also, Tavix, the AN discussion is news to me. Would you mind linking it when you're off of mobile? Tazerdadog (talk) 06:53, 28 August 2016 (UTC)[reply]
I made a X1 section because when I was looking to delete them via Twinkle, there was no X1 Criteria. I didn't know how I could nominate them as X1, because when I nominated them with db-rationale (X1), i believe they were moved to G6 housekeeping, which was where the Neelix redirects were originally nominated under. Therefore, I made the section so the users that could nominate them as X1 could do it since I couldn't.--MrLinkinPark333 (talk) 18:53, 28 August 2016 (UTC)[reply]
Neelix Disambiguations
Hello. I was going through the Hognose redirects by Neelix, and I came across a few redirects that were not on the list. These were because the redirects were converted into disambiguation pages: See Spread head and Spread-heads for example. I wanted to let you know in case there were some overlooked Neelix redirects --MrLinkinPark333 (talk) 01:36, 23 September 2016 (UTC)[reply]
Proposed "Keep" section at the top, like X1
@MrLinkinPark333: What I've been doing is marking them as "keep" (or redirect or whatever) with a Done tag to show I've looked at them. I realise this puts a little strain on others but makes sure that they're kept visible until added to the progress meter or otherwise looked over: @Tavix: should I perhaps move them to a "Keep; done" section? I really like the idea of "buffering" them now into an X1 section for delete, that's much better (for everyone, I guess) than having to take them individually to CSD for the more obvious ones, and I can see that the same could apply for a "keep" section until they are counted for the progress meter update (and a second-chance looksee by others), I tend to put a summary of the edit summary on the ones I keep, any I want retargeted I tend to take to RfD and just mark as "at RfD" unless bleeding obvious retarget). Should I be nominating them as X1 myself, too? I assumed a bot/admin runs over the X1 section periodically. @User:Sphilbrick is that the case? I missed out on the ANI discussion, but wholeheartedly support the "new" way of doing these. Si Trew (talk) 15:20, 28 September 2016 (UTC)[reply]
I note that User:Tazerdadog cleared out all that I had tagged as "done, keep" yesterday. Thanks for that. One reason I don't want simply to clear them out myself is to show that they have been looked at before they disappear from the list – partly so that people who say I and others are "neelix bashing" can be pointed to a historical version of the list that shows lots of "done as keep" tags and that they are not just tipped into the dustbin nonchalantly. However I do realise that this puts a burden on others then to tidy them up, and I am wondering if we should have a holding section for "Done as keep", analagous to "X1", which is cleared out by any editor when clearing the X1 and updating the progress meter; this would reduce the burden on the person clearing them and the X1 out and also give a good historical version of which had been kept (as well as deleted) over the last couple of days without having to refer to individual edits. In the meantime I will keep marking them as done inline, but I realise this does add a burden on other editors which I should probably carry myself. Si Trew (talk) 05:05, 29 September 2016 (UTC)[reply]
In short, an editor deciding to keep (or boldly retarget) an item on these lists should:
Move the item from the section to a new "Done" or "Checked" section, analagous to X1 section
Optionally, add an edit summary of any change made (e.g. rcatted, retargeted)
The editor then clearing the X1 section should also clear the "Done" section and include the number of entries there in the progress meter total.
I think that more evenly balances the load between the editor qua administrator, doing the X1s and the progress meter, and the gnome checking individual redirects. (Of course one may be both.) I am trying to reduce, not increase, administrative or quasi-administrative burden, here. Today I have been doing it my "new way" at List 2 and List 4, and I think you'll agree it makes it easier for a casual glance to have all the keepers at the top of the article – in going through List 4 I found two I had already done back in May but had not struck or marked on the list. I presume it's no problem to add a brief repetition of the change made (e.g. rcat) since the person deleting the "Done: Keep" section just has to tot them up (i.e. look at the list number of the last entry) before deleting, to adjust the progress meter. Si Trew (talk) 05:10, 29 September 2016 (UTC)[reply]
Incidentally it would also give an opportinity for any I (or other editor) boldly marked as "keep" to be taken to RfD, rather than them simply disappearing off the list, probably never to be found again. Obviously as we pare down these lists, the ones that remain get less obvious boldy to take action with. Si Trew (talk) 08:06, 29 September 2016 (UTC)[reply]
Trying to address all of your points somewhat systematically:
The progress bar is updated on an ad hoc basis, and certainly not after each edit or cluster of edits. Basically, don't worry about the progress meter.
@Tavix: has generally been the admin who has been spot-checking the X1 redirects and then deleting them. If you feel that a redirect would get SNOW deleted at RFD, just move it to the X1 section.
The lists are generally viewed as to-do lists. If someone wanted to see what percentage of the neelix redirects got deleted, or check for indiscriminate deletion ("Neelix bashing"), I'd point them to the original lists, so that they could view the number and distribution of redlinks for themselves. [1].
Once a redirect has gone to RFD, we generally delete it from these lists, as RFD will handle it.
My removal of your tagged redirects were done essentially as housekeeping, I did not scrutinize them at all. The hope for these lists is that one editor can look at all of the redirects, having to double up seems optimistic.
I'd advise you to just remove redirects from the list if you feel they should be kept, but a section at the bottom where kept redirects are placed for 24 hours isn't a big deal, and certainly beats inline tagging. Thanks' for helping out with the work, it really does make a difference. Tazerdadog (talk) 08:10, 29 September 2016 (UTC)[reply]
I'll carry on for a day or two putting them in a bunch at the top of the list (and thanks for clearing the last load and giving me an {{ec}}!). I don't expect others to do likewise, but for myself I find it helpful I note from your "sweep" a few minutes ago, which presumably because of time differences etc I was the only contributor to either, I had about twice as many keepers as X1 deleters (about 25 to 49 I think it was): but that is partly because it's easier to make an obvious keep than an obvious delete or burden RfD with the less obvious ones. It is certainly much easier for both nominator and admin to have the X1 criterion rather than having to take them all individually to CSD. I am aware of the history of arguing whether "all neelix redirects should just be blank deleted and start again" etc and as far as I took out of that, there is not carte blanche to mass delete things just because they are Neelix, i.e. they have to have the minimum of a strawman argument that had anyone else created them they would have gone the same way.
Sorry for not being particularly systematic in my arguments here, these Rs tend to get you into a very unsystematic way of thinking, and I tend to dot about a bit because after a while one becomes a bit word blind and literally can't see things like yesterday I listed one at RfD that had a typo in it, without noticing it had a typo, that is just a physical brain/eye thing I can't think of the name for it but lierally the brain or optic nerve no longer sees things when the pattern matching makes them look the same as too many others one has seen (desensitization does not seem to cover it, but it's something like that). So I tend to switch sections seemingly at random from time to time to avoid that: and that probably accounts a bit for my seeming lack of coherence too. Also cos I am writing in a hurry so I can plough on through the lists! Si Trew (talk) 08:35, 29 September 2016 (UTC)[reply]
2:1 in favor of keeps is about where I'm at as well. Rfd would not be happy with us if we had to take every one through them. Do you want me to do anything other than spot/sanity check the deletions, or was that what you had intended? Tazerdadog (talk) 08:44, 29 September 2016 (UTC)[reply]
(edit conflict)Quite happy with that. Incidentally you may notice I convert the redirect links to {{-r}} section by section, as that is useful to me to check the history of the redirect etc rather than the target, but I can't do it per-list as it's too big for the inbuilt editor. The find/replace terms I use are "# [[" to replace with "# {{-r|" and "]] -->" to replace with "}} -->". That seems to work quite well. To check the target, one still has the target link there as well of course. I did suggest we did this ages ago but nobody took up the suggestion. A one-off bot run over all four (three?) remaining lists ould be good, but I am hopeless at suggesting that kind of thing. Si Trew (talk) 08:50, 29 September 2016 (UTC)[reply]
I've done that before (I created the NeelOldLan template that is currently in use), and I can make some updates if needed. Could you explain exactly what the change that you want is - it might be as simple as changing a couple of characters in a template. Tazerdadog (talk) 08:54, 29 September 2016 (UTC)[reply]
Seven years ago, less two months, consenus was achieved at RfD for deleting every multifarious name for a noble. That consensus still stands, as far as I know, as it has never been discussed since. We do not need every single possible combination of every noble's name, that is what the search engine is for. Neelix has maybe read that, perhaps I am after all as the nominator of that because that was Dec 2019 and Neelix started proliferating the combinatorial explosion of redirects in early 2010, although never got as far as Alf Freddie Candide of Windy-Grates.
We do not need or want seventeen thoussand variations of kings and queens. A few, four or five, is fine. As I argued seven years ago, what it leads to is that editors pick the first thing that matches so an enormous inconsistency across articles. I am not trying to straitjacket people into writing one way, but on a series of related articles, use the same language. Xerox PARC learned this with early machine translation systems. By restricting the language (is it a "button" or a "pushbutton"? they found not only did the quality of the machine translation go up, the quality of the english went up by any measure. And I am a technical writer and engineer, and I believe that is true, too. Limited vocabulary can help express things more clearly. Esperanto is a good or bad example.
I need to make this plain: Neelix made a mistake to add these in bulk, and I for one do not believe that he is lightning fingers and added them individually, because I am on a quick computer and I know how long it takes me to mark on as keep and rcat it, and I am lightning fingers too. I've been touch typing since i was seven, thirty-seven years ago, on a proper keyboard where you have to hit the sodding things and you get a noisy response back if you get it wrong. Or am I confusing it with a piano? But I do believe that Neelix created this combinatorial explosion in good faith when the search engine was not as good as it now is. I do not believe for a second that Neelix created any of these redirects maliciously or for any kind of personal gain, money or status. Having gone through thousands of them, Neelix has I would characterise as a somewhat Christian bent, perhaps Evangelical but would be a good Anglo-Catholic if he lived in England, but he is not a bad man and I will sodding well defend him if anyone says he is. He made a mistake, that's all. A big one, but he didn't kill anyone, he didn't steal from anyone, he didn't cause an earthquake or start the Irish Potato Famine (I am not Irish) because of not bothering to do something he could do, he just did too much. Si Trew (talk) 15:59, 4 October 2016 (UTC)[reply]
WOOHOO
I believe this job is finally done, having just deleted the last few hundred x1/g6 neelix redirects. It was a long tough slog, and I don't even know how many people were involved in total, but it looks like it is finally completed. Beeblebrox (talk) 00:31, 11 November 2016 (UTC)[reply]
Yep. Now we have to decide where to go from here. There are ~30,000 Neelix redirects which were edited by someone other than Neelix before this mess was discovered. I think these should be at least skimmed. @Anomie:, is it possible/practical to create this list? Secondly, we need to advertise the completion to a wider audience so that they can audit our work, and make sure the decisions made were reasonable. Thirdly, if we do not feel that we need to check the remaining 30K redirects, we need to talk about repealing X1. Pinging others who have worked hard on these redirects (not an exclusive list) @Tavix:, @Nyttend:, @Legacypac:, @SimonTrew: Please ping anyone whom I have forgotten. Tazerdadog (talk) 03:16, 11 November 2016 (UTC)[reply]
