Wikipedia talk:Cross-namespace redirects
From Wikipedia, the free encyclopedia
Comments and/or proposed amendments are appreciated. --Zoz (t) 15:05, 24 July 2006 (UTC)
- Have you added this to Centralized discussion? It's supposed to be a clearinghouse for changes to deletion standards, above all else, from what I understand. -- nae'blis (talk) 18:31, 24 July 2006 (UTC)
Feedback is welcome. If you don't like this proposal, please leave a comment and explain why, or alternatively, fix it right away. --Zoz (t) 16:19, 29 July 2006 (UTC)
[edit] Searching vs linking
This proposal seems to consider CNRs only from the aspect of the search bar. However, wikilinks behave differently and show up red no matter whether a similar term exists in a different namespace. Some CNRs are type-in (like cat:csd and the other uncontroversial shortcut redirects), but Articles for deletion has often been used as a direct link on talk or project space, not as a type-in shortcut. How should links to shortcuts work, especially if a term exists both in article and in shortcut space? Kusma (討論) 15:36, 24 July 2006 (UTC)
- People use Articles for deletion and AFD for shortcuts because they are there - it saves typing a few characters. If they noticed they were red links then they would fix them... There's no excuse really not to use the WP: shortcuts for this. /wangi 15:43, 24 July 2006 (UTC)
- AFD is not a shortcut. I agree that if CNRs should go, they need to be redlinks instead of {{deletedpage}} (because that creates situations like Userbox, which is a blue link: a self-referential page talking about Wikipedia in article space without the benefits of a cross-space redirect). Kusma (討論) 15:57, 24 July 2006 (UTC)
- I just want to add: I absolutely hate {{deletedpage}}. Deleted CNRs would stay as redlinks of course. --Zoz (t) 16:00, 24 July 2006 (UTC)
- AFD is not a shortcut. I agree that if CNRs should go, they need to be redlinks instead of {{deletedpage}} (because that creates situations like Userbox, which is a blue link: a self-referential page talking about Wikipedia in article space without the benefits of a cross-space redirect). Kusma (討論) 15:57, 24 July 2006 (UTC)
- I think the main usability problem here is when newbies can't find Wikipedia:Articles for deletion simply because they forget/don't know they should type Wikipedia:. As for linking, I do remember a discussion at WP:RfD about redirects from months like March 1426. There were consensus to delete them because when the editor notices they're redlinks, they would correct it to the proper target. I think making linking easier is not as important as making navigation easier (searching). Besides, links to Articles for deletion could be corrected by an automated bot even if Articles for deletion doesn't exist as a redirect. --Zoz (t) 15:53, 24 July 2006 (UTC)
- All of the links to Articles for deletion have already pretty much been corrected; there used to be a lot more links than there are now. Someone ran a bot on it (I don't remember who, could've been me). --Cyde↔Weys 16:02, 24 July 2006 (UTC)
- And what newbie is going to type in "Articles for deletion" - to know it's called that assumes prior knowledge. Someone new to WP is going to be searching for delete article, article deletion, delete a page and the like - not our esoteric term! /wangi 16:04, 24 July 2006 (UTC)
- There could be redirects from Wikipedia:Delete_article and Wikipedia:Article_deletion and Wikipedia:Delete a page. They would be able to find it. --Zoz (t) 16:14, 24 July 2006 (UTC)
- And every other possibility? There are hunners of them! /wangi 16:18, 24 July 2006 (UTC)
- Just like with other redirects, if it's implausible that anyone would type it, then we won't create them, but from a couple of plausible titles (like the ones you mentioned) we could create redirects. It would be much better for newbies compared to the current situation - when typing delete article doesn't take you anywhere. --Zoz (t) 16:47, 24 July 2006 (UTC)
- And every other possibility? There are hunners of them! /wangi 16:18, 24 July 2006 (UTC)
- Agreed, once someone already knows it's called "Articles for deletion" I think they will be able to find it. I just searched for "Articles for deletion" and the second search result was Wikipedia talk:Articles for Deletion, which immediately leads you there. The cross-namespace redirect is unnecessary. I really don't understand why people insist on putting non-encyclopedic content in the encyclopedic namespace. --Cyde↔Weys 16:06, 24 July 2006 (UTC)
- I suppose you found Wikipedia talk:Articles for Deletion because you have "search in Wikipedia_talk:" on. Newbies with default settings don't have that on. --Zoz (t) 16:14, 24 July 2006 (UTC)
- I don't know if proposing to change the defaults is a good idea; Wikipedia is first and foremost an encyclopedia, and the search should only return encyclopedic results by default. Maybe we can discuss modifying the default search options for logged-in users (pure readers probably aren't looking for Articles for deletion anyway). But I find it inexcusable that, when someone has setup their search options to specifically search just the encyclopedic content and none of the Wikipedia stuff, they will get hits on Wikipedia: stuff anyway because someone has made an XNR that effectively makes the search filter useless. --Cyde↔Weys 16:23, 24 July 2006 (UTC)
- I completely agree. Read the proposal: "Modify the default search filter for logged-in users [...] These settings could be changed anytime at Special:Preferences."; "The default search filter for users searching Wikipedia anonymously would stay as article namespace-only". If someone sets up their search options to search just the encyclopedia, it won't return WP stuff at all. That's what this proposal attempts achieve, anyway. --Zoz (t) 16:31, 24 July 2006 (UTC)
- This is putting the cart way ahead of the horse, but I can see two situations: either you reset everyone's Search Preferences to follow the new behavior, thus "breaking" years of habit, or you only implement the new Search Preferences for new users, in which case the impact of the helpful software change is reduced proportional to how many people will actually see it. I've got no strong opinion on this as such yet, but I figured it was worth mentioning... -- nae'blis (talk) 02:33, 25 July 2006 (UTC)
- My original idea was to implement this change only for new users. But yes, it might be better to change the search filters for all users who have the default settings at the time of its implementation and notify them once at the "login successful" page (at their first login after the change) with something like "Please review your search preferences at Special:Preferences/Search". Editors don't have the default (main namespace only) settings would not be affected by this change. (I think it wouldn't be a good idea to change the settings for those who have already altered them to according to their preferences.) --Zoz (t) 12:04, 25 July 2006 (UTC)
- This is putting the cart way ahead of the horse, but I can see two situations: either you reset everyone's Search Preferences to follow the new behavior, thus "breaking" years of habit, or you only implement the new Search Preferences for new users, in which case the impact of the helpful software change is reduced proportional to how many people will actually see it. I've got no strong opinion on this as such yet, but I figured it was worth mentioning... -- nae'blis (talk) 02:33, 25 July 2006 (UTC)
- I completely agree. Read the proposal: "Modify the default search filter for logged-in users [...] These settings could be changed anytime at Special:Preferences."; "The default search filter for users searching Wikipedia anonymously would stay as article namespace-only". If someone sets up their search options to search just the encyclopedia, it won't return WP stuff at all. That's what this proposal attempts achieve, anyway. --Zoz (t) 16:31, 24 July 2006 (UTC)
- I don't know if proposing to change the defaults is a good idea; Wikipedia is first and foremost an encyclopedia, and the search should only return encyclopedic results by default. Maybe we can discuss modifying the default search options for logged-in users (pure readers probably aren't looking for Articles for deletion anyway). But I find it inexcusable that, when someone has setup their search options to specifically search just the encyclopedic content and none of the Wikipedia stuff, they will get hits on Wikipedia: stuff anyway because someone has made an XNR that effectively makes the search filter useless. --Cyde↔Weys 16:23, 24 July 2006 (UTC)
- I suppose you found Wikipedia talk:Articles for Deletion because you have "search in Wikipedia_talk:" on. Newbies with default settings don't have that on. --Zoz (t) 16:14, 24 July 2006 (UTC)
- There could be redirects from Wikipedia:Delete_article and Wikipedia:Article_deletion and Wikipedia:Delete a page. They would be able to find it. --Zoz (t) 16:14, 24 July 2006 (UTC)
[edit] Software changes, but... in the short term?
This essay seems to focus on software/config changes which could be made to help resolve this problem. While this is good in the longer term it would be nice in the short-to-medium term to address the current issue.
Is there consensus to avoid cross-namespace redirs 99% of the time? Can we put together a guideline to that effect...
I'll get me coat! /wangi 15:48, 24 July 2006 (UTC)
- I don't think there's consensus to avoid CNRs. The 'keep' arguments are actually quite strong, and I don't think deleting all CNRs without these software changes would solve anything. Obviously I'm not a WP developer, but from my experience in programming in C++, I suppose that that these source code changes could be done relatively quickly. As for the related changes in the database - I don't know. But given a debate of this scale, I think there is no short-term solution. --Zoz (t) 16:09, 24 July 2006 (UTC)
- Where have you been looking? There is a consensus to avoid XNRs; most of the ones brought up on RFD end up being deleted. Also, there's WP:ASR, and despite people making the outrageous claim that "it's not policy, it's just a guideline, therefore I am free to totally ignore it", it makes a compelling case for why the separation of encyclopedic content and unencyclopedic content is important. I have a simple question that no one has been able to answer yet, and I'd be glad if you could. The namespaces were originally created because Wikipedia-specific stuff was being created and it was getting confused with the encyclopedic content. Once the namespaces were created, the non-encyclopedic content, you know, the stuff that doesn't belong in an encyclopedia, was moved to Wikipedia:. Why oh why should we violate that separation of namespaces by having a redirect to every Wikipedia:XX page from XX? It makes no sense and I've seen at least half a dozen examples now of XNRs camping on names that were later turned into valid articles once I went through and deleted the XNR. --Cyde↔Weys 16:14, 24 July 2006 (UTC)
- We should not ignore guidelines, but weigh their importance and their applicability to a specific topic. Per WP:ASR and the "purity" of encyclopedia space, we should (for example) delete all instances of {{deletedpage}} right now. We don't do that because {{deletedpage}} has some uses, and we have some blatantly nonencyclopedic pages like Brian Peppers in the encyclopedia namespace. Would you want to delete Brian Peppers? Kusma (討論) 16:25, 24 July 2006 (UTC)
- Look, Cyde, I think your arguments for deleting the CNRs are really compelling, but the fact is there are lots of editors who don't want them simply deleted (in short-term, without software changes), and there are a couple of valid and strong arguments (usability, WP:BITE) for keeping them. I think there is no consensus to delete all of them right now, but I hope with the help of this proposal they can be dealt with accordingly. --Zoz (t) 16:27, 24 July 2006 (UTC)
- I do believe I've already explained why WP:BITE isn't relevant here. --Cyde↔Weys 17:13, 24 July 2006 (UTC)
- Actually I agree with you there, but I think this proposal is still preferable over the "delete all CNRs right now" solution, because it's more friendly to newbies (Okay, let's call it "newbie-friendliness" or "usability" instead of WP:BITE.) and at the same time it keeps the namespaces separated, and thus solves the "mirror issue", etc; and pretty much everything from here. --Zoz (t) 17:34, 24 July 2006 (UTC)
- I do believe I've already explained why WP:BITE isn't relevant here. --Cyde↔Weys 17:13, 24 July 2006 (UTC)
- Where have you been looking? There is a consensus to avoid XNRs; most of the ones brought up on RFD end up being deleted. Also, there's WP:ASR, and despite people making the outrageous claim that "it's not policy, it's just a guideline, therefore I am free to totally ignore it", it makes a compelling case for why the separation of encyclopedic content and unencyclopedic content is important. I have a simple question that no one has been able to answer yet, and I'd be glad if you could. The namespaces were originally created because Wikipedia-specific stuff was being created and it was getting confused with the encyclopedic content. Once the namespaces were created, the non-encyclopedic content, you know, the stuff that doesn't belong in an encyclopedia, was moved to Wikipedia:. Why oh why should we violate that separation of namespaces by having a redirect to every Wikipedia:XX page from XX? It makes no sense and I've seen at least half a dozen examples now of XNRs camping on names that were later turned into valid articles once I went through and deleted the XNR. --Cyde↔Weys 16:14, 24 July 2006 (UTC)
[edit] Aid in accidental linking?
I've seen people bring this up occasionally but nobody's ever explained what it means or why it's a good thing. Care to explain? --Cyde↔Weys 17:33, 24 July 2006 (UTC)
- I think what Kusma means is when someone wants to link Wikipedia:Articles for deletion and they accidentally link it as Articles for deletion, they would still be able to arrive at Wikipedia:Articles for deletion without changing the link. That's simply an (admittedly not too strong) argument to keep CNRs. However, as I explained here, this proposal would not keep CNRs, and accidental linking would be corrected by an automated bot and/or users correcting themselves after they realized it's a redlink. --Zoz (t) 17:42, 24 July 2006 (UTC)
- Related to the above, there is an RFD underway right now for Wikibreak. Someone might well type the term into the search engine (as indeed I did) and hope to find a page explaining what the term means. In fact, there is no such page ... instead, the redirect goes to Wikipedia:Wikiholiday which is a term I rarely hear used. The apparent assumption is users will magically know to type in "Wikipedia:Wikiholiday". I actually saw someone post a comment (in another cross namespace debate) that Wikipedia shouldn't make things easy for its editors. I think that's utter bollocks, personally. We should make things as convenient as possible, and I've yet to read any argument that persuades me that having a redirect that, say, allows someone to type "3RR" into the search engine and be sent to the page on the Three Revert Rule, thereby saving time and keystrokes, in anyway harms Wikipedia. 23skidoo 18:21, 25 July 2006 (UTC)
- However, there is a redirect from Wikipedia:Wikibreak to Wikipedia:Wikiholiday, so a user searching the project space shouldn't have any problems. I don't think the accidental linking is a reason to support CNR's; the onsite search being broken so often is certainly more persuasive for me. -- nae'blis (talk) 22:26, 25 July 2006 (UTC)
- Related to the above, there is an RFD underway right now for Wikibreak. Someone might well type the term into the search engine (as indeed I did) and hope to find a page explaining what the term means. In fact, there is no such page ... instead, the redirect goes to Wikipedia:Wikiholiday which is a term I rarely hear used. The apparent assumption is users will magically know to type in "Wikipedia:Wikiholiday". I actually saw someone post a comment (in another cross namespace debate) that Wikipedia shouldn't make things easy for its editors. I think that's utter bollocks, personally. We should make things as convenient as possible, and I've yet to read any argument that persuades me that having a redirect that, say, allows someone to type "3RR" into the search engine and be sent to the page on the Three Revert Rule, thereby saving time and keystrokes, in anyway harms Wikipedia. 23skidoo 18:21, 25 July 2006 (UTC)
[edit] Reason removal
While it's nice to be quoted, and I think the majority of CNR's are harmful (for many of the reasons listed), I've removed one of the reasons referened to me as it has been subsequently shown (by Splash) that I mis interpreted the page in question. On further reading, the legally questionable bit of verbatiam copying is more aimed at cases where a mirror changes an article from "blah blah blah This nonexistent article blah blah blah" to "blah blah blah This nonexistent article blah blah blah". Regards, MartinRe 23:13, 27 July 2006 (UTC)
- That's ok, but do you think this proposal could work? Any suggestions on how to improve it? Since you were quite active in the discussions about CNRs, I'd like to hear your opinion on this. --Zoz (t) 16:19, 29 July 2006 (UTC)
- I think creating a namespace for short cuts is a good idea, as often when searching for something, many of the first results are all pointers to the same article, it would be nice to have a google ways of "ignore duplicate pages" (or filtering on the target of the redirect). However, I think the default search for logged in users should remain just the encyclopedia, as just creating an account shouldn't modify a user's search functionality. (To me, having a different default for anon and logged in users is not the least surprising option - log in, and your search function changes?). I think many of my reasonings against CNR's are in the proposal, except for maybe the comment that "when searching encyclopedia Bitianica, you would not expect to find minutes of a Bitianica board meeting, so why should it be that way for wikipedia?". Also, to reply to the WP:BITE suggestion above, I think the opposite is the case. Deletion of CNR's isn't biting newcomers, the fact that the encyclopedia and wikipedia are not interchangable is an important concept, but blurring the line with CNR's is giving newcomers the wrong impression from the start, and that is doing them a dis-service by creating confusion in the long run. Also, I believe that if someone has not selected "wikipedia" in their search, we should not return wikipedia results. Yes, they may have not selected it as a newbie, or it may be an experienced user who has forgotten to tick the box, but both have the option of ticking the box and searching. However, for the reader who has deliberately unselected the box because they don't want wikipedia results, it is unfair to return them anyway. With CNR's, a reader can choose to search wikpedia by ticking the box, but they have no option not to do so. Without CNR's, readers wishing to search wikipedia can do so, and those who do not want to can also do so. Thus, the existance of CNR's removes choice, for the minor convience of a subset of editors, which is not a valid trade off in my view. It is important to remember that we are creating an enclyopedia that is freely available, so entwining it with one particular host (wikipedia) goes against that principle in my opinion. I want the enclyopedia I contribute to to work just as well, no matter who it's hosted by, now, or in the future. Personally I think any non-WP: style enclyopedia redirects should either be re-targeted to remain in the enclyopedia, or, if no valid target exists, deleted. (retargeting any incoming links, as appropiate). I'd also support it becoming a speedy criteria, for the simple reason that currently if an editor, well meaning or otherwise, creates dozens of redirects out of enclyopedia space, they each have to spend the week on rfd, which seems a little overkill. As an alternative, something that might be considered is whether CSD G4 can apply to redirects. Surely if say, speedy delete is deleted by rfd, speedy deletion which contains the identical content/target is a valid G4, but the impression I've received is that this might be contravertisal to some, and each redirect needs to be rfd'd. Anyway, hope that expands somewhat, unfortunately I'm busier than normal so my contribution time has been limited recently. Regards, MartinRe 19:08, 29 July 2006 (UTC)
- Yes, we have a problem with the principle of least astonishment here: what's better, to have a clueless newbie enter "Articles for deletion" in the search bar and get no results whatsoever (even though they entered the correct phrase!), or to have different default search filters for logged-in and anon users? (Obviously changing the default search filter for both anons and logged-ins would be overkill.) Neither option is without disadvantages, however I'd favour having a different default for anon and logged-in users because I think the vast majority of newly registered users registered to edit the Wikipedia (and therefore interested in the encyclopedia-making process, so showing them the Wikipedia:Articles for deletion as default would be justified), and the vast majority of anon users very rarely (just typo fixes, etc) or don't edit Wikipedia at all (so showing them unencyclopedic content as default would not be justified). Of course, we're talking about the default values here, so if anybody wanted to modify their search filter, they'd be able to do that. --Zoz (t) 19:33, 31 July 2006 (UTC)
- I think creating a namespace for short cuts is a good idea, as often when searching for something, many of the first results are all pointers to the same article, it would be nice to have a google ways of "ignore duplicate pages" (or filtering on the target of the redirect). However, I think the default search for logged in users should remain just the encyclopedia, as just creating an account shouldn't modify a user's search functionality. (To me, having a different default for anon and logged in users is not the least surprising option - log in, and your search function changes?). I think many of my reasonings against CNR's are in the proposal, except for maybe the comment that "when searching encyclopedia Bitianica, you would not expect to find minutes of a Bitianica board meeting, so why should it be that way for wikipedia?". Also, to reply to the WP:BITE suggestion above, I think the opposite is the case. Deletion of CNR's isn't biting newcomers, the fact that the encyclopedia and wikipedia are not interchangable is an important concept, but blurring the line with CNR's is giving newcomers the wrong impression from the start, and that is doing them a dis-service by creating confusion in the long run. Also, I believe that if someone has not selected "wikipedia" in their search, we should not return wikipedia results. Yes, they may have not selected it as a newbie, or it may be an experienced user who has forgotten to tick the box, but both have the option of ticking the box and searching. However, for the reader who has deliberately unselected the box because they don't want wikipedia results, it is unfair to return them anyway. With CNR's, a reader can choose to search wikpedia by ticking the box, but they have no option not to do so. Without CNR's, readers wishing to search wikipedia can do so, and those who do not want to can also do so. Thus, the existance of CNR's removes choice, for the minor convience of a subset of editors, which is not a valid trade off in my view. It is important to remember that we are creating an enclyopedia that is freely available, so entwining it with one particular host (wikipedia) goes against that principle in my opinion. I want the enclyopedia I contribute to to work just as well, no matter who it's hosted by, now, or in the future. Personally I think any non-WP: style enclyopedia redirects should either be re-targeted to remain in the enclyopedia, or, if no valid target exists, deleted. (retargeting any incoming links, as appropiate). I'd also support it becoming a speedy criteria, for the simple reason that currently if an editor, well meaning or otherwise, creates dozens of redirects out of enclyopedia space, they each have to spend the week on rfd, which seems a little overkill. As an alternative, something that might be considered is whether CSD G4 can apply to redirects. Surely if say, speedy delete is deleted by rfd, speedy deletion which contains the identical content/target is a valid G4, but the impression I've received is that this might be contravertisal to some, and each redirect needs to be rfd'd. Anyway, hope that expands somewhat, unfortunately I'm busier than normal so my contribution time has been limited recently. Regards, MartinRe 19:08, 29 July 2006 (UTC)
[edit] Support
I fully support this proposal. It will allow people to find the pages they are looking for outside the article space easily without redirecting to non-encyclopedic content from the main namespace. Polonium 15:18, 29 July 2006 (UTC)
[edit] Major issue with removal
Some of these redirects are very old and were in many cases created before the namespaces existed. Extreme caution must be taken so as to not delete any of those (unfortunately, many of them have probably already been lost...recovering them might make an interesting WikiProject). For an example, see Requested articles. If the redirects were moved to the correct namespace and then deleted (e.g. Wikipedia:Requested articles/archive or something like that), that would be fine (right?). Ardric47 04:29, 31 July 2006 (UTC)
- Or better yet (if someone has the time), the edit histories could be merged. Ardric47 04:30, 31 July 2006 (UTC)
- I've made a change in the proposal to reflect this. --Zoz (t) 19:15, 31 July 2006 (UTC)
- A proposed clarification: I would suggest erring on the side of merging rather than deleting (regarding "useful" edit history). Keeping histories is important for GFDL/copyright purposes. Ardric47 06:16, 1 August 2006 (UTC)
- I've made a change in the proposal to reflect this. --Zoz (t) 19:15, 31 July 2006 (UTC)
[edit] Another reason to consider
Another argument against allowing redirects out of the main namespace is that they encourage original research. Editors think that because they hit an internal concept, peculiar to Wikipedia and not documented elsewhere, via a name in the main namespace, it is all right to add other such original research concepts to Wikipedia. For example: The existence of the redirect from Wikiholic to Wikipedia:Wikipediholic encouraged someone to write Wikibyss (AfD discussion). (From Talk:Wikibyss: "The process of being 'caught in a Wikibyss' as it is termed here is an actual phenomena, I believe fairly common to regular Wikipedia users. I just happened to coin the term for it. Because of the nature of Wikipedia, I thought this was the best way of popularising it, and encouraging people to use it.") Uncle G 09:46, 2 August 2006 (UTC)
[edit] I don't really like this idea that much ...
If someone is searching in the "search" box for "Pokemon", they should NEVER get WP:POKE or WP:PTEST among the results unless they specifically select the Wikipedia namespace. The default search really shouldn't include anything other than article space.
The WP: and CAT: links definitely need to be a shortcut or alias namespace for Wikipedia: and Category: ... just like Image is an alias for Bild on German Wikipedia. That doesn't really need a policy, I wouldn't think. It obviously needs to be done - a glut of them sometimes show up in searches.
Personally, I think the most FUNDAMENTAL Wikipedia pages need either dab links or cross-namespace redirects. That doesn't mean we need redirects for everyone's pet wikiproject ... but the most essential processes and policies would definitely be helpful. BigDT 01:51, 8 August 2006 (UTC)
- The big problem here is that the new users will edit with default settings. And they will search for "articles for deletion" and similar WP-specific phrases and they won't get results unless they go to the preferences and change the search filter settings. And you can't expect that from a newbie, just as you can't expect them to prefix "Wikipedia:" to their searches. On the other hand, readers who don't want to edit WP at all (and therefore didn't bother to create an account and log in) would have article namespace-only default search settings according to this proposal. --Zoz (t) 16:37, 9 August 2006 (UTC)
[edit] Status quo is fine
In the long term I'd like to see cross-namespace redirects removed, but the effort required to update links and keep new links updated may be substantial. On the other hand, some redirects clearly might refer to legitimate subjects and need to go. I think RfD does a good job of distinguishing these currently and there's no need for a change. In particular, I would strongly oppose a speedy deletion criterion in light of the dissent surrounding this issue - we don't need another T1. Deco 02:01, 8 August 2006 (UTC)
- So far RfD deleted each and every cross-namespace redirect and there is some controversy because of that. You see, RfD deletes CNRs - users create CNRs (see a recent example). Actually, the idea behind this proposal is to abolish the need for cross-namespace redirects: if newbies get to their targets directly without even touching a cross-namespace redirect (read the proposal, that's what it aims to accomplish), there would be no need to create them. In fact, since noone would ever use them anymore there would be 0 controversy about deleting all of them - so comparing this to T1 is unjustified. And status quo is not fine: currently some users complain that deleting useful CNRs is against newbies and usability, others say that polluting the encyclopedic article namespace with non-encyclopedic cross-namespace redirects is against the very goal of wikipedia. In my opinion status quo is really far from being "fine". --Zoz (t) 16:57, 9 August 2006 (UTC)
Establishing a speedy criterion on the basis of a technical wishlist is just ... weird. -Splash - tk 22:13, 9 August 2006 (UTC)
- Once noone uses CNRs, deleting them would be nothing more than housekeeping. The proposed addition to the speedy criterion is just to ensure that CNRs wouldn't have to go through RfD if this proposal is accepted. --Zoz (t) 14:58, 13 August 2006 (UTC)
- No, these will not qualify under speedy-deletion case G6 as "housekeeping". Please go re-read the speedy-deletion case and, in particular, read all the discussion and precedent that's in the archives about how that specific case is to be used. Even if you orphan every one of them, these will not qualify for speedy-deletion under any existing criterion. Rossami (talk) 04:32, 14 August 2006 (UTC)
- I know, and that's why I proposed a change in R2. What I meant above, is that these deletions would be "non-controversial maintenance tasks". --Zoz (t) 12:34, 14 August 2006 (UTC)
- Maybe we don't need a new CSD, just an application of the existing ones. Right now, if an article "rubbish article" get's deleted, then if the same content is at "rubbish article II" then it get's G4'd. However, if the content is a redirect, somehow a new discussion seems to be needed (c/f rfd of wikipedia is not and rfd of what wikipedia is not (point at the same place, with the same content) Right now, anyone could create a dozen redirects to WP:NOT, "what wikipeida isn't", "wikpieida isn't this", etc, and each and every one of them would be required to go though a week long discussion on rfd. That to me, makes no sense. Even if existing CNR's have to be rfd'd (despite idential ones being previously rfd'd) surely new CNR's should class as a speedy for that reason alone? Regards, MartinRe 00:24, 17 August 2006 (UTC)
- The point is that if this proposal is accepted and there is no need for CNRs anymore, they should be deleted. The deletion of these redirects would be non-controversial so I think it would be reasonable to delete them without listing them at RfD (there are lots of CNRs, imagine all of them listed at RfD at the same time). On the other hand, as Rossami pointed out, currently there is no CSD criterion under which they could be deleted. So I think it's sensible to expand one of the criteria to include explicitly this kind of situation. By the way, there is a discussion about applying G4 to redirects here, but I think that's not relevant here, because there are lots of CNRs which haven't been deleted before (so G4 wouldn't apply), but would have to be deleted anyway. --Zoz (t) 12:39, 21 August 2006 (UTC)
- Maybe we don't need a new CSD, just an application of the existing ones. Right now, if an article "rubbish article" get's deleted, then if the same content is at "rubbish article II" then it get's G4'd. However, if the content is a redirect, somehow a new discussion seems to be needed (c/f rfd of wikipedia is not and rfd of what wikipedia is not (point at the same place, with the same content) Right now, anyone could create a dozen redirects to WP:NOT, "what wikipeida isn't", "wikpieida isn't this", etc, and each and every one of them would be required to go though a week long discussion on rfd. That to me, makes no sense. Even if existing CNR's have to be rfd'd (despite idential ones being previously rfd'd) surely new CNR's should class as a speedy for that reason alone? Regards, MartinRe 00:24, 17 August 2006 (UTC)
- I know, and that's why I proposed a change in R2. What I meant above, is that these deletions would be "non-controversial maintenance tasks". --Zoz (t) 12:34, 14 August 2006 (UTC)
- No, these will not qualify under speedy-deletion case G6 as "housekeeping". Please go re-read the speedy-deletion case and, in particular, read all the discussion and precedent that's in the archives about how that specific case is to be used. Even if you orphan every one of them, these will not qualify for speedy-deletion under any existing criterion. Rossami (talk) 04:32, 14 August 2006 (UTC)
-
-
-
-
-
- Actually there is more to do than just accept the proposal - as it would require changes to the setup of wikipedia - you would need not only community support but also support from amongst those tasked with maintaining the system (of course they generally are helpful to the community and only object when there are technical reasons not to). This argument for speedy should wait until the proposal has approved - and implementation has been done - now it is moot - and deletion of CNR's now is just annoying those that use them --Trödel 15:30, 21 August 2006 (UTC)
-
-
-
-
[edit] Alternate solution: Stop deleting cross-namespace redirects
I think this page is a start at fairly laying out the advantages and disadvantages of cross-namespace redirects. I think, however, that the proposed solution is overly complex, error-prone and expensive. This is a solution in search of a problem. I remain unconvinced by the blanket arguments against cross-namespace redirects. Many such redirects are not only harmless but actively meet some of the specific reasons at WP:R for not deleting redirects. RfD has recently been deleting these redirects but I believe that they are doing so only by being very selective in their reading of Wikipedia policy and tradition. I'm trying very hard not to be inflammatory in my language but I honestly believe that damage has been done to the project through this misguided crusade. Rossami (talk) 04:10, 11 August 2006 (UTC)
- Although I support this proposal, I agree that some damage has been done—specifically, with respect to redirects with histories. For example, Be bold in updating pages was recently deleted. I have suspicions (but obviously can't check!) that it contained history, especially since Wikipedia:Be bold in updating pages has no history before 25 February 2002. Or did we not have "be bold" yet then? Ardric47 04:36, 11 August 2006 (UTC)
- History is a potential issue with any deletion. There are means of dealing with it and if it wasn't done in this case, then that's more an issue of improperly following the procedures for capturing history before deletion than about the deletion itself. -- JLaTondre 14:59, 11 August 2006 (UTC)
[edit] Recent additions to "Arguments for keeping CNRs"
It seems to me that this argument is mistaken in its basic premise. I know some mirrors contain Talk & Project pages. I thought some mirrors also contain User pages (why else the {{Userpage}} template?). I don't think a claim that mirror readers will never see them is correct. However, I don't think it's any worse for a mirror reader to see them than a direct Wikipedia reader. I recommend that this argument be re-written along the lines of "Redirects which are used exclusively on non-article pages are irrelevant as readers of the article-space only will never fall into this "pipework" because these "cracks" (or more accurately, "access ports") are only being left in the maintenance corridors." We can debate how strong that argument is, but it wouldn't be factually wrong. -- JLaTondre 14:59, 11 August 2006 (UTC)
- My addition was intended as a rebuttal to the last bullet in the prior section which says that these redirects are bad because they are "broken" when copied to a mirror. My point is that if the mirror is not bringing over the User, Talk or Wikipedia-spaces, that argument is largely moot. No reader will see them. If the mirror is bringing over all the project-spaces, then the redirect won't be broken - again, making that argument largely moot. If a mirror is, for some inexplicable reason, bringing over Talk pages but not Wikipedia-space pages, the reader of the mirror will be faced with many thousands of broken links regardless of whether these redirects exist or not.
If there's a better way to word the thought, please do so. Rossami (talk) 16:47, 11 August 2006 (UTC)- How about "Redirects which are used exclusively on User, Talk and other project pages are irrelevant as readers of the article-space only will never see them. This is especially true for mirrors which only duplicate the main article namespace ." Also, I think the last point under "Arguments for deleting CNRs" should be changed to "Some mirrors...". -- JLaTondre 18:05, 11 August 2006 (UTC)
- Grammatically, that sentence parses to "Redirects ... are irrelevant" which was not quite the intent. I think we are trying to say that "The arguments against redirects are irrelevant" in that specific case. How about "For redirects which are used exclusively on User, Talk and other project pages, the first and fourth arguments against CNRs are irrelevant as readers of the article-space only will never see them. This is especially true for mirrors which only duplicate the main article namespace."
That wording seems overly specific and could be disrupted if the bullets in the section above are reorganized. A cleaner solution might be to reword the bullets in the top section to include the necessary qualifiers. Rossami (talk) 19:00, 11 August 2006 (UTC)- Instead of "Redirects ... are irrelevant", how about ""Redirects ... do not cause confusion"? I also do not like using the bullet numbers. Depending upon the qualification, it may be problematic to add as not everyone may agree with it. -- JLaTondre 00:10, 12 August 2006 (UTC)
- Grammatically, that sentence parses to "Redirects ... are irrelevant" which was not quite the intent. I think we are trying to say that "The arguments against redirects are irrelevant" in that specific case. How about "For redirects which are used exclusively on User, Talk and other project pages, the first and fourth arguments against CNRs are irrelevant as readers of the article-space only will never see them. This is especially true for mirrors which only duplicate the main article namespace."
- How about "Redirects which are used exclusively on User, Talk and other project pages are irrelevant as readers of the article-space only will never see them. This is especially true for mirrors which only duplicate the main article namespace ." Also, I think the last point under "Arguments for deleting CNRs" should be changed to "Some mirrors...". -- JLaTondre 18:05, 11 August 2006 (UTC)
[edit] Absurd
The idea that cross-namespace redirects are so harmful that they should be banned is absurd. The entire Wikipedia project is based on the collective contributions of thousands of editors. Making it harder for new editors to find the resources they need to contribute productively is simply an asinine thing to do. Cross-namespace redirects exist because they are useful. Something useful should not be removed from the project without a good reason. No reason has thus far been offered. Not only that, but some folks apparently think this proposal is policy, or at least feel the urge to implement it immediately. This is a bad thing, because it short-circuits the process for determining whether these redirects should exist or not. IMHO, cross-namespace redirects are, at best, extremely helpful and useful, and at worst, merely annoying to a few easily-annoyed editors. There is no harm to the project caused by these redirects. It is absurd to pretend that they cause any significant damage. --71.36.251.182 17:43, 21 August 2006 (UTC)
[edit] Cracks in the Floor
The arguments for CNRs talks about people accidentally falling into cracks in the floor of a building. It says the cracks are there on purpose so it's easier for the engineers to get around. Thing is, no one really accidentally falls through those cracks - only the engineers know where they're located. -Zapptastic (talk) 21:02, 24 August 2006 (UTC)
- Not quite, the floor plan (search engine) also knows where they are, so a guest can ask for directions to the resturauant and end up in the kitchens. And they can't even say "I'm a guest, only tell me about guest things" (i.e. set the search filter up) because it's included in guest handbook, even though it's only applicable to the employee handbook. Regards, MartinRe 14:17, 30 August 2006 (UTC)