This is more for wikiHow staff, when they get time to code more bots
Basically, I think we can do with a bot that removes links to nonexistent pages and templates, and automatically updates double redirects so they point at the intended target.
These two tasks are very tedious and can probably be replaced with a bot. It would be nice because we then can have less WikiGnome tasks to do. The bot would remain asleep just looking at recent changes, Special:WantedPages, Special:WantedCategories, Special:WantedTemplates, and Special:DoubleRedirects and wake up as soon as a deletion/move occurs. Then, the bot removes all the red links and fixes double redirects and goes back to sleep, waiting to be woken up when another red link/double redirect comes by.
Unlike what we have right now, this would be almost instant (within minutes of a red link happening), so it will not be noticeable by anyone. Plus, we can have the bot’s edits autopatrolled so they do not fill up recent changes.
Sometimes nonexistent pages are linked to deliberately, such as in {{split}} tags. Also, sometimes pages are moved with no redirect, sometimes the pages are still relevant, even if they were moved, in which case updating the link might be better than removing it, even if there isn’t a redirect. People also sometimes accidentally break links. A common one I see is people adding a space in category tags, e.g. [[Category: Recipes]] instead of [[Category:Recipes]], having them fixed manually might work better. I can see why fixing double redirects could work, it would mean fewer tasks, and I can’t think of anything that could go wrong with that, especially as we have a RedirectsBot already.
This sounds like a good idea.
MediaWiki has a built in feature for automatically fixing double redirects whenever you move a page, but it’s disabled by default, because most wiki’s let anybody move pages. However, on wikiHow, since page moving is restricted to admins and boosters, we probably would not have an issue with it.
This feature is described in the documentation here
and here
.
I think that there shouldn’t be a bot that removes red links. Sometimes, there just a small capitalization error that can be easily fixed, and removing them would cost a link weaver some more time. And regarding what @RubyRoseRain
, it’s not really easy to remove red links from templates.
@R2_d2000
I agree. There aren’t many double redirects, and takes less than 5 minutes to fix them all.
That would probably be a good idea. See if we can get this enabled here. @JayneG
There is actually a magic word for detecting whether a page exists or not: {{ #ifexist:
{{{1}}}|[[{{{1}}}]]|{{{1}}}}}
I’m going to preface this by saying I’m running on four hours of sleep and am trapped in 80 to 110-degree temperatures with broken air conditioning, so I’m incredibly irritable right now and that’s going to leak into my response.
We don’t need this. We already have a bot that will eventually handle redirect links that aren’t updated when pages are moved - it’s not immediate but it gives people time to update or remove links on their own accord, which is useful when a page is retitled to the point that a link becomes unrelated and needs to be removed. And I’ve never seen so many title changes that the subsequent double redirects become unmanageable. (And again, sometimes double redirects need to be removed or are no longer interchangeable, and using a bot to fix these would potentially tie up non-interchangeable titles.) Some people also like doing these mundane tasks - I did it regularly for awhile and was actually disappointed I couldn’t do it on Wikipedia when I tried contributing there because it was one of my go-to tasks here.
Ruby’s already pointed out why automatically removing broken links wouldn’t work as well. If someone inadvertently breaks a link (which I know I’ve even done myself a few times), a bot firing to remove it could result in articles missing links that are actually useful for readers.
Aasim, I’m going to be extremely blunt: Stop suggesting changes for the sake of changing or automating things, or making us more like Wikipedia. We’re deliberately
not like Wikipedia and deliberately don’t use the same tools or bots as them at times to promote a friendlier community, and switching everything to being done by bots or software tools does not mesh with wikiHow culture. Our engineering team is also not that big and has to prioritize things based on what’s necessary
and widely benefits users
, readers and editors alike. I strongly
suggest taking a step back from suggestions for at least a month - if something is extremely necessary during that time, then another editor can suggest it.
@Galactic-Radiance
it is completely understood that we are not Wikipedia
. I have understood this from the start. I can tell that you are a bit irritated. I am a bit too (partly because my sleep schedule is thrown off (I normally am awake by 08:00, not 00:00), and partly because I am trying to figure out how to deal with an IP on Wikipedia). I can also say that I am not suggesting changes for the sake of suggesting changes. The reason why I have not been editing for the past several days is because I have been working on testing and writing user scripts on Wikipedia. I have also been busy with end-of-school-year stuff, particularly because I am a senior in high school.
I have still dropped in to vote in NFD guardian. And with regards to “I strongly suggest taking a step back from suggestions for at least a month - if something is extremely necessary during that time, then another editor can suggest it.”, I have taken a step back anyway… I only made this
because of a problem with a new editor using AutoWikiBrowser. And FYI, I have only made two suggestions in the past month.
I only made this as feedback for the wikiHow staff so that I and other good-faith contributors do not have to waste any time on something this tedious and do something more productive like patrolling recent changes or improving articles. If you disagree, that’s fine, I am open to feedback.
On a tangent, relating to your last statement, I think what @Chris-H
said a while back about suggestions to all of us may be an excellent idea; just resist the urge, note it down somewhere (maybe in our own userspace), and revisit them after this pandemic is over and when everyone is able to be more productive (i.e. when teams can meet in person and work on issues again):
So basically, in a nutshell, maybe it would be good if all of us just held off on proposals until everything gets back to normal. I would like to see if (at least temporarily) we should lock threads related to major changes (like this) until after the situation improves. Thoughts @JayneG
or @Chris-H
?
I completely understand that as well. But not everyone does. But don’t they get tiring after a while?
JayneG
11
Yes, stopping these sorts of feature requests is the plan for now. Though we haven’t made a specific standalone thread, it’s widely knownWe will not likely be locking threads, as I don’t think that’s necessary at this stage.
I remember at one meetup, me and another wikiGnomer spent a good 10 min grinding through a huge list of 75+ double and broken redirects. Then we both high fived each other and chugged a beer after feeling a pretty awesome sense of accomplishmentA thoughtful proposal for sure, but no matter how tedious a task is, human effort is always better than automation and scripts!
@Galactic-Radiance
: on a Wikipedia tangent, yes, there are still double redirects to fix on Wikipedia. It took a bit of scrolling to find one though (like on the second/third page of 500 redirects).
