Wikipedia:Bots/Requests for approval/JL-Bot 3
- The following discussion is an archived debate. Please do not modify it. Subsequent comments should be made in a new section. The result of the discussion was Approved.
Automatic or Manually Assisted: Automatic (operator initiated & supervised)
Programming Language(s): Perl (uses perlwikipedia)
Function Summary: Remove no longer valid orphan templates from articles
Edit period(s) (e.g. Continuous, daily, one time run): Daily to weekly
Edit rate requested: 10 accesses per minute (built in delay of at least 6 seconds between any read or write)
Already has a bot flag (Y/N): Y
Function Details:
This is in response to this bot request. It will iterate through articles tagged as orphans, check the number of other articles that link to it, and remove the tag if it is no longer applicable.
It uses 3 links as the minimum number required before removing the orphan tag. However, that can be easily adjusted up based upon community input.
It also remembers the list of pages checked. It will only re-check a page after a set amount of time has passed. It is currently using 15 days, but I anticipate adjusting that (up or down) as I get a feel for how often orphans are resolved.
Discussion
editWhat about disambiguation pages? — Werdna talk 01:20, 13 April 2008 (UTC)[reply]
- I handled that by upping the the minimum number of links to 3 from the recommended 2. That allows there to be a disambig page and 2 links. This seemed a better solution than retrieving each page as it reduces the complexity and the amount of page reads. If there is concern about multiple disambig pages linking, I could raise it to 4.
- If the preference is to do an actual check of the pages, I can do that. -- JLaTondre (talk) 13:15, 13 April 2008 (UTC)[reply]
Never mind, I was thinking it was adding the tags too, and that disambig pages were going to get a bunch of those tags. But it's removing the tags. I don't see any other problems, any objections to a trial run? — Werdna talk 00:18, 14 April 2008 (UTC)[reply]
- For automatic removal I think the incoming link count should be higher. When you do the trial, can you use a higher limit, say > 4 to remove the tag? And just to be clear, you're only counting incoming links from article space, right? Gimmetrow 21:10, 14 April 2008 (UTC)[reply]
- No problem. I'll switch to that not only for the trial, but also the ops if approved. I'm fine with being conservative. And yes, it only counts incoming article links -- what links here filtered to the main namespace minus named disambig pages [those with (disambig), (disambiguation), (name), & (surname) in the title] minus the pseudo-namespaces [those starting with CAT:, MOS:, P:, & Transwiki:]. Even though the latter shouldn't ever link to an article, I ignore them. -- JLaTondre (talk) 23:07, 14 April 2008 (UTC)[reply]
- If not already using it sounds perfect for the backlinks api call, makes it real easy to filter out namespaces. ex ample Q T C 23:17, 14 April 2008 (UTC)[reply]
- Agreed. However, I'm using perlwikipedia which has a built-in what_links_here call. It saves me from parsing the results myself, but it hasn't been rewritten to use the API for that call yet. -- JLaTondre (talk) 23:39, 14 April 2008 (UTC)[reply]
- If not already using it sounds perfect for the backlinks api call, makes it real easy to filter out namespaces. ex ample Q T C 23:17, 14 April 2008 (UTC)[reply]
- No problem. I'll switch to that not only for the trial, but also the ops if approved. I'm fine with being conservative. And yes, it only counts incoming article links -- what links here filtered to the main namespace minus named disambig pages [those with (disambig), (disambiguation), (name), & (surname) in the title] minus the pseudo-namespaces [those starting with CAT:, MOS:, P:, & Transwiki:]. Even though the latter shouldn't ever link to an article, I ignore them. -- JLaTondre (talk) 23:07, 14 April 2008 (UTC)[reply]
- Any objections to a trial run? — Werdna talk 02:55, 15 April 2008 (UTC)[reply]
- Given that the BOTREQs only real problems was the number of incoming links that it considered orphan, I'd think that to do a trial with the discriminator set to 4 incoming links would be a good and accurate test, I'll be bold and say Approved for trial (20 edits). Please provide a link to the relevant contributions and/or diffs when the trial is complete. Q T C 03:54, 15 April 2008 (UTC)[reply]
- Trial done. Looked through all 20 edits and didn't see a problem. -- JLaTondre (talk) 00:50, 16 April 2008 (UTC)[reply]
- I particularly like the clear edit summary. The only thing I noticed, is that you will pick up some articles (e.g. Barney Boko) that are linked to in templates (e.g. {{Dandy}}), but not necessarily in article text. I know that the definition of orphaned doesn't make mention of those cases, but it might be something that should be considered. I guess I would question the definition versus intent of "orphaned." This bot expertly enforces the definition, but does it address the intent (I don't know, just throwing it out there). It's not a deal breaker for me, but we might want to think about it. - AWeenieMan (talk) 03:46, 16 April 2008 (UTC)[reply]
- Correct, it counts article links that occur via template inclusions. This makes sense to me. The point of links is to provide navigation between articles. It doesn't seem like how the link occurs should matter. Template links still add to the web of linkage and allow people to find articles they are interested in. Template links are just as valuable a navigation element as in paragraph links in my opinion. -- JLaTondre (talk) 00:59, 17 April 2008 (UTC)[reply]
- I particularly like the clear edit summary. The only thing I noticed, is that you will pick up some articles (e.g. Barney Boko) that are linked to in templates (e.g. {{Dandy}}), but not necessarily in article text. I know that the definition of orphaned doesn't make mention of those cases, but it might be something that should be considered. I guess I would question the definition versus intent of "orphaned." This bot expertly enforces the definition, but does it address the intent (I don't know, just throwing it out there). It's not a deal breaker for me, but we might want to think about it. - AWeenieMan (talk) 03:46, 16 April 2008 (UTC)[reply]
- Trial done. Looked through all 20 edits and didn't see a problem. -- JLaTondre (talk) 00:50, 16 April 2008 (UTC)[reply]
- Given that the BOTREQs only real problems was the number of incoming links that it considered orphan, I'd think that to do a trial with the discriminator set to 4 incoming links would be a good and accurate test, I'll be bold and say Approved for trial (20 edits). Please provide a link to the relevant contributions and/or diffs when the trial is complete. Q T C 03:54, 15 April 2008 (UTC)[reply]
Reviewed, and I have no objections to a full approval. Does anybody have any problems with this bot? — Werdna talk 01:15, 18 April 2008 (UTC)[reply]
- Nope! Approved. SQLQuery me! 14:12, 19 April 2008 (UTC)[reply]
- The above discussion is preserved as an archive of the debate. Please do not modify it. Subsequent comments should be made in a new section.