Wikipedia:Bots/Requests for approval/AnomieBOT 50

The following discussion is an archived debate. Please do not modify it. To request review of this BRFA, please start a new section at

WT:BRFA. The result of the discussion was

Approved.

Operator: Anomie⚔

Automatic or Manually assisted: Automatic, unsupervised

Programming language(s): Perl

Source code available: User:AnomieBOT/source/tasks/ReplaceExternalLinks3.pm

Function overview: Replace {{cite web|url=http://www.nr.nps.gov}} with {{NRISref}}, update certain other NRIS links, and tag remaining NRIS links with {{NRIS dead link}}.

Links to relevant discussions (where appropriate):

Edit period(s): One-time run

Estimated number of pages affected: about 30000

Exclusion compliant (Y/N): Yes

Already has a bot flag (Y/N): Yes

Function details: The bot will make the following changes to correct obsolete links to the National Register of Historic Places (NRHP)'s NRIS database (and perform other small cleanup):

• Links beginning with "http://www.nr.nps.gov/multiples/" will be replaced with the corresponding link starting "http://pdfhost.focus.nps.gov/docs/NRHP/Text/".
• {{cite web}} templates with url http://www.nr.nps.gov/ will be replaced with {{NRISref}}, with |1= chosen depending on the value of the existing |date=.
• {{cite web}} templates with other urls under that domain will have {{NRIS dead link}} placed after them.
• Any remaining links to http://www.nr.nps.gov/ or subpages will have {{NRIS dead link}} placed after them.
• Usage of term "{{convert|.9|acre}}" within NRHP infoboxes will be converted to "less than one acre" to correct previous error in interpretating NRIS database field (where in fact .9 value has different meaning than .4 or any other)

Discussion

Can we first rename to "NSISref" to NSIS ref" or "NSIS reference" or anything less cryptic? -- Magioladitis (talk) 11:46, 9 January 2011 (UTC)[reply]

I personally don't care, although I personally question whether "NRIS ref" is really less cryptic than "NRISref". I've invited Doncram (who made the original bot request) to comment. Anomie⚔ 14:57, 9 January 2011 (UTC)[reply]

I dunno. I started to comment here that spelling out "NRIS reference" rather than "NRISref" for new insertions of this would be fine by me, but it occurs to me that there are now already a bunch of articles (a few hundred) showing NRISref, and the Elkman generator is being used to produce more, and I don't myself have access to change that. Elkman might or might not prefer this other version and be willing to change it, upon request. I suppose there could be or would have to be a redirect from one to the other (i think a redirect of template would work, right?) to allow both versions to exist.

I guess now seeing new, spelled-out "NRIS dead link" does sorta seem to make it more suitable to have this be less cryptic, too, if u are looking at the two of them at the same time. But the usage out in several hundred articles now, soon about 30,000 articles, and eventually in 55,000 more articles to be created in the next year or two, is all going to be with just the NRISref reference alone. After some dead links are cleaned up, the other template will not appear at all. BTW it is definitely not necessary to spell out NRIS, which once meant "National Register Information System" though, as i think that is a meaningful acronym on its own.

I am not aware of any comparable reference templates that this could/should be named like. I just found Template:Cite DANFS for ships, but that is different, it is an adaptation of cite web, while NRISref invokes cite web. I don't see the NRISref as being out of line, against anything else i know of.

The changeover to the "NRISref" template was discussed for a longish time at

wt:NRHP

and no preference for spelling it out came up. I guess i myself suggested NRISref and kept it short, out of my view that it is short and "elegant" maybe. It's the name for a template; it is not viewed by encyclopedia readers; it is technical in nature.

So i am not immediately supportive of changing the template name. I have been working on this changeover for many months, and am thrilled to have gotten this far with Anomie's willingness and abiility to take care of it, and partly I would just like it done. I grant there is no rush really, however, if both I and Anomie keep staying available. Aargh. Lemme suggest: Magioladitis, if u feel strongly about this, can u explain more analytically why you feel that way here, and would you bring it up in the main discussion at

wt:NRHP and at User talk:Elkman, to invite further comments here? Or maybe M doesn't feel strongly, and Anomie can just proceed as planned. --Doncram (talk) 16:31, 9 January 2011 (UTC)[reply

]

I am the bot operator who did the url rewriting last august. I replaced about 1920 links . I missed no links as Anomie guessed, but i skipped the replacement of all links to documents where the new one had no content. Not all documents were digitalized at that time and you got a pdf document exactly like this example (which is from nomination). Now it seems that more documents are digitalized in between.

My script also checked if e.g. http://www.nr.nps.gov/nomination/79002143.pdf could be replaced by https://npgallery.nps.gov/NRHP/GetAsset/NRHP/79002143_text but for nomination i always got that placeholder document. Because there are only three nomination links left now another person must removed nearly all of them.

btw: The webserver was performing very very bad.

issimo

02:24, 10 January 2011 (UTC)

Thanks for what u did before and for explaining. What you did sounds very sensible. Maybe some more MPS docs are being occasionally put into proper location over time. I am aware of some cases where digitized MPS doc once was valid i.e. was available at old URL, but did not get moved correctly by them so is unavailable now. Together with your explanation is adequate to explaining the 300 or so cases where the old URL is in wikipedia articles. It would be most helpful if these are labelled as dead links now. Thanks!

I proposed it because the current fashion in Wikipedia is to separate words when possible. (Deadend -> Dead end, Noreferences -> No references, etc.). Moreover, the dead link version of this one has the words already separated. This helps with code's readability. Anyway, I am good either way. -- Magioladitis (talk) 11:39, 10 January 2011 (UTC)[reply]

Well, I adjusted the bot to periodically check if {{NRISref}} is redirected somewhere else, so if you guys decide to move it the bot will go along with it automatically. Anomie⚔ 14:27, 10 January 2011 (UTC)[reply]

Thanks for explaining Magioladitis. As M is okay either way, and as I don't see benefits and do see some difficulty with a change now, i would rather then that we consider the NRISref name fixed and okay as is.

One last minute question: Anomie, could the bot possibly handle one other cleanup, to search the NRHP articles for occurrence of {{convert|0.9|acre}} in the infobox, and change that to less than one acre. As covered at User talk:Elkman#NRHP places having area of .9 acres, etc., questioned at User talk:Elkman#Areas of less than an acre, and explained by me, the .9 value in the area field is a data entry code that wasn't previously recognized by Elkman's generator. New articles get the "less than one acre" treatment from Elkman, properly, but it remains there are a couple hundred of the mistaken interpretation out in mainspace. If u can handle this noncontroversial cleanup, that would be great, but don't let it stop the main bot run. Thanks! Hoping all is good to go now. --Doncram (talk) 04:46, 12 January 2011 (UTC)[reply]

Sure, that's easy enough. Anomie⚔ 12:25, 12 January 2011 (UTC)[reply]

Great. Maybe that .9 acres change could be done within the NRHP infobox only, to avoid any possible other uses of {{convert|.9|acre}} out in the article. By searching/replacing on | area = {{convert|1|acre}} to change to | area = less than one acre. (Note it seems to be Elkman's preference to use that wording rather than to put in | area = less than {{convert|1|acre}}, perhaps to avoid some bad line-breaking that I've noticed the latter would introduce. So the "less than one acre" wording is good, IMHO.) Adding mention into statement of bot's function above; pls. feel free to revise that mention.

Then, is this just waiting for someone's approval? --Doncram (talk) 20:34, 13 January 2011 (UTC)[reply]

Approved for trial (50 edits). Please provide a link to the relevant contributions and/or diffs when the trial is complete. –xeno^talk 20:19, 20 January 2011 (UTC)[reply]

Trial complete. [1] Anomie⚔ 17:44, 21 January 2011 (UTC)[reply]

The link to trials edits doesn't work for me. I saw some of the Bot edits and noted there was a fix needed for one of the templates that the Bot put into place, but saw no problem with any of the Bot edits themselves. I noticed that it is worthwhile for me or other NRHP editors to review the Bot edit summaries, because those are informative, i.e. that in one article two different NRIS versions were edited (unanticipated, but correctly done). I think the trial worked fine and should be extended. --doncram 02:15, 24 January 2011 (UTC)[reply]

I changed the URL within Anomie's comment to use "w/index" rather than "wiki/wndex", following another example, now seems to work to show trial's edits. --doncram 02:23, 24 January 2011 (UTC)[reply]

This diff is one showing correct replacement of .9 acres by "less than one acre" phrase. This Bot edit inserted "no date specified" for a 2007 date that is invalid in terms of not being an expected version date. I think that is probably exactly what we asked for, what we wanted (but we want then to, ourselves, revise the NRISref template code to put that article into a cleanup category for our attention). Other edits insert a dead link notice, and insert version "2006a" as expected. As far as i can tell, the Bot ran fine. Thanks! --doncram 02:38, 24 January 2011 (UTC)[reply]

There was a slight hiccup with the bot output that wasn't including "version=____" as the first parameter; instead it was just putting a date code (e.g. "2006a"). I've modified the template code to accept either, so no further changes need to be made to the bot. Looks like a go for a full run to me.--Dudemanfellabra (talk) 03:16, 24 January 2011 (UTC)[reply]

Thanks for the template revisions, Dudemanfellabra! Thanks so much, Anomie! --doncram 03:23, 24 January 2011 (UTC)[reply]

• Trusted botop; necessary task; everything looks fine.  Approved. –xeno^talk 16:20, 28 January 2011 (UTC)[reply]

The above discussion is preserved as an archive of the debate. Please do not modify it. To request review of this BRFA, please start a new section at

WT:BRFA

.