Jump to content

Wikipedia:Bots/Requests for approval/RileyBot 9

From Wikipedia, the free encyclopedia

This is an old revision of this page, as edited by Pigsonthewing (talk | contribs) at 14:53, 30 March 2013 (Discussion: refuting more FUD). The present address (URL) is a permanent link to this revision, which may differ significantly from the current revision.

Operator: Riley Huntley (talk · contribs · SUL · edit count · logs · page moves · block log · rights log · ANI search)

Time filed: 03:05, Friday March 29, 2013 (UTC)

Automatic, Supervised, or Manual: Automatic

Programming language(s): Python

Source code available: N/A

Function overview: Looks at every page that transcludes {{Infobox NRHP}}, whenever it finds the | built = parameter, followed by a four-digit number, it encases the number in {{Start date}}. Whenever it doesn't find the parameter, or when it finds the parameter empty or followed by something other than a four-digit number, it does nothing.

Links to relevant discussions (where appropriate): Wikipedia:Bot_requests#Mark_a_lot_of_pages_for_microformatting.2C_round_2 and Wikipedia_talk:Bot_requests/Archive_2#RFC:_Deploying_.27Start_date.27_template_in_infoboxes

Edit period(s): One time run and whenever requested

Estimated number of pages affected: No clue, 1000+ (there is 40,000 pages that transclude {{Infobox NRHP}})

Exclusion compliant (Yes/No): Yes

Already has a bot flag (Yes/No): Yes

Function details: As stated in the function overview, make a list of pages that transcludes {{Infobox NRHP}}, if the bot finds a page that has the | built = parameter and is followed by a four-digit number, the bot adds {{Start date}} with the number merged in (e.g "| built = {{Start date|1989}}") If it doesn't find the parameter or when it finds the parameter empty or followed by something other than a four-digit number, it does nothing. -- Cheers, Riley 03:05, 29 March 2013 (UTC)[reply]

Discussion

Approved for trial (50 edits). Please provide a link to the relevant contributions and/or diffs when the trial is complete. MBisanz talk 04:49, 29 March 2013 (UTC)[reply]

Comment I believe this is the bot requested at wp:botrequests#Mark a lot of pages for microformatting, round 2, about applying "start" template to "built" field in NRHP infobox. I just noticed that and object. The bot request and this trial are being railroaded through, without notice to previous discussants, and over objections expressed by others too. I totally object. I suppose this should go to the Administrative Noticeboard, or to a new RFC, or something. It should not be railroaded through. --doncram 18:08, 29 March 2013 (UTC)[reply]

Followup About the sample, it is highly NON-random, being the first alphabetical items in the list of NRHP listings, so happens to be all items starting with numbers. And it omits, relatedly, all churches. There are many thousands of NRHP-listed churches, many started/founded at an earlier date than some later NRHP-listed building was built, e.g. many at First Presbyterian Church (disambiguation), for example.

For one of the items done by the bot, I have just removed the Start date template and corrected information in the article, in this diff. There, the NRHP infobox field was incomplete, showing 1929 when 1929-1931 would have been the correct information. It has not been settled what a start date should be, when a construction period spans years. Usually, perhaps most often, the NRHP infobox "built" field is incorrect in articles, for such cases. A category also was incorrect, stating that the building was completed in 1929 when it was completed in 1931, per corrected interpretation of the NRIS information.

I expect there are other problems in the first 60 done, but there would be more/different problems in a truly representative sample. I request that this go back to the bot request stage, or a general RFC specific to NRHP applications. The bot proponents have cited so-called community consensus on a previous RFC which was conducted with no mention and no awareness by any participant about NRHP application complications. --doncram 18:20, 29 March 2013 (UTC)[reply]

--doncram 18:20, 29 March 2013 (UTC)[reply]

Doncram, The randomness was in my selection from the articles used in the test; once again you've taken something out of context and made a molehill into a mountain. If infoboxes - or categories - contain bad information that's not the fault of this bot, the date template, or this proposal; and the errors can be corrected regardless of them. You have already been told taht an RfC has approved this task for all infoboxes with these datable microformats. You disrupted a previous version of this request with long screeds of opposition, demonstrating no understanding whatsoever of what was proposed or its benefits. Please refrain from doing the same here. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 18:47, 29 March 2013 (UTC)[reply]
The items done by the bot are NOT random. Look at it, those are the entries with names starting with numbers, and then a first few A acronym ones. That is not random selection from any list.
You, Pigsonthewing, brushed off valid objections in the previous bot request. At the current bot request, I ask the quite serious question to you, of whether you care or not about achieving accurate start date information. Seriously, there are some thousands of churches, schools, ships, libraries, other articles having NRHP infoboxes, where the infobox built date, even if accurate for its purpose there, is not clearly the relevant start date for the church, school, ship, library, etc. And is obviously the wrong date in the context of probably hundreds of these articles. That is different from the also-vexing problem of inaccurate content of the built date field. --doncram 19:15, 29 March 2013 (UTC)[reply]
Your first point is correct, but as no-one has claimed that the items done by the bot are random, tilts at windmills. Your tl;dr objections are not valid, and are about content, not this bot activity or the template used. Take them to a relevant venue. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 19:21, 29 March 2013 (UTC)[reply]
The bot says the 1935 Labor Day hurricane started in 1937! In this bot edit. Because the NRHP infobox correctly gives 1937 as the built date of a memorial covered in the hurricane article. Fairly often NRHP designations are included in articles in a secondary way, as here. This is not a "content error" in the NRHP built field, which is accurate. This is an error introduced by the bot.
The bot has not been approved by any general consensus, for the NRHP infobox application. A general RFC concluded that bot runs would have to be considered on a case-by-case basis: Wikipedia:Requests for comment/Microformats, as brought up by HellKnowz in Wikipedia:Bot_requests/Archive_51#Break for convenience. Pigsonthewing replied then with reference to anohter RFC, also not recognizing NRHP infobox content. I don't think it is right for a bot to march through 30-40,000 NRHP articles adding frequently-erroneous information. To the bot programmer, too, do you want that? --doncram 13:59, 30 March 2013 (UTC)[reply]
Your claim that "the bot says the 1935 Labor Day hurricane started in 1937" is false, as the subject of the infobox in which it was made (and therefore the name parameter of the relevant microformat) is "Florida Keys Memorial"; and you refer to the wrong, earlier RfC, which was about a different subject; and not bot runs. The bot in this BRFA will not add any information; it will mark up the information which is already present. Your continued objections, based on your false assumptions and gross misunderstandings, are disruptive. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 14:53, 30 March 2013 (UTC)[reply]