User talk:Trappist the monk

From Wikipedia, the free encyclopedia
  (Redirected from User talk:Monkbot)
Jump to: navigation, search
CS1 This user is responsible for those
CS1 error messages (help).
Comments are welcome. If your comments are about my work on a particular article, please make
them at the article's talk page so that everyone who has an interest in the article may participate.



Status of Monkbot activity on deprecated parameters?[edit]

Do you feel that Monkbot's tasks have done as much as they can to fix |coauthors= and |month= errors in Category:Pages containing cite templates with deprecated parameters?

A search indicates that there may be about 2,000 articles that still use |month=, and a random sampling of those shows month/year pairs that look easily fixable with a script.

I was observing the various Monkbot tasks' passage through the category to fix the coauthor errors, and it is not clear to me that each of the tasks has completed a full run through the category. The various tasks seemed to jump from one part of the alphabet to another.

Would you like help identifying patterns that might be fixable with adjustments to the bot tasks' code? If the bot thinks it has fixed all of the bot-fixable errors, should we ask the community for opinions on exposing the coauthors error messages?

Thanks for all of the work you do on the CS1 module and your bot. – Jonesey95 (talk) 02:37, 26 August 2014 (UTC)

Certainly getting there. If I recall correctly, the last runs were making less than 500 fixes. Not much, but 500 fixes here and there from time to time is 500 fixes. I'll start up Monkbot this morning and see what happens. It's been nearly a month since the last run.
Trappist the monk (talk) 12:21, 26 August 2014 (UTC)
Monkbot tasks 1–5 have just completed. There were 625ish edits among all of the tasks. At the time of the task 1 RFA Category:Pages containing cite templates with deprecated parameters contained 163,762 pages. As I write this, the category contains 31,217 pages.
It's possible that tasks 2–5 might be extended to handle |coauthor(s)= with more than 10 names. I haven't spent any time looking at what remains in the category to see why there are still |month= parameters lurking.
Trappist the monk (talk) 12:21, 28 August 2014 (UTC)
Fixed a bug in task 1 that prevented fixes when |month= is the last parameter before the citation's closing braces. So, rerunning task 1.
Trappist the monk (talk) 12:52, 28 August 2014 (UTC)
That explains it. I see that the bot is finding many more hundreds of month parameters to fix now, with version 1e of the task. – Jonesey95 (talk) 16:00, 28 August 2014 (UTC)
It looks like the bot has fixed a few hundred more month parameters, but there are still over 1,000 left (a crude search says 1,785, but I wouldn't bet on it). A search for incategory:"Pages containing cite templates with deprecated parameters" insource:/\|\s*month\s*=\s*[A-Z\d]/ will show you some articles that still contain |month=. Some of them are false positive search results that have |month= in a non-CS1 template; others will require manual fixing, like this one, which had a populated |date= along with |month= and |year=. But some are straightforward fixes. – Jonesey95 (talk) 20:33, 3 September 2014 (UTC)
What search tool supports regular expressions? I've been scanning through Category:Pages containing cite templates with deprecated parameters with an awb script that fixes some errors (|date= containing a properly formatted date followed by |month= and |year= – in either order – example). Mostly I just use awb to highlight |month= and then hand edit. Lots of these errors have |date= with a complete date, sometimes properly formatted, sometimes not, followed by |month= where the value is different from the month portion of the value in |date=example.
I've picked up about 500 pages so far with this script. There have been just a few that I think Monkbot should have caught but for the most part these are all errors that Monkbot couldn't / shouldn't fix because an editor needs to determine which 'month' is the correct 'month'.
Trappist the monk (talk) 21:22, 3 September 2014 (UTC)
The search example above works (for me) in the "New Search" available under Beta features (at the top of the page, next to Watchlist). There are supposed to be ways to do more complex searches, but they usually time out or give me an error. The one I pasted displays results pretty reliably. Some clever text processing of the search results, shown 500 results at a time, might result in a nice list of articles to feed to AWB (which I don't have the right OS to use, unfortunately). Search documentation here.Jonesey95 (talk) 21:50, 3 September 2014 (UTC)
I did a quick and dirty copy/paste/clean of the search results and put them here. Feeding that list to AWB might be fruitful. – Jonesey95 (talk) 22:07, 3 September 2014 (UTC)
I've been through that list and made changes to about 640 of the approximately 1520 pages (awb considers some of your list's 1700ish pages to be duplicates).
Trappist the monk (talk) 14:47, 8 September 2014 (UTC)
That's great progress. I'll chip away at the remaining list, which looks like it's under 1,000 by now. It shouldn't be too long before |month= can go the way of |day=. – Jonesey95 (talk) 18:02, 8 September 2014 (UTC)
I tweaked the script to add {{cite magazine}} which is an alias of {{cite journal}}. The modified script picked up 5 edits in the first 50 of your list. I think I'll make the same tweak to Task one and let Monkbot have a crack at it.
Trappist the monk (talk) 18:26, 8 September 2014 (UTC)
Yeah, so Monkbot got about 80 more from your list by adding {{cite magazine}} to task 1.
Trappist the monk (talk) 20:35, 8 September 2014 (UTC)

────────────────────────────────────────────────────────────────────────────────────────────────────I'm also seeing a non-trivial number of |month= in {{cite paper}}, {{cite document}}, and {{cite interview}}. It looks like the majority of the remaining instances have curly braces somewhere in the citation; my AutoEd script ignores them just fine and merges the month and year parameters. I am checking every edit visually before saving, though, in order to avoid false positives. – Jonesey95 (talk) 20:51, 8 September 2014 (UTC)

I added those three and Monkbot edited 90 pages. In the next days I'll add those four to tasks 2–5 an see where that gets us.
Trappist the monk (talk) 21:40, 8 September 2014 (UTC)

Coauthor parameter[edit]

I have just put a coauthor back into an article. I do not appreciate a bot removing coauthor when there is no consensus to do so.[1]. The solution of using editor2 is a particularly nasty one because it means that an editor has to have knowledge of how to change the ref= parameter if the long citation is linked to a short one , which is an unnecessary step if coauthors is used. Please remove this from your bot list until such time as you gain a consensus via a well advertised RfC to remove this parameter. -- PBS (talk) 21:01, 27 August 2014 (UTC)

Your reasoning for why Monkbot should not operate appears to be based on some false premise. The Monkbot tasks that replace |coauthor= or |coauthors= parameters with a series of |authorn= parameters (tasks 2, 3, 4, and 5) do not do the replacement if a CS1 citation template contains |ref=harv or if the citation template is {{citation}} which, by default, sets |ref=harv.
Your example citation uses |ref={{harvid|Neutzner|2010}}. Because the page also uses {{sfn|Neutzner|2010}} as the short-form citation, all of the expected links that were present and valid before Monkbot's edit were still present and valid after Monkbot's edit. The short- to long-form citation links for citations like your example, work regardless of whether the long-form citation uses |coauthor= or |author2=.
Monkbot did not break anything when it edited the article.
You can always place a {{bots|deny=<botlist>}} template.
Trappist the monk (talk) 23:22, 27 August 2014 (UTC)
The point is that the bot says it is depreciateddeprecated. The bot could just as easily write it is appreciated. Who has decided that coauthors is depreciateddeprecated"? -- PBS (talk) 08:23, 28 August 2014 (UTC)
The bot is fixing articles in Category:Pages containing cite templates with deprecated parameters. Note the word "deprecated", which is not the same as "depreciated". A deprecated function, in programming, is one that is no longer needed or supported. There is no value judgment associated with deprecation; it's just a standard programming practice.
|coauthors=, along with |month= and |day= are no longer needed, because the citation templates have been upgraded to function without them. Deprecation in code development happens all the time. The bot is simply performing what would otherwise be an onerous and tedious human task of converting these deprecated parameters into supported parameters. – Jonesey95 (talk) 16:00, 28 August 2014 (UTC)
I often make that spelling mistake. This project works on consensus, not on whimsical "deprecation in code development". "coauthors" and "coeditors" are both useful because they the do not invoke messing around with changing CITEREF values if author2 or editor2 are used insetead. Bots should not be making changes like these when there has not been an RfC to agree changes. -- PBS (talk) 21:50, 28 August 2014 (UTC)

September 2014[edit]

Hello, I'm BracketBot. I have automatically detected that your edit to List of Olympic medalists in short track speed skating may have broken the syntax by modifying 2 "{}"s. If you have, don't worry: just edit the page again to fix it. If I misunderstood what happened, or if you have any questions, you can leave a message on my operator's talk page.

It's OK to remove this message. Also, to stop receiving these messages, follow these opt-out instructions. Thanks, BracketBot (talk) 12:51, 5 September 2014 (UTC)

Hello, I'm BracketBot. I have automatically detected that your edit to Cancer Genome Anatomy Project may have broken the syntax by modifying 2 "[]"s. If you have, don't worry: just edit the page again to fix it. If I misunderstood what happened, or if you have any questions, you can leave a message on my operator's talk page.

List of unpaired brackets remaining on the page:
  • | author = G. Vasmatzis]], M. Essand, U. Brinkmann, B. Lee & I. Pastan

It's OK to remove this message. Also, to stop receiving these messages, follow these opt-out instructions. Thanks, BracketBot (talk) 16:21, 11 September 2014 (UTC)

Reference Errors on 18 September[edit]

Hello, I'm ReferenceBot. I have automatically detected that an edit performed by you may have introduced errors in referencing. It is as follows:

Please check this page and fix the errors highlighted. If you think this is a false positive, you can report it to my operator. Thanks, ReferenceBot (talk) 00:42, 19 September 2014 (UTC)

Deletion of content[edit]

Please do not make knee-jerk reversions of valid content as you did here. If you had made the slightest effort to check, you would have found numerous sources discussing this issue. I have added a couple now, but this sort of unthinking reversion just pisses people off and drives away new editors, so please be more careful in future. Thank you. — Preceding unsigned comment added by 31.51.131.96 (talk) 20:36, 19 September 2014 (UTC)

ISSNs[edit]

You seem to violate the AWB rule of use #4. Please do reconsider. Multiple incremental edits only flood watchlists and cover up vandalism. Regards. Materialscientist (talk) 21:17, 19 September 2014 (UTC)

The purpose of the ISSN changes I've been making is to move an ISSN number held within a CS1 template, but at the same time isolated from it, into the CS1 template's |issn= parameter so that the ISSN number is available to users by way of the citation's COinS metadata. Unlike the 'special' identifiers (ISSN, ISBN, PMC, etc), values assigned to |id= are not made part of the metadata.
This simple citation, which uses {{issn}}:
{{cite book |title=Title |id={{issn|14055066}}}}
generates this html and metadata:
<span class="citation book">''Title''. [[International Standard Serial Number|ISSN]]&nbsp;[//www.worldcat.org/search?fq=x0:jrnl&q=n2:14055066 14055066].</span><span title="ctx_ver=Z39.88-2004&rfr_id=info%3Asid%2Fen.wikipedia.org%3AUser+talk%3ATrappist+the+monk&rft.btitle=Title&rft.genre=book&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Abook" class="Z3988"><span style="display:none;">&nbsp;</span></span>
If I move the ISSN number from |id={{issn|14055066}} to |issn=14055066:
{{cite book |title=Title |issn=14055066}}
we get this html and metadata:
<span class="citation book">''Title''. [[International Standard Serial Number|ISSN]]&nbsp;[//www.worldcat.org/issn/1405-5066 1405-5066].</span><span title="ctx_ver=Z39.88-2004&rfr_id=info%3Asid%2Fen.wikipedia.org%3AUser+talk%3ATrappist+the+monk&rft.btitle=Title&rft.genre=book&rft.issn=14055066&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Abook" class="Z3988"><span style="display:none;">&nbsp;</span></span>
The key-value pair we're looking for is &rft.issn=14055066
In this case, {{issn}} did no harm but there are other templates that editors use that corrupt the COinS metadata because they include extraneous html and css markup. This markup is included in the metadata, most often in the title and author parameters – {{nihongo}}, {{au}}, {{smallcaps}} come to mind.
Making these edits also subjects the ISSN number to CS1's error checking and renders the number properly formatted.
To the naked eye, nothing much has changed. But now, users of the metadata have more complete citation data. Admittedly these are the minority users but, they are just as entitled to quality citations as those who consume Wikipedia with their eyes. I guess that the point from which you observe the edit results determines whether these edits amount to vandalism or wrongdoing.
Trappist the monk (talk) 00:12, 20 September 2014 (UTC)
On that note, you have cleaned up stuff that I did years ago in what is now the "wrong" manner; I did it that way because {{cite book}} didn't then support |issn= - it was added later. Good cleanup. --Redrose64 (talk) 10:45, 20 September 2014 (UTC)

Precious[edit]

Cornflower blue Yogo sapphire.jpg

templates
Thank you, helpful silent monk, for tireless quality contributions to templates, taking care of ships and launches, for cite repair, for pointing out problems precisely, for defining yourself here by your contributions alone, – you are an awesome Wikipedian!

--Gerda Arendt (talk) 07:35, 20 September 2013 (UTC)

A year ago, you were the 610th recipient of my PumpkinSky Prize, --Gerda Arendt (talk) 11:04, 20 September 2014 (UTC)

Julian Assange[edit]

Hi there, as a recent editor of the page in question, you may wish to contribute to the discussions: ==Merge discussion for Assange v Swedish Prosecution Authority ==

Merge-arrows.svg

An article that you have been involved in editing, Assange v Swedish Prosecution Authority , has been proposed for a merge with another article. If you are interested in the merge discussion, please participate by going here, and adding your comments on the discussion page. Thank you. prat (talk) 15:45, 20 September 2014 (UTC) prat (talk) 15:45, 20 September 2014 (UTC)

Queen Mary[edit]

I saw you reverted my revision on the RMS Queen Mary page to remove the definite article before the ship's name. I've been an ocean liner enthusiast for thirty years, and have shelves of books written by maritime experts, almost all of which use the definite article before a ship's name. However I know James Cameron made the choice in his 1997 film TITANIC to never use the definite article. It appears Wikipedia users have made a similar decision now. I'm not sure of the reason for this, but I wonder if this is perhaps a case of British English transitioning to general use in America as well?

Regardless, I just checked Wikipedia and saw that the preferred style here is not to use the definite article. Personally, I disagree with this, as it does not reflect the way writers in the field refer to ships. However I'll obviously go with the Wikipedia consensus.

However there were a number of other small revisions that were not to do with the definite article. I understand you couldn't undo only the definite article revisions, but I'll go back and redo the other changes. User:jamesluckard (talk) 20:34, 25 September 2014 (UTC)

Pretty much the whole guidance on the subject of the definite article is in Wikipedia:Naming_conventions_(ships)#Using ship names in articles and Wikipedia:WikiProject_Ships/Guidelines#Introductory_sentence. For the rational, you will have to troll the talk pages.
At the top of the Queen Mary article there is a {{Use British English}} template. See there? Did you see what I did? A definite article preceding a ship name. But take it out and the sentence would read like sammat written by country folk from the North of England.
Trappist the monk (talk) 22:59, 25 September 2014 (UTC)

Tools for using wikidata items as citations[edit]

Hi,

I've drafted a potential IEG on meta: Tools for using wikidata items as citations

A big part of the proposal involves writing a new template to format citations from a wikidata item, so any kind of input on the proposal from you would be very welcome! And if you wanted to join as a developer that would be even better :). Also let me know if there's anyone else I can contact or a good place to post messages in order to find people interested in participating, if you are not! Mvolz (talk) 19:56, 26 September 2014 (UTC)

I have no experience with wikidata except the bad taste in my mouth I get when I see interactions with it done through very user-unfriendly stuff like: {{#property:P214}} and local snaktype = entity.claims.p625[0].mainsnak.snaktype. We are now living in the 21st century, surely we can do better than this. To make wikidata readily usable by generic, non-technical editors, a raw citation template invocation must not look like {{cite book |wd-cite-id=p12345 |page=24}}. Editors reading that have no idea what it cites and that is just wrong.
Without having studied it closely, I don't see any obvious technical reasons why some mechanism that extracts citation data from wikidata can't be integrated into CS1.
You might try posting at Help talk:Citation Style 1 and or WT:CITE.
Trappist the monk (talk) 11:43, 27 September 2014 (UTC)

Link confusion[edit]

Any idea what's going on with this link http://www.pcquest.com/pcquest/news/176960/people-whove-shaped-the-internet ? — Cirt (talk) 00:49, 1 October 2014 (UTC)

Don't know. Looks like the server at pcquest is serving blank pages. Try again tomorrow. Perhaps by then they will have fixed it. Or not.
Trappist the monk (talk) 00:59, 1 October 2014 (UTC)
It's okay I found it confirmed in an archive database as well. Thanks for all your help, — Cirt (talk) 01:55, 1 October 2014 (UTC)