Wikipedia talk:Automated taxobox system/Archive 2

Archive 1

Archive 2

Archive 3

Archive 4

Archive 5

Disambiguation in taxonomy templates?

So I found Template:Taxonomy/Macroglossinae which refers to Macroglossinae, which is a disambiguation page pointing to Macroglossinae (Chiroptera) and Macroglossinae (Lepidoptera). There is currently Template:Taxonomy/Macroglossinae (Chiroptera) and Template:Taxonomy/Macroglossinae (Lepidoptera). Is there a way to make Template:Taxonomy/Macroglossinae disambiguate? Am i overthinking this? Nessie (talk) 16:47, 1 December 2017 (UTC)

@NessieVL: I think your overthinking this. As it stands, the link in the undisambiguated subfamily template is going to the disambiguation page. If anybody creates a creates a genus article with a template that has the undisambiguated subfamily template as a parent, they will create a link to the dab page that will show up in dab link reports (at which point it should theoretically get noticed and fixed by somebody). But it's probably best just to delete the undisambiguated template to prevent creating of ambiguous links in the first. Plantdrew (talk) 20:26, 1 December 2017 (UTC)

Sorry. I should have deleted that when I concluded it wasn't a good idea. William Avery (talk) 22:10, 1 December 2017 (UTC)

Related question: Alethopteris has | fossil_range = {{Geological range/linked|Pennsylvanian|Aptian}} in its automatic taxobox which shows up as a disambiguation link to Pennsylvanian. There is an appropriate link, Pennsylvanian (geology), but how can it be disambiguated in the infobox? Leschnei (talk) 02:21, 29 January 2018 (UTC)

@Leschnei: see the documentation at Template:Geological range/linked. Peter coxhead (talk) 22:06, 29 January 2018 (UTC)

@Peter coxhead: Perfect! Thank you, Leschnei (talk) 23:17, 29 January 2018 (UTC)

Automatic taxobox usage statistics update

The latest stats on automatic taxobox usage by WikiProject:

Project	Auto	Manual	Total taxa	Percentage auto	# auto added since 15 August
Algae	339	1371	1710	19.8	240
Amphibians and Reptiles	10960	2699	13659	80.8	461
Animals	819	3512	4331	18.9	91
Arthropods	324	5468	5792	5.6	38
Beetles	2484	23091	25575	9.7	24
Birds	6041	7626	13667	44.2	1162
Bivalves	80	1236	1316	6.1	8
Cephalopods	284	1012	1296	21.9	43
Cetaceans	319	9	328	97.3	17
Dinosaurs	1225	206	1431	85.6	52
Fishes	7942	12172	20114	39.5	3403
Fungi	193	11497	11690	1.7	50
Gastropods	7024	23349	30373	23.1	428
Insects	10574	44660	55234	19.1	231
Lepidoptera	16201	78413	94614	17.1	4575
Mammals	2466	4556	7022	35.1	227
Marine life	661	3804	4465	14.8	166
Microbiology	60	7924	7984	0.8	24
Palaeontology	6139	6431	12570	48.8	576
Plants	18518	40573	59091	31.3	7137
Primates	327	538	865	37.8	7
Rodents	1457	1614	3071	47.4	684
Sharks	686	48	734	93.5	541
Spiders	5099	2290	7389	69.0	116
Turtles	152	491	643	23.6	12
Viruses	NA	1415	1415	NA	NA
Total	91067	266198	357265	25.5	19784

Plantdrew (talk) 17:58, 18 December 2017 (UTC)

I'm real curious what those 9 cetaceans are. Nessie (talk) 13:46, 23 December 2017 (UTC)

Red pencil tool tip

Moved here from Template talk:Automatic taxobox as it's actually about all the automated taxoboxes.

Request to change {{{edit_link|e}}} [in {{Automatic taxobox}}] to read {{{edit_link|Edit}}}, as found in the | edit link = line.
Christopher, Sheridan, OR (talk) 09:03, 22 December 2017 (UTC)

Not done: please establish a consensus for this alteration before using the {{edit template-protected}} template.

I've had a look at your recent contributions history to try and find any discussion of this change, or find somewhere you've used this parameter and drawn a blank. Changing default values for parameters needs a bit more consideration. Cabayi (talk) 09:44, 22 December 2017 (UTC)

It was only my intent to call attention to 'e' when a mouseover is performed over the pencil icon. See Scientific classification section of {{speciesbox}} on plant articles (example, see Venus flytrap). It is clear the intent is "Edit" rather than simply "e", but I have neither the authority nor the interest in performing the edit myself (hence submitting the request). I'm only drawing attention to {{{edit_link|e}}}; thank you for acknowledging, that's all I wanted.

Christopher, Sheridan, OR (talk) 00:04, 23 December 2017 (UTC)

@DeNoel: I've been aware of this issue for some time (I do a lot of the maintenance on automated taxoboxes). There seems no good reason to show "e" as a tool tip for the red pencil icon rather than "edit". The slight complication is that all seven automated taxobox templates (see Wikipedia:Automated taxobox system) currently do the same. They could all be changed individually, but a better approach would be to fix this at a deeper level.

On looking further, {{Edit taxonomy}} is where the display actually happens. This can easily be fixed to show "edit" as the tool tip rather than "e", if this would be considered helpful. Peter coxhead (talk) 08:06, 23 December 2017 (UTC)

By the way, not all editors will see the tool tip. I normally have Peferences / Gadgets / Navigation popups turned on, when I see wikitext rather than tool tips where the link leads to an edit.

The red pencil icon at the end of this line uses a sandbox version to show "Edit" rather than "e" as the tool tip.

Is there any reason not to make this live? Peter coxhead (talk) 08:29, 23 December 2017 (UTC)

A minor quibbble. Users who aren't autoconfirmed are unable to actually edit taxonomy templates. Perhaps "View/edit taxonomy" would be better as the tool tip (even for autoconfirmed users, clicking the pencil doesn't immediately open an edit window). Of course, this isn't the only place on Wikipedia where users might see an "edit" link where they don't have permission to actually make edits. Plantdrew (talk) 15:39, 23 December 2017 (UTC)

Originally, as per the comment in the documentation at {{Edit taxonomy}}, it seems to have been thought that the tool tip should be short. I don't know why, or whether longer tool tips would be a problem for some editors. If length doesn't matter, then I agree that "View/edit taxonomy" is better. Peter coxhead (talk) 19:15, 23 December 2017 (UTC)

I agree with Peter coxhead; "View/edit taxonomy" makes more sense.

Christopher, Sheridan, OR (talk) 00:41, 25 December 2017 (UTC)

I like "View/edit taxonomy" as well. Kaldari (talk) 22:51, 23 March 2018 (UTC)

Due to the high impact of editing taxonomy templates, it has been decided to disallow new users to edit them.

Due to the high impact of editing taxonomy templates, it has been decided to disallow new users to edit them. when was it decided to disallow new users to edit them and where was the discussion 72.73.117.105 (talk) 21:26, 10 March 2018 (UTC)

Template-protected edit request on 7 May 2018

I want to change the parent taxon from the exclusive Dinosauriformes to the more inclusive Dracohors, based on Andrea Cau's classification. 4444hhhh (talk) 20:43, 7 May 2018 (UTC)

References

I've asked about this at WT:WikiProject Dinosaurs#Taxonomy question, as this is a taxonomic rather than a template editing issue. Peter coxhead (talk) 06:13, 8 May 2018 (UTC)

 Not done: please establish a consensus for this alteration before using the {{edit template-protected}} template. — Martin (MSGJ · talk) 10:48, 8 May 2018 (UTC)

Agreed. The immediate response at WT:WikiProject Dinosaurs#Taxonomy question is that this change shouldn't be made. Peter coxhead (talk) 12:25, 8 May 2018 (UTC)

Template-protected edit request on 7 May 2018 (2)

I want to change the parent taxon from the exclusive Toxicofera to the more inclusive Ophidiomorpha. 4444hhhh (talk) 20:44, 7 May 2018 (UTC)

Reference? I suggest you ask at WT:WikiProject Amphibians and Reptiles and/or WT:WikiProject Palaeontology, since this is a taxonomic rather than a technical template editing question. Peter coxhead (talk) 06:16, 8 May 2018 (UTC)

 Not done: please establish a consensus for this alteration before using the {{edit template-protected}} template. — Martin (MSGJ · talk) 10:49, 8 May 2018 (UTC)

Reverting to 'manual' Taxoboxes

Many thanks for all your work on the above Plantdrew - and your edits of my efforts with various plant and insect pages. The table reveals a particularly low take-up of Automatic Taxoboxes with fungi and I wonder if this partly due to resistance to change by editors of this group. My interest mostly relates to the Ganodermataceae and Clavicipitaceae (esp. potential biopesticide spp.). In the former, I was most recently checking the affiliations of Tomophagus cattienensis, which is of special interest here, and wrote a genus page for it - apparently being monotypic - I hope to provide a photo soon. Before writing this page, I felt that something was needed on Ganoderma colossus and found that pages for higher order taxa did not have automatic taxoboxes. I therefore edited the latter: only to find them reverted by Polyporales (who are you really by the way?), simply stating "revert to 'normal' taxobox" without giving any reasons. I gather that Nessie and others have had similar problems and I really think we ought to 'move on' to more important issues (such as including the many genera and higher orders that don't yet have pages for example!). I respectfully suggest that automatic taxoboxes are the 'new normal' and should be adhered-to in all but exceptional cases (as Plantdrew has discussed). In fact, there was a problem when I tried install an AT in the Ganoderma page (so didn't undo the revert), but in cases like this, wouldn't it be best to simply fix the 'bug'?
In my opinion, the whole automatic taxobox system and 'taxonbar' are wonderful additions to WP, simplifying the creation of new pages, enabling automatic updates from 'parents' as taxonomic changes are made ... and the many other advantages that others will understand better than me. They substantially strengthen the rigour of WP content by providing easy cross-checking with other databases (but some of which may also be wikis, by the way and I have now come across several conflicts between 'respectable' DBs). One of the main risks of course is vandalism: which again in my opinion, could be mitigated by strictly forbidding anonymous editing of Wikidata and Template:Taxonomy pages (forcing editors to fill-in the reference line would be good as well, but I know that can be inconvenient at times).Roy Bateman (talk) 07:33, 7 July 2018 (UTC)

I support the Auto Taxobox System. Plus having another set of eyes going through these articles to update other issues like say taxonbars improves the whole site. --Nessie (talk) 15:14, 8 July 2018 (UTC)

@Roy Bateman:, we should probably have an RfC at WP:TOL to establish a consensus to use automatic taxoboxes. I think it's basically a fait accompli at this point; automatic taxoboxes are the "new normal". But that happened without there really being any discussion, and there should have been discussion. I did ask about using them at WikiProject Plants before I started switching plants over, and had no objections, but not very many people chimed in there, and I'm not aware of any recent discussions in other Tree of Life subprojects. We've gone from 45,000 to 143,000 automatic taxoboxes since April 2017 (and I think most of the effort to get to 45k took place in six or so months prior to April 2017). It's all fairly recent. Up until March 28 2018, there was a message basically not to use automatic taxoboxes at all; I took it out in this edit, as it didn't reflect current practice.

My sense is that some of the previous objections to using automatic taxoboxes by gastropod, arthropod and fungi editors was that they were waiting to see if other Tree of Life subprojects were going to start using them as well. I don't want to force editors in those areas to use automatic taxoboxes, but maybe with some TOL-wide discussion they'll come around. Plantdrew (talk) 23:52, 10 July 2018 (UTC)

Ok @Plantdrew: and @Roy Bateman:, I did a thing: Wikipedia talk:WikiProject Tree of Life#Request for comments: Should the automatic taxobox system be the current recommended practice? --Nessie (talk) 14:21, 11 July 2018 (UTC)

Oh, I forgot to address Roy's concern about vandalism. Taxonomy templates can only be created/edited by confirmed users. While confirmation is a rather low bar (I think it takes 10 edits and 4 days), it does prevent drive by vandalism. The most highly used taxonomy templates (those with more than 1000 transclusions, I think) are fully protected and can only be edited by people with template editor rights. Plantdrew (talk) 15:26, 11 July 2018 (UTC)

Ranks

subform and subvariety are used (rarely) in botany. Lavateraguy (talk) 21:32, 14 July 2018 (UTC)

@Lavateraguy: yes, but we are very unlikely to have an article on a subvariety or a subform (will never have one?), so we don't need the taxobox system to cater for these ranks. Peter coxhead (talk) 09:14, 15 July 2018 (UTC)

Displaying the subgenus in the speciesbox

Sorry if this is answered somewhere: I really did look.

Is there a way to display the subgenus in a speciesbox or other automated taxobox? How would the Taxonomy templates work? --Nessie (talk) 14:45, 19 July 2018 (UTC)

@NessieVL:, there are two different ways to this.

1. Use |parent= and put in the name of the subgenus. Hit preview, and you'll have usual prompt to create a missing taxonomy template. This method can handle additional ranks between genus and species (e.g. section). All infrageneric ranks will be linked (and usually the links will be red). This method is used in 1368 articles. See Drosera capensis for a plant using this method (with a section as well), and Varanus timorensis for an animal.

2. Use |subgenus= and put in the name. No taxonomy template is needed for this method. Other ranks between genus and species are not supported. The subgenus can be linked or not; if you want it to be linked, you'll need to add the markup for a link. This method is used in 69 articles. See Colasposoma hajeki for an example using this method.

Note that article titles for animal subgenera are not standardized; see Category:Animal subgenera. Some are formatted as Foo (Bar) (more conventional in zoology, but kind of weird for Wikipedia where a parenthetical term is used for disambiguation), others are formatted as just Bar (which then necessitates "Foo (subgenus)" for the nominate subgenus). Plant subgenera are pretty consistently titled in the format "Foo subg. Bar"; see Category:Plant subgenera. Plantdrew (talk) 15:15, 19 July 2018 (UTC)

I can dig it. Thanks.

Subgenus, section, and subsection with same name

So, Pinus has a subgenus Pinus, a section Pinus, and a subsection Pinus. Wikipedia:Automated taxobox system/taxonomy templates § Title conventions recommends using the format genus (subgenus) for subgenera, but doesn't explicitly say what to do with sections and subsections (or series for that matter). It also mentions appending /qualifier to ambiguous names, and the example given for the qualifier is a rank name.

So I've combined those two patterns and created the taxonomy templates Pinus (Pinus)/subgenus, Pinus (Pinus)/sectio, and Pinus (Pinus)/subsectio. (I've stripped off Template:Taxonomy/ for brevity.) However, there is also the pattern shown by Pinus subg. Pinus and Pinus sect. Pinus, which date back several years. I created other templates with this pattern, like Pinus subsect. Strobus and Quercus ser. Virentes, before reading the documentation page.

So two questions. Is the title pattern Pinus (Pinus)/sectio what the documentation page is recommending? And if so, should pages with the other pattern, like Pinus sect. Pinus, be replaced and deleted? — Eru·tuon 07:16, 15 August 2018 (UTC)

The style X (Y) is definitely used for subgenus names under the ICZN, i.e. animals. ICNafp names, i.e. plants, fungi, etc., need a connecting term for ranks below species and one is often used for ranks below genus, such as subg., sect., subsect. I would always title the taxonomy templates accordingly. So all the titles you created on that pattern are correct. I wouldn't recommend any other pattern for 'plants'. I've set up the taxobox at Jack pine to appear in a self-consistent way, but we need a wider discussion on styling ICNafp names between genus and species, because at present styles are inconsistent (which I hadn't realized earlier).

P.S. I thought the ICNafp names with connecting terms at whatever rank were automatically italicized correctly, but they don't seem to be. This needs to be fixed. Peter coxhead (talk) 10:31, 15 August 2018 (UTC)

@Peter coxhead: I would be pleased if the genus were automatically abbreviated too (Pinus subg. Pinus → P. subg. Pinus).

I guess italicization is handled by the taxonLink function in Module:Autotaxobox, which just italicizes the whole taxon name (even if it contains subg., sect., and so on). My italicization function in Module:eFloras (see the testcases) uses a method that would work for cases like Pinus subg. Pinus or P. subg. Pinus: look for words ending in periods and un-italicize them if they are abbreviations of ranks. See Module:Autotaxobox/sandbox for a first draft of an improvement to taxonLink, though I don't know how to actually test it. ({{#invoke:Autotaxobox/sandbox|taxonomyList}} just uses Module:Autotaxobox's taxonLink function apparently.) — Eru·tuon 07:16, 17 August 2018 (UTC)

Sorry to be slow in replying; I've been away. There's a Lua module, Module:TaxonItalics, that has a function to handle this – it's used in Module:Taxonbar, for example. (It needs "subsection" adding, I see.) The taxonLink function in Module:Autotaxobox should use italicizeTaxonName for consistency throughout – this has been on my "to do" list for some time. It's important to ensure that any existing manual italicization doesn't produce erroneous output.

Abbreviating is something I've also thought about. There seems to be a consensus on abbreviating the genus name; for ranks below species, there seems less agreement on also abbreviating the specific name/epithet. Peter coxhead (talk) 06:27, 18 August 2018 (UTC)

@Erutuon: ah, I now see you were aware of Module:TaxonItalics. It deliberately doesn't handle cases like "Fragaria vesca subsp. vesca f. semperflorens" because this is not an acceptable name in the ICNafp. Infraspecific names here are Fragaria vesca subsp. vesca and Fragaria vesca f. semperflorens – only one infraspecific rank is allowed. The full string is a classification. Peter coxhead (talk) 06:38, 18 August 2018 (UTC)

I'd also like to handle not italicizing the hybrid symbol, ×. Some operating systems and fonts don't show this symbol differently when in italics, but others do, which is wrong. Again, a problem is dealing with all the different existing ways of getting round this issue. Peter coxhead (talk) 06:56, 18 August 2018 (UTC)

@Erutuon: I've set up a table at User:Peter_coxhead/Test#Module:Autotaxobox#taxonLink to test changes. The yellow shaded boxes in the live version needed fixing, the others needed to be left unchanged. The current sandbox version appears to behave correctly (your version fixed one of the subgenus cases but messed up the other). Peter coxhead (talk) 10:20, 18 August 2018 (UTC)

@Erutuon: after a lot of trial-and-error programming (sigh), I think I have now fixed Module:Autotaxobox and Module:TaxonItalics so that both connecting terms and hybrid symbols are correctly handled. The code might need a bit of polishing. Peter coxhead (talk) 19:02, 18 August 2018 (UTC)

Italicization of taxon names updated

Following some work by Erutuon and myself, the code that automatically italicizes taxon names in the automated taxobox system has been updated. It should now handle 'connecting terms' better in ranks between genus and species (e.g. Pinus subg. Pinus) and in ranks below species (e.g. Pinus sylvestris var. sylvestris). It should also not italicize the hybrid symbol, ×.

If you notice any new errors in italicization in automated taxoboxes, please leave a message here. Peter coxhead (talk) 19:16, 18 August 2018 (UTC)

Automated taxoboxes not using the matching taxonomy template?

Ok, found another mess under a rug. Hesperochernes was until today calling only

|taxon=animalia

. Is there a category or search that could show pages that don't use a taxonomy template that matches the page title? or something like that that would avoid common names? my PetScan-fu is too weak to figure it out. --Nessie (talk) 19:49, 27 August 2018 (UTC)

@NessieVL: I can't see how you could avoid English names, given that there's no consistent categorization of articles as using this form of title. Peter coxhead (talk) 07:24, 30 August 2018 (UTC)

Yea, figured it was nuts, but didn't hurt to ask.--Nessie (talk) 11:38, 30 August 2018 (UTC)

It's definitely not "nuts". It would be good to be able to identify for certain all organism pages at vernacular names. I'm wondering if this could be defined sufficiently well that a bot could run to ensure they were all put into some maintenance category. This would enable other checks as well. @Plantdrew: any ideas or comments? Peter coxhead (talk) 15:02, 30 August 2018 (UTC)

It's beyond my skills. PetScan is certainly not the right tool for this. Focusing only on Automatic taxobox (not Speciesbox or Taxobox), we'd want articles were |taxon= doesn't match the article title. But there are other reasons aside from vernacular titles where taxon wouldn't match title:

1. Disambiguated title, non-disambiguated taxonomy template; title collision with a topic that isn't an organism. Not an error.
2. Malplaced monotypic taxon; e.g. family title rather than genus for a monogeneric family. An error to be fixed.
3. Inconsistent titling of animal subgenera; subgenus Foo (Bar) may be at Bar or Bar (subgenus). Not presently regarded as an error.
4. Taxonomic inconsistency between title and automatic taxobox; e.g. title is a family name, but taxobox treats as subfamily (this situation will be more frequent with articles on species that have varying generic assignments). An error to be fixed.
5. Incorrect taxonomy template called (e.g. Hesperochernes). An error to be fixed.
6. Spelling difference between title and taxon. An error to be fixed.
7. |taxon= not specified. Not an error per se, but not good practice; occurs in a little over 3000 articles with automatic taxobox

Finding title/taxon mismatches will uncover many articles with vernacular name titles, a small number of errors, and at least one situation (case #1) with scientific name titles where the mismatch is OK. The errors can be fixed, but we can't automatically assume that a mismatch indicates a vernacular name title.

|name= is too inconsistently used to be helpful. Speciesboxes will have the same issues as automatic taxoboxes, but speciesboxes using |genus= and |species= need to have those values concatenated before comparing to the title.

Finding mismatches with manual taxoboxes would be very difficult. I guess a first pass would be to compare |binomial= (minus any markup for italics) to the title. There would be some errors; e.g. articles using |species=, but not |binomial=, or subspecies articles using |binomial=. For taxa above species with manual taxoboxes, I guess the first pass would be to find a bolded parameter value in the taxobox, strip off the bold/italic markup and compare that to the title. Cases of monotypy with multiple bolded values would be a problem, as would cases where the taxon was left unbolded.

It would be great if this could be done, but it certainly comes with a bunch of complications, and I don't what tools could used to run the necessary searches in the first place. Plantdrew (talk) 16:40, 30 August 2018 (UTC)

You've laid out the issues very comprehensively, making it seem very difficult or impossible to do, however useful. Pity. Peter coxhead (talk) 17:15, 30 August 2018 (UTC)

If we could figure out how to accomplish the search for taxon/title mismatches, it would be well worth doing. It'll be obvious at a glance which articles are using non-disambiguated scientific names as titles, and this will be indicative of an error. I'd estimate there are no more than 25,000 articles with vernacular name titles (there are ~16k bird+mammal articles which consistently use vernacular name titles, and ~32k herps+fish articles which I'd say use vernacular names less than 1/3 of the time, and neglible use of vernacular names for non-vertebrates). And most of those will be species, not higher taxa. A search for taxon/title mismatches for articles with {{Automatic taxobox}} would be quite useful. Plantdrew (talk) 18:18, 30 August 2018 (UTC)

This is yet another reason to replace manual taxoboxes with Automatic taxoboxes where possible. Loopy30 (talk) 18:53, 30 August 2018 (UTC)

It could be done with a module, if no bot owner wants to work on it. — Eru·tuon 21:58, 30 August 2018 (UTC)

Virusbox

In case you didn't see Wikipedia talk:WikiProject Viruses, @Bob the Wikipedian: has resurrected the Template:Virusbox (edit | talk | history | links | watch | logs) template and it seems to be fully functional. --Nessie (talk) 18:22, 9 October 2018 (UTC)

Template-protected edit request on 27 October 2018

Please put Dracohors as parent, as it appears to be the clade in which Dinosaurs emerged. See Dracohors. Jmv2009 (talk) 14:57, 27 October 2018 (UTC)

Done. Peter coxhead (talk) 15:35, 27 October 2018 (UTC)

Tetrapodomorpha and Pterosaurs

Tetrapodomorpha is now shown in all tetrapod articles. It's set to always display, which I assume is important for some stem tetrapods, and could be retained if the appropriate skip templates are in place at lower ranks. Template:Taxonomy/Tetrapoda, Template:Taxonomy/Teleostomi‎ and Template:Taxonomy/Sarcopterygii‎ were recently edited, but I'm not quite sure what needs to get skipped to fix the output at lower ranks.

Also, pterosaurs are no longer showing a class (nor any major clade around that rank). I'm pretty sure they used to have class Reptilia. I'm not sure what got changed there, and I'm nervous about making an edit in the wrong place that unskips Reptilia (possibly causing problems with other major tetrapod classes). Plantdrew (talk) 22:46, 31 October 2018 (UTC)

Template for deletion discussion for Virusbox 2018 November 27

Template:Virusbox (edit | talk | history | links | watch | logs) is being discussed at Wikipedia:Templates for discussion/Log/2018 November 27#Template:Virusbox, if anyone would like to chime in. --Nessie (talk) 20:10, 27 November 2018 (UTC)

17 December 2018

See Wikipedia talk:WikiProject Tree of Life#Large number of taxoboxes messed up. Peter coxhead (talk) 08:04, 17 December 2018 (UTC)

All ok now. Peter coxhead (talk) 15:23, 22 December 2018 (UTC)

Criteria for display of a taxon in a taxobox

Non-principal ranks above the target taxon

I'm not sure if it ever worked, but the old code underlying the automated taxobox system (e.g. here) had some logic that said:

IF a taxon wouldn't otherwise be displayed AND
   the taxon is 2, 3 or (in some versions) 4 levels above the taxon which is the target of the taxobox AND
   the taxa below it are all not at a principal rank
THEN display the taxon
END

I was never sure why this was appropriate, but I attempted to reproduce the logic when converting the system to Lua. What it should have meant was that if the hierarchy ran e.g. Clade A – Subclass B – Clade C – Infraclass D – Clade E then all of these would be displayed by default, whereas if it ran e.g. Clade A – Subclass B – Clade C – Infraclass D – Genus E only the target taxon (Genus E) and the immediate parent (Infraclass D) would be shown by default.

I can just about see the sense of this logic if the test were not just whether the taxa below were not at principal ranks but also whether they were not otherwise displayed, since this would provide some context that would otherwise be missing. However, as originally set up, given the hierarchy Clade A – Subclass B – Clade C – Infraclass D – Clade E and |display_parents=2, all 5 taxa would appear in the taxobox, whereas given Clade A – Subclass B – Clade C – Infraclass D – Genus E, only the last 3 would.

So while working at converting more of the system into Lua, I have removed this logic altogether. I'll be happy to put it back if someone can give me a good reason, preferably with an example taxobox where it would make a difference. It's always possible to manually provide more context by setting |display_parents=. Peter coxhead (talk) 15:25, 22 December 2018 (UTC)

Taxa with authorities given

In an automated taxobox, it's possible to specify authorities up to the greatgreatgrandparent of the target taxon. This is intended for use where there are multiple levels of monotypic taxa (e.g. Amborella). However, unless the corresponding taxon was displaying for some other reason (e.g. it was a principal rank), the authority would previously not have been displayed. This seems wrong to me, so now, all taxa that have an authority specified will be displayed in an automated taxobox by default. Peter coxhead (talk) 15:33, 22 December 2018 (UTC)

I've been looking at the logic for display and couldn't work out the infrataxon logic for parent 4. I don't think it worked as intended as Felidae didn't display Carnivoramorpha (parent 4) even though parent 2 was an infrataxon (suborder Feliformia). I think the change to force display if parent authorities are given is a sensible improvement. I can confirm that Carnivoramorpha now displays when the Felidae taxobox is given a greatgrandparent_authority, which is what you would expect when adding such an authority.

Specifying authorities in 'lower' automated taxoboxes

Technical discussion re specifying authorities

(Started a new subsection)
There is one other thing I don't think is working. If there is both a variety AND a subspecies in infraspeciesbox the parent authorities don't get shifted properly. It works for subspecies or variety but not both. With the new Lua code in p.taxoboxList() and l.showTaxon() I think there is a simple solution. Create a table of authorities to match the taxon and rank table and fill them in from the frame parameters with an appropriate shift. This would remove the need to shift the parent authorities in the templates (speciesbox, infraspeciesbox, etc) and could also handle the subgenus shift.   Jts1882 | talk  16:22, 22 December 2018 (UTC)

@Jts1882: the value of display_parents isn't adjusted correctly either. I will give a more detailed response, but can't do so right now. Peter coxhead (talk) 20:55, 22 December 2018 (UTC)

'Lower' automated taxoboxes (LATs, let's say) are those that handle genus upwards via taxonomy templates and ranks lower than genus manually in the taxobox template. I think that, as far as possible, LATs should maintain a clean separation between the automated and manual parts of the taxonomy; the more modular a system is, the easier it is to maintain.

• With hindsight, I think the best way of doing this would have been to use the genus as an explicit reference, so that instead of parent_authority, grandparent_authority, etc., LATs used genus_parent_authority, genus_grandparent_authority, etc. Then display_parents would have been display_genus_parents. It would be possible to change over to this approach, I believe, with the aid of a bot for {{Speciesbox}} uses, but it would mean editors learning new parameters.
• Without taking this route, only limited simplification is possible. Since LATs handle below-genus ranks in the taxobox, {{Infraspeciesbox}} still has to decide what rank parent_authority refers to in order to know whether to display it or pass it upwards to the automated system. The rank can be subspecies (if variety and subspecies are present); or subgenus (if only subspecies and subgenus are present); or genus (if only subspecies is present). The only simplification would be to handle greatgrandparent and upwards outside the taxobox, since these can't be ranks handled inside the taxobox – but how many such cases exist?

I'm still thinking about how it might be possible to simplify specifying the authorities and the number of parents to display in LATs. Ideas welcome! Peter coxhead (talk) 08:07, 23 December 2018 (UTC)

@Jts1882: see Module:Autotaxobox/sandbox|taxoboxList for an implementation of your elegant idea of an authority table. The offset is easily calculated inside LATs by counting the number of below-genus ranks present. It still leaves the complexity of matching the authority parameters within the template. Peter coxhead (talk) 10:15, 23 December 2018 (UTC)

My thinking was that authority, parent_authority, etc. get left unchanged in the LAT (auth would be authority, parentAuth would be parent_authority, etc). Speciesbox would pass offset=1 or offset=2 if a subgenus. Then you set authories with authTable[1-offset] = auth (which would be ignored), authTable[2-offset] = parentAuth (which would match the parent authority with the genus in row 1), etc.   Jts1882 | talk  10:32, 23 December 2018 (UTC)

(edit conflict) Over the last few days I've been experimenting with a modified version of p.taxoboxList() and a g.showTaxonRow() function, which like your new l.showTaxon() emulates Template:Taxobox/showtaxon. I was pleased to see your update yesterday as I was wondering why the infrataxon forcing was there for parents 3 and 4, although it would have been clearer if I'd waited three days and seen your update first. My experimenting leads to the following comments.

1. Now the automatic taxonomy is handled in Lua it should be easier to handle the shift of the authorities there, rather than all that parent=grandparent code in the LAT templates. You can create a table of authorities matching taxonTable and taxonRankTable for taxon and rank. Then you can pass a shift_authority parameter from the LAT templates. Speciesbox would pass one, subspeciesbox 2, infraspeciesbox 2 or 3 (depending on whether variety and/or subspecies). If there is a subgenus the value would be incremented by one. This shift can be used to place the parent authorities with the correct row. [this is essentially what you have just implemented with the offset parameter]
2. The shift_authority parameter isn't necessary, as the module could determine it from the template parameters based on whether there are species, subspecies, variety and subgenus parameters (each adding one).
3. A further simplification would be to load the entire taxonomic heirarchy as a 2-dimentional array (machine ocde = all) with an additional authority element. The table could be addressed directly, e.g. taxonInfo[i]['rank'], taxonInfo[i]['authority'], taxonInfo[i]['always_display'], etc. This would simplify l.showTaxon() as it could get most information just from a passed i parameter.
4. Function l.showTaxon() could be further simplified by using taxonInfo[i]['always_display'] as a forcing flag. The value would be set to true for i<display_taxa and for major ranks.
5. An addition step would be to add the variety, subspecies, species and subgenus rows at the bottom of the hierarchy to handle the whole taxonomy within p.taxoboxList().

Comments on your numbered points above:

1. We have implemented this independently, as per Module:Autotaxobox/sandbox|taxoboxList.

2. Um... I think that it's a design mistake to lose the modularity of the two parts of the LAT templates, so long as one is in the template language and the other in Lua.

3. I have actually implemented this (see makeTableAll in this version), but decided that it wasn't really necessary – at least at present – and backed off. (I'm reluctant to use name entries in Lua tables for efficiency reasons, but I guess that's my age, i.e. being used to having to optimize code for slower machines.)

5. Yes, I'd thought of that, and it would make some things easier (although there is the issue of the binomial/trinomial box, which currently is always handled inside the template). I think this is a step to take after the taxobox templates themselves are converted to Lua, which I think we both agree is a desirable goal.

Peter coxhead (talk) 11:00, 23 December 2018 (UTC)

I'm hestitant to point you to the code implementing some of these ideas as its new (since 19 Dec) and I was exploring a number of incompatible approaches, which haven't been fully resolved and is a bit messy. But it might help make my comments above clearer. If you look at Module:Sandbox/Jts1882/Biota_Infobox/auto, there is a function p.loadTaxonomyTable() (currently line 67) that loads the whole automatic hierarchy with additional taxon and authority columns. It also demonstrates inserting variety, species and subspecies at the bottom of the heirarchy. The function g.taxoboxList() is modified from your p.taxoboxList() and g.showTaxonRow() does a crude version of your l.showTaxon(). I should point out that the g.templateArgs[] contains all the LAT or automatic taxobox parameters after checking for a value, converting aliases and removing unrecognised parameters. Test examples are in User:Jts1882/sandbox/test/taxobox/auto.

Hopefully this explanation isn't too confusing and offers some helpful ideas.   Jts1882 | talk  10:20, 23 December 2018 (UTC)

No, not at all confusing, and certainly helpful. My plan was basically to work upwards, slowly removing all the template calls except for those involved in configuration, since (a) some are also used by the manual taxobox system, which I'd rather keep clear of Module:Autotaxobox and (b) it makes it easier for editors in future to change the configuration without messing with the Lua code. At every step I've been finding errors, odd behaviours and now-invalid assumptions built into the original templates – to be fair, these were designed a long time ago now and then often tweaked by multiple editors, based on the assumption that taxonomic hierarchies would mainly involve Linnaean ranks, and then fixed to work with clever work-arounds for the limited expansion depth of templates. Remembering how long it took me to do the core conversion to Lua, I wasn't willing to work top-down. But maybe you should start from the top, picking up the bits I've done?? Peter coxhead (talk) 11:00, 23 December 2018 (UTC)

Oh, and the other issue to watch is that editors have created taxoboxes with incorrect (or at least unexpected) parameter combinations, which worked, although not by design, or have deliberately used parameters in an unexpected way to get around bugs, which causes failures when the bugs are fixed. This has caught me out several times – the last when I started trying to sort out the automatic italicization of page titles and taxobox names in {{Speciesbox}}. This is another reason I've been cautiously working upwards, step by step, anxiously watching Category:Taxobox cleanup! Peter coxhead (talk) 11:11, 23 December 2018 (UTC)

2 July 2018 usage statistics update

July 2 update

Project	Auto	Manual	Total taxa	Percentage auto	# auto added since 17 December
Algae	966	859	1825	52.9	627
Amphibians and Reptiles	12042	2424	14466	83.2	1082
Animals	1999	3441	5440	36.7	1180
Arthropods	1320	5569	6889	19.2	996
Beetles	11598	23056	34654	33.5	9114
Birds	9348	4368	13716	68.2	3307
Bivalves	189	1134	1323	14.3	109
Cephalopods	870	654	1524	57.1	586
Cetaceans	356	5	361	98.6	37
Dinosaurs	1276	171	1447	88.2	51
Fishes	11728	9140	20868	56.2	3786
Fungi	477	11339	11816	4.0	284
Gastropods	7889	22905	30794	25.6	865
Insects	21498	43921	65419	32.9	10924
Lepidoptera	24863	72695	97558	26.3	8662
Mammals	2947	4233	7180	41.0	481
Marine life	1754	2951	4705	37.3	1093
Microbiology	230	8592	8822	2.6	170
Palaeontology	7045	6264	13309	52.9	906
Plants	27486	34100	61586	44.6	8968
Primates	668	214	882	75.7	341
Rodents	1870	1200	3070	60.9	413
Sharks	702	47	749	93.7	16
Spiders	6228	2272	8500	73.3	1129
Turtles	513	135	648	79.2	361
Viruses	NA	1431	1431	NA	NA
Total	143293	243063	386356	37.1	52226

I'd been doing these updates on a 4 month basis, but it's now been more than 6 months since the last one. Whoops. But I'll probably stick to a 6 month schedule going forward. Of the 26 projects I've been tracking, 13 now are using automatic taxoboxes on more than half of their articles, up from 5 projects as of last December. 5 more projects are between 33% and 50% automatic taxobox usage. Of the 8 projects under 33%, two (Insects and Lepidoptera) have a massive number of articles and while there is significant effort to switch to automatic taxoboxes it will take quite a while. Viruses aren't compatible with automatic taxoboxes. WikiProjects Gastropods (December 2016), WikiProject Fungi (November (2014) and WikiProject Arthropods (March 2012) have had proposals to switch to automatic taxoboxes that have attracted opposition. That leaves Bivalves and Microbiology as the two projects that haven't significantly adopted automatic taxoboxes yet, and which haven't had opposition to a proposal to use them. However, it may be worth revisiting the issue for the three projects that previously refused. There was a major revision of gastropod classification released in December, by the same authors of the classification previously accepted by WikiProject Gastropods. The new classification abandons the use of unranked clades; as a result, every single gastropod taxobox (including those now using the automatic system) will need to be changed if the new classification is to be followed.

I will say that I don't think 100% usage of automatic taxoboxes should be a goal at this time. There are some situations that aren't handled well by automatic taxoboxes, and I'm not sure it is worth the effort to change automatic taxoboxes to accomodate them. Two of the cetacean articles using manual taxoboxes use |binomial2=, which overrides linking to the genus in automatic taxoboxes. All living shark species now use automatic taxoboxes; the remaining sharks using manual taxoboxes are known from fossils. I'm reluctant to do much work on automatic taxoboxes with palaeontological taxa; there are various competing, mutually incompatible hypothesized clades for fossil, and I'm just not familiar enough with fossil taxa to be sure I'm not accidentally mixing different classification systems together. Plantdrew (talk) 20:28, 5 July 2018 (UTC)

@Plantdrew, Od Mishehu, Pvmoutside, Quetzal1964, and William Avery: I haven't attempted to update the detailed analysis, but the overall counts are interesting. {{Taxobox}} has about 226k transclusions, {{Principal rank}} (used only by automated taxoboxes) about 166k transclusions, suggesting that automated taxoboxes now make up about 42% of the total of 392k, rather than about 37% as in the table above. (Viruses now have a functioning automated taxobox; there are about 277 uses of {{Virusbox}} right now.) The really striking change is in the number of automated taxoboxes, from 143k to 166k, which I find impressive in a such a short time. Peter coxhead (talk) 17:54, 31 December 2018 (UTC)

I've been planning to do a detailed update of the stats tomorrow, to ring in 2019. Plantdrew (talk) 18:22, 31 December 2018 (UTC)

@Plantdrew: If it's not too much hassle, could WikiProject Tree of Life also be included? I know many taxa are "dustbinned" into this category when they don't fit neatly into more well-known projects. I think it would be interesting to compare, especially with the paraphyletic projects like WikiProject Marine life, WikiProject Microbiology, and WikiProject Palaeontology. And don't forget the new one: WikiProject Hypericaceae, which I know has a few pages not listed in WikiProject Plants. --Nessie (talk) 18:55, 31 December 2018 (UTC)

If you want to get super granular/anal, there's WikiProject Banksia, WikiProject Carnivorous plants, WikiProject Cats, WikiProject Dogs, WikiProject Equine, WikiProject Hypericaceae, WikiProject Mammals/Bats Task Force, WikiProject Poultry, WikiProject Pterosaurs, and WikiProject Tree of Life. These projects each have separate talk templates, if that's what you're using for the PetScan or whatever.

@NessieVL:, I'll add ToL to the stats. I might include some of the others you suggested, but they're mostly already accounted for. I've been adding Carnivorous plants and Banksia into the Plant stats the last 2 or 3 times I've posted updates (and I'll include Hypericaceae going forward). All Cats, Dogs, and Equine taxon articles were also tagged for Mammals (or another ToL subproject in the case of some pathogenic organisms) last time I checked, and are almost all using automatic taxoboxes (barring some paleontological taxa where sources are inconsistent about the taxonomic hierarchy). Plantdrew (talk) 07:50, 1 January 2019 (UTC)

1 January 2019 usage statistics update

January 1 2019 update

Project	Auto	Manual	Total taxa	Percentage auto	# auto added since 2 July
Algae	1014	852	1866	54.3	48
Amphibians and Reptiles	12757	2338	15095	84.5	715
Animals	2657	3645	6302	42.2	658
Arthropods	1547	6142	7689	20.1	227
Beetles	12121	22878	34999	34.6	523
Birds	11481	2306	13787	83.3	2133
Bivalves	197	1154	1351	14.6	8
Cephalopods	1017	1414	2431	41.8	147
Dinosaurs	1337	136	1473	90.8	61
Fishes	13274	7915	21189	62.6	1546
Fungi	1400	10775	12175	11.5	923
Gastropods	10015	21906	31921	31.4	2126
Insects	23165	43010	66175	35.0	1667
Lepidoptera	26498	71691	98189	27.0	1635
Mammals	4067	3296	7363	55.2	1120
Marine life	2428	2888	5316	45.7	674
Microbiology	736	9534	10270	7.2	506
Palaeontology	7876	6453	14329	55.0	831
Plants	36060	26536	62596	57.6	8574
Primates	812	83	895	90.7	144
Rodents	2893	179	3072	94.2	1023
Sharks	705	43	748	94.3	3
Spiders	6370	2185	8555	74.5	142
Tree of Life	51	178	229	22.3	?
Turtles	601	56	657	91.5	88
Viruses	276	1224	1500	18.4	276
Total	165819	226196	392015	42.3	22526

Mammal subprojects with articles tagged for both mammals and subproject:

Project	Auto	Manual	Total taxa	Percentage auto
Bats	630	781	1411	44.6
Cetaceans	384	8	392	98.0
Cats	186	25	211	88.2
Dogs	203	38	241	84.2
Equine	85	28	113	75.2

Presenting the latest statistics. As always, articles with multiple project banners are counted under each project. I've broken out 5 of the mammal subprojects into a separate collapsed table above; articles for these projects are universally tagged with both mammals and the subproject banner (articles under the other two mammal subprojects, Rodents and Primates largely lack a Mammals banner). Thanks to NessieVL, all articles with any kind of taxobox now have a banner from at least one of the WikiProjects in the stat table.

All the projects that are above 90% are effectively "done"; I've reviewed all the articles under these projects that are still using manual taxoboxes. The remaining manually taxoboxed articles for these projects are largely taxa where there seems to be some disagreement about their taxonomic placement (and most of these are paleontological taxa).

Great work on this, as always! Btw, what's up with Wikipedia:WikiProject Bivalves? --Nessie (talk) 01:59, 2 January 2019 (UTC)

Bivalves are just waiting for you to tackle them <smile>. Seriously though, I'm not sure what source the bivalve classification follows, nor whether it consistently follows any source. WoRMS spells the order names differently than we do, and cites more recent sources than the 2010 paper Wikipedia may be following. Bivalves just need a good source and some work. Plantdrew (talk) 02:55, 2 January 2019 (UTC)

Not "always display"ed, apparently

I updated some taxoboxes recently, and it seems some major ranks are not displaying. See Leishmania infantum just shows genus and species. --Nessie (talk) 16:55, 2 January 2019 (UTC)

The ranks seem case-sensitive currently. "Ordo" was capitalized. Lowercasing it ("ordo") made it display in the taxobox. —Hyperik ⌜talk⌟ 17:00, 2 January 2019 (UTC)

Seems like it shouldn't be so picky, especially when the template will say stuff like "Rank: phylum (displays as Phylum)" implying that caps are preferred. Is there a way to search for capitalized rank parameters in taxonomy templates? --Nessie (talk) 17:17, 2 January 2019 (UTC)

There is an argument that a lot of the taxobox templates have been too forgiving in parameter names. However, I think this one is easily handled by changing line 629 of function l.makeTable() in Module:autotaxobox to the following code.   Jts1882 | talk  17:49, 2 January 2019 (UTC)

if ok then taxonRankTable[i] = string.lower(rank) else taxonRankTable[i] = '' end 

Actually, I agree that most of the taxobox templates are far too flexible in the parameter names and values they allow. It makes the code over-complex, and is a fruitful source of error when maintaining them. It's always easy to add that little bit of extra code to cope with variation, but then miss the overall complexity that results. Editors have, for example, asked for English rank names to be accepted in taxonomy templates. Then you could end up with having to handle capitalized and uncapitalized Latinized and English rank names. I've spent a lot of time recently trying to make the system's coding simpler. However, if there's really a demand for this... Peter coxhead (talk) 12:12, 5 January 2019 (UTC)

@NessieVL: by the way, the "displays as" bit is really there to show the English form used, I think, e.g. "ordo" displays as "Order" – the capital is only because it's at the start of a row in the taxobox.

@Jts1882: if agreed, then I would add {{lcfirst: to {{Don't edit this line rank}} – it's the cleanest way. Peter coxhead (talk) 12:27, 5 January 2019 (UTC)

I created a tracking category Category:Taxonomy templates using capitalized rank parameters. So far few entries, but these categories update slowly. Peter coxhead (talk) 22:14, 5 January 2019 (UTC)

It's filling up now (829). Rather than adjust the code to allow more parameter flexibility, there is a case for being stricter and using more such tracking categories. It does mean more work correcting the taxoboxes, though. However, in this case, the mistake (capitalisation of the first letter in the rank) looks like it is quite common and is not just the result of carelessness. It looks like they are mostly for Genus, which is handled differently when constructing the binomials etc. As capitalisation of genus is effective allowed, it makes sense to cover other ranks. The change would also need to be made in {{Don't edit this line all}}.   Jts1882 | talk  08:18, 6 January 2019 (UTC)

Rather than adjust the code to allow more parameter flexibility, there is a case for being stricter and using more such tracking categories – I agree. (Most of the current code that deals with parameter/value flexibility attempts to replicate what has been there for a long time – another area of flexibility is yes/Yes/true/True for |always_display= and |extinct=.) It also makes sense to deal with such flexibility as low down as possible, so as to keep the main code clean. So I guess I should alter the two "Don't edit this line" templates (sigh). I'll also see about getting a bot to change the rank value to lower-case (requested at User talk:Tom.Reding#Another bot request). Peter coxhead (talk) 09:35, 6 January 2019 (UTC)

You could use {{Yesno}} or Module:Yesno to handle all the yes/no/true/false possibilities, although |extinct= has an unexpected behaviour when empty, iirc.   Jts1882 | talk  11:12, 6 January 2019 (UTC)

@Jts1882: re empty values, originally for all the parameters where yes/no/true/false would be expected, any nonblank meant yes, including "no". {{Yesno}} seemed rather overkill when I looked at it earlier. I think it's worth at least lower-casing on access, though. Peter coxhead (talk) 22:43, 7 January 2019 (UTC)

Approaching this from a usability perspective, I specified lowercase in the preload boilerplate for taxonomy templates at {{Automatic taxobox/editintro/sameas}}. What about having a warning display for editors entering uppercase ranks? Or is that just adding more complexity? --Nessie (talk) 16:18, 6 January 2019 (UTC)

@NessieVL: actually, there are four more "editintro" templates, including the main one. I've edited them all as per the one you did. Peter coxhead (talk) 22:32, 7 January 2019 (UTC)

Template:Taxonomy/Iguanodon was missing a value for |link= for close to a year. I occasionally come across taxon templates that are missing a value for |rank=; when not specified, rank defaults to alliance. We could use tracking categories (or warnings to editors on save?) for taxonomy templates missing essential parameters (missing |parent= already puts templates in the automatic taxobox cleanup category) . Plantdrew (talk) 01:46, 7 January 2019 (UTC)

@Plantdrew: defaulting to "alliance" is now fixed (caused by a stray "|"). Omitting a rank will now add the taxonomy template to Category:Taxonomy templates using unrecognized rank parameters; adding an error message to the displayed template would also be useful. Peter coxhead (talk) 17:32, 7 January 2019 (UTC)

If you go to Template:Taxonomy/Junkiaceae, remove |rank= and press "Preview", there's now a big red error message. Peter coxhead (talk) 18:08, 7 January 2019 (UTC)

I suppose I shouldn't be by now, but I was surprised to see so many taxonomy templates lacking a rank. Every time we add another tracking category, we seem to find long-standing errors. What surprises me more is that some of these templates were in use, so the relevant taxoboxes would previously have been displaying "Alliance:" when this clearly wasn't the right rank. It seems that many editors simply don't notice issues with taxoboxes. Peter coxhead (talk) 09:15, 8 January 2019 (UTC)

Extinct symbols

Hello all, I have a question (or perhaps a suggestion for future module programming) regarding the automatic generation of extinct symbols. In an "automated" taxobox for an extinct taxon, why is it also necessary to then manually enter the extinct status on each subordinate taxonomy template as well? This question also extends to the need for manual entries to include the extinct symbols for type species and subdivisions of extinct taxa in the taxoboxes on the article page as well. Curiously, Loopy30 (talk) 20:47, 11 January 2019 (UTC)

@Loopy30: it's a good question which I've thought about from time to time. It would certainly be possible to search up the taxonomic hierarchy encoded in the taxonomy templates and treat a taxon as extinct if any ancestor taxon is marked as extinct. One answer for the taxonomy templates is that it's not much work to set the extinct flag when setting them up, and it does guard against inadvertent error: if the system did behave as I suggested above, then one error in setting a higher taxon as extinct would propagate over the entire subtree. What I have intended to do for some time is to add a check for "extinct status consistency" when viewing a taxonomy template. So far whenever we've added extra checks, a surprising number of errors have appeared. Peter coxhead (talk) 14:51, 12 January 2019 (UTC)

Category for automated taxoboxes with no templates?

So i know there is a category for missing taxoboxes, but was wondering if there were one for this ugly error. I found a page with an automatic taxobox with no associated taxonomy template, so it was a mess, but didn't seem to be in any specific cleanup category. I looked around Category:Taxobox cleanup and didn't see anything. I'm hoping this is rare, but I feel it's a priority, especially now that automated boxes are the recommended taxobox. Is there a category for these? Can there be? --Nessie (talk) 02:59, 27 August 2018 (UTC)

It looks like Category:Automatic taxobox cleanup with the sortkey mu is where that sort of error is tracked. I just ran into this at Eutrochium steelei. — Eru·tuon 03:42, 27 August 2018 (UTC)

thanks! --Nessie (talk) 11:23, 27 August 2018 (UTC)

@NessieVL: I regularly monitor Category:Automatic taxobox cleanup, Category:Taxoboxes with no color and Category:Taxoboxes with the error color, which between them catch most (?all) major taxobox errors. (I've been offline for a few days.) Peter coxhead (talk) 07:21, 30 August 2018 (UTC)

Forgot to point out that very often the pages listed in these cleanup categories are actually ok. If editors save a page with an automated taxobox before creating the taxonomy template, then the page gets added to a cleanup category, and doesn't immediately disappear when the taxonomy template is created. It needs a null edit on the page to force the category to be updated. It's helpful if editors only use "Preview" to check taxobox changes until the taxonomy template exists. Peter coxhead (talk) 08:45, 30 August 2018 (UTC)\

Is there normally (or at least sometimes) a time lag between when the taxonomy template is created and when it gets indexed & starts working right? If so, maybe that could be documented at Wikipedia:Automated_taxobox_system/convert, so that newbies like me don't think they must be doing something wrong, get discouraged, & throw in the towel for good on any further attempts to create/convert taxoboxes! I tried converting a manual to automatic taxobox for the 1st time today. When I previewed, I got the nice message saying "no template; click here to create one". I did so, published the template, & then returned to the genus page & did another preview. But I kept getting a message saying that the template didn't exist -- even though when I clicked again on the link to create one, the template page it opened contained the info I had already saved. I did this a couple of times (probably ~5 min total), then went looking (fruitlessly) for documentation. Finally, when I returned & tried previewing yet again (feeling foolish for even trying it, but not knowing what else to do), it worked this time (after maybe 10–15 min total). Is a delay like this normal? Gould363 (talk) 20:24, 20 January 2019 (UTC)

@Gould363: If this is about Laothus then I think the problem was your forgot to include |taxon=Laothus. --Nessie (talk) 21:07, 20 January 2019 (UTC)

@NessieVL:Yes, that's the page I was editing. Thank you for fixing it up for me! But I was following the instructions (from Wikipedia:Automated_taxobox_system/convert) for genus level or higher, which say "Remove all the parameters which specify the taxonomic hierarchy. You only need a single parameter, taxon = (even this can be omitted if the title of the article is the same as the name of the taxon)." That was indeed the case here: the article title was "Laothus." Gould363 (talk) 21:38, 20 January 2019 (UTC)

@Gould363: no, there's no delay. As soon as you have correctly created the taxonomy template, a new "Preview" will show it in use. If not, assume you've made an error. Peter coxhead (talk) 21:18, 20 January 2019 (UTC)

Thank you, Peter. I guess it will just remain a mystery, then. The events most certainly happened as described: I tried repeatedly to preview the page & got an error message, then previewed one more time (with no intervening changes to the taxobox) and it worked. Maybe it could be some sort of browser cache issue. I'm using Firefox 64.0 under Win7E. Anyway, it's good to know that there's no lag normally. Gould363 (talk) 21:38, 20 January 2019 (UTC)

I do sometimes experience short lags when saving a taxon page immediately after creating a template. But a null edit within no more than ten seconds resolves it.Plantdrew (talk) 22:03, 20 January 2019 (UTC)

Inconsistent extinction status

Please add a category called Category:Taxonomy templates with inconsistent extinction status, to include any taxon labeled as not extinct, where the parent is labeled as being extinct. עוד מישהו Od Mishehu 13:51, 21 January 2019 (UTC)

@Od Mishehu: it's at the top of my to-do list to handle this. See the thread immediately above. I have some code sketched out off-wiki but haven't had time to implement it yet. Peter coxhead (talk) 20:05, 21 January 2019 (UTC)

@Od Mishehu: ok, Category:Taxonomy templates with inconsistent extinct values exists now, and worked on some test cases with errors (now restored). Peter coxhead (talk) 17:36, 22 January 2019 (UTC)

I made a mess of the coding the first time, but it should be ok now. Peter coxhead (talk) 09:04, 23 January 2019 (UTC)

Unused taxonomy templates

(Moved to a new section and outdented. Peter coxhead (talk) 07:17, 8 January 2019 (UTC))

As an aside, once invalid templates (such as Kniphofia nana and Lepilemur wrightae) have been identified in the category listed above, how are they best nominated for deletion? List at WP:TFD or CSD by WP:T3? Loopy30 (talk) 01:13, 8 January 2019 (UTC)

Good question. I've been adding to Category:Unnecessary taxonomy templates, but that doesn't get them deleted. I think it's worth trying T3 for species templates; they are basically hardcoded duplicates of genus templates. It'd be nice if there were some active admins familiar with automatic taxoboxes who were willing to boldly delete templates identified as unneeded, without needing to go through the bureaucracy of TFD. I'm not sure if anybody meets that description, but I'll ask the somebody who might.

Pinging Shyamal, an admin with experience editing taxonomy templates who may be able to help in boldly deleting invalid templates. Loopy30 (talk) 03:14, 8 January 2019 (UTC)

How are these different from the bulk of those in Category:Taxonomy templates for species? Besides that not all of them are unnecessary? --Nessie (talk) 03:08, 8 January 2019 (UTC)

Hi Nessie, To quote the category description:

"Such templates are only needed when the species name is not displayed in a taxobox as a straightforward binomial, in particular when the generic name is displayed other than as a single word. In all other cases, the template is redundant since the taxonomic hierarchy is picked up from the genus name or from the |parent= parameter in the taxobox."

The examples above (Kniphofia nana and Lepilemur wrightae) are redundant to their respective genera templates with no need for a special binomial display like the viruses or the sections above species level for the maple trees. Loopy30 (talk) 03:26, 8 January 2019 (UTC)

There is a lot of orphaned content "below" the article namespace (I remember something about journal references that were transcluded by some editors long ago) apart from below the "template" namespace that has not been deleted. I think it might need more discussion about how they should be handled. Some of them, I suspect, may not be appropriate for quick deletion. Shyamal (talk) 03:34, 8 January 2019 (UTC)

@Loopy30, NessieVL, and Shyamal: some points:

• Category:Unnecessary taxonomy templates contains templates that are not used. They should be blanked before the category is added; otherwise the template is likely to generate further error messages and error-tracking categories. I have tried to get these deleted without the bureaucracy of TFD before, without success, and I feel I have more useful things to do. If someone else is prepared to deal with TFD, great.
• Category:Taxonomy templates for species holds templates for which there is, at present, a legitimate reason for the existence of a taxonomy template at this level, often because the species are placed in taxa below the level of genus. {{Speciesbox}} will cope 'manually' with some of these (e.g. subgenus), but there is a better way which isn't fully worked out or documented yet, namely to specify the entry into the automated taxobox system via |parent=. At some point, most of the species taxonomy templates in the category won't be needed, and will be blanked and moved to Category:Unnecessary taxonomy templates, but this isn't the case yet.
• Quite separately, Category:Unused taxobox templates holds taxobox templates that are no longer used, almost all because their functions are now coded in Module:Autotaxobox. At some point I suppose all these will have to go to TFD, but as with taxonomy templates, I'm content just to categorize them as unused.

Peter coxhead (talk) 07:34, 8 January 2019 (UTC)

I believe it would be legitimate to delete the pages of your first category under Wikipedia:Criteria_for_speedy_deletion#T3. Shyamal (talk) 07:52, 8 January 2019 (UTC)

@Shyamal: that would be good; I'll try it. Actually the stated reason at T3 better fits Category:Unused taxobox templates – almost all are unused because their functionality is provided by Module:Autotaxobox. Peter coxhead (talk) 09:19, 8 January 2019 (UTC)

I think the separation of template (code) v. data (subpages) seems to be difficult here but I think as a combination it would still unused qualify T3 - using the spirit rather than the wording here. PS: I have (suo motu) deleted some of them but I see that you are tagging it with T3, so I will stop for now. Shyamal (talk) 09:35, 8 January 2019 (UTC)

So far I've only been tagging the species ones, since they are clearly redundant w.r.t. the genus template. You're welcome to go ahead with the others. Peter coxhead (talk) 09:58, 8 January 2019 (UTC)

@Shyamal: please look at Template:Add new taxon, which I've nominated using a tweaked version of {{Db-t3}}, which allows a module to be specified as the subject of the duplication, rather than a template. I'll raise this at Wikipedia talk:Templates for discussion. Peter coxhead (talk) 10:06, 8 January 2019 (UTC)

I don't fully understand why we got rid of species-level templates. If anything, shouldn't we get rid of leaf-level templates? i.e. taxon with no child taxa articles? But even then, why not just have one *kind* of place where we write extinct=, authority= etc? The current system is weird. ErikHaugen (talk | contribs) 18:06, 15 February 2019 (UTC)

@ErikHaugen: Well, {{Speciesbox}}, which removed the need for species taxonomy templates goes back to 2011, well before I started working on taxoboxes.

• We don't need species taxonomy templates, except for viruses and other special cases, because the genus name is enough. Why have redundant taxonomy templates, one for each species, each saying the parent is the genus?
• Authority is only in one place – in the taxobox for the taxon whose authority it is. The idea of putting it in the taxonomy template is sometimes raised, but it won't work because it should always be referenced, and we can't put references in templates because they might duplicate those already in the article or be in a different style, neither of which is allowed.

Some of the details of the automated taxobox system are complex, but not weird. Peter coxhead (talk) 18:37, 15 February 2019 (UTC)

Why have redundant taxonomy templates – b/c there's more than one way to express things like parentage and extinct=true. (That's what I mean by weird.) ErikHaugen (talk | contribs) 18:47, 15 February 2019 (UTC)

Unorthodox taxonomy templates

Is this an approved method: {{Taxonomy/Ursini (Ursinae)}}? I can't see a reason why it is not simply {{Taxonomy/Ursini}}, which exists as a redirect.   Jts1882 | talk  13:02, 16 February 2019 (UTC)

probably one could use |sameas=. --Nessie (talk) 18:10, 16 February 2019 (UTC)

There's a spider genus Ursa, from which could be derived the tribal name "Ursini", as from Ursus. But both don't need to be disambiguated, so {{Taxonomy/Ursini}} is fine for bears, with {{Taxonomy/Ursini (Araneidae)}} for spiders. Peter coxhead (talk) 19:53, 16 February 2019 (UTC)

Extinct incertae sedis?

NDB, just having a fastidious moment, but should |extinct=yes be included in incertae sedis taxonomy templates that are tethered to extinct taxa? Example: {{Taxonomy/Incertae sedis/Ornithocheirae}}, where Ornithocheirae is an extinct taxon. --Nessie (talk) 22:43, 16 February 2019 (UTC)

IMO, include |extinct=yes, mainly so that it's daggers all the way down. But it might be worth rethinking the status quo. Could/should extinct be inherited all the way down, and not explicitly stated in each subordinate taxonomy template? Should incertae sedis ranks be explicitly stated, or just omitted (and if any major ranks are omitted in taxonomy templates, could/should incertae sedis be automatically displayed for skipped ranks)? I've seen a lot of taxoboxes both manual and auto, and my impression is that it's more common to omit incertae sedis ranks than state them. Plantdrew (talk)

I agree re including |extinct=yes in these cases.

I'm very reluctant to consider inheriting information like this. The problem is that if one taxonomy template high up erroneously gets |extinct=yes, then all those below will be "fixed" automatically and wrongly. Taxonomy templates are checked for consistency of extinction status (try changing the extinct status in {{Taxonomy/Ornithocheirae}}), but at present "incertae sedis" ones are ignored. Maybe they too should be checked.

The problem with automatically inserting "incertae sedis" for skipped major ranks is that there are many skip templates that deliberately skip major ranks for consistency reasons, and where clade-based taxa are more common, as with "dinosaur taxonomy", there may not be any consensus major ranks. Peter coxhead (talk) 14:02, 17 February 2019 (UTC)

I like having incertae sedis displayed in the taxobox, to make it clear that the placement is not certain. I just wasn't sure about adding the dagger, as some readers might think all incertae sedis taxa are extinct, perhaps. Maybe that's something to clarify on the incertae sedis article. --Nessie (talk) 14:27, 17 February 2019 (UTC)

Yes, I agree; my point (which I've clarified above) is that it shouldn't be added automatically by the system but by editors. Peter coxhead (talk) 16:57, 17 February 2019 (UTC)

I've just discovered that there are quite a few taxonomy templates with extinction only marked in the link text, like this version for Baru. The taxonomy listing on the right and the resulting taxobox suggested that the taxonomy template had |extinct=yes, but it did not. Up to now, this was not flagged as an error; now such templates will be added to Category:Taxonomy templates with inconsistent extinct values.

This does illustrate a general point. Those who work on the automated taxobox system, from 2010 to the present, tend to assume that editors will use it as it was designed to be used. But not all editors do! Catching the resulting errors is made very difficult if the system automatically implements "sensible default behaviour" (such as inserting extinct status to ensure consistency), so I increasingly believe that correct use of parameters should be more strictly enforced. Peter coxhead (talk) 20:26, 17 February 2019 (UTC)

I think the way we've being going has been pretty sensible. As we work, we find new oddities, but they don't tend to overload the system. We can't prevent editors from doing strange things, especially with such an involved system. But the system is becoming cleaner and more resilient.--Nessie (talk) 02:24, 18 February 2019 (UTC)

Templates like Template:Taxonomy/Incertae sedis/Trilobita will now turn up in Category:Taxonomy templates with inconsistent extinct values if they do not contain |extinct=yes. I've been fixing those that appeared so far. Peter coxhead (talk) 16:52, 18 February 2019 (UTC)

Oddity with |always_display=

When the parameter is explicitly set to |always_display=yes or |always_display=no the display in the left-hand table is correct, but when the parameter is missing or left blank (|always_display=) the font for "No" is different from other parameters because it is not wrapped in the code tag. The behaviour seems fine, so this is just an appearance issue.   Jts1882 | talk  09:54, 18 February 2019 (UTC)

Yes, it's deliberate to distinguish the two cases: showing blank as "no" (or "yes") versus a value actually present. It's implemented for |extinct= too. I find it helpful sometimes when checking/correcting taxonomy templates (e.g. the recent cases I've found that have † in the link text but no extinct parameter). Peter coxhead (talk) 13:08, 18 February 2019 (UTC)

Look at Template:Taxonomy/Hexapoda, for example. Values actually present in the template are shown in code font, values supplied in normal font. Peter coxhead (talk) 13:11, 18 February 2019 (UTC)

Always display incertae sedis parents

Automated taxobox system/Archive 2 Temporal range:
Samburupithecus kiptalami fossils, Muséum national d'histoire naturelle, Paris
Scientific classification
Domain:	Eukaryota
Kingdom:	Animalia
Phylum:	Chordata
Class:	Mammalia
Order:	Primates
Suborder:	Haplorhini
Infraorder:	Simiiformes
Family:	incertae sedis
Genus:	†Samburupithecus Ishida & Pickford 1997
Species:	†S. kiptalami
Binomial name
†Samburupithecus kiptalami Ishida & Pickford 1997

The taxon above an incertae sedis rank should always be displayed, no? It seems misleading otherwise, and it's not like you can click on incertae sedis and see the missing rank. For example, the speciesbox for Samburupithecus calls {{Taxonomy/Incertae sedis/Hominoidea}}, but does not show Hominoidea (see here) unless |display_parents= is included (see article). --Nessie (talk) 17:17, 19 February 2019 (UTC)

Blank space above taxobox in certain cases

Converting {{automatic taxobox}} to Lua has caused it to display differently on some pages. If there is a {{DISPLAYTITLE:}} or {{italic title}} above it, there is extra vertical space at the top of the page above the taxobox where there wasn't before. This can be seen in Silphium (genus), for instance. This is because the MediaWiki software inserts a newline before the table-starting syntax ({|) if it appears at the beginning of the text that is output by a module. The display title is interpreted as an empty line by the parser, so there are now two empty lines, which generate an empty p tag above the table, which displays as blank vertical space.

This has also been an issue for {{Weather box}} (see Template talk:Weather box § Spacing). To prevent this, the table is enclosed in a div tag, with a newline before {| (see the main function in Module:Weather box). This both ensures that no extra whitespace is displayed and that the table syntax is interpreted correctly. — Eru·tuon 06:14, 2 April 2019 (UTC)

I think this is another case of the parse bug that we had with the {{clade}} module. Adding a <nowiki/> statement before the invoke statement was the workaround. A quick test shows this removes the line when added before the {{automatic taxobox}} in Silphium (genus).   Jts1882 | talk  08:18, 2 April 2019 (UTC)

{{done}} @Erutuon: I've added the <nowiki/> statement to {{automatic taxobox}} before the invoke statement. This removes the line in the example. The parser bug is discussed at T18700.   Jts1882 | talk  08:32, 2 April 2019 (UTC)

@Jts1882: thanks; this was, I suspect, a bug I re-introduced when converting the template to Lua. (Btw, DISPLAYTITLE isn't needed in cases like Silphium (genus); the automatically added {{Italic title}} handles simple disambiguation.) Peter coxhead (talk) 09:11, 2 April 2019 (UTC)

Always display incertae sedis parents

Automated taxobox system/Archive 2 Temporal range:
Samburupithecus kiptalami fossils, Muséum national d'histoire naturelle, Paris
Scientific classification
Domain:	Eukaryota
Kingdom:	Animalia
Phylum:	Chordata
Class:	Mammalia
Order:	Primates
Suborder:	Haplorhini
Infraorder:	Simiiformes
Family:	incertae sedis
Genus:	†Samburupithecus Ishida & Pickford 1997
Species:	†S. kiptalami
Binomial name
†Samburupithecus kiptalami Ishida & Pickford 1997

The taxon above an incertae sedis rank should always be displayed, no? It seems misleading otherwise, and it's not like you can click on incertae sedis and see the missing rank. For example, the speciesbox for Samburupithecus calls {{Taxonomy/Incertae sedis/Hominoidea}}, but does not show Hominoidea (see here) unless |display_parents= is included (see article). --Nessie (talk) 17:17, 19 February 2019 (UTC)

Always display for incertae sedis rank

This dovetails with the above, but I thought I's put it in it's own subheading so as to not get confusing. Should incertae sedis ranks always display in the taxoboxes? Example: Sahelanthropus will not display it unless {{Taxonomy/Incertae sedis/Hominidae}} has |always_display=yes. --Nessie (talk) 17:43, 19 February 2019 (UTC)

Two parent taxa?

What should be the general practice for manual taxoboxes with two parents. Examples: Australopithecus bahrelghazali, Australopithecus anamensis, Western pond turtle, Blanding's turtle, and I'm sure more. They cannot be converted to automated taxoboxes at present: Is that something that should be accommodated? Or do we just add |classification_status=disputed and move on? Do we want a tracking category for these, other than the proposed one for taxoboxes using |classification_status=? --Nessie (talk) 19:54, 13 February 2019 (UTC)

There can't be two parents, there can just be alternative choices. The automatic taxobox system requires a choice, with different interpretations handled in the text and/or the synomym section.

I've never seen such taxoboxes before. While I don't think accommodating such alternative thinking would be a good idea, some way of flagging them seems a good plan. How many such anomalous taxoboxes are there? I'd hope it were just a few, but we've seen such expectations confounded.   Jts1882 | talk  20:24, 13 February 2019 (UTC)

I've also found Agrochola bicolorago, Agrochola decipiens, Agrochola verberata, and Lamiophanes schroeteri, playing around with regex searches. I'm not very good at regex, but two parents looks pretty rare. Plantdrew (talk) 20:55, 13 February 2019 (UTC)

Agrochola bicolorago looks like a straight-forward fix. Two genus names, one that agrees with the article name and the rest of the taxobox, and the other genus name listed as a synonym binomial. Loopy30 (talk) 21:25, 13 February 2019 (UTC)

Found another one, Siroccopteryx, the last WikiProject Pterosaurs article left with a {{taxobox}}. I tried to use some regex to find more, but I couldn't even find the ones I knew about. --Nessie (talk) 19:08, 14 February 2019 (UTC)

Choose one, based on the best secondary source available, and relegate the other to a synonym. Peter coxhead (talk) 22:38, 13 February 2019 (UTC)

These cases show why the manual taxobox system should be replaced as far as possible. The articles can handle the uncertainties in the text. Admittedly, the Siroccopteryx example isn't clear without reading all the background references. An alternative would be to place the parent at suborder Pterodactyloidea or clade Ornithocheirae where there is agreement.   Jts1882 | talk  20:26, 14 February 2019 (UTC)

Another one: Kenyanthropus, which might have three parents. Also, I fixed Australopithecus bahrelghazali, Australopithecus anamensis, and Siroccopteryx.

Also Canaanimico, which @Tisquesusa: really wants to keep. --Nessie (talk) 23:28, 24 February 2019 (UTC)

And now I see Earthworm. This one is pretty strange. I can't make any sense of it.

@NessieVL: earthworms are not a taxon, and so should not, in my view, have a taxobox. Peter coxhead (talk) 10:28, 2 March 2019 (UTC)

How to delete taxonomy that is no longer valid?

For example, Template:Taxonomy/Guancha has been determined to be invalid. [1] Mattximus (talk) 20:59, 24 February 2019 (UTC)

This is one I keep needing reminding of. I think you just blank the template and add Category:Unnecessary taxonomy templates and then someone will take care of it. It might be best to wait for User:Peter coxhead to confirm this. --   Jts1882 | talk  14:48, 26 February 2019 (UTC)

Well, this is what I do. Really I would prefer to be able to nominate the template for speedy deletion, but there's been another discussion of this issue at Wikipedia talk:Criteria for speedy deletion#Should a 4th CSD for unused templates be added? which seems likely to uphold the present very bureaucratic process for getting templates deleted, so I don't bother any more. Just leave them blank in Category:Unnecessary taxonomy templates. Either an admin will delete them or they'll just sit there. Peter coxhead (talk) 15:39, 26 February 2019 (UTC)

I don't think I've ever bothered to tag any templates as unnecessary when they represent a taxon name currently considered to be a synonym. Unused taxonomy templates can be found in Wikipedia:Database reports/Unused templates/2 (non-redirects, some in the unnecessary category) and Wikipedia:Database reports/Unused templates/14 (redirects, not categorized as unnecessary). Plantdrew (talk) 16:09, 26 February 2019 (UTC)

I'm not sure how that list is generated, but the first only includes templates tagged with Category:Unnecessary taxonomy templates (the number matches approximately and I checked a dozen or so). There are many more unused taxonomy templates, i.e those that have no children and are not genera. For instance nothing links to Template:Taxonomy/Algae and it isn't in either of those lists.   Jts1882 | talk  17:16, 26 February 2019 (UTC)

{{Taxonomy/Mackaya}} is on list #2, but not in the unnecessary category. It's been awhile since I last looked at those lists, but I thought there were a lot more unused template than are showing up there now (many templates were created circa 2010, but articles continued to use manual taxoboxes, so the templates were unused; I don't think recent progress switching to automatic taxoboxes is sufficient to have brought all the old templates into use). Plantdrew (talk) 17:44, 26 February 2019 (UTC)

I only put taxonomy templates into Category:Unnecessary taxonomy templates if they are 'wrong' in some way, e.g. incorrectly set up, or for taxa that aren't used now. Merely because a taxonomy template isn't currently used shouldn't be a reason to delete it; templates are often set up before editors get round to converting the relevant articles from manual to automated taxoboxes. But if you're sure that a taxonomy template won't be used, then I would always blank it and put it into the category. Peter coxhead (talk) 22:56, 26 February 2019 (UTC)

There's also |sameas= for junior synonyms. --Nessie (talk) 23:31, 26 February 2019 (UTC)

sameas won't work in this case since it was divided into 3 existing genera. Hmm... Mattximus (talk) 00:24, 27 February 2019 (UTC)

The point I meant to make (but expressed poorly and kinda derailed the conversation) is that I only blank and categorize taxonomy templates as unnecessary when I think Wikipedia would be better of without them and they should be deleted (and since the title of the template is the only thing I can't change (without leaving a redirect), that means I think Wikipedia would be better off without a template with that title). There are other unused taxonomy templates across a spectrum of necessity. Templates for groups where manual taxoboxes are still being used; necessary. Templates in support of different points of view where taxonomic consensus is lacking; fairly necessary. Templates for former taxonomic POVs that unlikely to ever be readopted; not necessary but not necessarily harmful.

I treat Category:Unnecessary taxonomy templates as a initial half-assed step to mark that a particular template title should not exist (while not going all the way in taking it to the bureaucracy of TfD). Is that how everybody else treats it?

Mattximus kicked this off with the example of Template:Taxonomy/Guancha. Mattximus is new and doesn't understand when Wikipedia deletes things (no offense meant; there is quite a learning curve here, thank you for deciding to contribute). In Wikipedia:Articles for deletion/Guancha, I favored keeping as a redirect, which Mattximus later supported. The title Guancha is probably not going to be deleted. Should the taxonomy template for that title be deleted? Perhaps? I'm absolutely not going to bother to taking it to the TfD bureaucracy myself. I'm not going to even bother with the half-assed step of placing it in Category:Unnecessary taxonomy templates (which has been explained to Mattximus in this thread). Certainly some templates for obsolete taxonomic POVs need not be kept. But it's such a hassle to get them deleted, I'm not sure when it's worth bothering. Plantdrew (talk) 03:23, 27 February 2019 (UTC)

@Plantdrew: yes, I also treat [[Category:Unnecessary taxonomy templates]] as a initial half-assed step to mark that a particular template title should not exist (while not going all the way in taking it to the bureaucracy of TfD). Just replace all the content by Category:Unnecessary taxonomy templates and leave them until some admin(s) are willing to deal with them. TfD is just a waste of valuable editor time in this case; such templates do no harm to the automated taxobox system so long as they are don't contain {{Don't edit this line}}. See User talk:Peter coxhead#What is the story with Category:Unnecessary taxonomy templates? for more on this issue. Peter coxhead (talk) 08:40, 27 February 2019 (UTC)

If there is consensus here, it would be helpful to replace the text at Category:Unnecessary taxonomy templates with text that says something like "This category is a holding location for taxonomy templates that should uncontroversially be deleted. When a taxonomy template has an invalid title and should be deleted instead of moved to a correct title, replace the entire contents of the template with [[Category:Unnecessary taxonomy templates]]. When there are a few dozen templates here, notify an editor active in the Templates for Discussion process, who will be happy to nominate them for deletion on your behalf."

In this case, I'll be happy to nominate the current batch for deletion. It would be helpful if someone here could write a brief rationale for the deletion nomination, so that I don't get the words wrong. – Jonesey95 (talk) 09:15, 27 February 2019 (UTC)

@Jonesey95: I've edited the explanation at Category:Unnecessary taxonomy templates along the lines you suggested above.

All of the templates are:

• Unused – we know this because if an attempt was made to use them, the automated taxobox system would flag an error and put whatever tried to use them in one of the categories that track taxonomy template errors at Category:Taxobox cleanup. I monitor this most days, as do some other editors.
• Unnecessary – here we have to rely on the judgement of the editor that put the template in the category.

What is important for people at TfD to understand, and seems to have been difficult to get over in the past, is that taxonomy templates are not 'normal' templates. Template:Taxonomy/Eresoidea, for example, which is in this category, would never be used anywhere in an article or another template via the wikitext {{Taxonomy/Eresoidea}}. They are actually components of the database that underlies the automated taxobox system, as is explained at WP:Automated taxobox system/intro, and are used by Module:Autotaxobox to create automated taxoboxes. Blanking (and ideally then deleting) taxonomy templates is part of the necessary maintenance of this taxonomic database. Peter coxhead (talk) 17:52, 27 February 2019 (UTC)

Thank you all for your contributions here. I have submitted a TFD nomination for the 153 templates currently in the category. I don't hang out at TFD much, so I'm not 100% sure I did everything correctly. We'll see how it goes. – Jonesey95 (talk) 20:14, 27 February 2019 (UTC)

Can the templates at Category:Unused taxobox templates also be nominated for deletion? --Gonnym (talk) 10:58, 6 March 2019 (UTC)

@Gonnym: I don't think that would be a good idea. For example, an editor (I forget who, sorry!) set up all taxonomy templates for virus taxa down to family level. These are all valid taxa, and many are just waiting for the articles to get upgraded to {{virusbox}}. They are unused, but not invalid. I also saw a bunch for bivalves, which has a very slow rate of adoption for automated taxoboxes. Removing the templates would make that trickle even slower. --Nessie (talk) 16:17, 6 March 2019 (UTC)

I don't think that category contains taxonomy templates. I think those are all templates that were used with the taxobox system and are no longer used because they have newer versions or Lua replacements. Peter coxhead has greatly simplified what was an extremely complex web of templates and I think those in Category:Unused taxobox templates are the leftovers.   Jts1882 | talk  16:54, 6 March 2019 (UTC)

Oh, snap, you're right. Sorry. Get rid of them then. --Nessie (talk) 17:30, 6 March 2019 (UTC)

@Gonnym: as Jts1882 says, Category:Unused taxobox templates holds taxobox templates that have been rendered redundant by Module:Autotaxobox and more recently Module:Automated taxobox. I need to go through them, because some can definitely be deleted, but others are still of some value as they preserve the history of the logic of the code in the modules, although not the actual code itself of course. So for now, no, they should not be deleted.

@NessieVL: yes, taxonomy templates should not be deleted just because they are unused, only because they are both unused and unnecessary, as those in Category:Unnecessary taxonomy templates are. Many others await use, as the conversion of manual to automated taxoboxes proceeds – slowly but steadily. Peter coxhead (talk) 18:32, 6 March 2019 (UTC)

If you can create a sub-category of those that are safe to delete, I'll take care of the nomination. --Gonnym (talk) 18:34, 6 March 2019 (UTC)

@Gonnym: ok, thanks; I moved those to be deleted into Category:Unused taxobox templates/to be deleted. Peter coxhead (talk) 16:03, 7 March 2019 (UTC)

Nomination at Wikipedia:Templates for discussion/Log/2019 March 7#March 7. --Gonnym (talk) 16:52, 7 March 2019 (UTC)