Template talk:Taxobox/Archive 19

Archive 15

←

Archive 17

Archive 18

Archive 19

Archive 20

Archive 21

→

Archive 25

Auto-italicize titles

This might have been brought up before, but I was wondering why this hasn't been implemented. I propose adding the following to the template.

{{#ifeq: {{{genus|}}}|{{PAGETITLE}}|{{italictitle}}|
   {{#ifeq: {{{species|}}}|{{PAGETITLE}}|{{italictitle}}|
         {{#ifeq: {{{binomial|}}}|{{PAGETITLE}}|{{italictitle}}|}}
    }}
}}

That way, if the species, genus or binomial parameters match the page's title, the template automatically italicizes their title. Tim1357 ^talk 04:27, 27 June 2010 (UTC)

This does already happen, provided the name parameter is not provided. Note also that at least WP:ARTH agreed that automatic italicisation was likely to be a bit confusing for new editors (it confused me for a while when I first saw it), and that it was better to produce the italics in a more transparent way, by using {{italic title}} (to which {{italictitle}} redirects) directly. --Stemonitis (talk) 04:49, 27 June 2010 (UTC)

I'm not sure that I read the consensus in the same way as you; see also Wikipedia:Bot_requests#Italicising_article_names_for_species_and_genera for a bot request in this regard. Martin (Smith609 – Talk) 20:03, 12 July 2010 (UTC)

Only issue I see is this: the title for articles like gorilla would become italicized. You'd need a boolean parameter to handle special cases like this. Bob the Wikipedian ^{(talk • contribs)} 00:44, 15 July 2010 (UTC)

To see how this is resolved using current parameters, see the source code at Gorilla. Martin (Smith609 – Talk) 23:44, 15 July 2010 (UTC)

I had to play in the sandbox for a few minutes to figure out what you'd done, but it appears the simple inclusion of the name parameter prevented italicization of the taxobox title. I was under the impression this template is supposed to italicize the page title. Am I mistaken? Bob the Wikipedian ^{(talk • contribs)} 21:12, 17 July 2010 (UTC)

I've tried to improve the documentation (in two places) so that it's a little clearer in this regard. Perhaps you could let me know if it makes sense or whether it needs re-wording? Martin (Smith609 – Talk) 23:05, 17 July 2010 (UTC)

Updated script for Species microformat

Those of you using the Operator add-on in Firefox might like to know that there is an updated script to read the 'Species' microformat emitted by the Taxobox. More exciting developments soon! Andy Mabbett (User:Pigsonthewing); Andy's talk; Andy's edits 16:39, 8 July 2010 (UTC)

Without installing the add-on and script, I'm not able to tell what this is. Could you provide some description of what it does and why I need it? Thanks, Bob the Wikipedian ^{(talk • contribs)} 21:04, 15 July 2010 (UTC)

Scientific classification?

Shouldn't we change "Scientific classification" to something more accurate like "Biological classification" or "Taxonomic rank" instead?. This terms seems a bit flaky at best. Sjschen (talk) 20:37, 15 July 2010 (UTC)

"Taxonomic rank" doesn't make sense. "Biological classification" is fine with me, but I don't see how "Scientific classification" is problematic. Ucucha 20:41, 15 July 2010 (UTC)

Scientific classification is the term most commonly used these days, and most people know exactly what it means. I'd stick with that. Bob the Wikipedian ^{(talk • contribs)} 21:02, 15 July 2010 (UTC)

While saying it's scientific may not be wrong, it does seem a bit imprecise in definition. The classification here is based on specific biological traits of the organism, be it genetic, anatomical, or perhaps even physiological, and thus it should be stated as such (Biological classification). I suggested "Taxonomic rank" since it indicates that the classification imposed on the organism was done by someone who thought a certain ordering scheme was correct. -- Sjschen (talk) 21:10, 15 July 2010 (UTC)

"Taxonomic rank" is wrong, because the "Scientific classification" section of a taxobox lists more than just the rank of the taxon concerned (for example, at Lion, the "Taxonomic rank" would just be "Species"). "Biological classification" may be more precise, but I don't see much advantage, and I certainly don't see "Scientific classification" as inaccurate, as you implied it is. Ucucha 21:31, 15 July 2010 (UTC)

I am not disagreeing that "scientific classification" is rather accurate for describing this type of classification. However, the same term can be accurately applied to any sort of empirically and methodically constructed classification. Namely, anything can potentially be classified scientifically. Fact is, the classification listed here in this template is specifically for biological organisms (regnum, divisio, classis, ordo, familia, etc. ...) and not for say, linux distributions or the phylogeny of languages[1]. I'm not sure why you're resistant to this change, but a more precise term will certainly benefit the articles that use this template.-- Sjschen (talk) 16:25, 16 July 2010 (UTC)

I don't think we really need to worry too much about being precise. In my opinion, just "Classification" would suffice. The context makes it abundantly clear what type of classification is being presented. Martin (Smith609 – Talk) 16:32, 16 July 2010 (UTC)

"Classification" would also work well with the available context, however if the user wishes to click on it to get more info they should be directed to the right place. As we speak, the "scientific classification" link is being already redirected to the "biological classification", so why not just change the title to this? -- Sjschen (talk) 16:44, 16 July 2010 (UTC)

This confused me as well: I pointed out that the fossil ranges of theropods extends to the present day, only to be reverted by an editor who informed me that the taxobox refers to the Linnaean classification system rather than the clade, and the Linnaean suborder Theropoda does not include birds. Perhaps we should therefore use "Linnaean classification" instead, to highlight just what system is in use? Icalanise (talk) 17:43, 20 July 2010 (UTC)

"Linnaean classification" seem like a good substitute for the current header as well. I will make the change in the next few days if there are no further objections. -- Sjschen (talk) 18:57, 20 July 2010 (UTC)

No; many taxoboxes contain unranked, non-Linnean names. (I would also disagree with that editor at "Theropoda".) Ucucha 19:05, 20 July 2010 (UTC)

[ec] Well, except that a lot of our articles use distinctly non-Linnean items. Most plant articles now use unranked taxa like Eurosids or Euasterids I, while the gastropod articles are full of clades (Heterobranchia, Stylommatophora) and "informal groups" (Pulmonata, Sigmurethra). The one thing these are not is Linnaean, but they are still scientific. I am yet to be convinced that there is any reason at all to change the existing "Scientific classification"; sure, the term could be applied in other fields to mean other things, but that doesn't mean it's not right here. --Stemonitis (talk) 19:07, 20 July 2010 (UTC)

Once again I can only reiterate my position that (1) the term lacks precision and that (2) the corresponding link already redirects to Biological classification. To me it seem that if the definition can be made more precise (Scientific -> Biological) and the wikilink more direct, then this change should be made. It seems the counter to the proposal has been something to the essence of "Yes it can be made more precise but leaving it this way doesn't hurt, so we should let it be", but to me that doesn't work. Generality and ambiguity in terminology may be good thing when flexible definitions/descriptions are required, but when it isn't (as with here where the classification is solely biological in nature) the term should be tightened and rendered more precise. -- Sjschen (talk) 19:53, 20 July 2010 (UTC)

If there are no further comments I will change the heading to "Biological classification" -- Sjschen (talk) 21:00, 23 July 2010 (UTC)

No, you shouldn't. There's been no clear consensus. I, for one, would agree with Bob above that "scientific classification" is the most widely used term. I don't think it needs to be altered. The taxobox is used on many articles, including those about viruses. In some sense, biological classification does not apply; we even have a different article dealing with virus classification. The broadest term and the correct term is still scientific classification in this context. Rkitko ^(talk) 22:35, 23 July 2010 (UTC)

I admit I have not thought of the use of this template in virus articles, and certainly that might make the change from scientific classification to biological classification possibly less appropriate. However, it seems that the heading for the taxobox template changes to "virus classification" (as it should) upon specifying that the subject is a virus (using the "virus group" tag). With this in mind I think a change from Scientific -> Biological would not be problematic. If anything this example shows that precision the naming of a heading is a good thing! -- Sjschen (talk) 00:20, 24 July 2010 (UTC)

"Virus classification" in a virus taxobox would be redundant. You're still missing the point that "scientific classification" is the most widely used and understood term here. I'm still not convinced it needs to be changed. Rkitko ^(talk) 00:33, 24 July 2010 (UTC)

No, I get the point that "scientific classification" (SC) is "most widely used", but "widely used "does not necessarily mean more rational or that experts in the field would use it. To go further with your argument that we should use the "most widely used" term, then "classification" should be used, and that if anything is inadequate in this case. I also disagree with you that the term "SC" is most widely understood, I for one would certainly be surprised to find more people confused if you mention "biological classification" (BC) than if you said "SC". Do less people understand the word "biological" as opposed to scientific? If anything "BC" gets more quickly to the point. As for what you say about "Virus classification", first of all I'm not proposing to do it, it has already been done, and a good thing too. Second, it is definitely not "redundant" as you so stated, it informs the reader that viruses are categorized using a slightly different method as opposed to other biological creatures. Changing this would be a disservice to the readers. -- Sjschen (talk) 20:43, 24 July 2010 (UTC)

"TNC" vs "NatureServe".

Email copied below with permission from ticket:2010092210014122 (changed an URL to an internal link). -- Jeandré, 2010-09-25t00:16z

Hello, good afternoon,

I'm writing today to enlist your help in address a systematic error that appears on at least some Wikipedia species pages. My organization, NatureServe, is an IUCN Red List member program, and recent discussions about a partnership with Wikimedia UK has brought a small but, from our perspective, somewhat grating error to light. I'd very much appreciate your help in directing this inquiry to the appropriate staffer or volunteer.

The IUCN Red List conservation status is widely used on Wikipedia's species pages. However, there are pages that make use of NatureServe's conservation status ranking, such as Short's goldenrod Solidago shortii.

Note how in this example the Taxobox at right shows identifies this plant as a G1 species, signifying that its critically imperiled status under NatureServe's G-rank system. The troubling thing from our perspective is the persistence of the "TNC" reference; we did spin off from TNC, but we have been independent for ten years. However, the taxobox is protected content. We'd very much like to replace references to TNC status and update the reference to "NatureServe" system-wide, if possible.

It raises an interesting question, too, about whether Wikipedia might wish to consume NatureServe's species web service to pull current rankings for Western hemisphere species without a Red List rank. If you could point me in the direction of a contact who might be interested in discussing a more formal partnership of this sort, I'd be very grateful.

All best,

Kyle

[...]

NatureServe

I fixed this in the relevant subtemplate. The images that are actually displayed with the status assessment (File:Status TNC blank.svg) still have "TNC" in their filenames; if that is seen as problematic, someone should ask for them to be renamed on Commons. Ucucha 00:42, 25 September 2010 (UTC)

Thanks for fixing that, Ucucha. That would be my fault. I didn't know the relationship of the two orgs and treated them synonymously. —Pengo 07:00, 25 September 2010 (UTC)

Is non-breaking space useful in Taxobox?

Wikipedia:Content noticeboard#Is non-breaking space useful in Taxobox?. Follow the link. --Snek01 (talk) 12:49, 28 September 2010 (UTC)

Automatic taxonomy generation

A recent publication (mentioned above, I think) bemoans a weakness of Wikipedia as the inconsistency in taxonomic nomenclature (see section "classification").

I frequently add articles about new taxa and it is frustrating to have to copy out the entire taxonomy each time.

Taxobox/Archive 19
Scientific classification
Domain:	Eukaryota
Kingdom:	Animalia
Phylum:	incertae sedis
Family:	†Vologdinellidae
Genus:	†Vologdinella

One solution to this is to automatically generate a taxonomy. I have implemented a rudimentary version of this, for comment. The taxobox to the right is generated by the following code:

{{Automatic taxobox
| taxon = Vologdinella
}}

This is accomplished by the maintenance of separate pages for each taxon; for instance Template:Taxonomy/Dasycladales contains the code

{{#switch:{{{1}}}
|rank=ordo
|parent=Ulvophyceae
|link=[[Dasycladales]]
}}

This change might raise a number of issues; to keep the discussion simple, could we discuss each issue in its own sub-section below?

Do let me know your thoughts!

Martin (Smith609 – Talk) 02:14, 10 June 2010 (UTC)

Benefits

• Easier to create articles
• Easier to maintain articles
• Ensures consistency between articles
• Centralizes discussion on taxonomy
• Keeping articles up to date is easier

Creation of sub-pages

• The proposed method creates a sub-page for each taxon. A similar system is in place at Template:Cite doi, and despite original concerns that vandalism or performance could be a problem, in practise neither issue has caused any problems.
• If vandalism does transpire to be a problem, the uniform output of the template makes it easy for a bot to spot problems, and potentially to correct them.

Discussion

• It would be natural to discuss each node of the taxonomy either on the template's own talk page, or the talk page of the related article. A talk page header could direct the editor from the template's talk page to the talk page of the related article if this was deemed sensible.

Initial adoption

• The traditional 'specify-every-rank' approach still works, so I anticipate this change (if made) being gradually adopted by editors who are able to check that it produces the expected effects as it rolls out.
• It should be possible to program a bot to keep watch for cases where an editor has used the new method, but the corresponding subpages haven't been created yet; in most cases, the bot should be able to create the subpages by referring to the taxoboxes on the appropriate articles higher up in the hierarchy.

• Support, mostly, though it was very cheeky of you to unilaterally edit such a big change into a fully-protected high-visibility template.
  Question: if I see fit to provide a detailed infrageneric classification in a genus taxobox, but wish to omit that detail in species taxoboxes, is this supported?
  I am also concerned about the effect on obsolete taxa. If an arrangement is published that renders a taxon obsolete, we must take great care not to make changes that position that taxon in the new arrangement.
  Hesperian 03:15, 10 June 2010 (UTC)

• On reflection you are probably right to accuse me of being hasty; I wanted to allow editors to see the proposed framework in action, but it would perhaps have been more appropriate to keep that in a sandbox. Consider me chastened!
• Regarding flexibility, see the comment below; regarding the obselete taxon, it might help me to understand if you gave me a hypothetical example. One thing that the new system can manage very well is synonymy; if Taxon A is synonymized with Taxon B then the "taxon A" subpage can redirect to "taxon B" and all pages in Taxon A will be instantly reclassified. If you are envisioning a situation where taxon X is recognized as invalid and human intervention is necessary to reclassify them, it should be very easy to mark all taxa that still call taxon X into "Category:Articles in invalid taxa" or the like, allowing for cleanup. If I've misunderstood your point, feel free to elaborate. Martin (Smith609 – Talk) 13:03, 10 June 2010 (UTC)
  • It is a problem that already exists, but which would be greatly exacerbated by this new system. Many times I have seen the situation where a species article is created, and a taxobox generated, based on an outdated taxonomic arrangement, such as the Cronquist System. For example, the taxobox might say Magnoliopsida—Asteridae—Gentianales—Asclepiadaceae—Stapelia. It's outdated, but at least, within its outdated context, it is correct. Then someone comes along and mindlessly switches the higher taxa over to the modern APG system, making it Angiosperms—Eudicots—Asterids—Gentianales—Asclepiadaceae—Stapelia, without bothering to check whether Asclepiadaceae is even accepted in the APG system; it isn't. Thus the taxobox is changed from correct within an outdated context, to just plain wrong. And yet it looks righter than the previous version, because it has that gloss of modern scholarship on it. The proposed system is going to make these problems much worse, because rearrangements of higher taxa like in my example will be implemented simply by changing the /Gentianales template, and in one fell swoop thousands of article taxoboxes will change, without regard to the systems, circumscriptions or assumptions under which those taxoboxes were written. Hesperian 13:43, 10 June 2010 (UTC)
    • Although on the other hand, the new system will make it very easy to fix this situation by modifying the "parent" of "Asclepiadaceae". Perhaps the best thing to do is to leave a note on higher taxa warning editors of the potential issues that an edit could create. Alternatively, a toolserver tool could be designed to allow editors to see the lower taxa that will be impacted by their edit as they make it. I do agree that a case-by-case modiication of taxoboxes from the old system to the new sounds good, though. 192.75.204.31 (talk) 21:18, 14 June 2010 (UTC)

• Comment My concern here is flexibility. In an article by article basis, it is sometimes appropriate to include or exclude higher taxa relevant/irrelevant to the species being discussed. If a paper on a new dinosaur makes much of the fact that it's a member of Superfamily Hadrosauroidea, but lies outside Family Hadrosauridae, then clearly the taxobox should read as Family Hadrosauroidea Genus X. But it would be rather pointless and redundant to list Superfamily Hadrosauroidea in the taxobox of all the numerous Hadrosaurids--at that level of granularity, it's more informative to the reader which subfamily they belong to. How would these issues be addressed? Would all taxoboxes simply do away with sub- or superfamilies, -orders, -classes, etc.? Or would we have to decide to include all these by default if they're useful in x% of articles? MMartyniuk (talk) 07:29, 10 June 2010 (UTC)

  That's a great point, thanks for making it. Perhaps the best solution is to automatically generate the "skeleton" of the taxonomy (displaying nothing at the sub- or super- level), but to allow editors to "switch on" more detailed levels on an article by article basis (ideally by using |superfamily=show although an alternative would be to use |superfamily=Hadrosauroidea). Another idea would be to include granularity based on the rank of the taxon in question. So on genus pages we would display "subfamily" but not "superfamily", if either existed; on an order page we would display "superorder" and "subclass" but not "subphlyum". This would improve consistency and reduce editor effort, but at the cost of fine control. Do you think that the automatic approach would consistently provide the optimal output? Martin (Smith609 – Talk) 13:03, 10 June 2010 (UTC)

  I think the second approach (granularity based on rank) sounds like a good solution. I'm sure there are cases where we'd need finer ability to tweak, but I can't think of any right now and maybe having which ranks are displayed on which pages become standardized would outweigh those concerns anyway. MMartyniuk (talk) 00:28, 11 June 2010 (UTC)

  Actually this particular example doesn't seem problematic to me. Hadrosaurs will have parent=Hadrosauridae, which won't list the superfamily, but the new dinosaur would have parent=Hadrosauroidea, which would. Nonetheless the point stands that policy on which higher taxa to display needs to be flexible. Hesperian 13:24, 10 June 2010 (UTC)

  So if we are to try out this new system, sub- and super- taxa should only be displayed if they are specified by the parent= parameter in the template? The flexibility of higher taxa sounds like something that we need more data to clarify; using the new style of template on a few pages sounds like the best way to get some idea of the best way to get an idea of the sort of flexibility that is required. Should we proceed to this stage? (If so, then there is some substantial tweaking to be made to the prototype before it is ready for the prime time...) Martin (Smith609 – Talk) 21:22, 14 June 2010 (UTC)

• Work has begun at Template:Automatic taxobox. Feedback is welcome, although the template may not be stable enough for mass usage yet. Martin (Smith609 – Talk) 16:49, 8 July 2010 (UTC)

• Oh no, please don't do this. This makes it nearly impossible to extract infobox data from articles. —Preceding unsigned comment added by 91.64.30.249 (talk) 22:53, 8 July 2010 (UTC)

• Could you explain what you mean? Species microformats are still produced by the template, which makes all the information readily accessible. Martin (Smith609 – Talk) 22:57, 8 July 2010 (UTC)

• But it'd be necessary to render the page/template to get the microformats. currently one can extract the information from the wikitext without having to render the article, and many projects are doing that. the proposed template change would break that work. —Preceding unsigned comment added by 91.64.30.249 (talk) 23:04, 8 July 2010 (UTC)

  It's still possible to obtain taxonomy from the wikitext (details to follow at Template:Automatic taxobox/doc; this method has the advantage that the taxonomy generated will more often be the correct / up-to-date taxonomy. Martin (Smith609 – Talk) 13:45, 9 July 2010 (UTC)

• Love it! I can't wait to see this launched! Bob the Wikipedian ^{(talk • contribs)} 20:57, 17 July 2010 (UTC)

I like this idea in general, although I'm not certain I understand the details. I think the idea that a subfamily be displayed if it exists on a genus page is not a good one. In some plant families, for example, tribes are used, but in others they are not. If one is discussing legumes, for example, including the tribe would be important, but this is not the case with all plant families. A crowded taxobox with unused information seems the opposite of what the plant editors on wikipedia appear to be striving for: taxonomies that reflect what are actually used for that taxon. This isn't successful for all plant articles, of course, but I think when dealing with something as currently dynamic as taxonomy, in the age of rapid molecular sequencing, forcing inclusions negates the beauty of this tool.

I am concerned about the automatic taxonomies being used in certain groups, such as unicellular-things-which-once-were-algae groups. In large areas of photosynthetic single-celled marine organisms, taxonomies have been changed for only some specimens in the group. In the example above, this could result with potentially 100s of algae articles having inconsistent and wrong taxonomies due to higher level changes in automatic taxoboxes. In addition, wikipedia algae taxonomies are already full of mistakes, and adding more mistakes courtesy of an automatic taxobox could make a really bad situation much worse.

I like the idea, though, for chosen taxa. If an editor is certain that it will work for some taxon, then applies an automatic taxobox, that could make clean-up and editing much faster and easier, benefiting the reader eventually/

I don't understand the implementation, though. I'm trying to read as much as I can, but the examples just take me to pages that say things like "taxobox template."

How is it used? A specific example would help. --70.57.229.24 (talk) 06:46, 23 September 2010 (UTC)

In this stage of the template development, I am recommending that the template is only implemented in cases where it will certainly work. I'm currently waiting for approval for a bot task that will greatly facilitate the implementation; the editor won't have to edit any complicated templates but will be prompted to correct data extracted from the current Wikipedia taxonomy. See the current implementation and the bot request. The hope is that eventually, systematic mistakes in e.g. algal taxoboxes can be changed much more simply; if a genus is moved to a new family, for instance, then only one edit is necessary to update all the child genera's taxoboxes, rather than individual articles being updated on an ad-hoc basis. I'd certainly value your feedback as the template is rolled out across more-problematic articles, and if you can see any instances where the automatic taxobox would produce an undesired output, please let me know so that I can work on a resolution. Martin (Smith609 – Talk) 12:53, 23 September 2010 (UTC)

I think if you stick with editing the direct child's taxoboxes with the update, you may miss some of the big problems that exist in taxonomy today, but get incorporated as decided and true in wikipedia at times.

This is a major problem in wikispecies for example: the taxonomies as taxonomists are working on them today, are not settled to what wikispecies has adopted for its taxonomies. I don't want to implement anything in wikipedia that locks down taxonomies that aren't locked down. --70.57.229.24 (talk) 17:27, 23 September 2010 (UTC)

I'm interested to know how these problems are handled in the current taxobox system? Martin (Smith609 – Talk) 19:46, 23 September 2010 (UTC)

Often poorly. Ideally, and some of the plant editors are rather good at this, we careful write the taxobox with the known information, the taxon's name, for example, and we wikilink it, and we include text in the article explaining the omissions. I saw somewhere in this discussion where a couple of editors dismissed this technique. It's not possible to dismiss this technique, it's a reflection of the current dynamic state of taxonomy and phylogenetic systematics with great changes at higher taxonomic levels and more precise information at the genus and species levels due to the availability and speed of molecular methods, coupled with group endeavors in the sciences and high computing rates. You're not going to get all of wikipedia taxonomically organized when taxonomy is not.

With the automatic taxobox, flexibility will be the key that makes it usable, while simultaneously giving wikipedia articles some fast updating and consistency wherever possible. In my opinion, keeping this as flexible as possible will mean fewer problems in the short and long run, while we gain the benefit of tens of thousands of articles gaining consistent taxonomies with far less editing effort required. --70.57.229.24 (talk) 06:31, 24 September 2010 (UTC)

Thanks for that background. It sounds like the current system will be able to accommodate what you have described; making it intuitive to the user may be a little trickier. If you are able to provide a couple of examples of plants that would fall into that category, that'd be really helpful, so that I could see how they are produced and check that the format can be reproduced. It sounds like it might also be worth opening a discussion at WikiProject Plants or WikiProject Tree of life to generate consensus on the best way to represent taxonomic uncertainty. Martin (Smith609 – Talk) 17:41, 24 September 2010 (UTC)

Unranked clades

Does this template take account of unranked clades ? Look at the taxobox in the article Cerithium to see what I mean. JoJan (talk) 20:44, 9 July 2010 (UTC)

Yes, it can handle as many layers of clades as necessary, treating each clade as an "unranked" taxonomic rank. Martin (Smith609 – Talk) 21:36, 9 July 2010 (UTC)

And I've taken your suggestion of Cerithium as a test case, and implemented the automatic taxobox there. Please let me know if it has worked to your satisfaction. Thanks, Martin (Smith609 – Talk) 19:31, 14 July 2010 (UTC)

It works, but it have changed the taxobox from

(unranked): 	clade Caenogastropoda
clade Sorbeoconcha

to

(unranked): 	clade Caenogastropoda
(unranked): 	clade Sorbeoconcha

Imagine the situation for another real example Helix pomatia

this would change taxobox from

(unranked): 	clade Heterobranchia
informal group Pulmonata
clade Eupulmonata
clade Stylommatophora
informal group Sigmurethra

to

(unranked):    clade Heterobranchia
(unranked):    informal group Pulmonata
(unranked):    clade Eupulmonata
(unranked):    clade Stylommatophora
(unranked):    informal group Sigmurethra

This have added 5 unnecessary words. Consider, that we need all of its taxa to be displayed. A little improvement would be fine, but also try to keep the code simple. --Snek01 (talk) 10:44, 15 July 2010 (UTC)

Five (unranked) taxa between class and superfamily seems excessive to me. None of these clades are familiar to the lay reader, and only one is mentioned in the text (viz. Pulmonata). In almost all other groups, there would only be a single rank in that space, the Order; I would suggest cutting down your list of five clades to the single group Pulmonata, at least until someone produces a usable system within the Linnean system of ranks. That way, there is no problem of repetition of the text "(unranked)". --Stemonitis (talk) 11:22, 15 July 2010 (UTC)

Yes, it can seem to be too much taxa for this example. This is why the example was chosen. But it is either possible to delete all unranked taxa in this example or keep all unranked taxa. Nobody was able to choose some of them and there have to be something, how the species of gastropods are classified above superfamily level. The actual situation is, that all of these unranked taxa are used in the Wikipedia:WikiProject Gastropods. Even if some rows will be reduced, this very minor problem will stay for the taxobox in the article Sigmurethra and Aeolidida and so on. --Snek01 (talk) 12:13, 15 July 2010 (UTC)

One solution for this will be to hide the caption for lower level clades (by using rank=display_nothing, this rank is not implemented yet). — Ganeshk (talk) 12:38, 15 July 2010 (UTC)

Another option would be to replace "(unranked):" with "clade:" or "informal group:" in the left-hand cell.

However in this instance it is not clear that all the levels need to be displayed on every page. For instance, Stylommatophora (according to its article) was previously regarded as an infraorder; thus it is probably not necessary to show it on all of its child genera. Also, because Eupulmonata and Pulmonata are rather similar, I would argue that it would be possible to remove one level or the other without causing any confusion; anybody knowledgeable enough to understand the distinction will presumably recognize that a eupulmonate is necessarily a pulmonate. Of course, in the case of the Sigmurethra, its immediate parent (Stylommatophora) will be shown by default, and it is possible to over-ride the default behaviour and show all ranks if necessary in this case. Martin (Smith609 – Talk) 14:14, 15 July 2010 (UTC)

I've implemented this at Helix pomatia as suggested; there is something of an overburden of ranks, perhaps some (the informal groups?) could be removed? Martin (Smith609 – Talk) 20:24, 16 July 2010 (UTC)

• There is no need to unreasonably display a rank in italics.
• There is not possible to randomly reduce number of ranks in the very difficult taxonomy of gastropods although it seems a bit overburden. There exist neither universal nor widely known way how to do it.

This looks like this now:

clade:          Heterobranchia
informal group: Pulmonata
clade:          Eupulmonata
clade:          Stylommatophora
informal group: Sigmurethra

From taxonomic point of view of taxonomy of gastropods this can be acceptable. --Snek01 (talk) 00:21, 20 July 2010 (UTC)

Unsettled phylogenies

• Comment How would this handle situations like Lemur, specifically Lemur#Taxonomic classification and phylogeny? In the lemur community, arguing over superfamilies, infraorders, and other ranks has quieted down over the past decade, but certainly not because of agreement. Instead, many researchers refuse to touch higher-level taxonomy, instead preferring to discuss phylogeny and phylogenetic trees. Hence the "See text" in the taxobox. I'm sure this problem is not restricted to lemur research. – VisionHolder « talk »

• My solution to this situation would be simply not to list an infraorder in the taxobox. The current "see text" statement is not ideal as it will confuse any tool that extracts information from taxoboxes. In other situations where there is no formally accepted infraorder, the level is not specified. Is this solution unacceptable? Martin (Smith609 – Talk) 14:25, 12 July 2010 (UTC)

One issue that remains is: How do we define whether or not something is formally accepted? In the case of lemurs, I combed the literature and have spoken directly with several of the experts. What about more obscure taxa? Do we need to be careful before we start assuming that the taxonomy we promote is what is accepted. Then again, I'm guessing we can cross those bridges when editors get to those articles. – VisionHolder « talk » 17:47, 12 July 2010 (UTC)

By having a centralized taxonomy generation system, it then becomes possible to have a single discussion about the phylogeny of group x and have this decision implemented throughout the encyclopaedia, rather than piecemeal as per the current taxoboxes. Every page that displays a taxobox implicitly follows a certain taxonomic scheme, and I would be willing to bet that different taxonomic schemes are in place in different articles. Martin (Smith609 – Talk) 19:55, 12 July 2010 (UTC)

This I have no problems with! I'm just worried that we'll implement one taxonomic system in preference to another without any research to see which system is most widely supported. But then again, the same can be said of Wikispecies, which miraculously has one taxonomic system for everything that I've seen so far. – VisionHolder « talk » 02:19, 14 July 2010 (UTC)

• Comment - I think the use of this scheme for arthropods could be problematic. Much of the arthropod taxonomy is still unresolved or disputed. In many cases the unresolved portions are simply skipped over in the taxobox listings, which seems to be a good idea, IMO. In addition, the taxonomy for some arthropod groups (esp. insects and spiders) often involve numerous levels of unranked clades, some of which are important and some of which aren't. Showing the full tree would be obnoxiously long in most cases. Would we have to specifically turn off the 40 or so ranks that we didn't want displayed? This seems like it would complicate the taxobox code by an order of magnitude. Kaldari (talk) 16:45, 14 July 2010 (UTC)

Currently, only the levels of unranked clade that are immediately above the taxon of interest are displayed. Would it be necessary to display higher-level unranked clades on species- or genus-level taxa? Are there higher-level unranked clades that should always be shown on all of their child taxa? (This should be possible to handle.) It is also possible for the automatic taxobox to "jump over" disputed taxonomy. It would be really valuable, if you wouldn't mind, if you could suggest a couple of pages that you can foresee problems with; then I can see if the automatic taxobox template is capable of reproducing the existing taxoboxes. Martin (Smith609 – Talk) 18:45, 14 July 2010 (UTC)

I don't think that's true, Kaldari. I see no reason why the proposal should be any more problematic for arthropods than any other organism. As explained above, it would only include those ranks which were most important (usually kingdom, phylum, class, order, family, genus, species, but with the potential for other taxa which are always important, including, for instance, subphylum Crustacea and infraorder Brachyura). We would no more need to include every ranked taxon or unranked clade for an arthropod than we would include all the taxa leading to Homo sapiens (NCBI taxonomy: "root; cellular organisms; Eukaryota; Fungi/Metazoa group; Metazoa; Eumetazoa; Bilateria; Coelomata; Deuterostomia; Chordata; Craniata; Vertebrata; Gnathostomata; Teleostomi; Euteleostomi; Sarcopterygii; Tetrapoda; Amniota; Mammalia; Theria; Eutheria; Euarchontoglires; Primates; Haplorrhini; Simiiformes; Catarrhini; Hominoidea; Hominidae; Homininae; Homo"). I think this proposal would actually help to keep taxoboxes down to manageable sizes, simplifying matters enormously. The proposal is similar to the way things work at Wikispecies, where most of the taxonomy displayed on a page is produced from a template for the parent taxon, which in turn transcludes a template for its parent, and so on. --Stemonitis (talk) 17:09, 14 July 2010 (UTC)

I guess I just don't have a good idea of how this scheme is actually supposed to work. If it's limited to the major ranks by default, and you can turn any ranks on or off, I suppose that would work fairly well. Kaldari (talk) 18:54, 14 July 2010 (UTC)

The scheme will only display major ranks by default. In addition it will display the immediate two parent ranks to the taxon, whether they are major ranks or not. I have just been adding functions whereby the user can over-ride this, to display more or fewer parent ranks (e.g. if we need a tribe, subgenus, microgenus and nanogenus to be displayed).

In addition, in the light of your comments, I have added a function whereby a higher, unranked taxon can be considered a "major taxon" for display purposes: this may be useful if everyone agrees that X is (say) a pancrustacean, but the taxonomic rank of pancrustacea (class or superclass?) is disputed. Then "Pancrustacean" will always display, but as an unranked taxon. I hope that that makes things a little clearer; I'll set to work on documenting things a little more transparently as I go. Martin (Smith609 – Talk) 19:09, 14 July 2010 (UTC)

Overall goal

However, I do have a more general question. What is the overall goal? The discussion started with a reference to a paper talking about Wikipedia fulfilling E.O. Wilson's vision of an online species database, but the rest focuses on simplifying article creation and centralizing the taxonomic info. So which is it? Are we aspiring to eventually fulfill E.O. Wilson's vision? If so, and if this approach is a first step, then we might need to give this more careful consideration before we get too far along. For starters, I have been under the impression that these other attempts at online databases favor a phylogenetic approach to classification, not the old Linnean system. (I could be wrong.) This issue may tie in with handling unranked taxa. I guess before I can give my full support for this, I need to know what we are aiming for and how we plan to work with the scientific community in terms of classification. – VisionHolder « talk » 17:47, 12 July 2010 (UTC)

For me, the most important objective would be to ensure consistent classification within Wikipedia. I think that E.O. Wilson's vision is a worthy goal to which to aspire, and hopefully this system is a useful step in that direction – if it can be honed in this respect, then I agree that it should be. The system has no limitations on the number of unranked taxa that can be specified; the Linnean system is a phylogenetic system, albeit one where certain clades are assigned (somewhat arbitrary) ranks. It should address the problem raised in the paper.

Perhaps it would be an idea to trial the taxobox in a given taxon that may be problematic, to see in practise what issues may arise? This could give us ideas on where to go next, and would perhaps bring the template to the attention of other editors. Martin (Smith609 – Talk) 19:55, 12 July 2010 (UTC)

I would be willing to try this system out with the lemur articles, but since I'm the most frequent editor and most of the editors that view it are FAC reviewers, I don't know how effective of a test it would be. I agree with your statement about the Linnean system being phylogenetic with arbitrary ranks, but there is a reason why it is falling out of favor with some researchers. The "tarsier conundrum" or lemur taxonomy are classic examples, where people generally agree on who is most closely related to whom, but can't agree upon a taxonomic nomenclature for higher-level taxa because you can go either way. By giving up on naming the rankings, you remove a lot of unnecessary argument. Anyway... let me know what you have to say. I'll respond again when I haven't had anything to drink for the night. – VisionHolder « talk » 02:19, 14 July 2010 (UTC)

Perhaps a group with a larger editor base might be even better? We want to be sure that we have eliminated as many teething troubles as possible before using the template too widely. A [with an associated WikiProject] might be a good start to solicit input; candidates might include dinosaurs, gastropods, lepidoptera, sharks, cetacaeans or cephalopods.

Wikipedia:WikiProject Cephalopods might be a good place to start as the top item on their task list is "Ensure all Family articles are taxonomically consistent". The scope is small enough that it should be possible to roll out the taxobox across all families at least with relative ease, yet large enough that we might expect to hit any snags that are going to be widespread. If people agree that this is a good taxon to start with then I can solicit input directly from the wikiproject. Martin (Smith609 – Talk) 13:30, 14 July 2010 (UTC)

Comment: benefits of this template are "Ensures consistency between articles". But the question if it is "Easier to create articles" and "Easier to maintain articles". When somebody will want to change taxonomy of some species, then it is (meantime) very difficult to do it. The user have to manually find out the certain template of certain taxon. When the user does not known this, he/she is not able to update the taxonomy. This should be improved somehow. --Snek01 (talk) 12:13, 15 July 2010 (UTC)

Do we want to use only automatic taxoboxes in the future instead of standard taxoboxes? If yes, then the discussion should led to the task, how to add features of automatic generated taxonomy directly into {{Taxobox}}. If no, then there are always also other possibilities or mixed variants. For example, consistency of taxonomy can be checked, changed and updated automatically also on different principle. If something should be done automatically, then we can (theoretically) directly use taxonomy from Wikispecies so wikipedians will not need to take care about taxonomy in taxoboxes at all. If somebody mention "Centralizes discussion on taxonomy" as a benefit, then it can be noted that Wikispecies already is centralized discussion about taxonomy. Maybe decentralized discussion about taxonomy of each species is a feature of wikipedia. This is great challenge to keep all features. --Snek01 (talk) 12:13, 15 July 2010 (UTC)

Examples

If it helps to the template in operation; it has been implemented on the following pages:

• Blattellidae
• Vologdinella
• Lingulella and parent taxa
• Hadrosaurid (note handling of infra-order and super-family)
• Cerithium (note handling of higher-level unranked taxa that should (e.g. caenogastropoda) and should not (e.g. eumetazoa) be displayed)
• Cephalopod (class-level)
• Amanita muscaria (species-level)

Martin (Smith609 – Talk) 19:21, 14 July 2010 (UTC)

Initial feedback

• Hadrosaurid has the family wrong; it should say "Hadrosauridae" in the taxobox. (For that matter, the article itself should probably be called Hadrosauridae, but that's another discussion.) The others look good, except that I'm not sure that Lingulella should have the superfamily in the taxobox. Ucucha 19:35, 14 July 2010 (UTC)

  This issue has been bugging me for a while, so I brought it up for discussion here: Wikipedia_talk:WikiProject_Tree_of_life#Article_name_format_for_taxa MMartyniuk (talk) 00:13, 15 July 2010 (UTC)

  That was my error, which I have now fixed. It's highlighted the need for a more intuitive user interface, though. That's probably something for down the road. Martin (Smith609 – Talk) 20:20, 14 July 2010 (UTC)

  I've modified the behaviour so that by default, only the immediate parent taxon will display, thus hiding the "superfamily" in the Lingulella example. This can be manually over-ridden by specifying |display_taxa=number of parent ranks if it is desirable to illustrate more. Martin (Smith609 – Talk) 14:06, 15 July 2010 (UTC)

Resolved

• I see that in the species-level example (Amanita muscaria), we've lost the old "Binomial name" section below the main hierarchy. Is that a deliberate change, or perhaps a work in progress? I can see the potential logic behind dropping the redundancy of species= and binomial= (without expressing an opinion either way), but I think such a change should be explicitly discussed. --Stemonitis (talk) 19:51, 14 July 2010 (UTC)

  A deliberate change by me to remove redundancy, which I have reverted to avoid controversy. Martin (Smith609 – Talk) 20:20, 14 July 2010 (UTC)

Resolved

Categories

The template seems to be invoking categories, badly. At Maja (genus), it is trying to assign the article to Category:Majidae, which doesn't exist (nor should it, for the moment). Similarly, Octopus is given Category:Octopodiformes, and it shouldn't be. (And so on, for all the other examples implemented so far.) I can't work out where it's coming from, but it needs to be stopped. Categorisation is not something that can or should be automated. --Stemonitis (talk) 21:48, 16 July 2010 (UTC)

Agreed, on all points. I just noticed it at Levenhookia chippendalei as well, which is being categorized into Category:Coleostylis, the section of the genus that has three species and should never be a category. Rkitko ^(talk) 21:57, 16 July 2010 (UTC)

Found it and removed the code diff. Auto categories should be gone now. Rkitko ^(talk) 22:08, 16 July 2010 (UTC)

Just out of interest, why shouldn't categorization be automated? In particular, it is a useful way to determine the child nodes of a taxon. Martin (Smith609 – Talk) 22:27, 16 July 2010 (UTC)

There are too many problems with it. Not every parent node will be a category or should ever been a category (e.g., small genera are usually categorized in the family category, subgenera and sections almost never have categories of their own). If the template were to allow editors to choose the category instead of defaulting to the parent, it would be no different than letting the editor choose the category down near the others, where everyone already expects categories to be. Also, some taxa categories aren't located at the scientific name, e.g. it's Category:Cacti, not Category:Cactaceae (for now, at least). Can't child nodes be sourced from the transclusions of the Template:Taxonomy/foobar template? Seems an easier way to do it without creating unnecessary redlinked categories. That would be the argument against. Rkitko ^(talk) 22:46, 16 July 2010 (UTC)

Agree. Terrible idea. Martin, you really have to stop adding features without discussing them first. This is a high-use protected template. The way you are going about this is not acceptable. Hesperian 05:13, 17 July 2010 (UTC)

Note that these changes to Template:Automatic taxobox and its dependencies do not affect Template:Taxobox (except where the new "parent" parameter is used, of which I am aware of perhaps one). Perhaps discussion should be moved there, although it might get missed by other interested parties. Martin (Smith609 – Talk) 22:49, 17 July 2010 (UTC)

I think that there is a case for there to be some way to retrieve all child nodes of a taxon, independently of a potentially outdated editor-supplied list in the taxobox (via |subdivision=). Categories seem to be to be the only way to do this reliably; I am not sure that a bot could be programmed to accommodate the nuances of the non-uniform taxon specification in |subdivision=. Transclusions list all descendants, not just direct children (see Special:WhatLinksHere/Template:Taxonomy/Cephalopoda). It sounds like a better option would be to use a suite of hidden categories with names such as Category:Taxonomic categories/Cephalopoda, which would be easier to maintain by bot (i.e. the bot would just have to add {{Hidden category}} and the parent taxon to the new page). Martin (Smith609 – Talk) 22:49, 17 July 2010 (UTC)

What is that case? All it would recover is what existing articles are direct children of the taxon in question (which may depend on what ranks we've included; for some genera it would be subgenera, for some [plants] sections, and for others, species, for instance). To get verifiable information on what daughter taxa a taxon contains, the place to look should be the article text. I for one would need some convincing that a vast array of ghost categories largely duplicating the existing taxonomic category hierarchy would bring sufficient benefit to warrant the large number of edits and the large number of extra pages that goes with it all. If the list in the taxobox (or, better, the text) is outdated, then that is what needs to be fixed, rather than trying to sidestep it with automated categorisation. --Stemonitis (talk) 06:12, 18 July 2010 (UTC)

Usability

• A small question: Should we be bringing the new template "live" on articles already? I know a lot of new articles are created by just copying and pasting a taxobox from another article, then modifying. The number of articles is small right now, but won't the template be broken if it's used on a new article outside the scope of those prepared for it? Rkitko ^(talk) 23:16, 14 July 2010 (UTC)

  You are certainly invited to use the new template whereever you wish; real-world testing is the best way to identify potential pitfalls. But I would not yet recommend wholesale roll-out of the template to constrain the potential damage caused by future edits. It would be really helpful if you could give feedback on how intuitive the process is as I fear that the instructions might be a little obtuse; let me know how you get on if you add it to a new article! Martin (Smith609 – Talk) 23:23, 14 July 2010 (UTC)

  Hmm, yeah. Even for someone somewhat experienced with templates, the instructions are a bit hard to follow. And it's a lot more work than putting together just one taxobox! I'm sure once lots more taxa are filled in, new additions will be easier, but this is confusing to begin with. I wonder two things to begin with: 1) I tried it at Levenhookia chippendalei, but met problems when I wanted the parent taxon to be the section Coleostylis and I wanted it to be formatted just like that. I suppose the entry at Template:Taxonomy/Levenhookia chippendalei should be parent=Coleostylis, right? But the page will never be located at Coleostylis and never should (and something unrelated might be there). How do you format entries for the parent taxa? And 2) What are we to do with, say, a plant and animal genus with the same name, e.g. Leucoptera (plant) and Leucoptera (moth). I'm sure there's a way around that, but those are the special cases that should be described somewhere. Rkitko ^(talk) 23:48, 14 July 2010 (UTC)

  Oh, and I see the species parameter autofills to include the full name instead of an abbreviated genus, e.g. Levenhookia chippendalei instead of L. chippendalei. Is there a way to go back to the abbreviate form? I prefer that in the species parameter since the binomial parameter lists it in full. I've also run into places where even with the abbreviation or in the binomial parameter, the name is so long, it forces the name onto two lines. Any way we could make that autoinsert a no-break-space to prevent that? Rkitko ^(talk) 23:56, 14 July 2010 (UTC)

  Ok, I guess the most annoying thing is the link= part. Could we make that not automatically substitute the subpage name? It would allow for easier customization and would be better understood by the uninitiated. So I figured out the nbsp by editing that parameter. I must also note that it is excruciating when you have to wait for the templates to update with changes, but that's normal lag. Just more frustrating than creating an in-article taxobox... Rkitko ^(talk) 00:10, 15 July 2010 (UTC)

  Thanks for all your comments, I'll look into them fully when I have time. I am to produce a toolserver tool that can automate some of the laborious parts and remove the waiting, whilst making the user-interface more intuitive; perhaps this should be my next objective. And of course the higher taxa only need to be created once, so creating additional Asterales (etc) will now be easier... Martin (Smith609 – Talk) 14:33, 15 July 2010 (UTC)

  I've amended the edit notice in light of your comments; this hopefully keeps things as simple as possible without a toolserver tool. Martin (Smith609 – Talk) 23:43, 15 July 2010 (UTC)

Solutions of some these problems are mentioned bellow at #Broader implications. --Snek01 (talk) 00:24, 20 July 2010 (UTC)

I am a "primitiev" user, in that I just copy taxo-boxes from some related group and modify them whenever I need one. I have tried to get the new taxobox to work with the genus Cyanea (jellyfish) (it's a plant genus by the same name, hence the brackets). The taxobox reports that taxonomy is not available, though I have made sure the family Cyaneidae include Cyanea (jellyfish) in the "genera" section of the taxobox. What am I doing wrong? Petter Bøckman (talk) 19:54, 20 July 2010 (UTC)

Petter, what that means is that you have to create Template:Taxonomy/Cyanea (jellyfish), following the instructions in the [incomplete] taxobox. I have done that for you, now. Because the automated system is still in its infancy, I also had to make most of the rest of the hierarchy, too (Template:Taxonomy/Cyaneidae, Template:Taxonomy/Semaeostomeae, Template:Taxonomy/Scyphozoa and Template:Taxonomy/Cnidaria). Once more of the taxonomy is in place in these templates, there will be less work for each new article. --Stemonitis (talk) 20:36, 20 July 2010 (UTC)

Thank you for fixing my jellyfishes! I must admit I have read your explanation 3 times and I'm still sitting here gaping at it with blank incomprehension. How do I make a taxobox that work and anchor to the existing hierarchy by copying from another organism? The usability aspect of this new taxobox is somewhat ... lacking for the moment. Petter Bøckman (talk) 21:10, 20 July 2010 (UTC)

That's a fair point. Copying a taxobox from another article is no longer any guarantee of success, because all the details are behind the scenes in subpages of Template:Taxonomy. Some of those can be reused, but it's not as straightforward as the existing system. --Stemonitis (talk) 05:21, 21 July 2010 (UTC)

The concept of an automated taxobox is a great one, but it must be made so that it does not scare off potential editors. Remember most editors have extremely limited knowledge of how a template works (hence they'll copy an existing one), and a many are not native English speakers. Would it be possible to have a hidden (only visible in edit-mode) link in the taxobox to the templatepage, and a proper idiots guide on how to use the taxobox (included solutios to common problems,written in English, not Technicalese)? --Petter Bøckman (talk) 07:02, 21 July 2010 (UTC)

If anything needs translated from Technicalese to English, let me know on my talk page and I'd be happy to do it. Bob the Wikipedian ^{(talk • contribs)} 01:39, 25 July 2010 (UTC)

High taxonomy length causes failure

• I tried the autotaxobox on Xenophora solarioides. Super family and Kingdom are not showing up. What did I do wrong? Here is how it looked before the change. — Ganeshk (talk) 03:57, 15 July 2010 (UTC)

I just added a subgenus to the species. That made the colors go awry. — Ganeshk (talk) 11:39, 15 July 2010 (UTC)

The superfamily probably should not show up (to avoid nomenclatural overload); if you want it to appear on all descendants, see the comment below.

I've fixed it so that Kingdom does appear.

The colour problem is odd; it seems that some limit or other is being exceeded, as {subst:}ing the template does produce the colour. I can't presently work out what is happening; I imagine that it is a quirk of the MediaWiki software that I'll have to look into. Martin (Smith609 – Talk) 16:42, 15 July 2010 (UTC)

Perhaps WP:Template limits. Ucucha 17:09, 15 July 2010 (UTC)

Thanks, that gave me what I needed to resolve the problem. Martin (Smith609 – Talk) 23:32, 15 July 2010 (UTC)

It looks good now. Thanks. — Ganeshk (talk) 12:09, 16 July 2010 (UTC)

Resolved

Redirects and higher-level subtaxa

• I have given the new system a go at Maja (genus). One small point I notice is that using links to the same article to produce the boldface for the article's subject in the taxobox means that when an article is moved, the taxobox will contain a self-link through a redirect until the taxonomy template is updated. For example, if someone moved "Maja squinado" to "spider crab", the genus line in the taxobox would read "Maja squinado" (which would redirect back to the article now at "spider crab"), rather than "Maja squinado". (This is a minor problem, and just needs to be carefully documented, and probably worded better than I have done.) A bigger issue is the inclusion of important taxa at minor ranks. For every crustacean, I have included the taxon Subphylum Crustacea in the taxobox, and for every crab, I have included Infraorder Brachyura. These are both important taxa, because the average reader understands what a crustacean is, and what a crab is, but should not be expected to understand what Malacostraca means, for instance. In fact, a reader may well recognise only the taxa "crab" and "crustacean" and none of the others. I have not been able to see a way to force these ranks to be included; if none exists, I think it should be implemented, albeit with a note that it should be used sparingly. --Stemonitis (talk) 07:31, 15 July 2010 (UTC)

   Done – implemented; add |always_display=true to higher taxa that should be displayed on all their descendants. This has replaced the "displayed_unranked" taxon name used above. Martin (Smith609 – Talk) 16:34, 15 July 2010 (UTC)

Excellent! Many thanks. --Stemonitis (talk) 16:36, 15 July 2010 (UTC)

And I think I've addressed the bolding problem; see squid ( = Teuthida). Martin (Smith609 – Talk) 00:44, 16 July 2010 (UTC)

Broader implications

If I understand correctly, this system will eventually require a set of pages that list the immediate parent of each taxon that is used in a Wikipedia taxobox. That is, in fact, very similar to the content of Wikispecies pages. Is such duplication a good idea? Ucucha 18:13, 17 July 2010 (UTC)

Additional external databases that operate in a similar fashion are listed at Template:TaxonIds; I am aware of no way of making this information available to Wikipedia without first duplicating it. If the set of pages are to be generated en masse to make things as simple as possible for editors, then these external databases might be a good data source; a bot could recover the taxonomy listed there if the information is not available in Wikipedia and adopt it as a "starting point". This depends on the quality and availability of information in these external databases, and the extent to which automation is beneficial. Martin (Smith609 – Talk) 23:12, 17 July 2010 (UTC)

Beware of importing taxonomic data from external databases and/or Wikispecies. Most non-specialized external databases (and Wikispecies) are less taxonomically up-to-date than Wikipedia's existing taxonomic information. Of course there are many exceptions. Personally, I would prefer to see this system adopted organically, rather than automatically generated from questionable sources (including our existing taxoboxes). Kaldari (talk) 01:01, 20 July 2010 (UTC)

Alternative

How to make Automatic taxonomy generation easier? The first original proposed solution of Automatic taxonomy generation is/was based on doi template just to show how can such system work. The principle of this is using parent taxon. The solution based on doi template requires for each taxon a external template with fulfilled three rows like this:

{{taxonomic unit|{{{1}}}
 |rank=
 |parent=
 |link=
 }}

But this is much more difficult!

How is possible to do it in the easiest way?

• Not to use external template and add all necessary infomation in the article itself.
• Not to use unnecessary parameters (rows) and minimize number of them.

I would expect the most easy solution like this.

• To existing template {{Taxobox}} implement the new line

| parent = 

This will solve all two theoretically proposed simplification. The information is editable directly in the article and the number of rows is limited as much as possible.

Are there also other possibilities for simplification? Yes, of course:

• All species have its parent taxon automatically known, because first part of the binomial name of all species is the generic name. This is great advantage for automatization and for species (for all species that have fulfilled its binomial name) there is no need to write its taxonomic placement if there it is done for its genus. --Snek01 (talk) 23:31, 19 July 2010 (UTC)

• While it would be nice to keep the template coding as simple as possible for users, this approach overlooks a couple of problems. Firstly, the children of a genus will not always be species, and the parents of species will not always be genera. Often we will encounter subgenera, infragenera, species groups, and the like, so the rank of the parent cannot be assumed from the rank of the child (which is also unknown under Snek01's proposed alternative). Secondly, although the first part of the binomen is the genus, you also need to know which nomenclatural code you're working under. The parent genus of Oenanthe oenanthe is not the same as that of Oenanthe pimpinelloides. This can only be solved by directly mentioning the parent. In short, the additional fields cannot be removed. --Stemonitis (talk) 05:50, 20 July 2010 (UTC)

Problem 1: Homonymy. Either like this

 | genus = ''[[wheatear|Oenanthe]]''

or like this

 | parent = ''[[wheatear|Oenanthe]]''

Problem 2: subgenera and infragenera. For the majority of cases, the generic name is automatically detectable from its binomial name and thus can be defaultly displayed automatically. Where the subgenus or infragenus occur, then it have to be mentioned. --Snek01 (talk) 09:01, 21 July 2010 (UTC)

How would you extract the generic name from a binomial name using Wikipedia templates? Martin (Smith609 – Talk) 18:49, 21 July 2010 (UTC)

• Further, I do not see how one could display the full taxonomic hierarchy in this way. How would the taxobox template generate the full hierarchy? Martin (Smith609 – Talk) 16:35, 20 July 2010 (UTC)

The information about the "rank" should be taken from the taxobox from that field, that is compatible with the article name or the name which is the lowest in the classification. The "parent" should be the only necessary information. The "link" will became redundant, when the information will be in the article, where it belongs to. All necessary informations are automatically available so automatic systems should use them. Even if we will improve all of these details, it will not "Ensures consistency between articles" but it will "Ensures consistency between taxoboxes" only. This can paradoxly results in inconsistency where the focus on taxonomy will be on the taxobox only. Despite that we should try to use all advantages, that are available for hierarchical structure which classification is. --Snek01 (talk) 09:01, 21 July 2010 (UTC)

I still don't think I quite understand your proposition. How should the information be taken from the taxobox on the parent page? Martin (Smith609 – Talk) 18:49, 21 July 2010 (UTC)

Problem with non-formal names in higher taxons

I've fiddled a bit with the new system. I have a problem though. The taxoboxes insist on writing "Amfibian" where I want "Amphibia" for class. Is there a way to set display name in taxobox different from article name? Also, is there a way to have taxoboxes for certain groups include a comment after a unit? I'm working with Labyrinthodonts, and would like all child article to display class as "Amphibia sensu lato" (i.e "amphibians in the wider sense") rather than just Amphibia. Petter Bøckman (talk) 20:43, 21 July 2010 (UTC)

The solution to the first part of the problem (display "Amphibia" but link to "Amphibian") is this. I'm not sure about the sensu lato idea. It might require a fork of the taxonomy templates, which would seem to undermine the benefits they might bring. Maybe someone else will correct me. --Stemonitis (talk) 20:49, 21 July 2010 (UTC)

Thanks for ironing out the amphibians! I see your point of not forking systematics. It forces me to slap the cladisticans somewhat harder over the snout than I intended (assigning all kinds of early tetrapods to Amphibia without comment), but I guess that is the nature of systematics. The sensu lato thing would in my view only be acceptable if it could be implemented so that it was a display only thing, and restricted to the Labyrinthodonts. The idea was to have only Amphibia as a link, like this:

Class: Amphibia senso lato

A question/request: Would it be possible to have all vertebrates have the subphyllum displayed? While not formally correct, it is common in textbooks to list "Vertebrata", and it is rather more commonly known than Chordata. Petter Bøckman (talk) 21:10, 21 July 2010 (UTC)

Bot request

Editors with a technical bent may wish to contribute to a request to authorize a bot to assist with the creation of automatic taxoboxes. Martin (Smith609 – Talk) 23:16, 17 July 2010 (UTC)

A simplified and revised request is now awaiting comment. Martin (Smith609 – Talk) 16:50, 16 September 2010 (UTC)

I've created an additional request proposing that the bot helps editors who wish to convert existing taxoboxes to Template:Automatic taxobox. Comments are welcome. Martin (Smith609 – Talk) 20:08, 21 September 2010 (UTC)

Automatic taxobox cleanup

Can someone explain to me the meaning of Category:Automatic taxobox cleanup? At the moment, it appears to hold all articles that use {{Automatic taxobox}}, even where no cleanup is required. If it's just meant to be a tracking category for those categories that use {{Automatic taxobox}}, then it's pointless, because Special:WhatLinksHere/Template:Automatic taxobox does that job already. If, on the other hand, there is something to be cleaned up, then some indication of what and how is imperative. I would suggest using the existing taxobox cleanup categories where possible (there are, after all, quite a few), and renaming Category:Automatic taxobox cleanup, as it might suggest that the cleanup will be automatic. --Stemonitis (talk) 06:07, 26 July 2010 (UTC)

I think it's just a temporary tracking category to make requested changes like those listed above. I think Martin was going to remove the code when he's satisfied that the implementation of the code for every page is stable. I just got tired of staring at the red category, so I created it and hid it. I suspect it would be deleted when he's done. Rkitko ^(talk) 12:01, 26 July 2010 (UTC)

Following up on the above, would it be possible to make a hidden category for those taxa which have been set to always display, despite being at minor ranks? I can foresee a temptation among many authors (myself included) to consider taxa in their particular field to be of sufficient importance to be always included. I think that each instance of always_display=true should be confirmed by the community in some way. The automation of the taxoboxes gives us a valuable opportunity to make sure that decisions are based on the same standards across all groups, and to centralise any discussion that unfolds. A category would allow everyone to see what sorts of taxa are considered important enough, and would make it harder for taxoboxes to accumulate more and more minor ranks without anyone noticing. (With the taxobox generated automatically, watching a page will no longer be enough to see whether the information has been altered.) --Stemonitis (talk) 10:38, 31 July 2010 (UTC)

Maintenance

Ideas regarding how to maintain/edit the taxonomy generated by the automatic template are being solicited at Template talk:Automatic taxobox#Taxonomy_maintenance, where discussion should continue. Martin (Smith609 – Talk) 18:48, 15 October 2010 (UTC)