Template talk:Automatic taxobox/Archive 9

From Wikipedia, the free encyclopedia
Jump to: navigation, search
Archive 5 Archive 7 Archive 8 Archive 9 Archive 10 Archive 11 Archive 14

Pitta

Someone replaced the perfectly functional taxobox at Pitta with one of these new ones. Of course since the taxobox used to be about the genus and the family it used to display the genus and family authors. It no longer does, although taxobot helpfully left a note about this and asked for someone to chose the correct one. In English, how does this get fixed? Sabine's Sunbird talk 04:24, 12 January 2011 (UTC)

Wow, I really messed that up, sorry about that. ErikHaugen (talk | contribs) 04:28, 12 January 2011 (UTC)
Wait, no, I didn't mess it up. Taxobot messed it up - I reverted it. I did add the superfamily/etc - if you want we can take those out. ErikHaugen (talk | contribs) 04:35, 12 January 2011 (UTC)
Not really bothered about superfamily, ta for fixing it! Sabine's Sunbird talk 04:37, 12 January 2011 (UTC)
Follow up question. How do we turn off italics? The commmon name here is the same as the generic name, but the article name is about the family in toto, not the genus, so should not be in italics. Sabine's Sunbird talk 04:37, 15 January 2011 (UTC)
You add |name=Pitta – this parameter is treated as a common name that will be the title of the taxobox/article, so it won't be italicized. I did it. When I made the autotaxobox I guess I was thinking the title was the genus name, this is pretty common for monogeneric families. ErikHaugen (talk | contribs) 06:58, 15 January 2011 (UTC)

Speaking of this, is there consensus for going around changing taxoboxes to the automatic version? Right now these actions seem to be creating more problems than it is fixing. Ucucha 10:33, 15 January 2011 (UTC)

I'm seeing situations where editors have altered the taxonomy for the worse as part of implementation. This happens when (1) the editor installing the automatic taxobox is unfamiliar with the edited group, and (2) not all taxoboxes for organisms related to the group are in agreement. There are still issues to be worked out. --EncycloPetey (talk) 20:13, 15 January 2011 (UTC)
I thought there was. I'll stop replacing for now. But I want to point out that these two problems wrt Pitta had nothing to do with automatic taxoboxes. ErikHaugen (talk | contribs) 05:00, 16 January 2011 (UTC)
No, the problems are because these automatic taxoboxes are not intuitive to ignoramuses like myself. I've been reading the documentation trying to work out how to stop Pitta displaying the genus, because the family is now split into three genera. If these things are so much damn work are they really worth it? Sabine's Sunbird talk 03:10, 24 January 2011 (UTC)
Okay I worked it out eventually. Might I suggest some education and "for dummies" documentation for us dummies? Sabine's Sunbird talk 03:22, 24 January 2011 (UTC)
I meant that the title of Pitta was italicized because I thought it should be not because there is something wrong with automatic taxoboxes. And, the extra authority param was a taxobot problem, not one with automatic taxoboxes. But yeah, some things are pretty tricky. I think they're worth it... ErikHaugen (talk | contribs) 07:44, 24 January 2011 (UTC)
Sabine, please see "Automatic Taxobox for Dummies" and see if that helps. BOB THE WIKIPEDIAN (talkcontribs) 04:06, 26 January 2011 (UTC)

Automatic taxobox Broken! again on major articles including FAs

Please STOP! editing this template until you know what you are doing, everyone. And please STOP! adding it to articles until you know what you are doing, everyone.

I am tired of clicking on articles like Bird (a FA), Bryozoa and Agnostid and finding a huge red glaring error message instead of a taxobox. The three articles, out of the 100+ with broken taxoboxes that I have looked at, have unique errors.

Please stop experimenting with wikipedia taxa articles. Please find a way to use the sandbox. --Kleopatra (talk) 05:32, 28 January 2011 (UTC)

It doesn't look like it was experimentation in this case. An IP vandal added text to a taxonomy template high in the hierarchy, which broke every child taxobox (all animals). So, while it wasn't a fault of anyone developing the template system, it does expose a weakness in the system. Not only is it now easy to destroy very large numbers of articles very quickly, it can be very difficult to find where the problem is. (Who would have guessed that {{Taxonomy/Holozoa}} was to blame for a message about "Opishthokontanematoda" on our article on birds?) By adding the text "nematoda" to that template, outside {{Don't edit this line}}, that text got added to the end of the passed parameter "Opisthokonta". Perhaps some fix can be found to prevent this happening in future, but, like Kleopatra says, please test it somewhere other than the live pages. --Stemonitis (talk) 06:14, 28 January 2011 (UTC)
Edit: NM. Can we protect these templates from IP edits? MMartyniuk (talk) 06:16, 28 January 2011 (UTC)
I have semi-protected all the taxonomy templates from {{Taxonomy/Animalia}} upwards as a temporary measure, but I'm not entirely happy with it. If this system of automatic taxoboxes were ever to be rolled out across all the ToL articles, then even taxa relatively low in the hierarchy could be called thousands of times (think of some insect families, for instance). Permanent protection of all frequently-transcluded taxa is hardly viable, and also flies in the face of Wikipedia policy of being the encyclopaedia that anyone can edit. An IP editor may well have constructive changes to make to the taxonomic hierarchy, and we shouldn't be holding them back. What we need is a better system for finding vandalised pages and reverting them, or making the system more robust to vandalism. Or both. --Stemonitis (talk) 07:52, 28 January 2011 (UTC)
Personally, I would set all highly used taxonomy templates to full indefinite protection. This is what full protection is for, even though we want the encyclopedia to be edited by all. We can add a noinclude note at the top of each one directing editors here if they want to use {{Edit protected}} and suggest a change. Rkitko (talk) 12:56, 28 January 2011 (UTC)
Update: Bryozoa is still broken, and I can't work out what's wrong with it. In this instance, it looks like Bob's new purge button may be to blame. --Stemonitis (talk) 06:28, 28 January 2011 (UTC)
Yes, reverting that change fixed Bryozoa. Sadly, Category:Automatic taxobox cleanup is still growing, despite the main problem being fixed, and will soon breach 3,000 entries. It seems it takes a while for the floodwaters to subside after heavy rain. --Stemonitis (talk) 06:33, 28 January 2011 (UTC)
The category is almost back to normal now. The process seems to take about two hours at the moment. In this particular case, perhaps there's no real harm done, but it would have been a lot worse if a vandal had added something defamatory. Displaying a libellous claim on thousands of pages for hours at a time, and not being able to quickly remove it seems like quite a liability to me. --Stemonitis (talk) 07:56, 28 January 2011 (UTC)

Is this issue also causing other problems, like lack of color at African Bush Elephant? Rkitko (talk) 13:05, 28 January 2011 (UTC)

The reason for that appears to be that {{get regnum()|Loxodonta africana}} does not return anything. ({{get regnum()}} does work, and returns "animalia".) Why that function fails, I'm not sure. Ucucha 13:41, 28 January 2011 (UTC)
I've fixed this problem for Loxodonta africana. Martin (Smith609 – Talk) 15:39, 28 January 2011 (UTC)

The category appears not to be filled with an editor who made well over 100 articles in a very short amount of time, User:Abyssal, sometimes creating over 10 articles in a minute. Again, these editing faster than you can look at the articles and not previewing leads to the same results: filling wikipedia with articles with screaming red error message outweighing the text. 100+ taxa articles that are entirely overwhelmed by their error messages are not a welcome addition.

Yes, how can the templates be monitored in a way that allows for a when something like this happens. I've edited taxonomies for years as an IP, correcting many egregious errors in taxoboxes. This is one area where wikipedia has a lot of IP editors working well to keep articles accurate. Yes, I don't see any way around blocking IP editors from the taxonomy templates. I don't like full edit protection because that leads to administrators only editing, which I strongly disagree with. Can we semi-protect class and above? Is there a better way to signal IP edits to the templates, especially higher order ones impacting a lot of articles? --Kleopatra (talk) 15:21, 28 January 2011 (UTC)

The vandalism I've seen thus far has all involved the addition of text that changes the syntax of the template, stopping it from working. This would be easy to detect via an algorithm, so watching for this sort of vandalism seems the ideal task for a bot. It won't catch more intelligent vandalism, but will hit most IP "noise". (I wonder whether anyone could be persuaded to code one...) Martin (Smith609 – Talk) 15:44, 28 January 2011 (UTC)
Aside from the introduction of syntactical errors, I seem to recall a second type of vandalism-- one template's display text (Eukaryota?) was changed to "this animal has a big penis". Thankfully, display text vandalism is far easier to track down. I'm all for Cluebot or some other bot monitoring these things.
Isn't there some sort of bot on Wikipedia that goes around protecting all templates that appear on 40 or more pages? I find it odd we haven't seen the likes of it yet... Bob the WikipediaN (talkcontribs) 23:08, 28 January 2011 (UTC)

A few notes:

  • If consensus is for protecting everything and all the pages can be matched by a regular expression (e.g. all are subpages of some page), MediaWiki:Titleblacklist could be used to auto-protect all current and future pages just as is done for editnotices.
  • I have a script that helps in finding template vandalism, it adjusts the list of "Pages transcluded onto the current version of this page" on the bottom of the edit screen to include information on the most recent edit to each template. See User:Anomie/previewtemplatelastmod for details.

Anomie 16:12, 29 January 2011 (UTC)

So long as it's only semi-protected, I'd go for that. Also-- that script looks interesting...I think I might start using it. That would certainly be helpful in tracking vandalism. For example, if the Bird article were vandalized again, you'd simply need to go into edit mode and boom-- most recent edit summaries right next to all the transcluded templates! Sounds very handy! Thanks! Bob the WikipediaN (talkcontribs) 16:56, 29 January 2011 (UTC)

/inactive

What's this all about? Bob the Wikipedian (talkcontribs) 04:45, 18 January 2011 (UTC)

I guess a more descriptive question is in order here. Martin, I see you created this page, and I'm wondering what it is. It's appearing in the cleanup category, which indicates to me that it's either unfinished or test or something different. Bob the WikipediaN (talkcontribs) 04:09, 2 February 2011 (UTC)
The clade Rhabditophora isn't currently used in WP articles, but it had been incorporated into a taxobox. When I moved its child out of Rhabditophora and into Platyhelminthes, I left Rhabditophora but didn't want it displaying in the child list of Rhabditophora. Hence, "/inactive". Martin (Smith609 – Talk) 14:41, 2 February 2011 (UTC)
Oh. Are we doing that with empty taxa? Bob the WikipediaN (talkcontribs) 16:34, 2 February 2011 (UTC)
Didn't mean to set a precedent - just seemed the best solution to that specific problem. Set /inactive's parent to "life", if you like, to keep it out of the cleanup cat. Martin (Smith609 – Talk) 16:42, 2 February 2011 (UTC)
That might be the way to go. I gotta run now, but I'll do that later if it isn't done yet. Bob the WikipediaN (talkcontribs) 16:49, 2 February 2011 (UTC)
I thought some more about it. What it boils down to is that we've got a placeholder taxon, correct? Do we want to treat placeholders differently? Bob the WikipediaN (talkcontribs) 03:31, 10 February 2011 (UTC)

{{resolved}}

Template shows up in Google searches

Is this intended behavior? I noticed it after googling for reference in a belemnite genus I recently converted to use autotaxobox, Belemnotheutis. Googling for it returns its taxobox template on the second page. It could be confusing for people who might think that the hits on their templates are the articles on the taxa themselves.--ObsidinSoul 13:09, 29 January 2011 (UTC)

Google is responsible for "The Algorithm", which is updated at least daily if not multiple times per day. I've had success before several times in notifying Google of various things and seeing action taken; I'll see if I can't contact them and let them know about this. Bob the WikipediaN (talkcontribs) 16:41, 29 January 2011 (UTC)
Hm... doesn't wikipedia have a way for rendering a page invisible to search bots already? For example, in Template:User page, it has a noindex option which causes it not to show up in external searches. Something like this could help taxobox templates remain invisible and only accessible through the small 'e' edit button or internal searches. This might also help prevent the templates vandalism discussed above by making the templates themselves a bit more inconspicuous.
On the other hand, not sure if you guys intend it to be completely unsearchable by google though.--ObsidinSoul 17:07, 29 January 2011 (UTC)
Why not just leave the templates accessible to Google? They are probably becoming a taxonomically useful resource themselves. Let's allow people who want to find information about a taxon to find that information. Ucucha 17:25, 29 January 2011 (UTC)
Problem is, the template pages have rows upon rows of deliciously inviting 'edit' links for vandals. And vandalizing one of those at even family level can lead to hundreds of articles getting broken. I must admit though, they are actually very helpful, though obviously not intended for regular user browsing.--ObsidinSoul 17:37, 29 January 2011 (UTC)
Those deliciously inviting 'edit' links exist on a lot of Wikipedia articles; why not noindex those too? Pages that are overly vulnerable to vandalism should be semi-protected, not hidden. Ucucha 17:53, 29 January 2011 (UTC)
If you want to remove it from google searches, add __NOINDEX__ to the doc page. I agree that there seems to be no major harm in having it indexed. Semi-protection can be added if it starts becoming a vandalism target. Plastikspork ―Œ(talk) 19:50, 29 January 2011 (UTC)
They won't actually be hidden as they can still be accessed inside wikipedia and searched for. Point is, they weren't meant to be read as is, they are templates after all. If there were provisions for them to become different resources in their own right (which I support, as they really are very useful by themselves) then maybe the proposal to hide them from indexing would be absurd, but there aren't. As it is, they are quite obviously only for those who knows what they're doing. The template pages themselves have the header "Not sure why you're here?" and then provide links to instructions on how to set up auto taxoboxes, not on how to browse for other taxa.
If a person were looking for, say Marginocephalia, and stumble into that page instead, he has no way of knowing if that was Wikipedia's article itself or even if there was an article on the taxon whose template he's looking at currently.
Much more for taxa which do not have articles yet like Eugnathostomata, as searching for them on google only returns their template taxoboxes (no article page after all), and google lists them in the first page itself (being in Wikipedia). See here http://i.imgur.com/Vo1Ru.jpg. And since it's Wikipedia, I would assume people would click on it thinking it's Wikipedia's article on it and instead find themselves in the innards of templates.
This is what it looks like when I google the others I've mentioned at the moment: http://i.imgur.com/tAgcF.jpg and http://i.imgur.com/vW73O.jpg
Maybe if they could be turned into quasi-articles themselves like categories and be a bit more browsing-friendly for non-editors?--ObsidinSoul 20:28, 29 January 2011 (UTC)
As a start, I've modified the templates (at Template:Taxonomy key) to include _NOINDEX_ if an article exists, and to add a "what is this page" message if it doesn't. Martin (Smith609 – Talk) 18:04, 31 January 2011 (UTC)
Wonderful! Bob the WikipediaN (talkcontribs) 03:14, 2 February 2011 (UTC)

{{resolved}}

Anatoma proxima

Automatic taxobox/Archive 9
Scientific classification e
Kingdom: Animalia
Phylum: Mollusca
Class: Gastropoda
Clade: Vetigastropoda
Superfamily: Scissurelloidea
Family: Anatomidae
Genus: Anatoma
Species: A. proxima
Binomial name
Anatoma proxima
(Dall, 1927)
Synonyms[1]
  • Anatoma americana Bandel, 1998
  • Scissurella proxima Dall, 1927

What is wrong in the taxobox Anatoma proxima ? JoJan (talk) 16:44, 31 January 2011 (UTC)

There needs to be appropriate content at Template:Taxonomy/Anatoma proxima. Since the article had been at Anatoma americana, the template that should have been producing the taxonomy was at "Template:Taxonom/Anatoma americana" instead. I have now moved (and updated) that template, and I think the taxobox is now fine. --Stemonitis (talk) 16:50, 31 January 2011 (UTC)
Oop. That was my work. Are they synonyms? If so why did they have separate articles before? Have also removed the entry A. americana in the list of species in Anatoma--ObsidinSoul 17:02, 31 January 2011 (UTC)
Ok, you just moved it. Anyway yeah. Moving a page with an autotaxobox and renaming its contents requires moving and updating its template as well (or the deletion of the old template and the creation of a new one).--ObsidinSoul 17:08, 31 January 2011 (UTC)
Indeed, that's probably the biggest flaw with this system. By including the |taxon= parameter in the article, it will always link to the template, but the template will still need its link updated since redirects don't work well in the taxonomy template. More on that topic can be found at #Taxon parameter, moved pages. Bob the WikipediaN (talkcontribs) 22:39, 31 January 2011 (UTC)
I just have made during the last couple of months more than 500 moves or redirects and I've still have more than 250 to perform (see : Wikipedia:WikiProject_Gastropods/Unaccepted). If I'm going to have trouble fixing every taxobox, this will result in a serious headache for me. I'm not familiar with the fine intricacies of programming. And I have to manage more than 22,000 articles about gastropods where almost every day several changes in taxonomy occur. We have a bot tracing these changes and producing regularly a new list of "unaccepted" species. Up to now this was hard work managing these changes, but everything went smoothly. And suddenly I'm confronted with this problem with taxoboxes. Isn't there a way in the programming that when an article is moved to its accepted name, the taxobox just has to be changed with the right names without having to touch the template ? If not, can you explain exactly, in layman's terms, what I have to do ? And another question, If I delete an article with a taxobox (e.g. wrongly spelled species name), what happens to the template of the taxobox ? JoJan (talk) 09:23, 1 February 2011 (UTC)
Yes, there is a way: return to the old, manual taxobox. Any method implementing the automatic taxobox will require an additional page move. On your second point, if you delete an article, the taxon template remains untouched, which probably isn't a huge problem. --Stemonitis (talk) 10:00, 1 February 2011 (UTC)
I'm not sure I agree; if, as Bob mentioned above, you add the line |taxon=(name of taxon before move) to the automatic taxobox, then the corresponding taxonomy template won't need moving, and the taxobox should continue to display correctly. Martin (Smith609 – Talk) 13:34, 1 February 2011 (UTC)
But it'll be in the wrong place, increasing the overall confusion and opacity by another needless degree. The last thing we need is for the taxonomy of Anatoma proxima to be at [[Template:Taxonomy/Anatoma americana]], or any similar example. Setting |taxon= is useful for monotypic taxa and the like, where the title of the article needn't be the same as the core taxon in the taxobox (although they cover the same set of organisms), but not in the case of synonymy. The only good solution I can see in that case is to move the template. --Stemonitis (talk) 13:43, 1 February 2011 (UTC)
Oh, I see; I guess I'd misunderstood. Yes, you're quite right, it won't do to have the taxonomy at a different place.
I've been frustrated with adding auto-taxoboxes for species in the past... creating the taxonomy templates feels redundant, given that the genus name is part of the binomen. Perhaps the solution to both of these issues is Template:Speciesbox, which creates an automatic taxobox based on the "genus" parameter, the fills out the rest based on a "specific epithet" parameter (called |species=). I've just created a very quick knock-up, which you can see at Template:Speciesbox/sandbox; if it seems a good idea, I'll modify it so that there's a "species" entry in the taxonomy list and we can use this for species. Martin (Smith609 – Talk) 14:12, 1 February 2011 (UTC)
It would be nice to have the taxonomy templates for the species anyway, though. ErikHaugen (talk | contribs) 22:30, 7 February 2011 (UTC)
That may be a step in the right direction, but I suspect it's solving a slightly different problem. The issues of synonymy and page moves could equally apply at any rank, although such changes are naturally rarer in higher taxa. Adding a species template does not help the taxonomy template follow a genus article when it is moved to a new title. --Stemonitis (talk) 14:20, 1 February 2011 (UTC)
How will the |species= parameter be implemented? Will this affect the way the taxonomy is generated, aside from allowing the genus to be displayed correctly regardless of reassignment?
You can see the implementation in the example at Template:Speciesbox/sandbox (now also copied here), which uses |genus=Anatoma to generate the higher taxonomy, and |species=proxima to generate the binomial (and, anon, to add the line "Species: A. proxima" to the taxonomy list). Martin (Smith609 – Talk) 16:49, 1 February 2011 (UTC)
Also, to answer JoJan's question, if the scientific name of the species has been revised, you can, as Martin suggested, perform the following steps. The example here would apply to Geniosus speciosa being moved to Paragenius speciosa.
  • Ensure the following code is present in the automatic taxobox: |taxon=Geniosus speciosa
  • Update the binomial name in the taxobox to to new name (I assume you know how to do that)
  • Using the [e] edit link in the taxobox, edit the taxonomy at the species level. Replace Geniosus speciosa|G. speciosa with Paragenius speciosa|P. speciosa.
I agree, this is another annoyance that needs some ironing out if possible, although it could be far worse. Bob the WikipediaN (talkcontribs) 16:28, 1 February 2011 (UTC)
Thanks Bob, I'll try this out as soon I encounter another gastropod article with an automatic taxobox to be moved (luckily most gastropod articles still have traditional taxoboxes). As to the objection of Stemonitis about genera, when a genus A becomes a synonym of genus B (with an already existing article; if not, create one first), a redirect is made. And redirects don't have taxoboxes. So, the objection is actually a non-problem. All the species in genus A are then moved to genus B (or to several genera). The problem only arises in moving an article with an automatic taxobox. JoJan (talk) 18:17, 1 February 2011 (UTC)
My observation (not objection), was that if a genus article with an automatic taxobox is moved to a new name, then its taxonomy template will also need to be moved, just as was the case for a species article with an automatic taxobox (the implication being that the problem cannot entirely be solved by a new template specific to the rank of species). For articles with manual taxoboxes, there has never been a problem. --Stemonitis (talk) 18:33, 1 February 2011 (UTC)
The conversion of the Anatoma taxoboxes were actually my fault, so I apologize for that. But they were listed under the taxoboxes which used unranked_familia instead of unranked_superfamilia, and I opted for the second option (convert so they are removed from the list, rather than just change the naming). I do feel however that automated taxoboxes, while daunting at first with all the gobbledegook, will make future moves involving higher taxa a lot easier in the long run and more consistent.
For example, in manual taxoboxes, if an order has been reclassified, you would have to change each of the manual taxoboxes of the order, families, genera, and species articles (in addition to suborders, infraorders, superfamilies, etc. etc.) affected to reflect the change, and that can mean hundreds to thousands of articles that would need to be changed manually. While with auto taxoboxes, you would just need to relink the immediate children of the order, (the families involved, or even better, superfamilies or suborders if they exist), and every other article below that using the autotaxobox would update automatically.
The problems I see currently seem to involve genera/species which are still linked in the templates. Higher taxa using automated taxoboxes would have no such problems as then you'd only have to change the taxon in question and its accompanying template. No need to change every taxobox of the affected subtaxa to reflect the change (that is, if they use automated taxoboxes).
Manual taxoboxes may not need their accompanying templates moved, but then you also need to change every species article's taxobox under them anyway, which is more work. Anyway just my two cents. And yeah I agree, separating the species from the genus (so you wouldn't need to retype the genus specifically for each species' template and changing its genus would be as simple as changing the |parent= field) would be a nice addition. --ObsidinSoul 18:51, 1 February 2011 (UTC)
(After 2 editconflicts) As an aside to Bob, the [e] edit link in the taxobox doesn't show up in Firefox in my computer (it lies almost imperceptibly upon "Scientific Classification" and it is as good as impossible to click upon). But it does show up clearly in Internet Explorer. This means that whenever I have to move an article with an automatic taxobox, I'll have to switch from Firefox to Internet Explorer. No big deal. And @ Steminitis : indeed, in articles with manual taxoboxes, there has never been a problem. The task we're facing in Wikiproject Gastropods is enormous. There are almost a 100,000 accepted gastropod names and 400,000 to 500,000 synonyms (some even say much more - no one really knows). Many of the accepted names then turn out to be actually synonyms of other accepted names. And then, a great many snails (some think another 100,000) aren't even discovered, as most snails live on the sea bottom and the sea bottom is still mostly "terra incognita". If our workload is increased by the automatic taxobox, then this automtic taxobox may turn out not to be an improvement. Everything should be done to avoid this. JoJan (talk) 18:57, 1 February 2011 (UTC)
Hmmm....not sure why you aren't seeing the edit link. What operating system are you using? What version of Firefox?
The reason the species-level templates don't need moved is because typically the species is the lowest ranking taxon. In the event that there are subspecies/infraspecies, you'll need to move the template and update the taxon parameter if the article has that parameter. The automatic taxobox was developed with taxonomic revisions in mind, but not with pagemoves in mind. Of course, it'd be wonderful if you did move the templates; however, I'm fairly certain a tracking category could be developed to detect such modifications and add them to the cleanup category. Please keep in mind, however, that all subgeneric and higher-level taxa should have their templates moved and their daughters' |parent= parameters updated accordingly.
I think Obsidian's right, though...the average amount of reworking per taxonomic revision (including pagemoves) is significantly reduced in at least supergeneric cases. Perhaps streamlining the page-movability will be a good summer project in the event anyone comes up with a solid idea. I'm open to any, but I'm a bit too busy right now (and likely until summer) to work on any substantial coding. Bob the WikipediaN (talkcontribs) 03:43, 2 February 2011 (UTC)
I'm now recommending Template:Speciesbox (and Template:Subspeciesbox) for species-level articles. It uses the automatic system to generate the higher taxonomy, but the genus and species name are manually specified - making it easier and more intuitive to rename a taxon. Martin (Smith609 – Talk) 22:07, 4 February 2011 (UTC)
Awesome! :)--ObsidinSoul 22:35, 4 February 2011 (UTC)
Great! I recommend we convert existing species automatic taxoboxes to this where possible, although if smeone discovers a bug, please stop. Bob the WikipediaN (talkcontribs) 23:14, 4 February 2011 (UTC)

{{resolved}}

Virus taxa

I've begun working on structuring the virus database and just now learned that all virus taxa are italicized (except the group). Before I take a stab at this, any suggestions on what might be effective and efficient for identifying virus taxa? Or, perhaps, should we diverge virus support into a new template that automatically italicizes all taxa? Bob the WikipediaN (talkcontribs) 03:35, 26 February 2011 (UTC)

A couple of representative pages would help me to get an idea of what might work. A separate {{virusbox}} might be a good solution, performancewise. Martin (Smith609 – Talk) 04:02, 26 February 2011 (UTC)
See Flavivirus and Adenoviridae. Bob the WikipediaN (talkcontribs) 04:19, 26 February 2011 (UTC)
Thanks. As we're not really dealing with taxonomy per se, and because virus classification has its little quirks, I'd argue that a separate template is the way to go (although still using taxobox/core, of course!). This will improve performance and make upkeep of the automatic taxobox more straightforwards. Martin (Smith609 – Talk) 04:47, 26 February 2011 (UTC)
Alright, I'll add that to my list of things to do Face-smile.svg. I just signed up at the virus WikiProject to get them rolling with automatic taxoboxes; I'm actually somewhat surprised that this didn't pop up sooner, but the virus WikiProject has been dormant for a good several years now. Bob the WikipediaN (talkcontribs) 04:52, 26 February 2011 (UTC)

Just fyi, PhyloCode Article 6 recommends italics for all clade names. So this may become the standard. It notes that the ICZN does not recommend italics for ranks above genus, but given that only animal family-group names are covered there, it seems like a minor nitpick to argue that ALL taxon names shouldn't start getting italics. So I'd say don't work too hard on splitting off a separate template. MMartyniuk (talk) 06:06, 28 February 2011 (UTC)

Yes, but it may take a long time to become standard. The ICBN has been recommending italics for all taxa (and it does cover higher taxa, unlike the ICZN) for some time, and almost no publications follow that recommendation (Taxon being one notable exception). Wikipedia's policy is to follow the tendencies of the wider world of publishing, and that means not italicising the names of higher taxa of animals and plants for the foreseeable future. --Stemonitis (talk) 07:26, 28 February 2011 (UTC)

{{resolved}} Bob the WikipediaN (talkcontribs) 21:18, 12 March 2011 (UTC)

Taxa defaulting to always display when parameter omitted

I noticed that automatic taxobony templates which omit the |always_display= parameter, are defaulting to a value of true. (For examples, see this edit and this edit, both of which are currently needed to prevent inappropriate use of the always_display flag.) This is clearly the wrong way; the default value when the parameter is omitted should be the same as when the parameter is named, but no value is given, and in this instance, both should very much be defaulting to false. --Stemonitis (talk)

"Family" is a major rank and so displays by default. You may wish to specify |rank=unranked (etc). Martin (Smith609 – Talk) 18:00, 31 January 2011 (UTC)
Yes, but that wasn't the problem. I haven't seen it repeated in other, to my eye similar, templates, but at least those few were apparently being manually over-ridden to always_display, rather than simply displaying as major ranks. If you look at the state before I edited, you will still see "Yes (manual override)" against always_display. It is fortunate that these cases are [being treated as] major taxa, and accordingly are not appearing in taxoboxes where they shouldn't be, but a bug which allows always_display to be accidentally invoked could be rather serious, and needs investigation. --Stemonitis (talk) 18:26, 31 January 2011 (UTC)
Template:incertae sedis taxon invokes always_display where it needs to. I've re-worded the key that appears on the taxonomy/ pages to "manual/automatic override". Are there any cases where a taxon without |Always_display=true or a major taxonomic rank is always displayed? Martin (Smith609 – Talk) 21:12, 31 January 2011 (UTC)
Ah. I hadn't spotted that those pages were using the derivative {{incertae sedis taxon}}, and not {{Don't edit this line}}, although I'm not sure how that excuses the unexpected behaviour. How many further layers of complication are there likely to be before this system can be considered complete? This setup is already so uncommonly complex that it is likely to collapse under its own weight. Are there any instances yet where inconsistencies in taxonomy have been corrected (which was, after all, the point)? Are they inconsistencies that couldn't have been picked up by a clever spider and reported on a single page, for a human to fix, or request specific bot assistance if there were too many articles affected? I am coming to think that that method would be vastly preferable to this. --Stemonitis (talk) 21:35, 31 January 2011 (UTC)
The migration is very slow, but many, many taxonomies have undoubtedly been modified to reflect consistency. In fact, several deprecated taxa that have been replaced in automated taxonomies can be found at Category:Deprecated taxon templates. In the cases of several of them, including Sauropsida and the Latin versions of the upper plant taxa, RfCs were held and the appropriate WikiProjects were notified of the RfC before the taxonomies were merged. Others were determined based upon WikiProject guidelines or the articles about the taxa themselves. Bob the WikipediaN (talkcontribs) 22:34, 31 January 2011 (UTC)
(edit conflict) :::I believe the point of this template is the ease with which classifications can be changed on thousands of articles with just a few edits instead of the thousands of edits necessary. E.g., I'm running through all flowering plant articles with AWB and updating the old Cronquist system to the APG III system. This would have been so much easier if each taxon had been automated already. Hesperian and I have been working on this intermittently for about two years now and my current effort has been fairly strong for weeks with no other work. Future changes in classifications will be infinitely easier to edit; for this reason, the automatic taxobox has advantages over every other taxa database out there, which have become stale with old classification systems and would take so much effort to update. But to answer your question, yes, automating flowering plant taxoboxes will immediately correct an outdated Cronquist system to the APG III system. Rkitko (talk) 22:42, 31 January 2011 (UTC)
Whilst automating taxoboxes I've come across several genera that have been listed in the "subdivision" list of two different families. By using the |display children= parameter (and assigning the genus to the correct family), the problem is not just fixed, but won't occur again. Martin (Smith609 – Talk) 13:44, 1 February 2011 (UTC)
I don't see that an inconsistencies report and taxon-specific bot tasks could not achieve the same results, and that was the yardstick I suggested. The ability to change large numbers of articles in few edits is equally a strength and a weakness, as the recent vandalism proved. To my mind, that leaves us weighing up the intended increase in consistency against the increased template complexity, particularly for newer editors, and that was what I was trying to ascertain. But I fear this thread is veering off-topic. I think the time will come quite soon when the community needs to decide which way to go with this. While keeping two systems would be feasible, it would seem foolish to maintain both if there is a significant advantage to one over the other. --Stemonitis (talk) 23:05, 31 January 2011 (UTC)

References

Why are references not showing up in automatic taxoboxes? See Tortilicaulis as an example. Peter coxhead (talk) 21:07, 28 February 2011 (UTC)

Because they're not meant to. The references in Template:Taxonomy/Tortilicaulis are only meant to support the data contained within that template.
I'd suggest that it is more useful to the reader if the references are situated in the article text, where it is clear to what they refer. If they were positioned in the taxobox (besides technical limitations), it wouldn't be clear whether the reference referred to the person that placed (say) Tortillicaulis in Polysporangiophyta, or whether it refers to a definition of the polysporangiophyte clade, or whether it was the person who assigned Horneophytopsida to the polysporangiophytes... etc.
Another alternative would be to specify, in the taxobox on "Tortillicaulis", something like |parent authority=taxon established by <ref.../>.
I hope that one of those solutions will work for you. If not, we'll have to see what we can work out. Martin (Smith609 – Talk) 21:19, 28 February 2011 (UTC)
There are two different issues here, I think.
  • Surely automatic taxoboxes should not behave differently from manual ones except for features essential to the automation of the taxonomic hierarchy? I see that you've restored the reference by changing the place in which it appears; it would be good if the need for such 'tricks' can be avoided.
  • There is the interesting issue of whether editors should, in general, put references in taxoboxes. This is something I'm not sure about – maybe it needs discussing more widely. On the one hand, if I present a taxonomy in the body of an article without a reference then this violates WP:Sources and if I present only one such taxonomy when there are well-attested alternatives then this violates WP:NPOV. So why should presenting a taxonomy in a taxobox be different? On the other hand, as you say, it's difficult to see exactly how to make it clear which bit of a hierarchy is being sourced to the reference.
The only case which does seem clear is when the reference is to the publication in which the taxon name itself was established. This is only needed at the lowest level, and should be reasonably easy to work into the article, if it's agreed not to put such references in the taxobox. Looking around though, it seems to me that almost always the only place the botanical authority appears is in the taxobox, so this is the obvious place to put a reference to the publication which established that authority.
I should add that I'm a strong supporter of the idea of automating taxoboxes. But we've seen from recent discussions that if switching from a manual to an automatic taxobox changes what is presented too much, editors won't be happy and hence won't use them. Peter coxhead (talk) 23:58, 28 February 2011 (UTC)
No, it is not needed only at the lowest level. It is needed for every level that is covered on that page. For a taxon like Leiosporoceros, with a single species in its own class, there will be a single page for all the ranks of class to species and thus references for publication of all levels. If the automatic taxobox can't handle this efficiently, then we need to postpone (further) its implementation. --EncycloPetey (talk) 00:12, 1 March 2011 (UTC)
Ah, I was thinking that when a taxon is the only member of the next higher rank, the article is usually named as the higher rank. It was this that I meant when I wrote "at the lowest level". Thus for the Adoketophyton article which is at the genus level, the taxobox shows lower levels (species) as subdivisions, so that references are attached to the genus name and the species subdivisions. However, it turns out that it is possible to display references, as indeed the Leiosporoceros article does, but you have to be careful as to how this is done, using parameters like "greatgreatgrandparent_authority". I'm still finding the {{Automatic taxobox}} template tricky; you can't just change "taxobox" to "automatic taxobox".
We're in agreement about the desirability of referencing the authority for a taxon. Surely these authorities and references should be stored in the taxonomic hierarchy so that you don't have to enter them in the taxobox?
What about the more general issue of sourcing the taxonomy itself? E.g. we've agreed to use APG for angiosperms, but the taxoboxes which use this system don't say so or provide the reference. Peter coxhead (talk) 00:30, 1 March 2011 (UTC)
Ever since Kleopatra mentioned this about a month ago, I've been filling in the reference parameter on taxonomy templates with whatever my resource was for naming the parent taxon, even if the best I can do is a link to a Wikipedia article that (at that time) showed that parent taxon. My apologies for not doing this before then; originally I was only doing this where I included a taxon not already included on Wikipedia. Bob the WikipediaN (talkcontribs) 01:00, 1 March 2011 (UTC)
Bob and/or Martin: can you explain exactly what the present position is? When Martin wrote that references didn't show up "[b]ecause they're not meant to", this was obviously not quite right because (a) he fixed the article in question so the reference did show up (b) I read up about the "authority" parameters and saw the Leiosporoceros article. But if all those "-authority" parameters were removed, would the authorities and references appear? I'm still confused! (And still bothered about the APG system issue I raised above.) Peter coxhead (talk) 01:07, 1 March 2011 (UTC)

I'll clear up what I can--

There are several different "references" we all may be discussing...

  • Firstly and foremost, there's the "authority". The authority (denoted in {{taxobox}} by |species_authority=, |genus_authority=, etc, is denoted in {{automatic taxobox}} by |authority=, |parent_authority=, etc, all the way to |greatgreatgrandparent_authority=. This reference should indicate the taxonomic authority.
  • The second type of reference is the |refs= in the taxonomy template associated with a particular taxon. This reference should indicate the source of information used in linking the taxon to its parent. This information is not shown in taxoboxes and should also not use <ref> tags, just {{cite web}}/{{cite journal}}, etc. In many cases, the resource indicated here will simply be a link to a Wikipedia article, usually indicating the person who set up that template didn't want to alter the currently accepted classification scheme at Wikipedia and used whatever was listed on that article at the time.
  • The third type is the type which appears in the references section of an article. For those who don't know how to produce these, it's pretty simple-- add <ref> tags surrounding a citation template of some sort.

Hopefully that answers your question. Bob the WikipediaN (talkcontribs) 02:59, 1 March 2011 (UTC)

P.s. -- In converting from taxobox to automatic taxobox, you don't need to alter any ref-tag-style references. Bob the WikipediaN (talkcontribs) 03:01, 1 March 2011 (UTC)
By "reference" I've always meant "something in ref tags which should appear in the References section of the article". It seems a bit odd to me to be putting so-called 'references' in the taxonomic hierarchy which the reader doesn't see.
  • I think part of my problem is that more needs to be said at Template:Automatic_taxobox#Converting_from_a_Taxobox_to_an_Automatic_taxobox about how to use (not "remove") the authority parameters. As an example, Tortilicaulis old version has a manual taxobox which shows a reference for the genus authority. This version has an automatic taxobox with the same reference showing up. There are intermediate versions which don't display the reference. The correct display comes from converting the manual taxobox parameter genus_authority = Edwards 1979<ref name=Edwards1979/> to the automatic taxobox parameter authority = Edwards 1979<ref name=Edwards1979/>, which I didn't properly understand – the authority parameters are not explained at present at Template:Automatic taxobox/doc, although they do appear at Template:Automatic taxobox/doc/Step-by-step.
  • The second issue is how to store this information in the taxonomic hierarchy so that it gets displayed automatically. If, for example, I created articles on all the species of Tortilicaulis, I don't want to have to put parent_authority = Edwards 1979<ref name=Edwards1979/> in every taxobox on the species pages. The documentation available at present doesn't explain if it's possible to display such references automatically, and if so how to do it.
(As a side issue, it was unhelpful that in the middle of the process of getting the Tortilicaulis article right, User:AnomieBOT moved the actual citation out of the taxobox in favour of a named ref tag. This action makes copying and pasting this information between related articles more difficult.)
I'm afraid that that is the only solution at the moment: it's the same as what you'd have to do in a normal taxobox. Incidentally, you might find that Template:Speciesbox makes it a bit quicker to create species-level automatic taxoboxes, as you don't need to create a taxonomy/ template for each species. Martin (Smith609 – Talk) 15:44, 1 March 2011 (UTC)
Ok, thanks for the answer. You could usefully add to your doubtless long "to do" list the task of automating the storage and display of taxon authorities, so we don't need to add them manually to each automatic taxobox. Yours in hope, Peter coxhead (talk) 16:11, 1 March 2011 (UTC)
  • The third (and unanswered!) issue is whether and how a source for the general taxonomy, such as APG where used, can be shown.
I'd like to repeat that I think that all the work being done on automating taxoboxes is great. The problem seems to be that, as with most software (I used to teach this stuff), the documentation lags behind the code!! Peter coxhead (talk) 09:24, 1 March 2011 (UTC)
We held a rather lengthy discussion awhile back on potentially automating the authorities. The result of that discussion was not to do so, though I don't remember why we decided not to. Off the top of my head, however, an automated authority would definitely create problems if someone wanted to cite an authority in the reference section. Bob the WikipediaN (talkcontribs) 03:16, 2 March 2011 (UTC)
It took ma a little while to find the discussion on authorities in the archives, but here is the link! I still think that there isn't much reason to move the authority parameter to the back end to me honest though.--Kevmin § 07:36, 2 March 2011 (UTC)
Thanks for digging that up, Kevmin. Interestingly, authorities were mentioned in that discussion with the perspective of displaying the authorities of children, not parents. Now that I'm seeing it in a different light, it's somewhat tempting, provided we could get any reference-type bugs worked out. Bob the WikipediaN (talkcontribs) 08:08, 3 March 2011 (UTC)

Question marks in automatic taxoboxes

Another thing I can't quite figure out... If you look at the taxobox for Torticaulis, there are "?"s against both Horneophytopsida and Polysporangiophyta. This is correct; there is uncertainty as to how it should be classified because it's only a fragmentary fossil. It has been thought to be a moss, in which case it would be a member of neither Horneophytopsida nor Polysporangiophyta. Now I want the same for Nothia. It has been classified in the zosterophylls, but doubtfully, so the "?" after Zosterophyllopsida is correct. I managed to do this bit myself! But if it's not a zosterophyll then it's probably not a lycophyte either, so I want a "?" against Lycopodiophyta too. I can't work out how to do this. The "?" mechanism is very useful, but not documented. Peter coxhead (talk) 13:49, 1 March 2011 (UTC)

It sounds to me like the best solution is to set |parent= at Template:Taxonomy/Zosterophyllopsida/? to "Lycopodiophyta/?". This'll make anything assigned to Zosterophyllopsida (?) display a ? next to Lycopodiophyta.
Where do you think the best place for us to document this is?
Martin (Smith609 – Talk) 15:14, 1 March 2011 (UTC)
(P.S: nice work on the article! Thanks for all you're doing on early plants.}
To respond to the P.S. first: I got a bit annoyed by the predominance of "animals" all over the paleo stuff; just look at the Paleontology article – plants seem to get about 2 sentences on the angiosperms! Compare the coverage of, say, dinosaurs with all prehistoric plants. So having a copy of Taylor et al.'s 2009 Paleobotany textbook to hand, I thought I'd fill in some gaps, starting with species articles, and ideally moving upwards to some of the higher level articles, such as Paleontology. Unfortunately, I discovered that the textbook was just too sketchy (and sometimes not quite accurate) so I've had to work from original papers, which is a slow business as I'm sure you know. It would be nice if there were more people working on paleobotany, not just to share the work but also to exchange ideas with. I'm aware that I tend to write very 'academically' and find it hard to pitch articles at the right level for a Wikipedia reader. I should also say that this is a winter activity; as it warms up and my garden beckons, there'll be less Wikipedia work.
Back to the main issue. Ok, suppose Template:Taxonomy/Zosterophyllopsida/? is edited as you suggest. The problem then would be that if there's a plant which is doubtfully a zosterophyll but definitely a lycophyte the display won't be correct. (The real difficulty is that the classification of early land plants is a mess, and my reading of the original literature is that the more that detailed studies are done of individual species and genera, the less certain is the phylogeny and hence taxonomy.)
Then I suppose that we need a Template:Taxonomy/Zosterophyllopsida/Lycopodopsida/? template, with |parent=Lycopodposida/?, in addition to the existing Template:Taxonomy/Zosterophyllopsida/?. Martin (Smith609 – Talk) 20:18, 1 March 2011 (UTC)
Let me see if I've got this right. Given a taxon X, I can set up multiple alternative taxonomic hierarchies by having "Template:Taxonomy/X", "Template:Taxonomy/X/Var1", "Template:Taxonomy/X/Var2", etc. which have different parent links? So "Template:Taxonomy/X" might have the parent taxon as "Y", "Template:Taxonomy/X/Var1" might have it as "Y/Var1", "Template:Taxonomy/X/Var1" as "Z". Is that right? If so, it's in principle nice that you can be as flexible as this, but in practice a bit worrying, since if these 'chains' aren't set up carefully, surely they can intersect, producing unwanted (and unexpected) consequences?
If I have understood correctly, one use may be to have different but consistent hierarchies for extinct & extant plants, given the discussion earlier about the lack of a consistent Linnean system which covers both. Peter coxhead (talk) 12:45, 3 March 2011 (UTC)

Italics in display_children again

This was fixed a while back but seems to be happening again for some reason... that is, genera are not italicized in display_children when collapse is active. Any workaround? See Dromaeosauridae. MMartyniuk (talk) 05:59, 28 February 2011 (UTC)

Looks correct to me. Bob the WikipediaN (talkcontribs) 06:16, 28 February 2011 (UTC)
Retracting denial above...this is an issue. Bob the WikipediaN (talkcontribs) 21:20, 12 March 2011 (UTC)

Documentation

Where to document this stuff? It's difficult, I agree. If the main documentation at Template:Automatic taxobox is too long/complex, editors will just give up. Perhaps the "Advanced uses" section needs to be a separate page which covers all the 'tricky' stuff. Peter coxhead (talk) 16:34, 1 March 2011 (UTC)
Perhaps Bob and I need to sit down and think about this – I know that he made a start of making some guides. It's actually really useful to have all your feedback: after spending too long coding the template it's often easy to lose sight of what is and is not obvious!
Martin (Smith609 – Talk) 20:18, 1 March 2011 (UTC)
Sitting down together to discuss it would be nice...you're in Europe, right? And I'm in America...hmmmmm. And don't forget, Erik's helped write a lot of the documentation as well! (Where is he lately, anyway?) Well, anyway, I've broken this off (as I often like to do) into its own topic...
  • Let's start by rounding up all the pages we can find which have documentation on the automatic taxobox. Once we've located all of them, we can probably even get rid of some redundant stuff, or perhaps split them into transcludible sections. Post them inside the box below as you round them up. Have I missed any? Bob the WikipediaN (talkcontribs) 03:41, 2 March 2011 (UTC)
List of links to automatic taxobox documentation
Well, Canada rather than Europe, but I was more thinking of a virtual sit-down... (-:
I was thinking about this last night and wondered whether /doc should be no more than a "why are you here" page, with links to a dedicated doc page for each target audience. First-time users don't need technical information, and template maintainers don't need in-depth explanations.
I thought about offering the following links
  • I'd like a brief introduction to this template (perhaps this even can be on /doc)
  • I'd like an overview of how this template works
  • I'd like to create a new automatic taxobox
  • I'd like to convert a taxobox to an automatic taxobox
  • I'd like to create an automatic list of child taxa in a taxobox (automatic or otherwise)
  • I'd like to understand the parameters that are unique to the automatic taxobox (largely met by /Setup)
  • The taxobox is throwing an error. How do I fix it? (addressed in part in /Step-by-step; ideally the error messages should be self-explanatory)
  • I'd like to do something a little more complicated (extinction, question marks, references...)
  • I'd like a list of all parameters
  • I'd like to know the gory technical details of this template (partly addressed by inline comments in the template code)
  • I'd like a list of all the templates called by Template:Automatic taxobox (largely met by /map, which could be expanded with a little more info of what each template does)
Do you think that this is the best approach? Is there anything that I've missed? Martin (Smith609 – Talk) 13:36, 2 March 2011 (UTC)
Lol, I like the "I'd like to know the gory technical details". Don't forget one of our most popluar questions-- "how do I edit the taxonomy?" And then there's "I don't like the taxonomy offered"-- which ought to advise the client set up an RfC. Bob the WikipediaN (talkcontribs) 08:04, 3 March 2011 (UTC)
We really need "an idiots guide" with friendly wording for this stuff. Even fairly seasoned editors like myself aren't always technically inclined, and new editors will probably be really scared by it. Petter Bøckman (talk) 08:26, 3 March 2011 (UTC)
I've got a certificate saying that I'm a "Chartered Information Technology Professional" and the idea of editing the taxonomy still scares me at present! One problem seems to be that you can't create edits in user space to test them (or if you can, I can't see how). Compare this to creating a new template, for example, which is also quite scary, but can be done in user space sandboxes. As people learn best by doing, it would be nice if it were possible to create and test taxonomic hierarchies 'offline'. Peter coxhead (talk) 12:48, 3 March 2011 (UTC)
{{Taxonomy/Test-1}}-{{Taxonomy/Test-50}} are designed especially for testing. Test to your heart's content! Face-grin.svg Bob the WikipediaN (talkcontribs) 23:52, 3 March 2011 (UTC)
Right. I've replaced Template:Automatic taxobox/doc with an index, and made a start on one of the subpages (Template:Automatic taxobox/doc/advanced). Anybody, please feel free to improve my style, or to expand other pages! (I've copied what was at /doc to the appropriate sub-page.) Martin (Smith609 – Talk) 02:51, 5 March 2011 (UTC)
The more reading that editors have to do before they are comfortable with the template, the less willing they are likely to be to adopt it. So I'm currently putting an emphasis on walking the user through common tasks via active prompts such as editintros and clear error messages. You might notice some changes; if you can see any scope for improvement, please feel free to make edits or suggestions. Martin (Smith609 – Talk) 02:09, 18 March 2011 (UTC)

This has got to stop

There is no consensus yet to switch to the automated taxoboxes. I would therefore request that no further edits be made with the sole intention of changing from manual taxoboxes to automatic ones. The manual system has several real and important advantages over the automated ones (all the information a page needs are kept on that page; ease of editing; etc.), and no consensus has been demonstrated for the switchover. I disapprove very strongly of this apparent attempt to sneakingly switch over to the new system through manual editing. Any edit which replaces an existing manual taxobox with an automatic taxobox needs to have gained the consensus of the community before doing so. I think the disadvantages of the new system outweigh their advantages (and have proposed a simpler alternative that would solve all the same problems without introducing new ones); you can therefore assume that the consensus is against such edits, and they must cease until it is shown otherwise. --Stemonitis (talk) 06:56, 4 March 2011 (UTC)

Are you suggesting that if I edit an article on some critter and switch from the old to the automated taxobox, I should seek the consensus of the community before doing so? Petter Bøckman (talk) 07:51, 4 March 2011 (UTC)

I am saying that anyone who sets out to do that alone should make sure there is support for it (as for any such edit – this is not a new policy). Anyone writing a new article, for instance, can do as they like, but there has been a trend towards assuming consent for edits which are more contentious than a number of editors seem to realise. The automatic taxobox is not a perfect system, and is supposed to be still in its pilot phase. It is also supposed to reproduce faithfully the effects of the old taxoboxes, so any edit which only replaces one with the other is discouraged by existing guidelines, and by common sense. I don't want to see automatic taxoboxes forced upon the community as a fait accompli, which appears to be the current strategy of its proponents. --Stemonitis (talk) 08:03, 4 March 2011 (UTC)

Um...there's no sneaking around being done here. I've been leaving edit summaries stating "automated taxonomy" so that anyone reviewing the edit history can tell when it was automated. Also-- you say you've proposed an alternative...mind linking to it? This is news to me. Bob the WikipediaN (talkcontribs) 09:20, 4 March 2011 (UTC)
Also-- we're not in "pilot" phase...this thing's developed and released, so to speak. Bug-free, as far as we're aware. {{Speciesbox}} or {{virusbox}} might be what you're thinking of that's being tested right now. Bob the WikipediaN (talkcontribs) 09:25, 4 March 2011 (UTC)
"Sneaking" only in the sense that there was once a proposal to do a bot automation and a grand switchover. That didn't happen, and the subsequent edits have been small and unannounced. A big bot request would have opened up a full debate about whether we wanted to switch all taxoboxes to automatic or not. The current method just kind of assumes that there would have been consensus for it, and acts on that basis. Finished or not (I don't recall any announcement of completion), consensus is lacking. The appropriate place for that discussion to occur is, I think, WT:TOL rather than here, since this page is unlikely to get the necessary broad spectrum of editors.
My suggested alternative is just above; the idea is that a simple algorithm could find inconsistencies in taxoboxes on different articles (which is, after all, the problem that this template was supposed to solve) and report them somewhere, either on-wiki or off-wiki. Editors could then manually resolve the problems or, if there were too many to do manually, file a bot request to do a large number at once. I asked if there were any reason why such a system should not work and would not achieve everything that this template set out to achieve, and received no reply. You did read it, Bob, because you replied, albeit without addressing that point. --Stemonitis (talk) 09:51, 4 March 2011 (UTC)
I don't think a full bot-switchover would be a good idea. There are critters, particularly somewhat fragmentary fossil ones, where the uncertain phylogenetic position require the taxobox to be jury rigged so to speak, to illustrate that an animal can fall in either of two classes, or where the assignment should be followed by a question mark. Examples include Westlothiana and Conulariida, the latter where all the ambiguity that really should have been there has been removed by using an automated taxobox. While I strongly comment the use of automated taxoboxes to streamline taxonomy on Wikipedia, I believe the option to have manual taxoboxes should exist for just such cases. Perhaps they should be labelled differently, "taxoboxes for uncertain phylogeny" or something in that vein. Petter Bøckman (talk) 10:06, 4 March 2011 (UTC)
In fact, there is already approval for a bot to replace manual taxoboxes with automatic taxoboxes. Of course, only straightforward cases (i.e. where the "automatic" taxonomy is the same as what is already displayed in the taxobox) are suitable for upgrade using a bot. Martin (Smith609 – Talk) 14:47, 4 March 2011 (UTC)
I'm rather in favour of automatic taxoboxes, and I'm currently trying to use them on articles I create or substantially revise. However there are still problems of two kinds:
  • Technical: they are quite difficult to create and use, and the documentation is still "under development".
  • Substantive: inconsistencies between taxonomies are not simply errors to be fixed. There are deep issues around whether classifications which include extinct species can be made consistent with those for extant species only and whether they can be relatively stable.
As I understand it (see #Question_marks_in_automatic_taxoboxes above), it's possible to create alternative classifications for the same taxon, so it's not the case that automatic taxoboxes will remove all inconsistencies (which is good for editor freedom, but perhaps not quite what was originally intended).
So I wholeheartedly agree that a full bot-switchover would be a bad idea, but equally I encourage fellow editors to experiment with them. Peter coxhead (talk) 12:05, 4 March 2011 (UTC)
Problem is I need a taxobox saying "Amphibia or Reptilia" or "Amphibia/Reptilia" for class. Those situations are bound to be very rare. Perhaps we can have a separate type of taxobox (similar to the present non-automatic one) for anomalous cases like this? Petter Bøckman (talk) 13:39, 4 March 2011 (UTC)
Just set |parent=Amphibia/Reptilia. Martin (Smith609 – Talk) 14:41, 4 March 2011 (UTC)
I'm confused by your comment above, Stemonitis, that anyone starting a new article can do as they like. This doesn't seem to tally with your recent editing behaviour, where you replaced the automatic taxoboxes added by the article's creator with manual taxoboxes: [1] [2][3][4]
Your edit summaries say "Standardize taxobox". What was your motivation in doing this? What exactly were you attempting to "standardize" the taxoboxes with? Martin (Smith609 – Talk) 14:35, 4 March 2011 (UTC)
The main intention with those edits was to flag up the work needed to improve each of those stubs (all orphans, as I recall). The taxoboxes are standardised in that they are now the same as the vast majority of taxoboxes – manually-set, easily-updated straightforward taxoboxes. It is much easier for editors to deal with them, and it is the standard way to show an organism's taxonomic position in a Wikipedia infobox. Even if I had acted against my avowed recommendations (which seems to be the issue that concerns you), it would not undermine my argument, and you have not addressed that. As I have stated before, it would need to be demonstrated to the satisfaction of the community that replacing a well-known system with an unfamiliar one brings with it substantial benefits. That has not yet been done. A lot of people are already put off by the complexity of the manual taxoboxes; the potential benefits of the new system are real, but are massively outweighed by the overheads and the increased complexity. Do not assume that your proposed system will eventually be implemented across Wikipedia; there may be good reasons why the old system is better. Thus, any edit which seeks solely to further the spread of automatic taxoboxes is to be discouraged. --Stemonitis (talk) 14:52, 4 March 2011 (UTC)
A list of errors is already being updated automatically per several different algorithms. It's falsely reporting all new templates until we clear them manually, but it's been very beneficial so far in pointing to problems. I've been cleaning it out nearly daily and passing errors I'm not sure how to solve along to Martin.
Speaking of which, Martin-- we do have a template loop error currently on {{Taxonomy/Horneophytopsida/?}}-- would you check that out since you're the template loop expert? I mentioned it a few days ago but it's not had any attention yet.
Although the article which uses it, Torticaulis, displays the taxobox fine. Peter coxhead (talk) 17:52, 4 March 2011 (UTC)
As for notifying folks of the new template, bots, and several RfC's requesting input on classification strategies, that's been done countless times ever since October of last year-- I'm not sure how anyone checking the WP:TOL or WP:TAXOBOX talk pages even at irregular intervals could have missed this.
Was there consensus for replacing automatic taxoboxes with manual ones with no substance for doing so other than attracting editors?
I should add that my automations of taxoboxes are not being done "solely to further the spread of automatic taxoboxes"-- rather, they are being done for multiple reasons: to reduce future taxonomy maintenance; eliminate dated, invalid taxa; and to correct flaws in taxoboxes such as inconsistent rank names. In my eyes, it's a completely different set of rationale for automation.
Petter-- it's not a full-bot switchover. The bot performs only the edits a human editor requests it make. Its purpose is to enable streamlined edits and maintain templates that are critical for certain functions of the taxobox, not to automate things by itself. Think of it as a tool, not a bot. I wouldn't support a full-bot switchover myself if it were doing things under the covers without human guidance. Bob the WikipediaN (talkcontribs) 16:31, 4 March 2011 (UTC)
What is occurring now is not something that has been discussed. Advice was sought on how to implement automatic taxoboxes, but without any discussion on whether to implement them on a wider scale. I expected, and I imagine others did too, that once it was implemented, the community would be given a chance to approve or deprecate the system. I would not make sense to have such a discussion until there was practical knowledge of how it functioned, how well it functioned, how it differs from the existing system, and so on. So a failure to decry the system earlier cannot be taken as support for its current implementation. Having two systems in place – one off-puttingly complex, and the other several times more complex still – is too many; we should probably choose one and stick to it. My concern, and what starting this whole thread, was that it was being assumed by a fairly small group of editors that there was no reason not to implement the new system on any and all articles. This has not, to my knowledge, been asked. I didn't complain about the development of a possible alternative, and it would have been wrong to do so at that stage. We all needed to see how it would turn out before deciding whether it was beneficial or not. (If this is, as you say, the final form, then my mind is made up.) There are benefits to this new system, and there are benefits to the old one. It is not at all clear that the old one should be replaced by the new one. Thus, editors should not be replacing {{Taxobox}} with {{Automatic taxobox}}. If no more testing is required, then we already have all the information we need to decide between the two systems from the automatic taxoboxes already in place.
And all this still ignores an important point. What could not be achieved by an inconsistencies report and routine bot requests? I have asked several times, and I have heard exaclty nothing in reply. The cleanup category is not at all what I meant; that chiefly reports a failure to implement the automated system properly (missing elements in the taxonomy, etc.). My suggestion (I will repeat it again) was that a fairly simple algorithm could look through existing (presumably manual) taxoboxes and report at some other location what discrepancies there were between articles. It could report, for instance, that although the genus Fictivus is listed in the family Fictaceae, the family Fictaceae does not list Fictivus as a child. An editor would be able to examine the discrepancy and might choose to add the subfamily Fictivinae to the taxobox at Fitivus if that would solve the problem. If, on the other hand, someone found that the hundreds of articles on species of Alsofictivus listed an outmoded higher classification, then the editor could file a bot request to change all those articles to the same new format. It seems to me (I may be wrong) that this approach could find all the problems that the automatic taxobox set out to solve. It would have some advantages over {{automatic taxobox}}, and I dare say a few disadvantages. However, I would still like to know what it couldn't achieve in order to be able to compare the two (and the existing system) properly. --Stemonitis (talk) 17:53, 4 March 2011 (UTC)
To respond, here are a few advantages to the Automatic Taxobox system. These are just what trip of the top of my head:
  • Accountability: under this system, a reference can be provided for every node in the tree
  • Automatic listing of child taxa in taxoboxes (using |display children=
  • Shorter code in article pages
  • Inconsistencies only need fixing once (when the automatic taxobox is established)
  • No bot needs coding and maintaining (it's not entirely clear that a bot would be able to deal with every inconsistency)
  • Copy-and-paste errors avoided when creating new taxoboxes
  • Uniform format of parameters (e.g. extinction daggers)
I completely agree that the template has a steep learning curve at the moment: this is something that we are currently addressing, as we're revisiting the documentation (see above). Having waited this long to create the documentation means that we now have experience in the best way to deal with unusual cases, and experience of what editors need and struggle to do. We'd be very grateful for any input you might have in the documentation, as it progresses. Martin (Smith609 – Talk) 18:32, 4 March 2011 (UTC)
Martin, I have constantly acknowledged that the automatic taxobox has advantages. That really is not the issue here. I do not wish to discuss the ins and outs of the atuomated system here. My point is that (in addition to its advantages) this system has some serious disadvantages, and unless and until consensus can be demonstrated for edits such as [5][6][7][8][9] (basically, anything with an edit summary of "automate taxobox"), they must not continue. This is a proposed alternative, not a de facto replacement. Kindly treat it as such. --Stemonitis (talk) 18:50, 4 March 2011 (UTC)
I was trying to respond to your question "What could not be achieved by an inconsistencies report and routine bot requests?". Did I misunderstand what you were asking? Perhaps we should continue this discussion in the new section started below, to keep it in one place. Martin (Smith609 – Talk) 19:20, 4 March 2011 (UTC)
N.B. I just thought I ought to reassure you that you have been the first person to object to the Automatic Taxobox template, and that I will not convert any further taxoboxes until this discussion is resolved. Martin (Smith609 – Talk) 19:22, 4 March 2011 (UTC)
I find this demand for a community wide consensus before starting to use this really unnecessary. We are talking about thousands and thousands of pages and the reasons to hold this up amount to we have some unclarity here and there. So, if a few editors working ion a specific group think that it is best to switch to the automatic taxobox because it makes their lives of maintaining the phylogeny consistent in that clade, they by all means just should do so. We really do not need a many pages of discussion whether exception a, b and c that affect 5, 100 and 4 pages respectively should hold up a system that is already used widely and is beneficial in many places. The last thing we want is a months long community debate that results in a somewhat consensus to let people go forward where it works. because that is generally were these discussions end up at anyway. -- Kim van der Linde at venus 18:06, 4 March 2011 (UTC)
On the contrary, I expect that any controversial changes (and implementing the automatic taxobox in an article that previously used {{taxobox}} is certainly controversial) be performed with the backing of the community, or at least no explicit disapproval from the community. It may well be beneficial in some cases, but it is not the only way to achieve the desired result. If a group has determined that it would help their particular field, then that is consensus, and is of course entirely acceptable. Foisting an unwieldy system on areas where it is not wanted is not. --Stemonitis (talk) 18:50, 4 March 2011 (UTC)
Well, we obviously disagree on whether this is controversial, because I do not see it controversial at all. But that misses the point. My pointy is that we do not need community wide consensus to implement this in clades/groups where the editors agree it is fine. But we seem to agree, and I for that reason do not see why you demand a community wide consensus first. Anyway, I think it would be far wiser to treat this with the Bold, Revert, Discuss] cycle because that will zoom in rather quickly to where there is disagreement that to demand an halt to all implementation which mostly does not cause any issues. -- Kim van der Linde at venus 19:00, 4 March 2011 (UTC)

Discussion of alternative: find and list discrepancies

I would like some clarification about your proposal, Stemontis. You appear to be suggesting that a bot would go around and maintain taxoboxes according to some representation stored somewhere of the desired phylogeny. Other than being more confusing, I can't see how this is any different really from AT. What happens when I create a new taxobox that contains a phylogeny different from the blessed one? Does the bot "fix" it? How do I figure out where to change the "blessed" phylogeny? What if I go to a taxobox that I think is wrong, fix it, see that the saved page is like I think it should be, then go away? Will a bot come and revert my change? If so, how is this any less surprising or unclear than AT? If not, how will this solve anything? Thanks, ErikHaugen (talk | contribs) 19:02, 4 March 2011 (UTC)

No, you misunderstand. As I conceive it, bots would only be involved when an editor found some large task that he/she didn't fancy performing manually (just as bot requests function now). The system would not require any automated edits outside the project namespace; it might only be one edit per week to a talk page, or it might all be held off-site. All the power would be placed in the hands of the editors.
I am not really proposing my alternative as a well-thought-through system, and I'm very open to suggestions. In fact, it's more of a thought experiment than anything else at this stage. Basically, a spider would visit pages (read-only) and note any discrepancies between taxonomies on different pages, with discrepancies recognised according to criteria which I haven't really thought about. It would then report these, either at some project page erected for that purpose (such as this), or somewhere off-site (such as this). An interested editor could then examine the list (it would probably have to be divided up by taxonomic group to be useful) and make any changes he/she felt appropriate to rectify the apparent discrepancy. I would envisage most of the fixes being made manually, except in instances where a large amount of repetitive work would be easily farmed out to a friendly bot. I think a system like this would be able to find most problems. I can't think of a taxobox consistency problem that could not be solved in this way, but as I say, it is just a vague idea at the moment, and there may well be things I've overlooked. --Stemonitis (talk) 19:54, 4 March 2011 (UTC)
While Kim's certainly partially right about the level of controversy here, I must point out Stem's not too far off regarding controversy in a few areas. Namely, my automations of Cetacea and Aardvark proved to be controversial, as they altered the ancestral taxa which had been preferred by the contributors to those pages. When this does happen (and it does, trust me), a discussion follows with the goal of modifying the database to appeal to the contributors to that article. While it's not a problem if they prefer the manual taxobox, I do feel responsible to at least update the database to reflect their wishes so it is in line with whatever taxonomy they prefer. Automation's not just about standardization, it's also about compromises and keeping them as seamless as possible in order to appeal to those working with the said group.
Stem, as for your proposal, that's an intricate one that would definitely require a lot of programming, and it's one I hadn't even thought of (I like it, too!). Due to its intricacy and my current state of business, I'm unable to pursue that at the moment, however; though if someone does begin working on it, I'd be happy to help debug as I find time. Bob the WikipediaN (talkcontribs) 22:32, 5 March 2011 (UTC)
It would seem no taxoboxes have been swapped out recently per your request. Hopefully it will stay that way unless some consensus is established somewhere in favor of swapping the taxoboxes in either direction. At this time, I don't see any reason not to introduce automated taxoboxes to taxobox-less articles, so I'll continue doing that as I see fit. And in the meantime, it can't hurt to continue building the taxonomy database. Also per this discussion, I've added a warning to the top of this page requesting folks do not "swap out" taxoboxes without further consensus. Bob the WikipediaN (talkcontribs) 21:43, 12 March 2011 (UTC)
Thanks, Bob. I think this is a reasonable approach. --Stemonitis (talk) 07:09, 14 March 2011 (UTC)
I'm not entirely happy with the wording in the box Bob has put at the top of this page. I agree that where a whole set of articles has been worked on in the past and is provided with agreed and consistent manual taxoboxes, they should not (yet) be replaced wholesale by automatic ones. On the other hand, as always, there are special cases. User:Smith609 and I appear to be the only people who have recently created new articles on early polysporangiophytes. He used automatic taxoboxes (no surprise!); I initially did not, but now that I think I've mastered the template, I've changed all the articles I created to automatic taxoboxes. I have also changed the taxoboxes in a small number of articles which already existed covering closely related taxa. This seems sensible to me as it ensures consistency. Peter coxhead (talk) 18:18, 14 March 2011 (UTC)
I think the spirit of the notice is clear. In your case, where you are actively working on a group of articles, no-one should complain. The only problem is with rolling it out as if it were the natural successor to the familiar {{taxobox}}, which it is not (necessarily), and as if it were of direct benefit to the article, which it is not (necessarily). I don't want editors to think they are actively encouraged to use the automated system, but likewise, I don't want to discourage editors from making real improvements to articles. --Stemonitis (talk) 18:27, 14 March 2011 (UTC)
I strongly encourage you, Peter (and anyone else interested in cleaning up the controversy around migration), to discuss this issue in full with your respective WikiProject prior to engaging in any migrations. We don't need to stir up as much commotion as there was with the {{italictitle}} (which, if you recall, the collective WP:TOL had full consensus for adopting-- that's hope for any change-pushers right there Face-grin.svg, although the other projects had clear divides, which have since been overcome). Right now it would seem the project using this on the widest scale is the WP:DINO, and they're doing a wonderful job with rolling it out, from what I've seen, with no opposition yet (just a little resistance to learning how to use it[10]). I think that at this point, a proposal at that WikiProject for declaring automation of an uncontroversial taxonomy as an uncontroversial edit might go over well. Let the dinosaurs lead the way, and other WikiProjects may become inspired to adopt it as well. Bob the WikipediaN (talkcontribs) 07:16, 15 March 2011 (UTC)
I've tried to discuss taxonomy & taxoboxes generally and specifically at Wikipedia talk:WikiProject Plants, because there are some very serious problems in the way we are presently handling higher level taxa, but the number of editors currently active seems to be very small indeed, and I feel I've either been talking to myself or at most with 2-3 others. It's interesting that the main use of automatic taxoboxes is at WP:DINO, since I think their big advantage is in handling extinct organisms for which the literature uses cladistic rather than Linnean approaches. Sadly there is no "WikiProject Paleobotany"; I'm only aware of one other editor currently doing anything in this area (Martin Smith609) and his views are well known. Peter coxhead (talk) 10:40, 15 March 2011 (UTC)
Ach, Bob, you're once again assuming that it will eventually be adopted, and that that's a good thing. That talk page link you provided is very instructive. Editors are being made to feel that they "have to" use the automatic taxoboxes; they must not be made to feel that way. It's that mindset that I was trying to challenge, and I see now that my message hasn't got through. --Stemonitis (talk) 07:30, 15 March 2011 (UTC)
I'm afraid you've misunderstood. I'd like to remove the cloud of controversy surrounding automation (and yes, I do feel it's a major, long-awaited step forward, even if it has costs). How would you propose going about it? I mean, imagine you have this great invention that can help everyone; you want the world to try it out and see how they like it; but you're not allowed to let them try it, because, let's say, it's illegal in all countries due to laws that didn't forsee the invention coming? (Actually, I just heard about such an invention today on the news...) Bob the WikipediaN (talkcontribs) 07:55, 15 March 2011 (UTC)
You can't just "remove the cloud of controversy" without addressing the controversy itself. The system is not a "great invention that can help everyone"; it is seriously flawed. It has some advantages but not (at least in my opinion) enough to outweigh the very real problems that it begets. You ask how I would propose going about it. I would propose not going about it. That is the point that I have been trying to make for some time, and is how this discussion began. You feel that it is a "major, long-awaited step forward", but others (myself included) disagree, often quite strongly. That is the controversy, and it is not something that can be removed by a single editor for his or her convenience. You can try to argue your case, but pushing it through one project at a time and then presenting it to the community as a fait accompli is extremely bad form. --Stemonitis (talk) 08:23, 15 March 2011 (UTC)
Then let's attempt to address issues. I'm starting a new topic below shortly where everyone can throw out all the issues they have about it, and we'll do our best to address any that can be addressed within reason. Bob the WikipediaN (talkcontribs) 15:20, 15 March 2011 (UTC)
RE Peter's last comment-- All it takes to start a project on Wikipedia is one man and some determination. You might be able to interest me in it to some extent...I'm into paleobiology. Bob the WikipediaN (talkcontribs) 23:57, 15 March 2011 (UTC)
    • ^ Cite error: The named reference WoRMS was invoked but never defined (see the help page).