Template talk:Automatic taxobox/Archive 10

Archive 5

←

Archive 8

Archive 9

Archive 10

Archive 11

Archive 12

→

Archive 14

Oddities in display at Template:Taxonomy/X pages

A couple of oddities I noticed when setting up some taxonomic hierarchies for use by {{Automatic taxobox}}. They concern the display of Template:Taxonomy/X pages.

• If taxon X is extinct, Template/Taxonomy/X/? seems to display the † twice. See e.g. Template:Taxonomy/Horneophytopsida/?. There's no problem with the display from {{Automatic taxobox}}, see e.g. Horneophyton.

   Fixed, but may have caused other quirks in these templates? Martin (Smith609 – Talk) 13:42, 17 March 2011 (UTC)

• If I set the same as parameter, I would expect that all the fields would be copied over except those I change. However, the "Taxonomic references:" is not. Compare e.g. Template:Taxonomy/Euphyllophytina/Subdivision (for treating Euphyllophytina as a subdivision) to Template:Taxonomy/Euphyllophytina (for treating Euphyllophytina as a clade). Is this deliberate? There could be an argument for it, but I found it unexpected.

   Fixed: this was originally deliberate, but I think that the argument for it can be addressed by clarifying the source of the references. Martin (Smith609 – Talk) 13:42, 17 March 2011 (UTC)

There is still the strange behaviour that I always have to save twice (occasionally three times) to get the display to match the source. Peter coxhead (talk) 09:49, 10 March 2011 (UTC)

After saving any taxon template, you'll need to click the light blue link at the bottom of the template to completely update that page (that performs a null edit on the template). Likewise, following updating a taxonomy, the taxoboxes featuring that taxon need updated. Wikipedia is powered by cached information, so unfortunately this null edit on each page is what drives the revisions home.

Thankfully, Wikipedia automatically refreshes its article space cache at regular intervals (I've not seen evidence of it refreshing the template space cache, though), meaning that you don't have to provide that null edit on the articles themselves..they'll get updated on their own. It might take a day for this to happen, so remember that if you've left the page in a "critical" state you still need to perform that null edit. And yeah, I noticed that oddity of the extinction dagger. Thankfully, it's not causing front-end cosmetic issues and also not causing any error flags. Bob the WikipediaN ^{(talk • contribs)} 18:59, 10 March 2011 (UTC)

A null edit is fine if you save the page, but if I'm adding citations, etc. then I like to use "Preview" (e.g. not matching {{ and }} in a cite template in a refs parameter causes havoc!). Multiple uses of "Preview" usually seem to work; this is partly what I meant by "save" in my comment above. Peter coxhead (talk) 21:04, 10 March 2011 (UTC)

References in Taxonomy templates

The refs parameter in {{Taxonomy}} templates is important, but perhaps does not go far enough. Ideally three pieces of information are needed:

1. The authority for the taxon, in the style of the appropriate Code. Although there are perhaps problems in displaying this automatically (which would be ideal), an explicit field displayed on Template:Taxonomy/X pages would assist in consistency across articles.
2. At least one reference for the authority for the taxon, either the source in which it was formally named or a source which assigns that authority.
3. At least one reference to support the placement of the taxon in the parent taxon.

It is possible, although I think not ideal, to fit these into the refs, e.g. by using a layout like the following:

Authority: ....

Authority refs: ...

Placement refs: ...

Any views on this? Peter coxhead (talk) 10:25, 10 March 2011 (UTC)

Ehhhh perhaps the best way to handle it is to simply fill in the single |refs= with something like "parent - Smith 2006; placement - Johnson & Johnson 2007; taxon authority - Jacobs 1799; extant according to IUCN 3.0", just filling it in as needed, describing what refs were used for any piece of information. Bob the WikipediaN ^{(talk • contribs)} 19:05, 10 March 2011 (UTC)

• This could be confusing unless everyone agrees and uses a common format. Certainly when I look at some of the templates at present, it's not clear to me what the reference supports: the taxon or the placement.
• The easiest way to ensure that everyone uses a common format is via separate parameters.
• Putting over-complex stuff as a parameter value increases the chances of accidentally inserting something which interferes with the template functioning when expanded (unbalanced { or } for example). Peter coxhead (talk) 21:10, 10 March 2011 (UTC)

Vandalism watch

Well, thanks to the Taxobot watchlist and the task force watching it, I think we've solved most vandalism issues. Has anyone recently noticed any vandalism that lasted more than briefly? Bob the WikipediaN ^{(talk • contribs)} 03:32, 8 May 2011 (UTC)

Did someone just break major taxa or is it just my browser? See here.--Obsidi♠n^Soul 05:20, 11 May 2011 (UTC)

That would appear to be these three good-faith edits:

• Template:Taxonomy/Plantae
• Template:Taxonomy/Asterids
• Template:Taxonomy/Eukaryota

I've dropped the use a note, but it probably means the templates need to be fixed to allow non-included interwiki links (although how they they actually help in this case is currently beyond me). --Stemonitis (talk) 05:40, 11 May 2011 (UTC)

Actually, there were more than three, but you get the idea... --Stemonitis (talk) 05:42, 11 May 2011 (UTC)

I might add that neither this fix nor the previous vandalism I reverted (at Template:Taxonomy/Bilateria) used the "Taxobot watchlist" or involved the "task force". I don't mean to denigrate either of those, but I don't wish to give the impression that all our successes in combating vandalism and other unwanted changes are down to those innovations. They were found by simple, old-fashioned "Related changes". --Stemonitis (talk) 05:51, 11 May 2011 (UTC)

I thought it was that at first, but it didn't seem to break Angiosperms and Plantae which also had the interwiki. Only Eukaryota, Eudicots, etc. --Obsidi♠n^Soul 06:03, 11 May 2011 (UTC)

I can't explain the differences, but undoing those edits solved the problem. I think we can safely assume that some or all of them were causing trouble, and until we have a solution (I suspect anything on a new line breaks the templates. Bob? Martin?), we have to undo them. It will be easy to revert my actions once it's safe to do so. --Stemonitis (talk) 06:13, 11 May 2011 (UTC)

Yeah. The closing double pipes attempting to wrap around 2 lines is causing the break, probably.--Obsidi♠n^Soul 06:16, 11 May 2011 (UTC)

If you remove the new line between }} and <noinclude>, it seems to be ok according to my tests with Template:Taxonomy/Ipheion. As interwikis are likely to appear if other languages are copying the automatic taxobox system, I suspect it's better to have }}<noinclude><!-- message about adding interwikis, etc --></noinclude> in all the taxonomy templates. Sigh... Peter coxhead (talk) 06:43, 11 May 2011 (UTC)

Before we do that, we should investigate if there's an automatic way of stripping off trailing newline characters. The more experienced template coders might know of one. --Stemonitis (talk) 07:15, 11 May 2011 (UTC)

Since the Taxonomy/taxon templates are not the 'final' template which is used to generate HTML, then it IS possible to ensure that the templates which use the Taxonomy/taxon templates deal with any extra white space passed to them. However, if you look at Template:Automatic_taxobox/map, fixing this is not likely to be easy because of the very complex interaction between the templates. It took me about half a day to find and fix two erroneous space characters in Template:Speciesbox. However, Martin User:Smith609 or User:Bob the Wikipedian may be able to do it. In the meantime, removing the line break as above is, I suspect, the best solution, since it allows for later interwikis to be added safely (whatever use they are!). Peter coxhead (talk) 08:18, 11 May 2011 (UTC)

I'm on my way out the door for a good part of the day, but I'll certainly think about this. I think an interwiki parameter for {{Don't edit this line}} may be the solution. Bob the WikipediaN ^{(talk • contribs)} 14:54, 11 May 2011 (UTC)

Thanks for handling all those reverts, Stem. I was astounded to see 50-someodd changes on the watchlist this morning! I've surprisingly never noticed the "related changes" link; I'll have to check it out sometime and see how that puts its list together. I think we can safely modify the preload template to accommodate Peter's solution, and I'll attempt that right now. We'll still need to keep close tabs on interwikis being added to the already-existing ones, though. Maybe we can get a Taxobot task approved to set the comments up. If the new preload doesn't work, holler. Bob the WikipediaN ^{(talk • contribs)} 18:11, 11 May 2011 (UTC)

Sorry for the triple-post...do we want to inform the Macedonian guy that he may resume his helpful interwiki-linking if he adheres to this format? Bob the WikipediaN ^{(talk • contribs)} 18:29, 11 May 2011 (UTC)

I would say yes. Initially I couldn't see any value in interwiki links for these templates, but on reflection (a) it's interesting to see whether other languages are using the same taxonomy (b) they may have taxonomic references we don't have. Peter coxhead (talk) 19:48, 11 May 2011 (UTC)

{{resolved}} I've notified him and restored (with Peter's solution) the interwiki links on both wikis for Theraphosoidea so he can see what needs done. Bob the WikipediaN ^{(talk • contribs)} 20:08, 11 May 2011 (UTC)

Taxobox Task Force

Sign below if you're interested in helping form a taxobox task force

1. Bob the WikipediaN ^{(talk • contribs)} 22:58, 19 March 2011 (UTC)
2. Rkitko ^(talk) 23:58, 19 March 2011 (UTC) Yeah, sure. I rarely take wikibreaks and keep a vigilant eye on my watchlist. I'd be happy to revert vandalism on taxonomy templates.

Would anyone be interested in forming a task force from those of us who are maintaining, coding, and monitoring different aspects of the AT taxoboxes? Having great resources such as the public Taxobot watchlist is only useful, after all, if several people are watching it regularly. If there's enough interest, I'll set up a task force page where relevant resources et cetera can be posted without cluttering up this talk page. Bob the WikipediaN ^{(talk • contribs)} 22:58, 19 March 2011 (UTC)

Reworded my proposal just now...it only seems fair we should provide support for all taxoboxes, not just the automatic ones. After all, we can't look like agenda-pushers here. Bob the WikipediaN ^{(talk • contribs)} 02:33, 21 March 2011 (UTC)

Created task force at Wikipedia:Taxobox task force. Bob the WikipediaN ^{(talk • contribs)} 18:19, 2 April 2011 (UTC)

[ e ]

Happened across the Dytiscidae article and glaring back at me from the "Scientific classification" header of the automatic taxobox was a big [ e ] close to the right end of the header. I've been looking for the cause, and I thought I'd involve others in the search for why that [ e ] is messing up an otherwise professional looking tbox. So... HELP! – Paine Ellsworth ( CLIMAX )  08:39, 1 June 2011 (UTC)

• PS. I use IE8 - I'll check it in FireFox and get back to you.

It's supposed to be a small e, linking to the "edit" function. Petter Bøckman (talk) 08:40, 1 June 2011 (UTC)

Interesting, and it doesn't look nearly as ugly and unprofessional in FireFox as it does in IE. Also, rather than a link to the "edit" function, it links to Template:Taxonomy/Dytiscidae. Am I amiss? If that actually is the "edit" function, then it seems awfully editor-unfriendly. And did I mention how ugly and unprofessional it looks in my IE browser? – Paine Ellsworth ( CLIMAX )  08:55, 1 June 2011 (UTC)

Sorry, you're right, it is supposed to bring you to the Taxonomy template (which again can be edited). I guess Bob or one of the other template gurus will have to have a look at it. Cross-platform compatibility can be a bitch. Edit: It looks OK in my IE 8. What subversion are you at? Petter Bøckman (talk) 08:58, 1 June 2011 (UTC)

(edit conflict)So we've done a good thing here! Another reason I questioned it is that I didn't see the [ e ] in the tbox on the /doc page— you know the one— it has a great pic of a grinning, toothy chimp in it. (IE8.0.6001.18702 on WIN XP Prof.) – Paine Ellsworth ( CLIMAX )  09:14, 1 June 2011 (UTC)

What? Going from the Dytiscidae via the e I come to Template:Taxonomy/Dytiscidae, no grinning chimp there. You may have opened the taxobox template page in stead. My IE version is also 8.0.6001.18702, and I too run XP. This is very strange. Petter Bøckman (talk) 09:29, 1 June 2011 (UTC)

The grinning chimp's on this page, Mr. B. – Paine Ellsworth ( CLIMAX )  11:42, 1 June 2011 (UTC)

Also, the skin you're in can make a diff. I use Vector to edit...

– Paine Ellsworth ( CLIMAX )  14:35, 1 June 2011 (UTC)

• PS. The Dytiscidae tbox "Scientific classification" header pretty much suks across all the skins.

Okay, this might do it for you, Mr. B... I just decreased my text size to "medium" in the View menu, and the [ e ] no longer covered the "tion" in "Scientific classification". So it may be that usage of a smaller text size looks okay across skins, too. – Paine Ellsworth ( CLIMAX )  14:52, 1 June 2011 (UTC)

Quick thought...we could replace that ugly [e] with Bob the WikipediaN ^{(talk • contribs)} 17:18, 1 June 2011 (UTC)

A small picture file (whether the pencil or something else) would look nice. Petter Bøckman (talk) 20:25, 1 June 2011 (UTC)

Then if there's no objection, I'll do that tonight when I return home-- off to a rehearsal now. Bob the WikipediaN ^{(talk • contribs)} 21:42, 1 June 2011 (UTC)

I like it, as long as it won't clash with any major box colors to the point it would be hard to see. MMartyniuk (talk) 22:00, 1 June 2011 (UTC)

{{done}} Feedback? Problems? Bob the WikipediaN ^{(talk • contribs)} 01:22, 2 June 2011 (UTC)

Excellent! Even for old guys like me who like to use large print. The image is off to the right of the "Scientific classification" and looks very professional. It seems there was a question as to whether it linked to the right place. Was I wrong about that? Also, I'm curious as to why it is there in that particular header and not, say, at the bottom of the template where the usual v-d-e is found? – Paine Ellsworth ( CLIMAX )  06:54, 2 June 2011 (UTC)

There's just the small issue of licensing. That image is licensed under CC-BY-2.5, which requires the authorship information (i.e. image page) to be linked to. You will need to find a PD alternative, I fear, in order to be able to (legally) link to the taxonomy template rather than the image information. --Stemonitis (talk) 07:05, 2 June 2011 (UTC)

How about this icon ? A red pencil is more traditional for editing anyway. But I haven't checked the background colours for visibility. If it doesn't show with some, then maybe a black pencil ? A pencil icon is SO much nicer than the old [ e ].Peter coxhead (talk) 07:53, 2 June 2011 (UTC)

[ e ]

Happened across the Dytiscidae article and glaring back at me from the "Scientific classification" header of the automatic taxobox was a big [ e ] close to the right end of the header. I've been looking for the cause, and I thought I'd involve others in the search for why that [ e ] is messing up an otherwise professional looking tbox. So... HELP! – Paine Ellsworth ( CLIMAX )  08:39, 1 June 2011 (UTC)

• PS. I use IE8 - I'll check it in FireFox and get back to you.

It's supposed to be a small e, linking to the "edit" function. Petter Bøckman (talk) 08:40, 1 June 2011 (UTC)

Interesting, and it doesn't look nearly as ugly and unprofessional in FireFox as it does in IE. Also, rather than a link to the "edit" function, it links to Template:Taxonomy/Dytiscidae. Am I amiss? If that actually is the "edit" function, then it seems awfully editor-unfriendly. And did I mention how ugly and unprofessional it looks in my IE browser? – Paine Ellsworth ( CLIMAX )  08:55, 1 June 2011 (UTC)

Sorry, you're right, it is supposed to bring you to the Taxonomy template (which again can be edited). I guess Bob or one of the other template gurus will have to have a look at it. Cross-platform compatibility can be a bitch. Edit: It looks OK in my IE 8. What subversion are you at? Petter Bøckman (talk) 08:58, 1 June 2011 (UTC)

(edit conflict)So we've done a good thing here! Another reason I questioned it is that I didn't see the [ e ] in the tbox on the /doc page— you know the one— it has a great pic of a grinning, toothy chimp in it. (IE8.0.6001.18702 on WIN XP Prof.) – Paine Ellsworth ( CLIMAX )  09:14, 1 June 2011 (UTC)

What? Going from the Dytiscidae via the e I come to Template:Taxonomy/Dytiscidae, no grinning chimp there. You may have opened the taxobox template page in stead. My IE version is also 8.0.6001.18702, and I too run XP. This is very strange. Petter Bøckman (talk) 09:29, 1 June 2011 (UTC)

The grinning chimp's on this page, Mr. B. – Paine Ellsworth ( CLIMAX )  11:42, 1 June 2011 (UTC)

Also, the skin you're in can make a diff. I use Vector to edit...

– Paine Ellsworth ( CLIMAX )  14:35, 1 June 2011 (UTC)

• PS. The Dytiscidae tbox "Scientific classification" header pretty much suks across all the skins.

Okay, this might do it for you, Mr. B... I just decreased my text size to "medium" in the View menu, and the [ e ] no longer covered the "tion" in "Scientific classification". So it may be that usage of a smaller text size looks okay across skins, too. – Paine Ellsworth ( CLIMAX )  14:52, 1 June 2011 (UTC)

Quick thought...we could replace that ugly [e] with Bob the WikipediaN ^{(talk • contribs)} 17:18, 1 June 2011 (UTC)

A small picture file (whether the pencil or something else) would look nice. Petter Bøckman (talk) 20:25, 1 June 2011 (UTC)

Then if there's no objection, I'll do that tonight when I return home-- off to a rehearsal now. Bob the WikipediaN ^{(talk • contribs)} 21:42, 1 June 2011 (UTC)

I like it, as long as it won't clash with any major box colors to the point it would be hard to see. MMartyniuk (talk) 22:00, 1 June 2011 (UTC)

{{done}} Feedback? Problems? Bob the WikipediaN ^{(talk • contribs)} 01:22, 2 June 2011 (UTC)

Excellent! Even for old guys like me who like to use large print. The image is off to the right of the "Scientific classification" and looks very professional. It seems there was a question as to whether it linked to the right place. Was I wrong about that? Also, I'm curious as to why it is there in that particular header and not, say, at the bottom of the template where the usual v-d-e is found? – Paine Ellsworth ( CLIMAX )  06:54, 2 June 2011 (UTC)

There's just the small issue of licensing. That image is licensed under CC-BY-2.5, which requires the authorship information (i.e. image page) to be linked to. You will need to find a PD alternative, I fear, in order to be able to (legally) link to the taxonomy template rather than the image information. --Stemonitis (talk) 07:05, 2 June 2011 (UTC)

How about this icon ? A red pencil is more traditional for editing anyway. But I haven't checked the background colours for visibility. If it doesn't show with some, then maybe a black pencil ? A pencil icon is SO much nicer than the old [ e ].Peter coxhead (talk) 07:53, 2 June 2011 (UTC)

Stem, the thought hadn't even crossed my mind-- mostly because I've always made images that link to their page! Thanks for pointing that out! I'll stick the red one on there. Oh-- and thanks for the speedy creation of new public domain icons, Peter! Bob the WikipediaN ^{(talk • contribs)} 15:53, 2 June 2011 (UTC)

Another minor thing, the text popup when hovering the pointer over the icon should probably be something more descriptive than 'e'. -- Obsidi♠n^Soul 16:02, 2 June 2011 (UTC)

(Responding to Paine Ellsworth's comment.) Actually, it would be nice if the icon were always to the right of the text, but it isn't if the taxobox is sufficiently narrow (e.g. no image) and the text size sufficiently large. I don't see a solution to this if the text ("Scientific classification") is to be centred (it would look odd otherwise) and the icon is to be at the right. I'm wondering if there is somewhere else in the taxobox to put the icon which could ensure no overlap. Paine Ellsworth did suggest the bottom of the box. Anyone have any views on this? Peter coxhead (talk) 17:14, 2 June 2011 (UTC)

I fixed the tooltip so it now reads "Edit this classification" instead of "e". Hmmm. The bottom of the box is such a variable format, it might be difficult to program or even to anticipate problems with that. I hate to say this, but the text overlap is simply a risk you take when you change the default text size, and this problem isn't unique to this template-- it's a problem you'll find all over the Internet. Bob the WikipediaN ^{(talk • contribs)} 17:22, 2 June 2011 (UTC)

I agree that the bottom of the box is tricky; also the obvious place for an icon which allows you to change the scientific classification is next to the words "Scientific classification"...

Just because you find overlaps all over other people's web pages doesn't mean we should produce them on ours. :-) On the other hand, they are exceedingly difficult to avoid while maintaining browser independence. I hope we can keep thinking about whether there is a way to fix this. Any CSS experts out there with ideas? Peter coxhead (talk) 18:02, 2 June 2011 (UTC)

Not a CSS solution, but a possible way out nonetheless: since taxoboxes are created using table formatting, would it be possible to divide the cell that currently contains "Scientific classification [e]" into three cells, one on the right containing "" (or "[e]"), a centre one containing "Scientific classification", and a left-hand one containing a transparent image the same size as the pencil (or invisible text the same length as "[e]") to balance the formatting? In very narrow taxoboxes, the "Scientific classification" text might wrap over two lines, but it shouldn't ever overlap the pencil. --Stemonitis (talk) 18:08, 2 June 2011 (UTC)

Best idea I've heard all day! Bob the WikipediaN ^{(talk • contribs)} 22:53, 2 June 2011 (UTC)

Issues that have been brought up

Here are the complaints I've heard; if you know of other complaints you've heard, or have, add them to the list. My hope is that we can address at least some of these issues, although some of them are simply inherent of the system and cannot be addressed. Bob the WikipediaN ^{(talk • contribs)} 15:26, 15 March 2011 (UTC)

Issues

1. It adds a few seconds' time to server-end processing.
2. There's no report generated for telling what manual taxoboxes might contradict with the automated system
3. It's difficult to learn
4. It's difficult to put together certain "exceptions" without consulting the programmers
5. Usage of it in any article means the taxonomy has become inaccessible to those who don't know how to use it
6. Vandalism is less noticeable (watchlist, recent changes, page history)
7. Complete documentation is not yet available
8. It is too hidden
9. It discourages new editors
10. Links back to involved code-level templates aren't easily accessible
11. Article edit histories are more difficult to follow
12. Vandalism is heavier (affects multiple pages)
13. Vandalism is difficult to trace (which page got vandalized?)

Comments

Particularly, #5 is one that we won't be able to do much about. I'd comment more now, but I need to run off to class now. Bob the WikipediaN ^{(talk • contribs)} 15:26, 15 March 2011 (UTC)

I don't understand issue #2. Surely there's no telling which manual taxoboxes contradict with other manual taxoboxes, either? Martin (Smith609 – Talk) 13:46, 17 March 2011 (UTC)

I think that the worries that some editors have expressed about #2, including me in the past, leading to Bob's listing of it, relate to replacing manual taxoboxes by automatic taxoboxes and the fear that this could change an editor's or project's preferred taxonomy for a group where more than one taxonomy is legitimate. Taxoboxes of any kind can (and do) contradict one another for different reasons, including:

• errors – which automatic taxoboxes will help to prevent
• consensus taxonomies for different taxa conflicting – which is much more of a problem than I had once realized, but which I now understand is a problem that automatic taxoboxes can handle via "Template:Taxonomy/TAXON/VARIANT" approaches
• the absence of any consensus taxonomies at all leading to difficult, perhaps arbitrary choices by editors – this is a huge problem at higher levels, and can't be resolved by any kind of taxobox.

At root, I think concerns are (hopefully now "were") that the simplistic view "inconsistency = bad = needs to be removed" might prevail. Peter coxhead (talk) 13:17, 18 March 2011 (UTC)

I mostly added #2 in response to Stem's comments in more than one discussion. I'm not convinced that it's a strong reason against using automatic taxoboxes, but it would certainly be a useful tool if we had the opportunity to piece it together in the future.

Ah. I didn't recognise it as such. My point was not, however, that there were things the automatic taxoboxes didn't do, but that the things automatic taxoboxes do achieve could also be achieved by simpler, more straightforward (at least from a user's point of view) means. I was not seeking problems with the automatic taxoboxes, but to show how their advantages could be replicated without the overheads. I do think that there are problems with the automatic taxoboxes, but they're not explained by point #2, which, as Martin rightly points out, is fairly impenetrable. I also think that #5 is too weak. It is a problem that the taxonomy which appears on a page is encoded by material held elsewhere, possibly on several dozen indirectly linked pages. That's not just a problem for the uninitiated as currently implied; it makes the edit history of an article much harder to follow, since to know how an article used to appear, one would have to know what state dozens of other pages were in at the same time, some of which may no longer be linked. Transcluded templates already pose this problem to a certain extent, but changes to them are unlikely to represent changes in, or disputes over, content. Preserving the edit history is an important part of how wikis function, so convoluting that link serves to undermine the editing process. Similarly, I think #6 misses the point a little, and this is related to my comments on #5. The point is not that vandalism is less noticeable (in terms of changes appearing on watchlists), but that it may be quite hard to trace (and could actually be very visible, since it could affect thousands of articles at once), making it harder to undo, and leaving it there for longer (the trouble at bird a few weeks ago – or is it months? – springs to mind). --Stemonitis (talk) 18:10, 18 March 2011 (UTC)

For #5: That is already a problem with all templates at the moment. It's not unique to autotaxobox and has happened before (where someone vandalized, say an infobox person template, leaving well-intentioned new editors confused as to what was causing them and affecting multiple pages). It happens quite rarely though as most vandals don't understand the concept of transclusion.

I think the actual problem is that the template pages themselves can't easily be watchlisted. i.e. they do not inherit watchlisting the article page and most editors don't actually bother patrolling templates. Is there a way to inherit watchlists into the taxa templates transcluded into a page?

That way more obscure taxa can get patrolled by more people as soon as they put an article it is in on their watchlists. Most of them are quite static so they won't result in watchlist spam. But it would mean that any vandalisms to them would be noticed at the same rate as their parent articles. And more, it would mean the higher the taxon, the more people would be watching its template (as it would be transcluded in more pages) and the less likely for vandals to cause widespread disruption by changing higher taxa as what happened in birds.

As for the other problems mentioned in #2, i don't think it's actually a problem. Autotaxoboxes forces convention and consensus, and that's good right? Manual taxoboxes are done on an article to article basis which means there are far more chances for manual taxoboxes to contradict each other than there would be for autotaxoboxes contradicting manual taxoboxes.--Obsidi♠n^Soul 18:36, 18 March 2011 (UTC)

I believe I have demonstrated that consistency among manual taxoboxes can be achieved without many of the other problems engendered by the automatic taxobox system. --Stemonitis (talk) 07:32, 19 March 2011 (UTC)

It can be, but (a) it's easier with an automatic taxobox system (b) if a new consensus taxonomy appears for some of the currently very problematic higher level groups, it will be much easier to change over to it with automatic taxoboxes. Peter coxhead (talk) 10:31, 19 March 2011 (UTC)

Yeah. imho autotaxoboxes aren't that much complicated than say... creating pushpin maps in geographical articles. Anyone who has at least started to understand the concept of an infobox would find the instructions for creating autotaxoboxes just a little bit more difficult. But not by that much.

While manual taxoboxes seem more 'straightforward' in terms of technical know-how required, they are also far more labor intensive than autotaxoboxes in terms of involvement, time, and number of editors. And WP:TOL articles aren't exactly overflowing with participants. Some wikiprojects probably have editors doggedly updating taxa every time they change, requiring days to go through lists. Sure they achieve consistency just as well with manual taxoboxes, but at what cost? That time could have easily been spent improving textual content rather than having no choice but to obsess over phylogeny/taxonomy.

Bots might also do the trick, but you'd again need a coder, permissions, and it's far less reliable than the autotaxoboxes which are still somewhat 'manual'. And the use of bots is arguably far less new-user friendly as well and a bit more arcane in that there are after all, only a few users with the expertise to make them.

With autotaxoboxes you'd only need one editor. If you can't do it yourself you can always request someone else to do it or teach you to do it. Sure, it'd have to be someone a bit more knowledgeable with autotaxobox templates but he won't be anyone special. It's easier to learn how to edit autotaxobox templates than it is to write bots after all. And as I've said, it encourages consensus. Brings editors together to discuss changes in taxonomies, rather than purely unilateral actions as in the past (a model which is more 'article-ownership'-y, in that original authors can prefer different taxonomies for pages they write than that of the rest).

Noobs may also find it a bit more opaque but so will vandals. The opacity, imo, is just right in the way that while it doesn't actually forbid new editors to edit, it makes it so that they'll have to at least learn more about editing before they can change something as nontrivial as taxonomy. Something like due and undue weight, I suppose.

Autotaxoboxes though are more suitable for the more static/stable taxa. Especially those which have been forgotten, the only editors which had them on their watchlists long since gone. Or orphan articles gathering dust in some corner of Wikipedia. They are often the ones which remain hopelessly outdated when taxonomies are updated elsewhere. With autotaxoboxes, they are kept up to date without having to go hunt for them.

That was actually the reason why I've replaced some articles I've come across with autotaxoboxes (in addition to simply replacing them because their info didn't match other taxoboxes/didn't match textual descriptions). While it may seem crazy to replace manual taxoboxes on random articles for no reason at all than to replace them (and no I wasn't awate that I was doing it 'sneakily' either), what happens if its original creator (who in most cases, again, are the sole people who have it on their watchlists) leave? It only takes one updating of a family article, for instance, in which the user forgot to update the child taxa as well, to render some genus/species articles orphans. Not to mention that article chains are often 'broken'. An example of that is an article Stemonitis and I have edited in the past: Polydesmida. The article lists Suborders which were all redlinks before, leaving families, genera, and species belonging to the order 'orphans' with no apparent connection to it. That again, could have been easily circumvented by autotaxoboxes where you could have seen and updated articles under it without having to create articles on more obscure 'in-between' taxa.

Nonetheless I don't quite recommend enthusiastically replacing manual taxoboxes with autotaxoboxes in articles with taxonomies that are constantly being revised/are controversial. As I've discovered before when I mistakenly replaced Anatoma manual taxoboxes with autotaxoboxes resulting in more problems for the wikiproject concerned (and I apologize for that, heh).

So yes, I don't consider #2 an actual problem, though I agree there are still stuff that needs more discussion on. A bit more restraint of course, but replacing them should not be considered anathema. I'll stop though, for now, but I'll still use them in new articles I write or significantly add to.--Obsidi♠n^Soul 12:46, 19 March 2011 (UTC)

Couldn't have said it better myself, Obsidian, and you even raised a very insightful few points I hadn't thought of. (By the way....contrary to the Wiktionary entries (which have been reverted each time they've been fixed due to lacking scholarly sources), there's a difference between "newb", short for newbie, and "noob" (also "n00b"), a 1337 troll who disguises himself as a newbie; I'm guessing from the context you meant "newb" here). Bob the WikipediaN ^{(talk • contribs)} 20:26, 19 March 2011 (UTC)

In response to concern #6, it's actually possible to create a public watchlist for any user (under Preferences there are details about sharing your watchlist). Is there any way we could have Taxobot automatically add all taxon templates to its watchlist? That way, anyone who wishes to serve on a sort of AT task force could monitor that in addition to watching the error categories. Bob the WikipediaN ^{(talk • contribs)} 17:50, 18 March 2011 (UTC)

Sounds feasible! Is there documentation pertaining to this? Martin (Smith609 – Talk) 19:58, 18 March 2011 (UTC)

Oh and I somehow missed Bob the WikipediaN's comment lol. Edit window was opened for quite a while before I saved it. Anyway yeah his idea is more feasible. Watchlisting the generated watchlist should do the trick. --Obsidi♠n^Soul 21:14, 18 March 2011 (UTC)

Martin, you can share the watchlist by logging into Taxobot's account, and then navigate to Special:Preferences#preftab-5. Take note of the watchlist token (if it's not there, make one up). The watchlist can then be accessed by anyone with the link and may even be subscribed to as an RSS feed! The link will be in the format http://en.wikipedia.org/w/api.php?action=feedwatchlist&wlowner=Taxobot&wltoken=TOKEN where "TOKEN" is whatever the watchlist token is. If that isn't clear, details are also available at WP:Syndication#Watchlist feed with token.

Thanks; I've done that here. Should I include anything starting "Template:Taxonomy" on the list? Martin (Smith609 – Talk) 05:04, 19 March 2011 (UTC)

That would be wonderful, yes! Thanks for tackling that! I've subscribed to the RSS, and I highly advise anyone else concerned about vandalism of these templates do the same. Bob the WikipediaN ^{(talk • contribs)} 06:07, 19 March 2011 (UTC)

Also-- would you mind setting that watchlist to display all changes instead of only the most recent? Bob the WikipediaN ^{(talk • contribs)} 20:51, 19 March 2011 (UTC)

Just add "&allrev=1" to the end of the URL. Martin (Smith609 – Talk) 21:39, 19 March 2011 (UTC)

I've added the concerns raised above to the list of issues above. It would seem we're in accord that #2 isn't really an issue, so I'll strike that. Bob the WikipediaN ^{(talk • contribs)} 04:11, 19 March 2011 (UTC)

Regarding issue 1, it's possible to speed things up by switching to an extension to generate the template output, as currently being discussed at [[Template_talk:Citation/core#Looking_for_examples.2Fsuggestions_for_extension_to_replace_this_template|]] for citation templates. Martin (Smith609 – Talk) 21:08, 22 April 2011 (UTC)

Conventions: special chars

I just noticed that Template:Automatic_taxobox/Conventions suggests the format Template:Taxonomy/Citrus_×_floridana for hybrids. Wouldn't it be better to suggest "x" instead of "×", given that it appears on keyboards and cannot be mistaken for &cross;? Martin (Smith609 – Talk) 03:17, 18 March 2011 (UTC)

I think it's better to remain consistent with article titles (WP:FLORA). The ICBN suggests that if we are able to use ×, we should use it preferentially over x. Rkitko ^(talk) 15:40, 18 March 2011 (UTC)

I agree with the above. We should try to follow the ICBN where possible. Peter coxhead (talk) 18:21, 18 March 2011 (UTC)

If it's difficult to type, why not create a subst'able template similar to {{extinct}}? Bob the WikipediaN ^{(talk • contribs)} 04:15, 19 March 2011 (UTC)

Yeah, I mentioned {{Hybrid}} when I wrote the convention. I have no super-strong opinion about this convention, by the way, I was just trying to provide examples so things would be consistent. I think having hybrids show up properly was a complaint someone had over at template talk:Taxobox or someplace, so I tried to figure out how to make this easy/correct. But IANAB, as they say, so I probably got a lot of stuff wrong. ErikHaugen (talk | contribs) 23:18, 22 April 2011 (UTC)

Link to preload will only take you to genus page

Noticed this twice when creating autotaxoboxes for species. Under the "Automatic taxobox help" is a link that will take you to where you edit the templates. It works fine in higher taxa, but in species, the link only leads to the template page for the genus.

For example creating a taxonomy for Nauclea orientalis will show me this message in the help box: "Click orientalis&preload=Template:Taxonomy/preload here to enter the taxonomic details for "Nauclea orientalis"."

Yet clicking said link instead takes me to http://en.wikipedia.org/w/index.php?action=edit&title=Template:Taxonomy/Nauclea

I manually appended the species name at the end of the url for it to work, but might confuse beginners. e.g.: http://en.wikipedia.org/w/index.php?action=edit&title=Template:Taxonomy/Nauclea_orientalis Or is this intended behavior?--Obsidi♠n^Soul 12:56, 2 May 2011 (UTC)

I'm pretty sure that's how the {{speciesbox}} is intended to work, but Martin designed it, so he should know better than me. Bob the WikipediaN ^{(talk • contribs)} 15:19, 2 May 2011 (UTC)

Ah, I was using {{Automatic taxobox}}. Gotcha, thanks.--Obsidi♠n^Soul 00:02, 3 May 2011 (UTC)

Deprecated taxa

Okay, we need a new system, because I'm getting tired of providing rationale every time someone nominates three of these at once (or in this case, six) for deletion.

The purpose of keeping these templates is simple--

When a user enters a taxon that is outdated or otherwise not used in the automatic taxobox database, the taxobox displays a large error that shows a suggestion and also provides a link for the user to check to see if there are any similar pages referencing that taxon. In many cases, two or three taxa are offered as suggestions, depending on the nature of that taxon's disuse.

Unfortunately, they keep getting nominated for deletion because they are not transcluded anywhere (by their very nature).

Anyone have a better idea on how to handle deprecated taxon templates? These duplicate TfD's are getting slightly annoying. Bob the WikipediaN ^{(talk • contribs)} 15:27, 2 May 2011 (UTC)

Bob, sorry, I missed this post and responded on your talk page with much the same point. It's not only the deprecated taxon templates which could be an issue. I've created some taxon templates in order to set up a hierarchy before I got round to creating articles which used them. These could easily have got nominated for deletion. We need some BIG notice: don't mess with these templates! Peter coxhead (talk) 15:54, 2 May 2011 (UTC)

These are two different (potential) problems. Templates that are created shortly prior to deployment shouldn't be a big issue. If, in practice, attempts are made to delete them, then simply adopt a work flow where the templates are created either just before or shortly after the article that transcludes them is created. The bigger problem is with deprecated taxa, and here, I rather take the point of the proponents (as described above – I haven't seen the actual discussions). This project is not primarily a taxonomic database; our purpose is to produce an encyclopaedia. A template which exists only to tell people not to create that template will never be a useful part of the front-end encyclopaedia. Could it not be replaced with a project page which simply lists the templates that should never exist, perhaps with a short explanation of what taxa should be used instead? Any blue links appearing would then indicate cleanup work to be done. Admittedly it would be less obvious to the creator what they had done wrong, but it could be explained in a short edit summary linking to the page of bad taxa. How big a problem is the creation of inappropriate automatic taxobox taxa templates anyway? Do we need to take preventative measures or could we just deal with problems as they arise? --Stemonitis (talk) 16:07, 2 May 2011 (UTC)

What is the function of most templates? To make editing easier. In this sense, surely templates are never "part of the front-end encyclopaedia"; they are tools for editors. A template which only 'comes to life' when an editor creates a deprecated taxon is a useful 'help tool' and its cost in terms of storage and resources is utterly trivial.

"This project is not primarily a taxonomic database; our purpose is to produce an encyclopaedia." Absolutely! So editors need tools which reduce the time they have to spend on creating the taxonomic connections which are an intrinsic part of taxoboxes, so they can spend time on creating article content. The automated taxoboxes are not a total solution, and they have their limitations, but in areas where there is a clear taxonomy (e.g. APG3 for angiosperms), in my experience they really do help. Peter coxhead (talk) 18:11, 2 May 2011 (UTC)

I wasn't questioning the utility of the overall automated taxobox system, though. I was questioning the utility of taxon templates for taxa we will never use. You did not address my question: how many times has someone tried to use a deprecated taxon in an automated taxobox, only to receive the warning message? We may not be able to find that out, but I'd be interested in getting some idea of the scale of the problem. Has anyone reading this had that experience? What would have happened without the warning message? Could a template-free solution have achieved the same? Perhaps a more pertinent question: how much time and effort is likely to be wasted on repeated TfDs which could have been better spent? These templates do appear to fulfil the third criterion for template deletion, in that they will never be transcluded or substituted. While I recognise the frustration of going through the same arguments repeatedly, and I fully recognise that the templates were created entirely in good faith, I suspect that there are better – and more co-operative – solutions than claiming special status (e.g. "some big notice: don't mess with these templates!"). --Stemonitis (talk) 18:37, 2 May 2011 (UTC)

As for their usefulness, I myself have run into them while setting up taxonomies on multiple occasions (which is actually the reason I started "deprecating" instead of deleting them), though I can't speak for anyone else. Stem...your first post has a very intriguing solution-- Redirect these deprecated templates to a master list of deprecated templates which can then explain what action to take for each one (which probably makes more sense anyway). For reference, you'll find discussions at the following links:

Closed:

• Wikipedia:Templates for discussion/Log/2011 January 6#Template:Taxonomy/Lipotyphla (three discussions ultimately were condensed into one)

In progress:

• Wikipedia:Templates for discussion/Log/2011 May 2#Template:Taxonomy/Trimerophytina
• Wikipedia:Templates for discussion/Log/2011 May 2#Template:Taxonomy/Heterokonta
• Wikipedia:Templates for discussion/Log/2011 May 2#Template:Taxonomy/Albuginales
• Wikipedia:Templates for discussion/Log/2011 May 2#Template:Taxonomy/Dictyoptera (2nd nomination)
• Wikipedia:Templates for discussion/Log/2011 May 2#Template:Taxonomy/Insectivora (2nd nomination)
• Wikipedia:Templates for discussion/Log/2011 May 2#Template:Taxonomy/Lipotyphla (2nd nomination)

Bob the WikipediaN ^{(talk • contribs)} 22:01, 2 May 2011 (UTC)

I just observed that if you try to create a taxon template atop a deleted one, you see the pink warning (see Template:Taxonomy/×Amarcrinum). The warning won't display in the page you're working on, but on the "create" page you get after clicking "fix". Provided the warning contains a link to the template that should be used, I think this will suffice. If there's no objection to this protocol, I'll begin systematically deleting the deprecated taxa and leaving links in the pink warning box to the templates to use instead. Bob the WikipediaN ^{(talk • contribs)} 00:04, 15 May 2011 (UTC)

Hybrid genera

I was updating the article × Amarcrinum to APG III, as agreed, via an automatic taxobox and ran into problems with italicization. The ICBN allows a space between the × and the genus name, but says that the × should not be italicized.

• {{Italictitle}} italicizes the ×, which it shouldn't. However, the page title can be fixed via {{DISPLAYTITLE:× ''Amarcrinum''}}.
• The title of the taxobox is also incorrectly italicized if done automatically. However, this can be fixed by specifying |name=× ''Amarcrinum'' as well as |taxon=× Amarcrinum.
• The taxobox displays the genus as × Amarcrinum, when it should be × Amarcrinum. There seems to be no workaround for this.

All three incorrect default italicizations need fixing some time. Peter coxhead (talk) 18:39, 14 May 2011 (UTC)

Use {{hybrid}} instead of × for the display name. Bob the WikipediaN ^{(talk • contribs)} 00:07, 15 May 2011 (UTC)

Ok, this is a workaround for the third issue above. However, in Template:Taxonomy/×_Amarcrinum it causes the genus in the "Ancestal taxa" table to show up as not italicized. This matters much less than in the substantive article, but still suggests that these are work-arounds, not real fixes. I think the real fix is to ensure that any automated italicization ignores the × character. Peter coxhead (talk) 06:54, 15 May 2011 (UTC)

More Specific Clade Labels

I was just thinking, rather than having a long list of (unranked) or clade labels stuck in there among the Kingdoms and Families it would be nice to have more specific labels that designate what type of clade you're dealing with. For example, Node:, Branch:, Stem:, Crown:. Would this be doable? MMartyniuk (talk) 15:05, 18 May 2011 (UTC)

It would in principle be doable, but what exactly would be the advantage? (Btw, I'm not sure what a "stem clade" would be; a "stem group" is by definition paraphyletic and so not a clade.) Peter coxhead (talk) 20:19, 18 May 2011 (UTC)

An example would be Avemetatarsalia and Ornithodira. As it stands these clades contain the same content but have different concepts/definitions so they're not really synonyms. It would be nice to have Branch: Avemetatarsalia Node: Ornithodira to make the distinction clear. I think paraphyletic stem groups are technically allowed but discouraged under the current draft PhyloCode but that could be left out for now. Also this is a bit aesthetic as it's monotonous to have taxoboxes with long lists of clade: clade: clade: (any reason for the lowercase and italics there, btw?). Oh, and Apomorphy would be another one (Apo.:)? MMartyniuk (talk) 14:55, 24 May 2011 (UTC)

There's also "Plesion" which some sources use...

Lower-case and italics to distinguish the label clade from a Linnean rank, I think.

My serious answer to your basic question is that there's nothing to stop adherents to the PhyloCode writing their own template "Automatic PhyloCodeBox" with whatever labels they like. (Non-serious point: then we can have some nice edit wars swapping between "Automatic Taxobox" and "Automatic PhyloCodeBox".) Serious point: I don't think it's sensible to complicate a template which is already difficult to maintain/debug with labels from both Linnean and PhyloCode taxonomy. Peter coxhead (talk) 18:04, 24 May 2011 (UTC)

For me, it would be useful to have the existing taxobox function for both systems, by simply adding extra labels. This way we can maintain a coherent navigation/link system between articles in fields which use different systems. You seem to be coming from a view that the taxobox is inherently Linnean, but there are already numerous clades and unranked taxa in the auto taxobox system, and so far this hasn't caused any problems. Rebranding the Taxobox to a Linneaobox would require deletions of whole swaths of data. MMartyniuk (talk) 20:13, 24 May 2011 (UTC)

I was being more than a little mischievous above; time to come clean! :-) There is nothing whatsoever to prevent any editor using any labels for any taxon in the automatic taxobox system. One key difference between {{Taxobox}} and {{Automatic taxobox}} is how the hierarchy of parental taxa is created.

• The {{Taxobox}} system 'knows' the possible 'rank labels' and then puts them in the right order in the displayed taxobox (which caused real problems when this ordering was wrong, leading to the need for a bot to fix thousands of pages when the template was corrected). So you can't make up your own rank labels if you use {{Taxobox}}. The template would need to be changed and I am very much against this.
• The {{Automatic taxobox}} system doesn't work like this. It 'knows' nothing about the ordering of rank labels. You can create a displayed taxobox which shows (upwards) 'Phylum:' preceded by 'Kingdom:' preceded by 'Genus:' (although hopefully no-one ever would). This is because the ordering of parental taxa is obtained by following the "Template:Taxonomy/..." templates. So if you put on a page {{Automatic taxobox|taxon = Junkus}}, the system looks for "Template:Taxonomy/Junkus". If this exists, it will contain rank = ..., which will be displayed as the rank label. (The system capitalizes such labels, and italicizes "Clade".) You can put any rank label you like here and it will be displayed (capitalized) in the displayed taxobox. "Template:Taxonomy/Junkus" will also contain parent = ..., say parent = Junkaceae. So the system then looks for "Template:Taxonomy/Junkaceae" and so on. The ordering is determined by the |parent= values in the successive "Template:Taxonomy/..." pages regardless of what rank these are said to be.

So if you really want different rank labels, you can create them. Whether other editors will leave them is another matter – unless a consensus could be reached at WT:TOL or the appropriate subproject. Peter coxhead (talk) 10:43, 25 May 2011 (UTC)

Binomial name in Species Box

In the speciesbox template, as well as in others I think, the species name displays twice: Species : H. sapiens Binomial name: Homo sapiens. Could this be combined, replacing the binomial name heading with Species? Seems a little redundant as it is. I know I recently asked this elsewhere, somewhere, but I can't seem to find that thread :X MMartyniuk (talk) 14:48, 30 May 2011 (UTC)

See Template_talk:Speciesbox#Species_vs._Binomial, where I gave my explanation. Peter coxhead (talk) 17:19, 30 May 2011 (UTC)

{{automatic cladogram}}

Speaking of phylogeny, I'm hoping to start a MediaWiki template project soon as part of the Global University Program with Wikimedia. Does an {{automatic cladogram}} sound like a good programming project for a computer science senior project class? Bob the WikipediaN ^{(talk • contribs)} 19:42, 24 May 2011 (UTC)

That sounds like a good idea in theory, but the fact is that cladograms differ from study to study so much it may be impractical to maintain, and/or we'd need multiple independent cladoboxes for each article. MMartyniuk (talk) 20:13, 24 May 2011 (UTC)

Perhaps it would be a good idea for us to iron out the theory before proposing this project. Bob the WikipediaN ^{(talk • contribs)} 20:38, 24 May 2011 (UTC)

I think you need to explain a little more about how you think this would work. If you mean a system like {{automatic taxobox}} where a cladogram is generated by following some thing like a "Template:Clade/..." hierarchy, then I don't think this is the right way, because, as MMartyniuk says, the cladograms used from study to study differ so much, that you would actually need "Template:Clade/.../study1", "Template:Clade/.../study2", etc., with, I suspect, little sharing.

On the other hand there is a need to make cladograms easier to produce and easier to share parts of. I managed to subdivide the cladogram in Polysporangiophyte by composing it of templated subcladograms, but it's not a simple process. Ideally in the 'middle' of a cladogram you want to be able to say 'now put in the cladogram from ...'. Exactly how to achieve this I'm not sure; except that, as noted above, I don't think that storing every single child-parent connection in a separate template as per the automatic taxobox system is quite the right approach.

But the idea is worth pursuing. Peter coxhead (talk) 10:52, 25 May 2011 (UTC)

Here's what I have in mind. Let's say I want a cladogram ranging from brown recluses at one end to pigeons on the other end, favoring studies done by Jack Horner and Paul Sereno (in that order). So I would enter something like this:

{{automatic cladogram|Loxosceles reclusa|Columba livia|favor=horner|favor2=sereno|avoid=mendeleev}}

In my conceptual example, the already-existing taxon templates would have added to them parameters that might look similar to this:

|clad_parent1=Vertebrata
|clad_authority1=linnaeus
|clad_parent2=Theropoda
|clad_authority2=horner

Where a preference is expressed, the code would pull the first-indicated favorite (in this case, "horner" would be favored over "sereno", and both would be favored over all other authorities, and "mendeleev" would be avoided wherever possible). The cladogram would be constructed in a similar manner to the automatic taxobox, tracing both taxa back to their MRCA and cutting out irrelevant details as appropriate. Ideally, it should also support a third node. Bob the WikipediaN ^{(talk • contribs)} 17:30, 25 May 2011 (UTC)

Won't that require a lot of extra data to be added to all existing taxons? It's not that I think it is a bad idea per se, but what would be really neat is a WYSIWYG cladogram editor for Wikipedia. The current system is a bugger to edit. Petter Bøckman (talk) 17:58, 25 May 2011 (UTC)

It'd be one of those things where if you found a cladogram and wanted to enter the data, it'd mean editing the templates for all taxa involved, which would come nowhere close to "all existing taxa". Of course, cladograms are not nearly as common as taxoboxes, so these would likely be more shallow in most areas, only heavily concentrated in the sauropsidans. And I agree...the {{cladogram}} is rather discouraging to try to produce. Actually, I think an external Java applet to produce code for a cladogram may be pretty simple to do (though I have no server and wouldn't be able to host the applet if I were to develop it). Bob the WikipediaN ^{(talk • contribs)} 19:44, 25 May 2011 (UTC)

(Responding to Petter)

Yes, and yes!

On the first issue (extra data, how such a system would work):

• Another problem is that cladograms contain anonymous (i.e. unnamed) nodes, often many, which can't be dealt with in the way Bob suggested. The cladogram at Polysporangiophyte#Phylogeny, for example, has 12 internal nodes of which only 4 are labelled and so would potentially have a Taxonomy template.
• Another is that not all children of a parent are needed in a particular cladogram, so not only will entries be needed for alternative parents of a clade/node, there will have to be some way of selecting which children will be shown. This looks difficult to me because cladograms will be defined by the root and then by working downwards, whereas the taxonomic hierarchy in taxoboxes is defined by the terminal taxon and then by working upwards.

On the second issue (a cladogram editor), this seems to be a much more useful tool. I've used several different ways of generating cladograms in the required syntax, none brilliant, and all requiring tricky editing if an extra node is needed. Peter coxhead (talk) 19:51, 25 May 2011 (UTC)

Alright then, sounds like an external Java applet for producing MediaWiki code is in order. My power's out right now, and I've got a few other projects needing attention, but I'll put that on my back burner (or offer this as a potential class project for the upcoming Wikimedia Global University semester).

What would be the most intuitive way to construct a cladogram onscreen-- by typing it? by drawing it? by using buttons? Be creative... Bob the WikipediaN ^{(talk • contribs)} 17:41, 27 May 2011 (UTC)

Some graphical way. Perhaps something that starts with a single bifurcated branch, and make it so that when you right-click on a branch or node, you can add a branch to it or add a label. Ucucha 17:47, 27 May 2011 (UTC)

I think a good start would be a "start" button, which will give you a question of where to anchor your cladogram. If you fill inn e.g. Dinosauria, the basal branch will be named as such, and end with a simple bifurcation as Ucucha suggested. If you click/mark the split, you will get the alternatives of a two-way or "enter number of daughter clades" alternative. Typing in a e.g. "3" gives you a three way split and so on. For each terminal you get the choice of a taxon or a furcation (bifurcation or higher) in the fom of boxes to click on/fill in. For taxons, it should be an option to add pictures, see Labyrinthodontia#Phylogenetic_overview for an example). It should be possible to mark up the lines as full or stippled (for uncertain phylogenies, currently this is only possible for the horisontal lines I think) and choice of line colour and/or background colour (see Osteolepiformes for an example where this would be usefull).

Plese try to make it useable for Opera users too! Petter Bøckman (talk) 18:05, 27 May 2011 (UTC)

Sounds doable. I think we'll actually be able to pull it off easiest if we use the layout Wikipedia uses - that is, horizontal, instead of triangular. That way, we can represent lines using the vast assortment of piping symbols. Also-- Java is not platform-dependent, so hopefully we shouldn't have issues. Bob the WikipediaN ^{(talk • contribs)} 19:22, 27 May 2011 (UTC)

Something which is impossible (as far as I can tell) with the current method of using table borders to draw the cladogram lines is to bracket and label the leaves, which is a very common requirement, seen all over scientific articles on phylogeny. The diagram at Lilioid monocot is a good example: I was forced to draw it rather than use {{clade}}. So you could add this to your requirements list! Peter coxhead (talk) 19:42, 27 May 2011 (UTC)

That could probably be accomplished with piping symbols (obviously it would require a feature to be added to {{cladogram}}, as the Java applet would actually output the {{cladogram}} code)-- there are piping symbols, I believe, that cover those areas. Bob the WikipediaN ^{(talk • contribs)} 21:01, 27 May 2011 (UTC)

I tried to understand how {{clade}} actually draws the lines (by making a copy of the code and then forcing all the table borders to show). The problem is that it's produced by tables within tables and I couldn't find anything that lines up in the right place across all the tables to insert a vertical line at the right, whether generated by cell borders or symbols. The template is a really clever use of tables (the creator, User:Josh_Grosse, doesn't seem to be around Wikipedia now), but I suspect not very flexible when you want to do something a bit different. So my 'requirement' may not be possible given the way that {{clade}} works. Peter coxhead (talk) 21:42, 27 May 2011 (UTC)

When I went back to the work I did on this, I realized what I wrote above (now crossed out) was wrong. It's easy to change {{Clade}} to produce the right-hand side vertical lines (see User:Peter coxhead/Test/Clade which uses User:Peter coxhead/Template/Clade – ignore the colours). The problem is to put text against these lines. I haven't yet found out how to do this, if indeed it's possible. Peter coxhead (talk) 21:34, 30 May 2011 (UTC)

I'm not sure how much scripting MediaWiki will render properly, but you might have a look at this: [1] Bob the WikipediaN ^{(talk • contribs)} 21:51, 30 May 2011 (UTC)

There's a better version here; the one you gave a link to only works in IE. This:

<p style="writing-mode:tb-rl; -webkit-transform:rotate(90deg); -moz-transform:rotate(90deg); -o-transform: rotate(90deg);
white-space:nowrap; display:block; bottom:0; width:20px; height:20px;">Here is some vertical text</p>

works in Wikimedia pages in Safari and Firefox under Mac OS; I can't test Windows browsers.

I was pretty sure that vertical text wouldn't be a problem (in the last resort, one could just display a 'key number'). The real problem is that each invocation of {{Clade}} generates a new table, placed inside a cell of the last invocation. Provided there's no table margin or padding, the right borders all line up, so you can 'draw' a line by showing the right border of the relevant cells (which is what my version does via |r1=, |r2=, etc.). To display text, it seems to me that I need to add a cell beyond the border just drawn, with a rowspan of twice the number of leaves which have the bar added (since a leaf is put in a cell with rowspan=2). Although this is possible (in theory, I haven't tried it), the problem is that the user will have to specify this rowspan, since a particular invocation of {{Clade}} can't tell how many cells are going to be marked with a right border in the next invocation or the one after that. This seemed so yucky, I gave up. It's possible that you don't need a cell, but can just place the text with CSS, but this approach has caused me such trouble in the past with different browsers that I'm reluctant to try it. (I used to teach this stuff for my sins, and had great fun marking students' work, required to be browser-independent but almost always done with IE under Windows, by testing it with Firefox on a Mac. A bit cruel I guess...)

I think that any more sophisticated tree drawing algorithm needs two passes over the cladogram: one to collect information needed in the next pass which 'draws'. This is a bit tricky with the template language! Peter coxhead (talk) 17:03, 31 May 2011 (UTC)

Even if we have to do with a vertical line with either a colour or a number for legend, it will be leaps and bounds over what we have today! Do keep up the good work, Gents! Petter Bøckman (talk) 18:26, 31 May 2011 (UTC)

Great! Could you link to a complex example coded with the template using this newfound vertical text? I'd like to begin working on a MediaWiki-to-Java parser, as the applet would ideally be able to import an existing cladogram. Bob the WikipediaN ^{(talk • contribs)} 18:36, 31 May 2011 (UTC)

Let's continue this discussion at the more relevant {{clade}}. Bob the WikipediaN ^{(talk • contribs)} 20:19, 31 May 2011 (UTC)

Edit log

• 01:58, 2 June 2011 (UTC) - Replaced "[e]" in {{Edit taxonomy}} with pencil icon to eliminate text sizing issues and ugliness. Bob the WikipediaN ^{(talk • contribs)}
• 15:55, 2 June 2011 (UTC) - Replaced Creative Commons file with public domain file as attribution could not be achieved efficiently. Bob the WikipediaN ^{(talk • contribs)}
• 17:19, 2 June 2011 (UTC) - Fixed tooltip to say "Edit this classification" instead of "e". Bob the WikipediaN ^{(talk • contribs)}

Image/text overlap

Cluebot's set to archive anything with {{done}} in it on this page. Due to the nature of the topic, however, an archival wasn't needed...

Stemonitis had proposed that we could eliminate text/image overlap by using multiple cells to represent the "Scientific name" and edit button, as well as an empty third cell to the left the same size as the edit button's cell, and then span all the other cells across three columns. I think this is a great idea. Bob the WikipediaN ^{(talk • contribs)} 01:31, 3 June 2011 (UTC)

Problem

Something's wonky with the taxobox at Confuciusornis, but I can't figure out what... Seems to be an issue with everything south of Aves. MMartyniuk (talk) 22:17, 5 June 2011 (UTC)

The problem is not limited to the birds. The ammonite group Tetragonitidae is also wonky as is the plant genus Leiosporoceros. This is a problem in the core template coding, something it calls, or the way the coding is implemented. --EncycloPetey (talk) 22:24, 5 June 2011 (UTC)

This doesn't look right.. mgiganteus1 (talk) 23:06, 5 June 2011 (UTC)

Related to this edit? The template is protected so an admin would need to revert. mgiganteus1 (talk) 23:10, 5 June 2011 (UTC)

Looks like an admin has done so and the templates are now behaving properly, it seems. --EncycloPetey (talk) 23:38, 5 June 2011 (UTC)

Sorry for the problem. I could swear I was editing Template:Taxonomy/Antilopinae, but something odd must have gone wrong. Ucucha 06:55, 6 June 2011 (UTC)

{{resolved}}

Import/export of all Taxonomy/ templates

Hello! I've been messing with this template and trying to localize it to the basque wikipedia. The result now is good but, of course, it needs tons of work building up the database. Do someone know a way to take all the Template:Taxonomy/ templates and make an import/export to another Wikipedia? This would be great, so we can have a centralized data system for all Wikipedias (as commons is) and make little Wikipedias stronger, as all the information mustn't be typed twice. -Theklan (talk) 00:08, 6 June 2011 (UTC)

A bot is probably your best bet. I wish I knew a good way to transclude a template across wikis-- I'd suggest that if I knew the answer. Please note when adding interwiki links to the taxonomy templates that these need to be done carefully-- there should be no line breaks between the code and the interwiki links. Bob the WikipediaN ^{(talk • contribs)} 02:17, 6 June 2011 (UTC)

Interwiki template transclusion is a project that the MediaWiki developers are working on; see bugzilla:9890. We might get it sometime in the more or less near future. Ucucha 07:39, 6 June 2011 (UTC)

Yes, but it would be very useful before this interwiki transclusion is available to have a good import/export system. I own a bot, but I'm not very good in python programming. If someone knows how to, it would be great! -Theklan (talk) 12:38, 6 June 2011 (UTC)

Thanks, Ucucha...I voted for it, although it appears the fix is very close to being released. Bob the WikipediaN ^{(talk • contribs)} 16:31, 6 June 2011 (UTC)

Well, this is the code for a pywikipedia bot developed by Gomà from the catalan Wikipedia to use in the Basque Wikipedia:

# -*- coding:utf-8 -*-


import pagegenerators, wikipedia

lloc = wikipedia.getSite("en","wikipedia")
x = pagegenerators.AllpagesPageGenerator("Taxonomy/", 10, True,lloc )
for plantilla in x :
	print plantilla
	titol = plantilla.title()
	if titol[9:18] != "Taxonomy/" : break
	titol = 'Txantiloi:'+ titol[9:] # per treure Template: i posar Plantilla: (canvia-ho al nom en euskara)
	try :
		text = plantilla.get()
	except :
		continue #Para pasar de los redirects.print text
	pagina = wikipedia.Page( wikipedia.getSite('eu',"wikipedia"), titol)
	pagina.put(text,"Taxonomia txantiloiak esportatzen")

If someone wants to use it it's easy and very very fast. It can be used on a regular basis as creates new Templates when they're found, without changing the other templates. -Theklan (talk) 14:30, 8 June 2011 (UTC)

Reptiliomorpha won't show

I've been trying to tidy up the taxonomy at the lower end of tetrapods, and have discovered that all genera in Diadectomorpha fail to show Reptiliuomorpha in the automated taxoboxes. I have tried putting Reptiliomorpha as an order (I'd prefer it as "unranked" for the time being) to force it to show, but to no awail. The only way it seems to display is if I force Reptiliomorpha to allways display, but then it shows up in all Amniote articles too, despite me having put in a skip template to take Amniota straight to Tetrapoda. Help, please? Petter Bøckman (talk) 17:54, 16 June 2011 (UTC)

Does "|display parents=4" (in the taxoboxes of these genera) achieve the desired effect? Martin (Smith609 – Talk) 18:34, 16 June 2011 (UTC)

You may have to introduce a shim taxonomy template to accomplish this. See Template:Taxonomy/Incertae sedis/Physeteroidea – its parent is Template:Taxonomy/Physeteroidea/displayed, which is set to always display, because we want everything in the incertae sedis group to show Physeteroidea in the taxobox. But we don't want sperm whale's taxobox to include Physeteroidea, so Template:Taxonomy/Physeteridae's parent is Template:Taxonomy/Physeteroidea. Note that Template:Taxonomy/Physeteroidea and Template:Taxonomy/Physeteroidea/displayed are templates representing the same taxon, just one has the always display bit on. Petter, what amniote articles show Reptiliomorpha when the always display bit is on? I don't see how this could be true for all of them? Thanks, ErikHaugen (talk | contribs) 19:09, 16 June 2011 (UTC)

I see some examples of where it would show up, nevermind. It does seem to be showing up on Tseajaia; isn't that what you want? The problem now – and it is serious – is that it is showing up on Turtle, etc. I think the solution here is shim templates like I suggested. Reptiliamorpha should not be an ordo (unless it is??) and template:Taxonomy/Diadectomorpha should be parented to a new Template:Taxonomy/Reptiliomorpha/displayed, I think. Sound good? ErikHaugen (talk | contribs) 19:18, 16 June 2011 (UTC)

It seems to be in order now for the Diadectomoprh genera. It might have been a cash-related problem at my end on my small portable machine. I have reset Reptiliomorpha to "Unranked" again to keep it out of Turtle, but depending on what taxonomy I want to follow, it could well be an order (labyrinthodont taxonomy is an absolute mess, prferred taxonomy is really a matter of taste). I'm afraid I haven't heard of a "shim" template before, it that what Bob calls a "skip template"? I'm tempted to anchor Amniota directly in Tetrapoda in stead of routing it through Reptiliomorpha for phylogenetic correctness sake. That would make me free to fix things at my end without upsetting the various amniote articles. Petter Bøckman (talk) 09:40, 17 June 2011 (UTC)

I think it's implemented the same way as a skip template, but it is not for skipping anything. I guess there is no official name. That's why I included a detailed example of exactly what to do. Maybe we can call them same-as templates. It would be nice to have a reptilomorpha in there so that the taxobox looks right on its children; see Amniote. Also it's nice to have the "correct" (I realize our understanding of it is always changing) tree there so that people can build tools to see or browse the tree. WRT making Reptiliomorpha an order, the problem is that then two orders will show up on a bunch of taxoboxes. If the automatic taxobox system did not automatically display duplicate major ranks, it would solve this problem, which shows up a lot - just think about Saurischia! Again, though, the solution is a same-as template. Either to skip the offending major ranks, which is probably the right way to go; or an unranked version of the rank that the other taxa can use. ErikHaugen (talk | contribs) 16:59, 17 June 2011 (UTC)

Currently Amniota is parented to a /skip template; would you say that everything with the non-skip Template:Taxonomy/Reptiliomorpha in its ancestor list ought to display Reptiliomorpha in its taxobox? If so why not just set Template:Taxonomy/Reptiliomorpha to always display and Template:Taxonomy/Reptiliomorpha/skip to not always display? Does that work? If we want reptiliomorpha to be an order, then maybe we can make Template:Taxonomy/Reptiliomorpha/skip be unranked? (Perhaps abandonding same-as at this point?) ErikHaugen (talk | contribs) 19:48, 17 June 2011 (UTC)

I went ahead and did this: always_display=true in the base template and =false in the override. It seems to be working; let me know and revert if there are any problems! We could do something similar if you want to make this an order? ErikHaugen (talk | contribs) 20:20, 17 June 2011 (UTC)

And I got the comments backwards :) ErikHaugen (talk | contribs) 20:22, 17 June 2011 (UTC)

Sorry about late reply, had to put the kids to bed. Yes, the solution you suggests seem the best, and now it looks like it is working across the board, thanks! I had hoped to not split the automated phylogeny at any point, but when we're knee deep in Paleozoic amphibians it seems we don't really have any other choices if we want an even marginally coherent taxonomy. One thing: It seems I have done something wrong with either the template or the taxobox for Reptiliomorpha. Could you have aquick look? Petter Bøckman (talk) 20:42, 17 June 2011 (UTC)

I think this fixes it. Don't think of it as splitting; the "same-as" templates are all the same taxon so you can think of it as just an "ugly" syntax for getting it to show up right. ErikHaugen (talk | contribs) 22:12, 17 June 2011 (UTC)

Thanks for helping my out, I really appreciate it! Petter Bøckman (talk) 06:55, 18 June 2011 (UTC)

You bet; thanks for working on these!! ErikHaugen (talk | contribs) 15:22, 18 June 2011 (UTC)

Docodonta

I have had to revert Docodonta to using the manual taxobox, because the automatic taxonomy is broken. It showed two different orders, one supposedly a subtaxon of the other (Order: Therapsida; Superfamily: Chiniquodontoidea; (unranked): Mammaliaformes; Order: †Docodonta). I suspect that someone has been mixing taxonomies that place the same taxa at different ranks. I expect this error is repeated across a number of articles, and it must be fixed immediately. I don't understand enough about fossil and extant vertebrate taxonomy to know what solution is appropriate, whether both can be accommodated in a single taxonomic hierarchy, or whether we need separate templates for extant and extinct taxa; I hope someone reading this will. --Stemonitis (talk) 06:52, 21 June 2011 (UTC)

The problem here is the mic of phylogenetic and Linnaean taxinomy. Both Docodonts and Therapsids are are well established and well published orders. You could solve this by making a skip template to rout Docodonta around Therapsida an anchor it directly in Synapsida. If you really want Therapsida in there, you could make a "shim" template with an unranked Therapsida. Petter Bøckman (talk) 07:06, 21 June 2011 (UTC)

Docodontidae has two orders, also. What should that taxobox look like? I tried parenting Template:Taxonomy/Docodonta to Template:Taxonomy/Mammaliaformes/Amniota, but then Synapsida, etc won't show up. We may need a few skip/same-as templates if we want more of these to show up on those taxoboxes. If we can decide here what they should all look like I'll make the extra templates. It would sure be nice if the automatic taxobox system just didn't show redundant 1st rank taxa! That would solve a lot of problems. ErikHaugen (talk | contribs) 17:04, 21 June 2011 (UTC)

Maybe we just need a skip template Template:Taxonomy/Mammaliaformes/Synapsida, so the "class Synapsida" shows up on the taxoboxes. Does that seem like it would work? ErikHaugen (talk | contribs) 17:06, 21 June 2011 (UTC)

The problem here is the use of "order". In principle we can choose Benton, classical and phylogenetic, but if we introduce ranks we are stuck with the two former. I'd personally go fully classical (order Docodonta -> class Mammalia -> Amniota) but I'm probably in the minority here. Petter Bøckman (talk) 18:09, 21 June 2011 (UTC)

Is someone going to make the change soon? For my part, I don't care which system is used for these animals, but it evidently needs to be just one. --Stemonitis (talk) 08:49, 26 June 2011 (UTC)

Ghost transclusions

This is surely an issue with the code...if someone gets a chance, please have a look at the code and see if you can figure out why Florideae, Coelomys, and Toothed whale all transclude deleted taxonomy templates with the same name. This is a highly unanticipated predicament, and it's showing up in some database reports. I learned of this from Plastikspork this evening. Bob the WikipediaN ^{(talk • contribs)} 05:25, 25 June 2011 (UTC)

I think that may be because of the "|{{#ifeq:{{Taxonomy/{{PAGENAME}} }}|[[:Template:Taxonomy/{{PAGENAME}}]]" line in Template:Taxobox/taxon; MediaWiki counts that as a transclusion I believe. I don't think this should be considered a problem. Ucucha 12:35, 25 June 2011 (UTC)

Can this sort of thing be flagged? Bob the WikipediaN ^{(talk • contribs)} 18:22, 25 June 2011 (UTC)

In what way? Incidentally, the examples Plastikspork gave are not the only ones: Template:Taxonomy/Amphibian is also "transcluded" on Amphibian. Ucucha 18:37, 25 June 2011 (UTC)

Flagged so that they don't show up in the bad tranclusions backlog that is. Bob the WikipediaN ^{(talk • contribs)} 19:51, 25 June 2011 (UTC)

I can't think of a way to do that. Ucucha 20:10, 25 June 2011 (UTC)

If you replace {{Taxonomy/{{PAGENAME}} }} with {{#ifexist:Template:Taxonomy/{{PAGENAME}}|{{Taxonomy/{{PAGENAME}} }} }}, then the problem should go away. Would this work? This is much easier than creating an exceptions list. However, if this shouldn't happen at all, then we would want to track these. Thanks! Plastikspork _―Œ^(talk) 20:20, 25 June 2011 (UTC)

Checking the template code, it appears as though there was some thought that "ifexist" might be too expensive. I can't see why this would be the case, and would be a bug in the backend software if a different parserfunction hack would be cheaper than using the specialized one. I put what should be an equivalent version in Template:Taxobox/taxon/sandbox if someone wants to check it out. Thanks! Plastikspork _―Œ^(talk) 20:29, 25 June 2011 (UTC)

Looking at the parser reports for User:Ucucha/sandbox (transcluding Taxobox/taxon) and User:Ucucha/sandbox2 (transcluding Taxobox/taxon/sandbox), it seems that some measures go down in your version and others go up, so it's not clear which method is better performance-wise. Ucucha 20:44, 25 June 2011 (UTC)

Yes, they look very similar. The only clear difference is that my version has "1/500" expensive parsefunctions. I don't see this as having a large impact on performance, will keep these pages from having an inviting "redlink" at the bottom of the article in edit mode, and will keep these out of the database report. So, in my opinion, it's worth trying. Thanks! Plastikspork _―Œ^(talk) 21:08, 25 June 2011 (UTC)

Docodonta

I have had to revert Docodonta to using the manual taxobox, because the automatic taxonomy is broken. It showed two different orders, one supposedly a subtaxon of the other (Order: Therapsida; Superfamily: Chiniquodontoidea; (unranked): Mammaliaformes; Order: †Docodonta). I suspect that someone has been mixing taxonomies that place the same taxa at different ranks. I expect this error is repeated across a number of articles, and it must be fixed immediately. I don't understand enough about fossil and extant vertebrate taxonomy to know what solution is appropriate, whether both can be accommodated in a single taxonomic hierarchy, or whether we need separate templates for extant and extinct taxa; I hope someone reading this will. --Stemonitis (talk) 06:52, 21 June 2011 (UTC)

The problem here is the mic of phylogenetic and Linnaean taxinomy. Both Docodonts and Therapsids are are well established and well published orders. You could solve this by making a skip template to rout Docodonta around Therapsida an anchor it directly in Synapsida. If you really want Therapsida in there, you could make a "shim" template with an unranked Therapsida. Petter Bøckman (talk) 07:06, 21 June 2011 (UTC)

Docodontidae has two orders, also. What should that taxobox look like? I tried parenting Template:Taxonomy/Docodonta to Template:Taxonomy/Mammaliaformes/Amniota, but then Synapsida, etc won't show up. We may need a few skip/same-as templates if we want more of these to show up on those taxoboxes. If we can decide here what they should all look like I'll make the extra templates. It would sure be nice if the automatic taxobox system just didn't show redundant 1st rank taxa! That would solve a lot of problems. ErikHaugen (talk | contribs) 17:04, 21 June 2011 (UTC)

Maybe we just need a skip template Template:Taxonomy/Mammaliaformes/Synapsida, so the "class Synapsida" shows up on the taxoboxes. Does that seem like it would work? ErikHaugen (talk | contribs) 17:06, 21 June 2011 (UTC)

The problem here is the use of "order". In principle we can choose Benton, classical and phylogenetic, but if we introduce ranks we are stuck with the two former. I'd personally go fully classical (order Docodonta -> class Mammalia -> Amniota) but I'm probably in the minority here. Petter Bøckman (talk) 18:09, 21 June 2011 (UTC)

Is someone going to make the change soon? For my part, I don't care which system is used for these animals, but it evidently needs to be just one. --Stemonitis (talk) 08:49, 26 June 2011 (UTC)