Template talk:Automatic taxobox/Archive 6

Extended nesting support is gone....new limit 27 taxa
See the colorless Saurischia... Bob the Wikipedian (talk • contribs) 03:59, 9 November 2010 (UTC)

Is there a limit on how deep the tree can be? Tyrannosaurus has a truncated taxonomy section. ErikHaugen (talk) 05:08, 9 November 2010 (UTC)


 * The limit is supposed to be around 40 or 50 taxa...I don't recall off the top of my head. Martin may remember. Bob the Wikipedian (talk • contribs) 05:16, 9 November 2010 (UTC)
 * Yikes...it's chopping it off at 27. Martin.... you sure the new updates haven't knocked out the extended nesting support? Bob the Wikipedian (talk • contribs) 05:19, 9 November 2010 (UTC)


 * A quick check at Saurischia confirms the extended nesting support has indeed been dropped to 27 taxa. Bob the Wikipedian (talk • contribs) 05:22, 9 November 2010 (UTC)


 * I don't recall tinkering with anything that should have this effect. It looked like everything was working when I finished yesterday; I'd've thought I'd've noticed all colours disappearing.  Perhaps not; or perhaps the mediawiki software's been tweaked?  Martin  (Smith609 – Talk)  12:57, 9 November 2010 (UTC)


 * First I noticed it was last night...I thought the taxobot pages looked a little broken, but figured it was just the taxobot, but when I went to check on a taxonomy someone had modified, I noticed that it and all of them I'd been doing are now being truncated at 27. So something, somewhere has changed, and it's up to us to get it working. Step one is to do sandbox testing and find out if one of the recent edits caused it...step two, if that wasn't the problem, might be to implement a parent_27 (26?) calculation. I've got a tragic amount of coding needing done today, so I can't be of much help coding, but I can certainly do the check to see if the code got messed up. Bob the Wikipedian (talk • contribs) 17:56, 9 November 2010 (UTC)


 * A quick check confirms that at least one code modification by one of us has broken it. I'm working on figuring out just which change that was now. Bob the Wikipedian (talk • contribs) 18:09, 9 November 2010 (UTC)


 * Update-- the color was lost in the last modification of automatic taxobox. I'm still working on finding the cause of the 27 taxa truncation. Bob the Wikipedian (talk • contribs) 19:14, 9 November 2010 (UTC)


 * After exhaustive testing, I've arrived at the following conclusion: We made a mistake (which killed the taxobox color, see above), and the Wikimedia software seems to have changed, causing displayed taxonomies to truncate after the 26th parent (not sure how we can verify that they changed it, but we still need to adapt regardless).
 * I've also made the following observations: get regnum is still working. That means the trick we used to grab the super-grandparent 16 or so levels higher before calculating the kingdom might also work for displaying the taxa in the taxobox.
 * Unfortunately, it's the second half of the semester at college, so I've got a ton of coding I need to do for school and won't be able to look at this anytime soon.
 * In the meantime, please stop automating taxoboxes until the issue is resolved. Making taxon templates can't hurt right now, but implementing broken taxoboxes can. (That's the main reason I've been making templates but not implementing them.) Bob the Wikipedian (talk • contribs) 19:51, 9 November 2010 (UTC)

A closer look at the system reveals the flaw-- Taxobox/taxonomy is hard-coded to allow no more than 27 taxa to a list. So that means that a the lost taxa bug didn't suddenly surface with "new Wikimedia code"-- rather, someone must have added a few intermediate upper-level taxa that finally exposed the glitch nearly globally. Bob the Wikipedian (talk • contribs) 23:01, 9 November 2010 (UTC)
 * That doesn't explain why colours are messed up, nor why they've gone in multiple taxa (Hildenbrandia is as far from dinosaurs as I can think yet has no colours). Have you tried adding more hard-coded levels and seeing what that fixes?  I've put in three more now. Martin  (Smith609 – Talk)  04:46, 10 November 2010 (UTC)
 * Got it. When reverting to use Template:Taxobox (not Template:Taxobox/core) I neglected to remove the   line.  The current taxobox template can't hack a blank colour parameter.  Really sorry to have left all and sundry confused.  Martin  (Smith609 – Talk)  04:58, 10 November 2010 (UTC)
 * No hard feelings :) As for the truncation, I did play around in my sandbox, and you can see what I've been up to as far as attempting to implement the same strategy used at get regnum. Unfortunately, no luck yet. Bob the Wikipedian (talk • contribs) 05:48, 10 November 2010 (UTC)
 * I've decided that it would be best to experiment with the implementation of Template:Taxobox/core in the sandbox. I think that it works now, with all headers coloured in (although I can't say I entirely understand why it didn't before; Mediawiki is very bizarre / poorly documented);  I'd be very grateful if someone more diligent than me were to test it out (especially noting combinations of rarely-used parameters).  Thanks! Martin  (Smith609 – Talk)  06:09, 10 November 2010 (UTC)
 * State of the template-- seems to work alright now, with the exception of a hard truncation at 30 taxa (29 parents). We're working on a solution. Bob the Wikipedian (talk • contribs) 15:25, 10 November 2010 (UTC)

Seeing as how the hard truncation at 30 taxa is affecting an unknown but considerably uncontrollable number of articles, I propose we add a hatnote (which would display at the top of the article) to the automatic taxobox template until the bug can get worked out. I feel obligated to work on the bug since it's becoming a vital template and is already in use on more articles than I care to count, but quite honestly, I need to focus on non-Wikipedia things at the moment, so a notice such as the one below might serve as a good temporary error notice.

Bob the Wikipedian (talk • contribs) 03:05, 11 November 2010 (UTC)


 * Is there any way in Wikimedia source code to pass by value instead of by reference? Or, can we store a string value to be recalled later in the template? The only way this is going to work is if we can pass an upper level taxon by value only. Bob the Wikipedian (talk • contribs) 20:41, 12 November 2010 (UTC)

Proposed solution
It requires a little bit of maintenance, but it's better than nothing at all, I think.

As bug reports are filed, it will become evident which taxa are problematic. Currently, we know that the taxa above Reptilia can be problematic. By adding an optional parameter true (abyssal signifying the deepness of the daughter tree), the taxobox/taxonomy template can be modified to look out for abyssal taxa. If a taxon is abyssal, then the taxonomy is pulled from a page Taxonomy/TAXONNAME/Cache. The cache page would contain the necessary code to fill in the top of a taxobox, but this code would be hard-coded. The code would be obtained by subst'ing TAXONNAME. A notice should be added to all its parent taxon pages stating that its cache will need updated if the parent is changed, and hidden categories could help out with that list. Upper-level taxa aren't very dynamic, so this shouldn't cause too many issues.

The logic for the Taxobox/taxonomy will need to be modified so that it checks for an abyssal taxon in the same way get regnum checks for the kingdom. Here's some pseudocode:

It's really ugly and will require we maintain caches, but it's the only thing I've been able to come up with after a week of brainstorming and testing. Bob the Wikipedian (talk • contribs) 20:13, 13 November 2010 (UTC)
 * Some of that code could even be put into taxobox/taxonomy cell instead of taxobox/taxonomy to save on space. Bob the Wikipedian (talk • contribs) 20:22, 13 November 2010 (UTC)

Proposed solution #2
I'm on a much-needed-for-classes wikibreak at the moment but just got an idea...


 * In the same way we are now using dummy taxa for the parents of Mammaliaformes and Avialae, perhaps we ought to set up truncated dummy parents for the problematic ones. This would only involve a few trees here and there. For example, you'd link the problematic daughters of any major taxon (for example, the immediate daughter of Reptilia in the dinosaur lineage) to a dummy taxon that has a dumbed-down parentage. This would reduce the number of parents for that daughter taxon to two, eliminating altogether the problem on all dinosaur articles. The same would be done for any branch that causes problems.

This solution would be much simpler to implement than the abyssal taxon solution, and it would also require significantly less upkeep. Bob the Wikipedian (talk • contribs) 06:20, 19 November 2010 (UTC)


 * I've implemented this at Taxonomy/Neodiapsida, although it appears we need to add color functionality for the regnum "Animalia (quickTax)". I made an attempt, but I don't see it working yet. However, the full taxonomy is being displayed now at Tyrannosaurus rex. resolved

On suffix standardization....

 * Thanks Bob! We can almost start adding more automatic taxoboxes. How about using quickTax for Mammaliaformes/Avialae? It would be kind of nice to have one convention for these shortcuts, even though obviously they solve completely different problems. ErikHaugen (talk) 20:15, 19 November 2010 (UTC)
 * Ehhhhhh...let's only implement it where needed. I.e., if you spot a truncation, then we'll deal with it...but until then, there's no need to do it. Remember that the quickTax method may require (highly unlikely) maintenance, so the sparser its use is, the more efficiently we can use the database. But be sure and report any truncations you spot...we don't want to give ourselves a worse reputation than we already have (after that two-week-long bug)! Bob the Wikipedian (talk • contribs) 21:15, 19 November 2010 (UTC)
 * I'm not sure we're talking about the same thing – Mammaliaformes is already shortcutted; I'm not sure how what I'm talking about will cause any additional bugs? I'm talking about how there are two taxonomy templates for Mammaliaformes: Template:Taxonomy/Mammaliaformes_(Amniota) and Template:Taxonomy/Mammaliaformes, each with different parents, just like how Template:Taxonomy/Diapsida and Template:Taxonomy/Diapsida (quickTax) have different parents. It's the same solution for the two different problems – in both cases, you have a bogus "shortcut" node representing the same taxon but it skips a long chain of parents. Can we use the same naming convention for those two solutions? ie, just move Template:Taxonomy/Mammaliaformes (Amniota) -> Template:Taxonomy/Mammaliaformes (quickTax)? My concern is that the way it is now makes it look like there is some other taxon named Mammaliaformes directly under Amniota, or perhaps it reflects some dispute about its phylogeny or something, because we are currently using the convention that we would be using if that were the case. (quickTax) might make it more clear that this is not the case, since there is no taxon called quickTax (at least I hope there isn't). Also, a convention like quickTax will make it easier for tools to figure out that the nodes should be merged if the graph is to be used for purposes other than taxoboxes. Thanks, ErikHaugen (talk) 22:23, 19 November 2010 (UTC)
 * Actually, there's a difference in the usage-- here's a diagram showing the difference:

Life \_Eukaryota \_Unikonta \_Opisthokonta \_Holozoa \_Filozoa \_Animalia \_Eumetazoa \_Bilateria \_Nephrozoa \_Deuterostomia \_Chordata \_Craniata \_Vertebrata \_Gnathostomata \_Eugnathostomata \_Teleostomi \_Tetrapoda \_Reptiliomorpha \_Amniota \_Mammaliaformes (Amniota) \_Mammalia

A quickTax version of this one would look like this:

Life \_Animalia (quickTax) \_Chordata (quickTax) \_Vertebrata (quickTax) \_Gnathostomata (quickTax) \_Tetrapoda (quickTax) \_Reptiliomorpha (quickTax) \_Amniota (quickTax) \_Mammaliaformes (quickTax) \_Mammalia

Since there are a large number of quickTaxes involved here, it would be more difficult to maintain.

QuickTaxing is a more ideal solution for taxa just below a major taxon, as in the case below. Here, a very large number of taxa are reduced to five taxa. The maintenance on this won't be as dynamic as the previous one would be, since far fewer taxa are involved.

Life \_Animalia (quickTax) \_Chordata (quickTax) \_Reptilia (quickTax) \_Diapsida (quickTax) \_Neodiapsida

Hope I'm making sense here. Bob the Wikipedian (talk • contribs) 22:55, 19 November 2010 (UTC)


 * It's certainly a workable solution, but all the maintenance makes it rather inelegant. (It might be possible to use "slowTax" for the first display_taxa ranks, then to automatically link in to the quickTax.) But I wonder whether there's a way to tackle this head-on.  There are three undocumented errors I've encountered in making this template; this; the missing colours in some cells of the taxobox when calling Taxobox/core; and the quiet failure of } when 1.  I wonder whether these could be related; since they are all dying silently rather than returning error messages, perhaps there's a bug with the Wikimedia software. If we can work out whether the errors are replicable, I'd suggest filing a bug report with the developers to see whether this is something that they can fix – or something that we could work around more stealthily.  Martin  (Smith609 – Talk)  03:42, 20 November 2010 (UTC)
 * "A quickTax version of this one would look like this:" - Bob, I certainly didn't want to do that. I meant just have Template:Taxonomy/Mammaliaformes (quickTax)'s parent be template:Taxonomy/Amniota? I am only talking about naming here. Does "quickTax" have some significance to some template? ErikHaugen (talk) 05:24, 20 November 2010 (UTC)


 * Yup, "quickTax" refers to the fact that the taxonomy load is faster and takes shortcuts. It works on a cache that has to be manually updated, if, say, chordates are suddenly found not always to be animals, or reptiles found not always to be chordates, or diapsids found not always to be reptiles or (by some really highly unlikely twist of logic), animals found not always to display life at some point during their existance....you get the idea. The reason this would get really ugly if we implemented it somewhere like Mammaliaformes is that the accepted classification of mammals and reptiles within Chordata is so fluctuant that it could change every five years or less. In fact, all it would take would be for us to suddenly adopt Sauropsida (which I would like to, but the WP Amphibians and Reptiles have said no) for the Aves and Reptilia to need an update; this would involve maintenance in multiple places if we jumped to the quickTax solution for Avialae. No bugs would be introduced, just a little bit of extra maintenance-- defeating the purpose of automating taxoboxes. Sorry if I misled you to believe bugs would be involved. Bob the Wikipedian (talk • contribs) 18:19, 20 November 2010 (UTC)


 * I still feel like we're talking past each other; I'm not suggesting anything that would take any more maintenance than what we already have. "has to be manually updated" Right, but this is also true of Template:Taxonomy/Mammaliaformes_(Amniota), right? I mean, what would be different if we moved Template:Taxonomy/Mammaliaformes_(Amniota) to Template:Taxonomy/Mammaliaformes_(quickTax)? That's it, just that one move? We're using these bogus taxonomy nodes for various purposes - we have things like Template:Taxonomy/Mammaliaformes (Amniota) for the purpose of skipping redundant primary ranks, we have Template:Taxonomy/Physeteroidea_(displayed) for fine-grained control of always_display, and we have Template:Taxonomy/Diapsida_(quickTax) for faster loads/depth issues. All I'm suggesting is some kind of convention for marking nodes as duplicates. And pointing out that "Mammaliaformes (Amniota)" is problematic/confusing for other reasons. ErikHaugen (talk | contribs) 19:33, 20 November 2010 (UTC)


 * Really the difference in usage is that one (the dummy-style shortcut) links to pre-existing templates, requiring maintenance in only one place, but the other (the quickTax-style shortcut) requires maintenance all the way from that taxon to the top. I suppose we could use the same naming convention if we wanted, but the suffix "quickTax" implies all its parents are also "quickTax", so it might confuse people trying to quickly interpret a situation based on the template name. I would encourage some sort of convention, however, for the dummies...though "dummy" is a bit derogatory, so there is probably a better word. Or, perhaps, "shortcut" might be an appropriate uniform suffix for the dummies. Bob the Wikipedian (talk • contribs) 06:18, 21 November 2010 (UTC)


 * It's true, quickTax would be confusing for those other uses. Maybe shim instead of dummy? Maybe we can put them all in a category or tag them in some other manner? Thanks! ErikHaugen (talk | contribs) 07:28, 21 November 2010 (UTC)

Martin-- you fixed it! I counted Velociraptor 's taxa and saw...43!!!! QuickTax is no longer needed until April 15th, but by then we'll move on to TurboTax. Bob the Wikipedian (talk • contribs) 06:56, 21 November 2010 (UTC)
 * zing! ErikHaugen (talk | contribs) 07:28, 21 November 2010 (UTC)
 * Erik-- Shim doesn't mean much for me...How does "simplified" grab you as a suffix? Bob the Wikipedian (talk • contribs) 19:34, 21 November 2010 (UTC)
 * You're right, shim is terrible. simplified doesn't really work for some of the ways its used. Maybe the title of the template is not the place to do this, perhaps some kind of template or something, or maybe a guy instead of don't edit this line? So maybe Template:Taxonomy/Physeteroidea_(displayed) could contain just . See, only replace the items that you need to replace, inherit the rest from the "real" taxon's template. Template:Taxonomy/Mammaliaformes (Amniota) could be  and so on. I think I can see how to implement Same as taxon here, does this seem misguided? If not and nobody has a better idea for a name I can write it. ErikHaugen (talk | contribs) 23:06, 22 November 2010 (UTC)

Implemented solution
Quite by the grace of God I happened to mention the parameter in the nested extension-template. This, for some reason I simply cannot fathom, allowed the nesting to work. So the quickTax system is no longer necessary. Martin  (Smith609 – Talk)  05:20, 20 November 2010 (UTC)
 * Lol....accidentally nowiki'd it? I see we are still using quickTax, though following a heavy bug reporting session....not to be harsh, but let's use sandboxes for testing things in the future. There's a reason the taxobox has a red lock on it. Bob the Wikipedian (talk • contribs) 18:21, 20 November 2010 (UTC)
 * I've been thinking about how to implement a testing regime with these templates. Producing and maintaining a sandbox for each of the ~40 templates called doesn't seem terribly practical.  Some edits need sandboxes for a complete taxonomy too – the Template:Taxonomy/Test-50 series has some value but can't replicate everything in the wild.  I think that the 'display long taxa' is the last major amendment that's needed, as most of the back-end is working pending WP:BRFA; non-structural edits are of course easier to sandbox.  Until everything's robust, though, it does seem perhaps a little premature to use a beta-stage template in featured articles.
 * I'd also welcome suggestions for a straightforward way to rigorously test edits; I noticed yesterday that several edits that were successful in the first three taxoboxes that I tested caused errors in others. I'm no software tester so any ideas are welcome.  Martin  (Smith609 – Talk)  22:48, 20 November 2010 (UTC)
 * What sort of tests are you hoping to do that can't be done with the 1-50? Also-- typically I just load a taxonomy into my sandbox by setting the taxon equal to something that already exists. Bob the Wikipedian (talk • contribs) 06:22, 21 November 2010 (UTC)

resolved

Now we can all take a deep breath and get back to whatever we were doing before...BIG thanks, Martin. Bob the Wikipedian (talk • contribs) 19:19, 21 November 2010 (UTC)

Edit link
I'm surprised no one's said this so far-- on my screen, "edit" is overlaid on top of "Scientific classification", making it difficult to see. Bob the Wikipedian (talk • contribs) 15:25, 10 November 2010 (UTC)
 * I noticed that just today. And it's only in taxoboxen that don't have an image or a long taxon title to force the taxobox wider. Maybe the words "Scientific classification" could be left-justified instead of centered? Rkitko (talk) 15:53, 10 November 2010 (UTC)
 * Would it help if the link just said "[e]" or even just "e", as in some other templates? (It looks okay on my screens.) Otherwise we might have to think of a different place to put it (perhaps in the bottom left corner of the taxonomy list). Martin  (Smith609 – Talk)  20:27, 10 November 2010 (UTC)
 * "[e]" would fit, sure. Anyone who doesn't know what the "[e]" means would probably do best to leave a template alone, anyway. Bob the Wikipedian (talk • contribs) 03:05, 11 November 2010 (UTC)
 * I've modified Template:Edit_taxonomy accordingly. How does it look? Martin  (Smith609 – Talk)  16:18, 11 November 2010 (UTC)
 * Much better, thanks! Bob the Wikipedian (talk • contribs) 21:53, 11 November 2010 (UTC)

resolved

Dinosaur truncations
Hey guys, on WP:Dinosaurs talk Bob marked the truncation issue as resolved but some low-level taxa are still truncating. See Velociraptor. MMartyniuk (talk) 02:09, 20 November 2010 (UTC)


 * . I've no idea why what I did worked, but it did.  Martin  (Smith609 – Talk)  05:18, 20 November 2010 (UTC)

resolved

Oversize
Oh dear. It looks like I've taken the template over the maximum permitted size, e.g. at Apatosaurus: Post-expand include size: 2048000/2048000 bytes Template argument size: 2048000/2048000 bytes

Too late at night to try to resolve this now, I'm afraid; things seemed to be working fine at least until I implemented the necessary ucfirst/lcfirst check (see Special:Contributions/Smith609). Sorry I'm not able to look into this now! Martin  (Smith609 – Talk)  06:56, 20 November 2010 (UTC)


 * I've returned the non-automatic taxobox until the automatic one is fixed. --Kleopatra (talk) 07:15, 20 November 2010 (UTC)
 * It appears to be completely broken on every page. I just see redlinked "Automatic taxobox". Giant panda is an example; but I can't find any that work. ErikHaugen (talk) 08:09, 20 November 2010 (UTC)
 * You can't just walk away after breaking every single transclusion of the template (actually, it's more than that – it also broke all templates transcluded after it, which is pretty much everything since this is one of the first templates to be transcluded on a page) and expect others to clean up the mess. I have reverted your edits today to Template:Taxonomy list/list and Template:Taxobox/taxonomy. Next time please use a sandbox. T. Canens (talk) 09:43, 20 November 2010 (UTC)


 * Feel free to revert my manual additions of taxoboxes. Articles like Human, Dinosaur, Bird, Reptile, etc., are some of the most highly visible articles on wikipedia, leaving them bleeding red broken and untranscluded templates was ugly. --Kleopatra (talk) 09:47, 20 November 2010 (UTC)
 * Thanks for resolving this. Sorry again about my error and sorry further that I didn't have time to fix it there and then.  I'll use sandboxes as far as I can do; I'm glad you found it so easy to restore a less-broken version.  Martin  (Smith609 – Talk)  12:37, 20 November 2010 (UTC)

resolved

common names vs binomial
Should it be Template:Taxonomy/House mouse or Template:Taxonomy/Mus musculus? ErikHaugen (talk) 20:27, 8 November 2010 (UTC)
 * I'd advocate taxonomic rather than common names where-ever possible. Martin  (Smith609 – Talk)  23:23, 8 November 2010 (UTC)
 * PLEASE!!!! Doing so makes it much easier to find parent taxa, should the species become parent to a subspecies. Bob the Wikipedian (talk • contribs) 03:59, 9 November 2010 (UTC)

resolved

Beaked whale
I tried to add an automatic taxobox to the Beaked whale article but I did something wrong apparently. Can someone take a look at the article and help?-- Kev min  § 20:11, 10 November 2010 (UTC)
 * When the official taxon is not the name of the article, you can use the taxon parameter to tell the automatic taxobox where to find the taxonomy template. — Preceding unsigned comment added by ErikHaugen (talk • contribs) 25 November 2010, 14:25

resolved

That darn colorlessness
Evidently, the color bug must have a doubly recessive genotype, because every time we think we've eliminated it, it pops up somewhere else...

See Apatosaurus. We've got a new breed of colorlessness bug. Bob the Wikipedian (talk • contribs) 08:41, 12 November 2010 (UTC)


 * Fixed. That solves the mystery of why things were listing Template:T^8Eukaryota in their 'called templates' list... Martin  (Smith609 – Talk)  15:35, 12 November 2010 (UTC)


 * Yeah, I guess I noticed that with a different affected taxon a couple days ago but didn't think to mention it. Bob the Wikipedian (talk • contribs) 16:28, 12 November 2010 (UTC)

The "Binomial name" still has a white background at Acrophyseter but I can't find any other pages exhibiting this. Any ideas? ErikHaugen (talk | contribs) 04:02, 22 November 2010 (UTC)
 * It may well be due to the presence of }}}}}s -- without spaces between the groups, the software sometimes has trouble interpreting them (just like when using 's for bold and italics). I tried adding the correct spacing at Template:Taxobox/Sandbox, but this caused an oversize error when I tested it out.  If you fancy taking a look at it, please go ahead... I won't be templating much until the weekend.  Martin  (Smith609 – Talk)  14:53, 22 November 2010 (UTC)

resolved

Unwanted italics
The article title and taxobox title of Mastodon shouldn't be italicised as it's used as a common name. mgiganteus1 (talk) 16:48, 12 November 2010 (UTC)
 * See Template:Taxobox. Martin  (Smith609 – Talk)  16:57, 12 November 2010 (UTC)
 * Yes, but in this case "the value of |genus=, |species=, or |binomial=" does not exactly match the title of the page. Is there something I'm missing? mgiganteus1 (talk) 17:12, 12 November 2010 (UTC)
 * I specified Mastodon as recommended and it worked.
 * However, you have raised an interesting point. Presently, the page is italicized if the target of the link equals the pagename, and if the rank is a genus (etc.).  Perhaps it would be more appropriate to test whether the displayed text (in this case "mammut") matches the page name; however this won't work if extinct is passed in the "displayed text" parameter.  One solution would be to invent a new parameter "true" which would automatically add the dagger if necessary.  But might there be other cases where the display text doesn't exactly match the page name, and it needs italicizing (etc)?  Martin  (Smith609 – Talk)  17:32, 12 November 2010 (UTC)
 * I'd love to see extinct, so I will volunteer to help with/do the migration if you implement it. ErikHaugen (talk) 19:07, 12 November 2010 (UTC)
 * Supported, in Template:Don't edit this line only. See http://en.wikipedia.org/w/index.php?title=Template:Taxonomy/Mammut&action=historysubmit&diff=396377184&oldid=394251727 for an implementation example.  If you want to migrate it, feel free!  Otherwise I'll try to do this whilst converting the to use "don't edit this line".  There may be documentation that needs updating. Martin  (Smith609 – Talk)  20:15, 12 November 2010 (UTC)
 * Great! Should we make the preload thing use Template:Don't edit this line now? Should we migrate to it? thanks, ErikHaugen (talk) 21:50, 12 November 2010 (UTC)
 * I wouldn't recommend putting time into a wholesale migration just yet; I might yet think of a way to improve the editing experience that involves further modifications to the code or the name of the template. But certainly it should be used over "Taxonomic unit" for the time being, and you may as well amend the preload.   Martin  (Smith609 – Talk)  22:12, 12 November 2010 (UTC)
 * I notice that true doesn't put the dagger on higher taxa – look at Hadrosaurid; notice "Hadrosauroidea" has no dagger. ErikHaugen (talk) 06:15, 14 November 2010 (UTC)
 * Martin  (Smith609 – Talk)  16:14, 14 November 2010 (UTC)

resolved

Clades vs. Linnaeus
I notice Aves' parent is Tetrapoda. I'd like to make it Avialae, but doing so would put Theropoda on all birds' taxoboxes because it is set to be "always displayed". I'm assuming this is undesirable. Should I leave Aves' parent alone? Should I make Theropoda not always visible? It would be nice if it could be selectively visible, but I'm not sure if there is a good way to do that. ErikHaugen (talk) 19:04, 12 November 2010 (UTC)


 * Thanks Erik....I've notified the folks at WP:BIRDS.

Attention members of WP:BIRDS
Aves, being a daughter taxon of Sauropsida, poses a unique problem. Here are our options:


 * Cladistic: If we link Aves to Avialae, then the taxa Reptilia, Dinosauria, Saurischia, and Theropoda will display above Aves in every automated bird taxobox.
 * Linnaean (simplified): If we leave out Avialae, the taxonomy will be incorrect in that Aves will be left out of Avialae.
 * Linnaean (complex with dummy): If we link Aves to a duplicate Avialae that displays as Avialae, then the dummy Avialae can be linked to Tetrapoda and the reptilian taxa will not display in automated bird taxoboxes, but, should the placement of Avialae change, the dummy Avialae may also need updated (just a matter of someone remembering to do it).

Please voice your thoughts. Oh, and the reason we're using Reptilia instead of Sauropsida is because that's what the Reptiles and Amphibians WikiProject has requested. Bob the Wikipedian (talk • contribs) 19:31, 12 November 2010 (UTC)
 * I'd go with the third of those options, I believe. It seems to make the most sense. – The Bushranger Return fire Flank speed 20:00, 12 November 2010 (UTC)
 * You can link the dummy template to the real template to save on maintenance: e.g. rank. Martin  (Smith609 – Talk)  20:06, 12 November 2010 (UTC)


 * There are at least two problems here. The parent (or further ancestor) of a class (e.g. Aves) cannot be at a lower rank (e.g. suborder Theropoda, the parent of [unranked] Avialae). These are conflicting classifications, and cannot be combined. I would also argue that the superclass Tetrapoda is not worthy of being set to always display. It is not a significant rank for taxoboxes about species and genera, which make up the vast majority. Of the three options presented, only the second one seems feasible. It is not incorrect to leave out certain ranks (indeed, it is very strongly recommended), so I see no problem. The third option is just asking for trouble; we really don't need taxonomic forks, and there are guidelines against it (WP:CFORK). --Stemonitis (talk) 20:09, 12 November 2010 (UTC)


 * Martin-- the parent is actually the part that would differ.
 * I've modified my example to "rank". You can use the same syntax for any parameter.  Martin  (Smith609 – Talk)  20:25, 12 November 2010 (UTC)
 * Stemonitis -- Tetrapoda would only be displayed in the Aves and Avialae taxoboxes, not their daughters, due to its inferiority. So don't let that influence your decision. Bob the Wikipedian (talk • contribs) 20:22, 12 November 2010 (UTC)
 * Sorry – my mistake. I meant the suborder Theropoda, which is indeed set to always display (cf. Template:Taxonomy/Theropoda), for which I can see no justification. --Stemonitis (talk) 21:32, 12 November 2010 (UTC)
 * I agree – I think the whole problem goes away if we mark Therapoda to not always display and reparent aves to Avialae. Then the bird folks won't have to see Therapoda and crap on their taxoboxes and we can still have our nice taxonomy tree. Although I think it would be nice if Therapoda could be visible for selected dinosaur articles only somehow. ErikHaugen (talk) 21:49, 12 November 2010 (UTC)
 * Okay, taxonomy is what it is, a classification. The second option is taxonomically correct, even if is cladistically incorrect. Phylocode has not yet been officially launched, and once it is, we can see how much traction that gets. But for now, lets stick to traditional taxonomic classification. -- Kim van der Linde at venus 20:46, 12 November 2010 (UTC)
 * Erik-- I believe Reptilia, Dinosauria, and Saurischia would still display. Also, Theropoda is a very important taxon to paleontologists that they've expressed should always be displayed on theropods. Theropoda is the taxon that contains all bipedal carnivorous dinosaurs. At the moment, I believe the complex solution with the dummy will be fairly simple to manage and should keep both parties happy (dinosaurs and birds). Bob the Wikipedian (talk • contribs) 02:20, 13 November 2010 (UTC)
 * The simplest option would simply be to make Theropoda an order, following Bakker 1984. This is how it's used in the literature (You'll never see a paper titled "...of Tyrannosaurus rex (Dinosauria: Saurischia)" but (Dinosauria: Theropoda) is almost universal). But that would screw up other perts of the tree and is a bit OR. MMartyniuk (talk) 02:53, 13 November 2010 (UTC)
 * Assuming we go with option 3, we could do something similar for mammalia and make a Template:Taxonomy/Mammaliaformes (dummy). Is that a good name for this kind of thing? ErikHaugen (talk) 20:19, 13 November 2010 (UTC)
 * I'm not sure it's needed for Mammalia. Is it? I've not seen anything that says Mammalia is now part of Sauropsida. Bob the Wikipedian (talk • contribs) 21:35, 13 November 2010 (UTC)
 * Maybe rather than (dummy) it would be clearer to state the parent rank in the brackets? i.e. Avialae (Tetrapoda).  Martin  (Smith609 – Talk)  22:36, 13 November 2010 (UTC)
 * Sounds logical to me. There's little chance that tetrapods evolved twice, so it sounds like a safe name. Besides, even if it does need renamed, that should be pretty simple to do with page moving and the new handy way of finding parents (view ancestry) and daughters (what links here, filter templates) of a taxon. Bob the Wikipedian (talk • contribs) 00:04, 14 November 2010 (UTC)
 * "I'm not sure it's needed for Mammalia" – well, it's not really needed for birds, is it? Is there a difference? ErikHaugen (talk) 06:11, 14 November 2010 (UTC)


 * I can't see any problem with the current taxobox. We don't actually need anything between the phylum and Aves Jimfbleak – talk to me?  19:36, 14 November 2010 (UTC)
 * If it's needed for birds, it's quadruply needed for mammals. There are presently four paraphyletic 'forks' in the stem-mammalia lineage. The parent of Mammaliaformes is currently Synapsida rather than Probainognathia, The parent of Mammalia is Amniota rather than Mammaliaformes, the parent of Therapsida is Synapsida rather than Eupelycosauria, etc. That's the only way I could keep its parent taxa, some of which are which are variously ranked as orders and suborders and even families, from screwing up the box for Mammalia itself (i.e. it would have two classes listed--Synapsida and Mammalia). MMartyniuk (talk) 03:45, 15 November 2010 (UTC)
 * Marty, assuming we adopt this dummy taxon strategy, I think we'll be counting on you to set the mammal lineage up for us as invisibly as possible...you seem to understand the cladistics pretty well. Bob the Wikipedian (talk • contribs) 04:28, 15 November 2010 (UTC)
 * "The parent of Mammaliaformes is currently Synapsida rather than Probainognathia" etc – this is just because nobody has bothered to flesh out the whole "non-mammalian synapsid" section of the tree, right? ErikHaugen (talk) 08:03, 15 November 2010 (UTC)
 * They have actually, it's just that the traditional ranks require a lot of paraphyletic groups within other paraphyletic groups. I'll try my best to sort it with dummies. MMartyniuk (talk) 12:04, 15 November 2010 (UTC)
 * I'm confused; why not just reparent Mammaliaformes to Probainognathia? ErikHaugen (talk) 17:49, 15 November 2010 (UTC)
 * I thought this wouldn't work because of conflicts in rank. Mammaliaformes includes an Order (Docodonta) while Probainagnathia is an infraorder. But I'll give it a shot. MMartyniuk (talk) 23:31, 15 November 2010 (UTC)
 * Also I keep forgetting that there are major ranks that we don't want to show (which is strange since that is the whole issue here). I suppose one way these sort of shenanigans might work is if the system ignored lower taxa after a higher taxa had already been displayed – eg; if the taxonomy tree said: Ursus->Ursidae->Carnivora->Mammalia->Therapsida(isn't shown b/c it's an order and we already showed a class)->Synapsida(don't show b/c it's a class and we already showed a class)->Chordata(shown b/c it's a phylum and we haven't shown anything that high yet)->etc Would that be hard to implement? ErikHaugen (talk) 00:37, 16 November 2010 (UTC)
 * That sounds like a good solution if it's possible. I made Probainognathia the parent of Mammaliaformes, and here's how Docodonta looks. I can imagine a lot of griping on the talk page that an order an't be within an infraorder within another order, and justifiably so as it's a little weird. MMartyniuk (talk) 02:11, 16 November 2010 (UTC)
 * So long as we don't disturb the peace of the extant mammal pages, I'm fine with it. I'm sure the cladists paleozoologists would love it. Bob the Wikipedian (talk • contribs) 04:58, 16 November 2010 (UTC)

Resetting indent. I tried making Mammalia's parent Mammaliaformes, and it's doing the same ting as Dcoodonta. Listing a lot of other classes, infraorders, and suborders as containers for Mammalia. This will upset the modern mammal taxa without a dummy Mammaliaformes. MMartyniuk (talk) 05:34, 16 November 2010 (UTC)
 * Thanks for testing Marty. "I'm sure the cladists paleozoologists would love it." – yeah, I was getting worried about the purity of our little tree. I'll poke around but these templates are pretty intimidating. ErikHaugen (talk) 05:45, 16 November 2010 (UTC)

mitigating confusion with dummies
If we go with Template:Taxonomy/Avialae (Tetrapoda) would there be a way to figure out that there aren't two taxa named Avialae or would you just have to know? ErikHaugen (talk) 17:49, 15 November 2010 (UTC)
 * A  disambiguating hatnote couldn't hurt on that template's page. Bob the Wikipedian (talk • contribs) 01:00, 16 November 2010 (UTC)
 * Why not Template:Taxonomy/Avialae (Amniota)? This would connect it to its most immediate parent taxon that doesn't run the risk of causing rank conflicts. MMartyniuk (talk) 23:34, 16 November 2010 (UTC)
 * Just a note that the new naming convention (for technical reasons) is to use Template:Taxonomy/Avialae/Amniota. Thanks! Martin  (Smith609 – Talk)  14:54, 9 December 2010 (UTC)

resolved

Family incertae sedis
I just came across a situation I am not sure how to deal with. There are several extinct genera of Raphidioptera which are not placed to family, such as Archiinocellia. Thus, when I created the pages initially I used a link the the article on incertae sedis on the taxobox. I was converting the articles I have created to automatic taxoboxes but I do not know what to do with the ones that have incertae sedis as a placement at some parent level. -- Kev min  § 09:16, 15 November 2010 (UTC)


 * Is there any reason why one can't make Taxonomy/Raphidioptera incertae sedis? --Stemonitis (talk) 09:28, 15 November 2010 (UTC)
 * I think it should be Taxonomy/Incertae sedis (Raphidioptera) – see Template:Automatic_taxobox/Conventions for more; or update it if you disagree, please. Or you could just parent your genera to Raphidioptera? ErikHaugen (talk) 17:42, 15 November 2010 (UTC)


 * Simply setting the parent to the next certain rank above the taxon gives the impression of an incomplete taxobox, rather than reflecting the uncertain placement. Including incertae sedis is a very good clarification in such instances. --Stemonitis (talk) 18:05, 15 November 2010 (UTC)
 * Fair enough. Please comment: Template:Taxonomy/Zygophyseter, Template:Taxonomy/Incertae sedis (Physeteroidea) (here we have familia, and extinct is not true although they're all extinct – what do you think?). Note for Zygophyseter I had to adjust display_taxa to get "Physeteroidea" to show up in the taxobox. It would be nice to be able to have a little more control over displayed taxa; for example, I think any taxon that uses Template:Taxonomy/Incertae sedis (Physeteroidea) should probably display Physeteroidea, but a lot of taxa under Physeteroidea probably don't need to, but that might be a problem if display_taxa can't be used. Maybe this scenario is too rare to worry about. ErikHaugen (talk) 21:39, 15 November 2010 (UTC)
 * I set the parent to"Physeteroidea (displayed)" and updated Template:Automatic_taxobox/Conventions. Martin  (Smith609 – Talk)  01:48, 20 November 2010 (UTC)
 * Very clever! ErikHaugen (talk | contribs) 17:32, 24 November 2010 (UTC)

resolved

Protosialis casca taxonomy
Does anyone know why the family and order are not displaying on the Protosialis casca page?-- Kev min  § 23:25, 16 November 2010 (UTC) resolved
 * Same thing is happening at Sialis lutaria-- Kev min  § 23:31, 16 November 2010 (UTC)
 * I think it's because the missing taxonomy templates were not using the latin ranks – ie, Family instead of familia. ErikHaugen (talk) 04:12, 17 November 2010 (UTC)
 * Ahhhhh! I didnt think about the fact I used family and Order rather then familia and ordo!  Good to note, thanks. -- Kev  min  § 04:19, 17 November 2010 (UTC)

Another problem, autotaxobox pushes text down
The automatic taxobox on the Bird article makes the article text appear below the bottom of the screen on my computer. I restored the manual taxobox so that this does not happen. I think that when you add an automatic taxobox to the article, please label your edit summary appropriately, and please check the results of your edit. I don't know if this happened due to the templates, and therefore, editors who added the taxobox could not see that this happened, but please monitor the appearance of articles as you add the automatic taxoboxes and as you edit the templates. Thanks. --Kleopatra (talk) 11:37, 20 November 2010 (UTC)
 * Interesting, it was working fine for me. Oh well. Are you sure it's not a problem with your browser settings? What browser are you using? T. Canens (talk) 11:49, 20 November 2010 (UTC)
 * Yes, it is browser specific, it occurs in IE, but not in Mozilla, my regular browser. There are many users of IE out there, and the code should not be such that it is problematic in one browser, particularly if it is destroying the appearance of an FA in a major browser. --Kleopatra (talk) 17:30, 20 November 2010 (UTC)
 * I assume that with the reversions this issue was resolved? Looks good to me in both Firefox 4 and IE 9 (both of which are betas). Bob the Wikipedian (talk • contribs) 18:30, 20 November 2010 (UTC)

resolved

extraneous pipe vs. italics
If you have a final pipe before your closing double-curly braces, ie |}} this seems to defeat the automatic italicization of speces/etc. See this diff. ErikHaugen (talk | contribs) 00:06, 24 November 2010 (UTC)
 * Nice! Any unwanted side effects? Bob the Wikipedian (talk • contribs) 04:05, 24 November 2010 (UTC)
 * p.s. Wait-- is this useful? Bob the Wikipedian (talk • contribs) 04:06, 24 November 2010 (UTC)
 * Heh, no, I don't think so; this was supposed to be a "bug report". :) ErikHaugen (talk | contribs) 05:31, 24 November 2010 (UTC)
 * I'd say it's more of a typo than a bug. Is there supposed to be support for the extra piping symbol for some reason? Bob the Wikipedian (talk • contribs) 06:50, 24 November 2010 (UTC)
 * I don't know if extraneous pipes are supposed to be allowed, but this is not optimal. A lot of taxoboxes have this, for some reason, so when people convert or add new ones they'll probably do this and not know what they need to do to get proper italics. Not a huge deal, of course. ErikHaugen (talk | contribs) 07:53, 24 November 2010 (UTC)
 * Parameter "1" was not explicitly "if'd"; it now is, in this revision.  Martin  (Smith609 – Talk)  15:00, 24 November 2010 (UTC)

resolved

Seemingly extraneous italic_name needed
For example, Lycaon sekowei. Should I need italic_name in this taxobox? thanks! ErikHaugen (talk | contribs) 01:12, 30 November 2010 (UTC)
 * It appears so, although I'd agree...it's a bit redundant. Bob the Wikipedian (talk • contribs) 03:16, 30 November 2010 (UTC)
 * I think that this is resolved in this edit. I've tested what I can think of but would be grateful if you had a quick check of any potentially-problematic taxa that you can think of, just to be sure it's not had any unintended side-effects. Martin  (Smith609 – Talk)  14:53, 30 November 2010 (UTC)
 * Looks good – thanks! My attempt to add subspecies in there didn't work, I guess – I still need italic_name at Ursus maritimus tyrannus – but whatever. Thanks, ErikHaugen (talk | contribs) 17:44, 30 November 2010 (UTC)
 * But Ursus maritimus tyrannus is still in Category:Automatic taxobox cleanup, so something somewhere is wrong. --Stemonitis (talk) 17:58, 30 November 2010 (UTC)
 * I don't know why it is there. Is there a way to find out? ErikHaugen (talk | contribs) 19:25, 30 November 2010 (UTC)

Martin  (Smith609 – Talk)  20:00, 30 November 2010 (UTC) resolved

Display immediate parent taxon?
Is it possible to display the immediate parent taxon when dealing with taxa in monotypic families? For example, in Ornitholestes it would be nice to have Maniraptora displayed above Ornitholestinae rather than jumping right to the order level, which makes navigation by taxonomy impossible. MMartyniuk (talk) 23:57, 4 December 2010 (UTC)
 * Is there no named family? Should it be incertae sedis? In any case, display_taxa seems to do what you're asking for, see my change. ErikHaugen (talk | contribs) 05:12, 5 December 2010 (UTC)
 * Thanks, that's the kind of thing I was looking for. Ornitholestinae was named as a subfamily of Allosauridae, but it isn't placed there anymore and no monotypic replacement family has ever been coined in the literature. It should properly by in Ornitholestidae but I thought since that form has never appeared in print it would be OR (even though ICZN says the nominal author is considered to have already coined it...). MMartyniuk (talk) 08:15, 5 December 2010 (UTC)

resolved

Two parents for incertae sedis?
Just to throw an idea out regarding the best way to handle incertae sedis, would it be possible to have a taxon appear as a child for more than one taxon? I guess this could be done with dummies or a redirect of some kind, but it would be cool in the future to be able to have say, Herrerasauridae appear with a question mark as children of both Saurischia and Theropoda (the two groups it tends to jump between in modern analyses). Maybe this scheme could even automatically generate the incertae sedis line on Herrersauridae itself. MMartyniuk (talk) 00:56, 8 December 2010 (UTC)

Try creating these:

Taxonomy/Possible Saurischia

ordo

Saurischia

parent

Taxonomy/Possible Theropoda (Possible Saurischia)

subordo

Theropoda

Possible Saurischia

Bob the Wikipedian (talk • contribs) 01:36, 8 December 2010 (UTC)


 * Ew....That didn't work so well. (Added the result to this page for convenience.) Bob the Wikipedian (talk • contribs) 01:59, 8 December 2010 (UTC)
 * See Template:Taxonomy/Theropoda/Saurischia/? and Template:Taxonomy/Saurischia/?.  Note that the slash convention has some advantages to the bracket convention.  Martin  (Smith609 – Talk)  16:05, 8 December 2010 (UTC)


 * Thanks, Martin. Marty, care to double-check Herrerasauridae? Bob the Wikipedian (talk • contribs) 02:00, 9 December 2010 (UTC)
 * Thanks Bob, that looks like it's working properly. However could we make the question mark only display for Theropoda? No recent studies have doubted that they're saurischian, only they're placement within Saurischia is controversial (as immediate daughter clade of Saurischia or within Theropoda). Sorry if this was unclear earlier. MMartyniuk (talk) 03:22, 9 December 2010 (UTC)
 * Done. Bob the Wikipedian (talk • contribs) 03:30, 9 December 2010 (UTC)

resolved
 * Sweet, thanks. I'll try to familiarize myself with this and implement it in similar cases down the road. MMartyniuk (talk) 03:38, 9 December 2010 (UTC)

Trial site
Since yesterday, the trial site has been extermely slow. If it does load, it does not complete the operation and stays in a wating mode. Are others having the same problem? — Ganeshk  ( talk ) 11:34, 6 November 2010 (UTC)


 * It's been doing this to me too, on and off. I think it's a problem with the toolserver.  Anyone who wants to inspect the bot's source code and suggest optimizations is more than welcome to!  Martin  (Smith609 – Talk)  23:38, 6 November 2010 (UTC)

Cleanup
Something's up. Category:Automatic taxobox cleanup is awash with entries, filed under "ϰ". --Stemonitis (talk) 16:17, 6 December 2010 (UTC)


 * Some dummy taxa use a method of getting ranks that is no longer supported. See this edit for the fix. Martin  (Smith609 – Talk)  16:40, 6 December 2010 (UTC)

resolved

Italics for binomial
It seems that the automatic taxobox is trying to italicise whatever is passed to it through. This causes problems where the parameter already includes the wikicode for italics, as as Cucurbita pepo or Welwitschia. For better compatibility with Taxobox, I feel that Automatic taxobox should not assume prior italicisation of the parameter, especially as this is damaging pages already. --Stemonitis (talk) 13:29, 8 November 2010 (UTC)
 * Are there any instances where the binomial should not be italicized? Martin  (Smith609 – Talk)  13:44, 8 November 2010 (UTC)


 * Is there any reason to force people to change their behaviour, and complicate the transition from the old taxobox to the new one!? The simplest, and best, solution, is to fix the template, which is currently broken. One can never predict what might be included in the field ; there might be cause to start one with a question mark, or extinction symbol, for instance, neither of which should be italicised. Let the editors decide what to italicise. That is not a decision for a template to make. --Stemonitis (talk) 13:54, 8 November 2010 (UTC)


 * I agree with Stemonitis. The binomial parameter can often include an "x" to indicate hybrid origin of the species, even though we prefer the × symbol, the letter x is common enough and shouldn't be italicized. There's also curious things like graft-chimaeras (e.g. +Crataegomespilus). Rkitko (talk) 14:53, 8 November 2010 (UTC)


 * I disagree. The same argument could be made for not automatically italicizing the genus/species parameter. Almost 100% of the time you just want it italicized and don't want to have to bother typing a bunch of single quotes. noitalic can be used here, a-la Extinct – perhaps a hybrid to display × ? In any case, it appears this was reverted and now the background color of the "binomial" etc headers is white again. Thanks, ErikHaugen (talk) 18:09, 8 November 2010 (UTC)


 * There are just too many weird cases, for all taxa. There are several cases that Polbot created with temporary names, e.g. Helichrysum sp. nov. A. I'm not confident a bot could identify these. I'd rather not take the ability to easily format such entries away from editors. noitalic isn't a real solution since most editors would be entirely unaware of it. It's better to leave the default be unformatted so the transition from taxobox to automatic taxobox is smoother. What did the reversion of auto-italicization of binomials have to do with background colors? Rkitko (talk) 18:32, 8 November 2010 (UTC)
 * "too many weird cases" – what about the species parameter? Don't the same issues crop up there? Why a different solution? ErikHaugen (talk) 19:00, 8 November 2010 (UTC)
 * "What did the reversion of auto-italicization of binomials have to do with background colors?" – template bug? who knows. ErikHaugen (talk) 19:01, 8 November 2010 (UTC)
 * I'm not arguing for different solutions. I don't think the template should be italicizing any parameters. You're correct that the same issues crop up in genera, species, etc. I think the color problem is unconnected. When I checked this morning at home, Velociraptor had all the necessary colored bars (even behind "Species"). Later this morning, Lobatus gigas had no color at all, but now it looks fine. Rkitko (talk) 19:26, 8 November 2010 (UTC)
 * OK – maybe the color thing is unrelated; sorry for the distraction. "I don't think the template should be italicizing any parameters." – should we stop italicizing genus/species? This transition does seem like a good opportunity to restore order if that is desirable. I guess the main reason I like binomial auto-italicization is that it is consistent with the genus autoitalicization. Although it does help reduce wikitext clutter to not have all the single quotes! ErikHaugen (talk) 19:46, 8 November 2010 (UTC)


 * I volunteer to fix all of the places like Cucurbita pepo where this is a problem. It should be pretty easy going forward for folks to get it right as they create new automatic taxoboxes. ErikHaugen (talk) 18:17, 8 November 2010 (UTC)


 * But we don't have to make the problem in the first place. In the meanwhile, the template has been fixed (I assume), because all the articles I mentioned now appear correctly, without having been edited in the meanwhile, while ones set up to work with the automatic italics (Lobatus gigas) have been edited to manually add the italics. If we just leave it this way, then there's no new problem, and everything's fine. --Stemonitis (talk) 18:56, 8 November 2010 (UTC)


 * I'm just saying, if the reason for objecting is that it messes up a bunch of pages like Cucurbita pepo that already use the automatic taxobox, then I want to take that off the table. If there are other issues, fine, but please don't object to this because of Cucurbita pepo. Thanks, ErikHaugen (talk) 19:00, 8 November 2010 (UTC)


 * Auto-italicization of the binomial parameter was only added 5 November. The vast majority of automatic taxoboxes made before that would have the correct italicization. Certainly those made between 5 November and today need to be fixed to include italics. Rkitko (talk) 19:26, 8 November 2010 (UTC)


 * Sorry if I'm not being clear. I would like to keep auto-italicization. One argument being put forward for getting rid of it is that it messes up articles that already use Automatic Taxobox, such as Cucurbita pepo. This is true. I am saying that I would be happy to go through and fix all such articles that are broken by auto-italicization in this way, thus removing this particular disadvantage of auto-italicization. ErikHaugen (talk) 19:41, 8 November 2010 (UTC)


 * My point is that there is no advantage in attempting to automatically italicise binomials. It is not always appropriate to do so (as discussed above); it makes it harder to implement non-italics when they are needed (I wasn't aware of noitalics, and I'm reasonably experienced); it makes taxobox and automatic taxobox less directly compatible. It all serves to take control away from the editor, and I don't see that as a good thing. Just as it is easy to go through and change all faulty occurrences one way, so it is easy to change them the other way. That's no argument for either system. I have yet to see any reason for automatically italicising the binomials, and quite a few reasons for avoiding it. (I'm not counting "don't want to have to bother typing a bunch of single quotes", because it seems like such a weak argument. It isn't a contentious issue with the standard taxoboxes, so I don't see why it should be now.) I don't know if the November 5 change was discussed; I imagine not. Changes like that really do need to get some kind of consent before being rolled out. These templates are live right now, and affect the appearance of articles that real people are using to get real information. Screwing them up reflects badly on the whole project, and is entirely avoidable. Unless someone can come up with a very persuasive reason for this change (and subsequently fixing all the articles that it would break), I can't see any reason why I or anyone else should support it. --Stemonitis (talk) 22:00, 8 November 2010 (UTC)


 * Well, I agree that it is confusing, but unfortunately the editor has to figure out how to make it work anyway, since the genus/species parameters are automatically italicized. Treating them differently just adds confusion, I think this is the worst possible scenario. Sure, they didn't cause contention before; it's really not that big of a deal – I doubt anyone cared or thought about it much until now. I also agree that if a change like this is ever made that breaks articles, that those articles that are broken should be fixed asap – I actually thought that had happened, since the article I happened to be looking at a few days ago was updated. "Just as it is easy to go through and change all faulty occurrences one way, so it is easy to change them the other way. That's no argument for either system." – exactly! ErikHaugen (talk) 23:10, 8 November 2010 (UTC)


 * Editors are already used to using single quotes to italicize entries. I can imagine someone creating an automatic taxobox with italics (like the taxobox requires) and being utterly confused why it looks weird. They may also not have the foggiest idea how to fix the special cases where noitalic would be required. I'd rather have it be easier and more intuitive to edit than not. Rkitko (talk) 23:39, 8 November 2010 (UTC)

I'd tried to add more automaticization as part of the move to the AutoTaxobox template; probably, the presence or absence of auto-italicization should be consistent between the two templates, so I suppose that it ought really to be discussed primarily at (the more widespread) Template:Taxobox; a partial adoption of the code in Automatic Taxobox only might be part of an implementation strategy, but it seems better not to hide away the discussion here. Martin  (Smith609 – Talk)  23:29, 8 November 2010 (UTC)


 * It had been briefly discussed earlier here: Template talk:Taxobox/Archive 10. Not many opinions, but a lukewarm reception anyway. Rkitko (talk) 23:39, 8 November 2010 (UTC)

Parent taxa
Is there any reason not to force all boxes to display at the the immediate parent? I'm having to add the display_taxa = field on nearly every taxonox, including ones I'd already automated that didn't have this issue in the past. For example, the displayed parent for Superfamily Ornithocheiroidea is showing up as Order Pterosauria by default, rather than its immediate parent taxon Suborder Pterodactyloidea. Forcing it to display parent taxa is also displaying the grandparent and ggranmdparent unranked clades, which isn't really necessary. I think for ease of navigation among articles, immediate parent taxa are crucial to keep people from jumping around the tree seemingly at random. MMartyniuk (talk) 02:46, 20 December 2010 (UTC)
 * Oops, I knocked out the default when modifying the code to support the "display parents" pseudonym for display_taxa. I've restored the default value to 1.  Thanks for catching this.  Martin  (Smith609 – Talk)  11:27, 20 December 2010 (UTC)

resolved

Editing assistance
I've implemented a new system at Template:Taxonomy/Marrella that makes it easier to access parent taxa from each page. In order to roll this out, it is necessary to modify the page code from {{taxonomic unit|{{{1}}} to {{don't edit this line {{{machine code|}}}|{{{1}}} This avoids template loop problems. The simplest way to convert existing pages would be to use a bot; I just thought I'd mention this here before posting at WP:BRFA so that people know what's going on and can comment. Martin  (Smith609 – Talk)  21:20, 6 November 2010 (UTC)
 * Truly righteous. Some templates still have the {{#switch:{{{1}}} style first line, I'm assuming your bot will convert those also? Thanks! ErikHaugen (talk) 22:43, 6 November 2010 (UTC)
 * Ooh good idea! I'll look out for those too.  (Do you happen to remember any that do, so that I can get an exact match for the formatting?)  Martin  (Smith609 – Talk)  23:08, 6 November 2010 (UTC)
 * Here is one: Template:Taxonomy/Eukaryota. ErikHaugen (talk) 02:35, 7 November 2010 (UTC)
 * Thanks. Martin  (Smith609 – Talk)  13:42, 8 November 2010 (UTC)
 * {Now listed for comment at WP:Bots/Requests_for_approval/Taxobot_5.) Martin  (Smith609 – Talk)  23:17, 6 November 2010 (UTC)
 * I see more characters in that string of text. Will this hurt the taxobox load capacity? Bob the Wikipedian {{sup|(talk • contribs) }} 03:20, 7 November 2010 (UTC)
 * It's the template depth that affects the capacity; character count has a negligible impact on performance. Martin  (Smith609 – Talk)  13:42, 8 November 2010 (UTC)


 * There's some kind of problem – Template:Taxonomy/Rotheca_serrata doesn't show the genus and subfamily. Template:Taxonomy/Ajugoideae shows a weird error. ErikHaugen (talk) 04:48, 7 November 2010 (UTC)
 * Hmm. It looks like there's a problem in the way that WP parses parameters passed in a template name; the template works fine when subst'd but not otherwise.  I'm not sure how to get around this impasse.  Let's not roll out the template any further until we can come up with a way... Martin  (Smith609 – Talk)  14:34, 8 November 2010 (UTC)
 * I believe that I've resolved this error (caused by unbroken chains of }s being misinterpreted by the parser); the {machine code|} section should again be used by default. Martin  (Smith609 – Talk)  15:03, 24 November 2010 (UTC)

{{done}} & {{tl|resolved}}Martin  (Smith609 – Talk)  18:59, 7 December 2010 (UTC)}}

Edit button
Please point the edit button on the scientific classification box to the taxonomy page of the current taxon. Now that the ancestory is showing on the right side, this will allow the user to pick the templates. —  Ganeshk  ( talk ) 17:54, 7 November 2010 (UTC)
 * I'll do that once the ancestry is showing on the right side of all automatic taxoboxes (i.e. once WP:BRFA has been run). Martin  (Smith609 – Talk)  13:22, 8 November 2010 (UTC)
 * ✅. resolvedMartin  (Smith609 – Talk)  18:59, 7 December 2010 (UTC)}}

API based browser
I made a little ajaxy page for quickly finding child taxa, etc. It has a lot of issues (I'm a javascript noob) but seems helpful for checking for certain kinds of errors: taxobrowser.erikhaugen.com. ErikHaugen (talk | contribs) 17:35, 24 November 2010 (UTC)
 * Thanks, Erik! Does it use a template-filtered "what links here"? I've found that method is amazingly helpful...but minimizing the clicks would be even more helpful-- in fact, is a link to the template-filtered "what links here" something we could implement directly on the taxa's pages? Bob the Wikipedian (talk • contribs) 03:35, 26 November 2010 (UTC)
 * By "template-filtered what links here", do you mean a list of child taxa? There's already a "list immediate children" link on taxon pages. (It's red, but don't let that put you off!)  Martin  (Smith609 – Talk)  04:44, 26 November 2010 (UTC)
 * Bob – yes, it does an API request to query for a list of backlinks, like this for Cetacea, for example. As far as I can tell, yes, this is identical to a "what links here" filtered by namespace:Template. Martin, your category scheme would probably be less flaky, maybe I should do this instead – but I think Bob is asking for those results to be transcluded onto the taxonomy template page directly so you don't even have to click anything? I don't know how to transclude categories so that a list of members shows up, let alone do it when the category doesn't even technically exist. I suppose we could do some javascript there to hit the API to get the items in the "immediate children" category? ErikHaugen (talk | contribs) 07:28, 26 November 2010 (UTC)
 * As far as I can tell, such a list would have to be hard-coded. Down the line, I am thinking of coding a bot to maintain such lists; these could also be displayed in taxoboxes where "subdivision" is unspecified.  For now, your tool is probably the best thing we have.  Martin  (Smith609 – Talk)  13:26, 26 November 2010 (UTC)


 * Actually, I just meant a link. No need to transclude it. Actually, I just noticed that link already appears on the page, although it's in red. Looks good. Bob the Wikipedian (talk • contribs) 17:52, 26 November 2010 (UTC)


 * As if to prove my point, the child taxa thing just added broke my backlinks search. I'll try to change it to the category search. ErikHaugen (talk | contribs) 21:37, 7 December 2010 (UTC)
 * And fixed. ErikHaugen (talk | contribs) 21:57, 7 December 2010 (UTC)

resolvedand linked on doc

Taxobox core
I've just modified the template to use Template:Taxobox/core. If you notice any quirks associated with the transition, please let me know. Otherwise, in a couple of days, I'll make Template:taxobox link there too. Martin  (Smith609 – Talk)  23:12, 27 November 2010 (UTC)
 * ✅ resolvedMartin  (Smith609 – Talk)  19:00, 7 December 2010 (UTC)}}

Automatic generation of synonyms?
Following on Martin's work with adding authorities and child taxa to the template, why not go all the way and add synonyms as well? This way we have all relevant taxonomic data on the back end, and keep the page code to content-related issues like range, time span, pictures, etc. MMartyniuk (talk) 01:29, 8 December 2010 (UTC)


 * An interesting idea, pushing the synonyms into their own taxon templates and linking them to their successors....It's intriguing, that's for sure! Bob the Wikipedian (talk • contribs) 01:39, 8 December 2010 (UTC)


 * Why is the goal to move information about the taxon out of the page code? ErikHaugen (talk | contribs) 08:23, 8 December 2010 (UTC)


 * Erik's got a good point...will this simplify anything or make anything easier? Bob the Wikipedian (talk • contribs) 09:12, 8 December 2010 (UTC)
 * I agree; synonyms only need to appear on the article page, so that's the most logical place for them to be entered. Martin  (Smith609 – Talk)  14:55, 8 December 2010 (UTC)


 * I also think its not really needed, and was wondering about the same thing with the migration of the authorities information into the backend. How inclusive is the database that he authors would be pulled from?  Does it cover paleoentomologists, paleobotanists, etc... and how well?.-- Kev  min  § 23:36, 8 December 2010 (UTC)
 * I had written up a similar question in response to the authorities thing, but I didn't end up posting it because in some cases the authority will be used on two pages: a taxon's taxobox and the taxobox of its parent. Compared to the amount of duplication in the parent-child relationships that the taxonomy template system saves us this is obviously small potatoes, but there is something. (Also, for browsers of this tree like taxobrowser.erikhaugen.com, including authority information is more compelling than synonyms, I think, but that's not necessarily a valid argument so I'll put it in parenthesis.) ErikHaugen (talk | contribs) 23:49, 8 December 2010 (UTC)
 * How much duplication of authorities are there for parent/children? and how well does it deal with the different conventions between ICBN, ICZN, etc for authorities of taxa that have been moved or redescribed.  Plus that still leaves the question of how extensive the authorities database is that would be used, and how many would have to be moved by hand and corrected by hand.-- Kev  min  § 18:18, 9 December 2010 (UTC)
 * Can you give examples of cases where it doesn't deal successfully with the different conventions? Can you give examples where anything would need correcting or moving by hand?  There isn't much duplication but keeping automated edits out of mainspace seems to me to be more straightforwards; it makes the data readily available via API and keeps automatically-generated "child taxa" lists up to date.  This'll make maintenance more straightforward. Martin  (Smith609 – Talk)  18:49, 9 December 2010 (UTC)

I dont know of any situations as of yet, but I know that there are differences in the formatting of the authority names, and inclusion/exclusion of year etc... which at this point are handled fine with the taxoboxes the way they are. I haven't seen detailed explaination as to why the authorieies need to be in teh back end of the template. Usually they are only shown in one to a couple of articles at most, so unlike the taxonomy itself it should be very easy to make changes if they happen. I feel at this point that too much may be being moved out of the generally accessible area of the main editing space and moved into the more complex and less accessible template area without strong need to do so. I love the taxonomy part, but moving everything to the back end does not seem needed, or a good idea if we want people to use the taxoboxes. -- Kev min  § 20:13, 9 December 2010 (UTC)
 * Sounds like you're saying that the best solution would be to keep the authority information in the main article. The bot could then extract this information as it needed it elsewhere.  A bit more coding, but this sounds manageable to me. Martin  (Smith609 – Talk)  06:45, 17 December 2010 (UTC)
 * Just a thought; do we really need the authority information on the parent page, or just the taxon itself? Are there any strong arguments for its duplication? Martin  (Smith609 – Talk)  07:57, 17 December 2010 (UTC)
 * In circumstances of extreme doubt, I suppose, but those are few and far between, surely. I'll stand with Kevmin here. And yes, the formatting of the authority often includes parentheses to indicate the authority's name and year don't apply to the currently accepted taxon's name description. Bob the Wikipedian (talk • contribs) 02:45, 18 December 2010 (UTC)
 * resolved – authority names will only be generated in mainspace. Martin  (Smith609 – Talk)  21:56, 1 January 2011 (UTC)

Automatic generation of child taxa
I have a prototype system under construction to automatically generate a list of child taxa; this can be viewed at, for example, Template:Taxonomy/Cephalopoda. Once this is complete, I'm planning to roll this out so that automatic taxoboxes with no subdivision specified will display the children automatically. Of course, it will sometimes be necessary to over-ride this manually, and this option will remain. Any comments on the implementation or suggestions are welcome. Martin  (Smith609 – Talk)  18:52, 7 December 2010 (UTC)
 * Do you think there is any hope for dealing with dummy taxonomy templates? eg, Template:Taxonomy/Mammaliaformes_(Amniota), Template:Taxonomy/Physeteroidea_(displayed)? I had tried earlier to think of some kind of convention for making it clear what was going on with these guys, but I couldn't think of anything. But now that you are trying to automate the listing of children, it might be worthwhile to revisit this. Oh and nice work! ErikHaugen (talk | contribs) 20:13, 7 December 2010 (UTC)
 * I like where you're going with Template:Taxonomy/Physeteroidea/displayed. ErikHaugen (talk | contribs) 22:56, 7 December 2010 (UTC)
 * See the new template "Taxon variant" (i.e. "same as taxon"); see Template:Taxonomy/Test-47/override. Comments welcome. Martin  (Smith609 – Talk)  23:09, 7 December 2010 (UTC)
 * There we go. Our convention should be to use a /dummy-details after the dummy taxon; see Template:Taxonomy/Test-48, who is correctly categorized in Category:Immediate children/Test-47 even though the parent is Test-47/override.  Template:Taxon variant is a nice additional twist but not now necessary.  Martin  (Smith609 – Talk)  23:15, 7 December 2010 (UTC)
 * Great idea Martin. But could we change the heading to something like "Subgroups"? "Immediate children" seems a little inaccessible of a term for the average reader. MMartyniuk (talk) 23:32, 7 December 2010 (UTC)


 * Done at Template:Taxonomy list. Martin  (Smith609 – Talk)  16:14, 8 December 2010 (UTC)
 * This looks great. Nice work – test-47's children include the children of test-47/override – perfect! Do you have any thoughts about what the convention should be for the name of the part after the last slash? Some name describing the purpose, like Template:Taxonomy/Physeteroidea/always_display and Template:Taxonomy/Mammaliaformes/skip or something more regimented? Thanks! ErikHaugen (talk | contribs) 23:34, 7 December 2010 (UTC)
 * A couple issues. 1.) It would be nice if Template:Taxonomy/Mammaliaformes showed Mammalia as a child and if Template:Taxonomy/Physeteroidea/displayed had incertae sedis as a child. 2.) Template:Taxonomy/Physeterida has a "template loop" detected thing on it; I'm not sure why. ErikHaugen (talk | contribs) 00:48, 8 December 2010 (UTC)
 * These lists only update when I run a script, which I currently do every now and then. I'll set up an automated schedule to run every... day? hour?  How often is appropriate?  Martin  (Smith609 – Talk)  16:14, 8 December 2010 (UTC)
 * Oh; I see. In that case, it probably doesn't matter too much how often. I guess it would depend on how expensive it is to run. Is it possible to run it just on one taxon, so people can poke around after templates are made to verify that it was done correctly? ErikHaugen (talk | contribs) 18:08, 9 December 2010 (UTC)
 * I'm not sure it's the best idea to make display the default when  is not used. There are about 30,000 (?) Polbot taxa articles and none of them used the subdivision parameter. They could have too many child taxa to fit neatly into an automatic taxobox. Would there be a way for the template to know if child taxa number greater than, say, 15?, then don't automatically display them when subdivision is unused? Rkitko (talk) 23:33, 7 December 2010 (UTC)
 * Perhaps the template should default to not displaying the subdivisions, and allow the templatist to set the value to true if they so choose. This would prevent incomplete listings from displaying without someone's knowledge of it. Bob the Wikipedian (talk • contribs) 01:24, 8 December 2010 (UTC)

Feel free to experiment with, on a "this could break at any time" basis, the new parameter child depth. Setting it to  will display immediate children;   or   will show grandchildren and beyond. Please let me know of any cases that you find where this doesn't work. (Note that taxobot is currently modifying the back-end syntax and has got to taxa beginning with D at time of posting; this means that Cephalopod is a good place to experiment, and you can check Special:Contributions/Taxobot_1 to see its current progress.) Martin  (Smith609 – Talk)  18:10, 18 December 2010 (UTC)
 * Ooooo....collapsible....I love it! But can we wikify it? Let's give it a 0px border and kill the bold lettering. Bob the Wikipedian (talk • contribs) 07:39, 19 December 2010 (UTC)
 * Sorry to confuse; Cephalopod doesn't use the new template yet; it's best not to use it in article space whilst its in its experimental phase. Collapsability would be (relatively) easy to implement when a certain size was exceeded, if desired.  Martin  (Smith609 – Talk)  14:13, 19 December 2010 (UTC)
 * display_children seems more intuitive, following the same format as display_parent. I'm placing the said taxobox in this discussion so it can be scrutinized more easily. Obviously, there's either a flaw with it, or I've misimplemented it. Bob the Wikipedian (talk • contribs) 19:19, 19 December 2010 (UTC)
 * Ah, it won't display outside of article space. I wonder whether I can amend this... Martin  (Smith609 – Talk)  19:56, 19 December 2010 (UTC)
 * display children it is! And it now displays in any namespace. Martin  (Smith609 – Talk)  20:34, 19 December 2010 (UTC)
 * Hmmmmm it appears to work rather nicely, but it could use some collapsible tree structure to it and also some indication of the rank of the descendent taxa. I'm not sure of the best way to indicate that, though. Bob the Wikipedian (talk • contribs) 05:57, 21 December 2010 (UTC)
 * How about we use the subdivision_ranks parameter to display the ranks? I've amended this taxobox accordingly.  I should imagine that setting 3 will be quite rare; most taxoboxes that I see display one or at most two levels of children.  Martin  (Smith609 – Talk)  12:15, 23 December 2010 (UTC)
 * You can also experiment with the newly supported 3 collapse to enable collapsing. Martin  (Smith609 – Talk)  12:35, 23 December 2010 (UTC)
 * Impressive! A second level of collapsing might make sense for 3 levels. As for the subdivisions, I think subdivisions would be a better choice (per the wording) than subdivision_rank. Great job, and also Merry Christmas to everyone, I'll be back after the holidays. Bob the Wikipedian (talk • contribs) 19:35, 23 December 2010 (UTC)
 * Thanks! subdivision_rank is what's currently used by Template:Taxobox, clunky though it is.  I guess that it avoids confusion with subdivisio and subdivision.  Double-collapsibility isn't straightforward to implement, and might be beyond me; are you sure that it'll be worth it?  I've not seen anything like it implemented on WP before, so I'm not sure whether there'll be a demand for it. Martin  (Smith609 – Talk)  23:50, 23 December 2010 (UTC)
 * You're right, double-collapsibility probably isn't necessary. In testing this out, it looks like it requires the client to generate a children template. I'm not sure that's making things easier. Is it? Bob the Wikipedian (talk • contribs) 04:51, 25 December 2010 (UTC)


 * The children templates are bot-generated. The bot only runs on my command at the moment, but will be set to automatically run once a couple more bot approvals have been passed.  Martin  (Smith609 – Talk)  19:09, 26 December 2010 (UTC)
 * Historical note: Mass generations of taxon articles by bot have often ended in mass AfD runs, especially with plant and algae articles. --EncycloPetey (talk) 20:18, 26 December 2010 (UTC)
 * I don't think anyone has suggested making new articles from this info; this is all back-end work on the Template/taxon. Rkitko (talk) 03:04, 27 December 2010 (UTC)
 * Taxobot task 7 has been approved, so this is now resolved and ready to go! Martin  (Smith609 – Talk)  03:45, 3 January 2011 (UTC)

Authorities
It is possible to automatically add authority information to many taxa that add it. I've coded a bot to do this. Comments are welcome at Bots/Requests_for_approval/Taxobot_6; comments on the implementation can be made here. In brief, I plan for taxoboxes with no authority information coded to consult the database to list it; this information would also be displayed in subdivision lists. Martin  (Smith609 – Talk)  19:20, 7 December 2010 (UTC)
 * Hi Martin, I'm testing the authority field on Herrerasauridae and it doesn't seem to be working. The information is in the database page but it's not displaying in the article on preview when I delete the authority fields there. Am I doing something wrong? MMartyniuk (talk) 01:37, 8 December 2010 (UTC)
 * I hadn't enabled it yet! It works now.  Thanks for your testing! Martin  (Smith609 – Talk)  15:04, 8 December 2010 (UTC)
 * Well that explains it, sorry for jumping the gun! I saw it appearing as an unused field on individual template pages so I dove in ;) MMartyniuk (talk) 00:46, 9 December 2010 (UTC)
 * In the light of the discussion above, this isn't beneficial, so I'm un-installing it. If you feel inclined, please do check that the information appears on the article page, and remove it from the template. Martin  (Smith609 – Talk)  16:40, 18 December 2010 (UTC)
 * Uninstalled? I'm still seeing it on the readouts for the templates. Bob the Wikipedian (talk • contribs) 20:02, 31 December 2010 (UTC)
 * I couldn't uninstall it whilst it was still in use. I've taken it out now &mdash; thanks to the kind souls who have removed all the depreciated parameters!  Martin  (Smith609 – Talk)  21:54, 1 January 2011 (UTC)

resolved

Taxobot
Now Database reports/Transplanted user templates has exploded with Taxobot pages. I suppose the bot that updates this page could ignore Taxobot, but it seems like there may be another solution. I just noticed this after seeing this comment on Taxobot's page. Plastikspork ―Œ (talk) 04:26, 18 December 2010 (UTC)
 * What does this mean, though? Is it a bad thing? Bob the Wikipedian (talk • contribs) 05:15, 18 December 2010 (UTC)
 * The usual convention is that we keep user content separate from articles, since having something in your own user space implies ownership. So, for that reason, we have this database report to trackdown people using personal stuff in article space.  Now, what Taxobot is doing isn't exactly the same thing, but I don't see why Taxobot's stuff has to be in it's own userspace, and couldn't be in the template namespace.  This would certainly make the database report more useful.  Plastikspork ―Œ (talk)  05:37, 18 December 2010 (UTC)
 * It's in userspace whilst I experiment. Editors' contributions will currently be overwritten by the bot, and I don't want to waste people's time.  Once it's stable, I'll file a bot request for it to operate in mainspace. Martin  (Smith609 – Talk)  16:19, 18 December 2010 (UTC)
 * I see. Sounds like perfectly sound idea, especially if the long term plan is to move it into template space.  Cheers. Plastikspork ―Œ (talk)  20:16, 18 December 2010 (UTC)
 * Any sort of timeline on this? If it's not long that's fine, but I don't want this template to essentially make at least 2 database reports useless for the next month or whatever. VegaDark (talk) 06:48, 20 December 2010 (UTC)
 * Very difficult to estimate, I'm afraid. Could easily be a month or more. (or less!) Martin  (Smith609 – Talk)  08:38, 21 December 2010 (UTC)
 * I've just filed a bot request; as soon as this is granted I'll be able to move the templates into article space. Comments welcome there. Martin  (Smith609 – Talk)  12:09, 23 December 2010 (UTC)
 * Approval granted, and pages being created in article space. Martin  (Smith609 – Talk)  04:05, 3 January 2011 (UTC) resolved

Request for comments
We'd be very grateful to any editors willing to take a look at the template's code to spot any glitches that we might have missed, and any other feedback; any suggestions received now will be much easier to implement than when the template is rolled out over thousands of pages. Please comment here if you're able to help. Thanks! Martin  (Smith609 – Talk)  15:05, 27 December 2010 (UTC)


 * I'm also available to assist users. Bob the Wikipedian (talk • contribs) 19:58, 27 December 2010 (UTC)

Hi, Wikid77 here. I noticed the following issues: Those are some issues to consider. -Wikid77 (talk) 02:36, 30 December 2010 (UTC) resolved
 * Internal HTML comments can be added: The template coding should be explained, inside, with more HTML comments. I have checked the formatted output from many templates, so I can confirm that templates can have many internal HTML comments (even doubling the size of the template), but those comments will be omitted from the formatted page, unless the template is subst'ed by a user invoking it as {&#123;subst:Automatic Taxobox|...}}. If you wish to have a complex template, it is imperative to help other editors to feel confident in maintaining that template, such as adding many explanatory comments inside.
 * ✅ Martin  (Smith609 – Talk)  22:10, 1 January 2011 (UTC)
 * Changing to an Infobox wizard: Should this template be changed into being an infobox-wizard? It could simply display the copy/paste template parameter list for invoking the regular {Taxobox} template, and perhaps also report other suggestions for a user. In that mode, the speed and size would not be an issue, because it would only be used to generate the one-time coding parameters for using the older {Taxobox} template. This also means that it would not have to "be perfect" because an editor would be expected to continue editing the suggest wizard output for invoking the {Taxobox} template.
 * I'm not sure that I understand what an infobox wizard is. Martin  (Smith609 – Talk)  22:10, 1 January 2011 (UTC)
 * Somewhere I've seen a template that you "subst:" and then it gives you instructions for completing it...if I recall correctly, it walks you through several subst: edits adding bits of text here and there. Bob the Wikipedian (talk • contribs) 03:26, 2 January 2011 (UTC)
 * I see. How would that enable the taxoboxes to rely on a central taxonomy?  Surely the point of this template is that it's not hard-coded.  Martin  (Smith609 – Talk)  14:02, 2 January 2011 (UTC)
 * I agree...this template is fairly straightforward. Anyone who takes the time to read the edit notice should be able to set up an accurate template. Wizard help would take longer. Bob the Wikipedian (talk • contribs) 00:48, 3 January 2011 (UTC)
 * Use more efficient string templates: There might be a need to use ultra-efficient string-handling templates, because some complex templates simply generate the wrong results when using deeply nested string templates, such as using str_find. I am working on a new strfind_short which can search for the same strings using a shorter method, as using only 5 levels of logic nesting, rather than the 18 of 40 total levels used by {str_find}.
 * This hasn't caused problems yet, but if you get strfind_short working, then we'll definitely be up for using it! Martin  (Smith609 – Talk)  22:10, 1 January 2011 (UTC)
 * Although the need to do so is rare, you can't put four of these on one page. Bob the Wikipedian (talk • contribs) 03:27, 2 January 2011 (UTC)

Category:Taxoboxes using display taxa
Is display_taxa being deprecated? Bob the Wikipedian (talk • contribs) 03:34, 2 January 2011 (UTC)
 * The user changing most of these is also removing the template -italictitle- from autoboxed pages for no stated reason. What gives? MMartyniuk (talk) 03:37, 2 January 2011 (UTC)
 * Yes, it's being re-named to the more intuitive "display parents", for consistency with "display children". (See ).
 * The taxobox automatically italicizes the template, therefore it is confusing to editors to add an additional, unnecessary "italic title" template, which would need removing if the page were moved to a common name, for example. Martin  (Smith609 – Talk)  13:50, 2 January 2011 (UTC)
 * The parameter may now be deprecated fully after someone double-checks that I've cleaned up the last of the violators. Bob the Wikipedian (talk • contribs) 00:52, 3 January 2011 (UTC)
 * Thanks, Bob! ✅ Martin  (Smith609 – Talk)  00:59, 3 January 2011 (UTC)

resolved

Template error from last update
The last edit is apparently causing an error where "Unexpected < operator Expression error: Unexpected < operator" is splashed across the taxobox in numerous articles (see Indri and Aye-aye) and on the template page itself. I looked at the code change, but couldn't see what would cause the problem. –  VisionHolder « talk » 00:39, 9 January 2011 (UTC)


 * Thanks, that was my fault. I'll get right on it. Bob the Wikipedian (talk • contribs) 01:02, 9 January 2011 (UTC)


 * It should be working now even better than before. The last edit was supposed to add support for underscores in certain parameters, but I managed to leave out a single piping symbol, throwing the whole thing out of whack. Amazing what a single piping symbol can do. Shame on me, and thanks again! Bob the Wikipedian (talk • contribs) 01:12, 9 January 2011 (UTC)

resolved

Deinonychus
I was looking for a good article to use as I am working on the step-by-step tutorial, and stumbled over Deinonychus. And what a fall it was when I stumbled over that one. Martin, I have a hunch the new fossil range code might have something to do with it. At first I thought it had to do with my latest revisions to the automatic taxobox to support consistent formatting of parameters, but it appears to be deeper than that. Bob the Wikipedian (talk • contribs) 04:59, 9 January 2011 (UTC)
 * resolved (diff). Martin  (Smith609 – Talk)  16:41, 9 January 2011 (UTC)
 * And I've implemented a check to categorize pages with erroneous colours into Category:Automatic taxobox cleanup in future. Martin  (Smith609 – Talk)  16:52, 9 January 2011 (UTC)

New deprecation template
I've put together a deprecated taxon template to make deprecation notices on taxon templates simpler and easier and also to provide an easy-access link to check for orphan taxa (although I'd still like to see a more effortless way to discover orphans). Bob the Wikipedian (talk • contribs) 23:23, 30 December 2010 (UTC)
 * What exactly do you mean by "orphan"? Can you give an example (perhaps in the Test-## suite)? Thanks, Martin  (Smith609 – Talk)  01:01, 3 January 2011 (UTC)


 * On Wikipedia, an orphan article is usually one that is not linked to by other articles. I assume then that an orphan taxon is one not linked via a parent or child taxon.  I come across that sometimes, where a species page was created, but the genus article does not yet exist. --EncycloPetey (talk) 01:06, 3 January 2011 (UTC)
 * Wouldn't all of these be listed at Category:Automatic_taxobox_cleanup under &mu;? (I've created Template:Taxonomy/Escherichia_coli as an example of this kind of orphan) Martin  (Smith609 – Talk)  02:35, 3 January 2011 (UTC)
 * Correct you are. I was actually referring to taxa whose parent has been killed, either via a move or a deletion, was the case with one of the bugs Kleopatra reported last week. I've set up a test scenario at Template:Taxonomy/Test-29 by linking it to a deprecated template (a template with a notice on it directing the user to use a different parent). Looks like it does indeed show up under the error category. Bob the Wikipedian (talk • contribs) 04:44, 3 January 2011 (UTC)
 * resolved (?) Martin  (Smith609 – Talk)  22:17, 17 January 2011 (UTC)

Preview please; delay release until error rate down
Please use preview every time you add an automatic taxobox.

The automation to Graptolite's taxobox gives this big ugly red error:

Taxonomy not available for Graptolite; please create it automated assistant Unrecognised rank: fix

Template:Child taxa/Graptolite"

As someone else pointed out, not all errors are catchable with a preview, but let's eliminate all errors that can be caught by previewing the article when the automatic taxobox is added.

I don't think automatic taxoboxes should be released fully until the experienced users get the template error rate down.

--Kleopatra (talk) 15:45, 4 January 2011 (UTC)


 * Sorry, I must have pressed the wrong button there; I've been previewing before saving as a matter of habit. I'd be interested to know how you are quantifying error rate and what level you would deem acceptable. From my recent experience, I've found the error rate to be higher in existing taxoboxes, many of which look to have been copied and pasted from a parent taxon without modification of e.g. the name parameter).  Martin  (Smith609 – Talk)  16:53, 4 January 2011 (UTC)
 * I can't run with the "other trash," on this. The selling point of automatic taxoboxes is they can be better, much better than the current method. So far, they're not, and the ugly red screaming bold font that declares they are not is seen far too often. Can we do something so the automatic taxobox is not changed when there is an error? I don't think so. I still can't figure out how to just fix them. I don't see the answer, but the problem remains. --Kleopatra (talk) 05:04, 5 January 2011 (UTC)
 * Would it be fair to interpret this comment as a criticism of my ability to use the preview button, rather than of the template? Martin  (Smith609 – Talk)  22:38, 6 January 2011 (UTC)


 * I was going to blame it on you, but that didn't seem like a very nice thing for me to do Bob the Wikipedian (talk • contribs) 22:58, 6 January 2011 (UTC)
 * Your failure to use the preview button is a discussion for your talk page. It would make the feature, automatic taxoboxes, appear more robustly programmed, though, if your failures to preview didn't show up on top of other errors.
 * There's simply no way to understand how to use this feature. If that was the only matter, failure to use a particular line of code, it seems a clear-cut instruction manual and commentary would avert that problem or at least make it easy for the editor to see the issue. How about a big bright red "missing taxon" tag like when you fail to close your html tags? Readily seen in preview mode, easily corrected by anyone. --Kleopatra (talk) 05:21, 7 January 2011 (UTC)
 * Just out of curiosity, what would HTML tags be doing in it? I've not had a need to use any. Bob the Wikipedian (talk • contribs) 21:11, 7 January 2011 (UTC)
 * When you edit a wikipedia page and you preview, if you have any missing HTML tags (nothing whatsoever to do with automatic taxoboxes), you get bright red error messages. So, how about a bright red error message such as missing taxon when you hit preview for automatic taxoboxes, like you can already get an obvious and easily catchable error message when you hit preview when editing wikipedia articles and omit an HTML tag. --Kleopatra (talk) 17:50, 9 January 2011 (UTC)
 * I'm not quite sure how this error message would differ from the error message created by Template:create taxonomy that displays when you try to create an automatic taxobox for a missing taxon (see this example)? Martin  (Smith609 – Talk)  18:18, 9 January 2011 (UTC)
 * Did an error message display? --Kleopatra (talk) 19:48, 9 January 2011 (UTC)
 * Looks good to me. A big, bright, red error that said there was no taxonomy available for anything called "Missing taxon". You can also see a very ugly error message if you use a deprecated taxon. ~Example~ Bob the Wikipedian (talk • contribs) 22:19, 9 January 2011 (UTC)

resolved

How to display synonyms
? --Kleopatra (talk) 21:05, 9 January 2011 (UTC)


 * Try adding this to the taxobox (before the closing braces)

|synonyms=
 * Genius Baggins 1934
 * Geniosus Harvey 1734


 * Bob the Wikipedian (talk • contribs) 21:48, 9 January 2011 (UTC)

resolved

Template:Taxonomy refs
Okay, I've basically just begun using this parameter (in honor of requests that I begin citing my sources for parent taxa). Would someone have a look at Taxonomy/Salicaceae? My goal is to allow the reference to appear in the Salicaceae article when the taxobox is automated, but it's giving me a weird error in the reflist when I preview that says the ref tag is empty...I think it's got something to do with the way I've included the reference in the template. Bob the Wikipedian (talk • contribs) 05:18, 17 January 2011 (UTC)
 * The intention of the ref parameter is so that the template pages display reference information; they are not intended to be displayed in articles. See fix. Martin  (Smith609 – Talk)  22:12, 17 January 2011 (UTC)


 * When citing parent taxa in Angiosperms, use APG III, not on-line databases. We use the APG III system for angiosperm classification at all major ranks of family and above. --EncycloPetey (talk) 22:26, 17 January 2011 (UTC)

resolved

Translator bot
Would it be possible to set up a translator bot for any other languages who might be interested in adopting the relational database we are using? I just realized the other languages don't have access to this powerful system we're enjoying. Bob the Wikipedian (talk • contribs) 18:19, 2 December 2010 (UTC)
 * What "relational database" are you talking about? ErikHaugen (talk | contribs) 18:40, 2 December 2010 (UTC)
 * The pages that begin with "Template:Taxonomy/" are all elements of a relational database. They're scattered across Wikipedia's servers, but it's still a relational database because they've got links to one another. Bob the Wikipedian (talk • contribs) 19:03, 2 December 2010 (UTC)
 * I just realized that this would require logic to be adapted to match the transwiki taxobox templates. However, once the appropriate templates are added to the other wikis to allow full compatibility with the taxon templates we use, translator bots could be implemented to simply copy the taxon templates. Bob the Wikipedian (talk • contribs) 01:20, 8 December 2010 (UTC)

display children
Species and genus names are currently not italicized. Also, call me oblivious but I'm not sure how to generate a children template for a given taxon. Bob the Wikipedian (talk • contribs) 05:45, 3 January 2011 (UTC)
 * Not sure whether this is related, but the appearance of the documentation suggests that a template limit has been exceeded at Template:Automatic taxobox. Martin  (Smith609 – Talk)  22:52, 3 January 2011 (UTC) (see below)
 * P.S. where not? Paradycheia looks fine in the taxonomy, title and page title.  Martin  (Smith609 – Talk)  22:54, 3 January 2011 (UTC)
 * P.P.S. Child templates can be generated by going to Taxonomy/Taxon-name and clicking "Update" beside the Subgroups subtitle (in the taxonomy list). Martin  (Smith609 – Talk)  22:55, 3 January 2011 (UTC)
 * Hmmmm. I'm not immediately recalling what I did to generate that error; sorry for the poorly documented error there. However, I spy something wrong with the logic generating this box:


 * Can you spot it? Bob the Wikipedian (talk • contribs) 05:10, 4 January 2011 (UTC)
 * Are you referring to the box's guessing of subdivision_ranks? This still requires manual input if the children aren't the obvious.  I could create another set of bot-maintained templates listing all the childrens' ranks but this might be overkill. Martin  (Smith609 – Talk)  13:26, 4 January 2011 (UTC)
 * A wise guess. Shall we give it a try? The rollback links seem to be great friends right now... Bob the Wikipedian (talk • contribs) 06:36, 5 January 2011 (UTC)

Any luck being able to italicize genera and species in mixed-rank children lists? MMartyniuk (talk) 04:13, 26 January 2011 (UTC)
 * Doesn't this happen automatically when you specify display children? (Perhaps an existing subdivsion parameter is overriding the automatic generation?)  Martin  (Smith609 – Talk)  15:21, 26 January 2011 (UTC)

I'm marking this as "resolved" since it was a false alarm... Bob the WikipediaN (talk • contribs) 16:18, 29 January 2011 (UTC) resolved