Dead links for taxonomic names

I think a rigid algorithm might be fruitful enough to be worth considering:

For entry with dead link

1. If project link is in Translingual and Translingual section contains {{taxon}} then
   1. If headword is not one word (ie no spaces) then
      1. Link to first word in headword
      2. If link to first word in headword fails then
         1. Link to parameter 3 in {{taxon}}
         2. If link to parameter 3 fails then
            1. Categorize as problem and next
   2. Else link to parameter 3 in {{taxon}}
      1. If link to parameter 3 fails then
         1. Categorize as problem and next

(I realize this isn't the simplest way to do the logic and omits some elses, mostly obvious ones, I hope.)

This should dramatically reduce the number of these that are categorized as problems for manual resolution.

WP does not consistently use taxonomic names for organism entries so there may be many fails. This type cannot be resolved automatically AFAICT. Some taxonomic names are homonyms of others. Species and WP both have varied ways of disambiguating. I don't think these are worth trying to resolve either. automatically. DCDuring TALK 13:44, 7 March 2016 (UTC)

If there is a dead species or WP link not in a Translingual section and the link is to what looks like a multiword taxonomic name, then trying to link to a sister project article whose headword the first word in the link would be worthwhile. Perhaps the simplest approach would be to see whether there was a species link and then see whether the WP link was to the same headword or to a truncation of it. DCDuring TALK 13:50, 7 March 2016 (UTC)

Most(?) Translingual entries do not have Hypernyms sections. I think sequentially testing those is not such a good idea as a too-distant point on the tree of life is not worth it.

Thanks for documenting that, I have been thinking about starting up a tools project for Wiktionary related stuff and this seems like a good candidate for that (so it would always be watching). I might have time to play around with that in the near future, I will be in touch again if I do to bounce ideas/questions off of you.

I have other tools ideas, most importantly, a KWIC display for any corpora we can get access to. We really need modern corpora.

I'm starting to use Special:Search/.... at WP and Species for terms they use but have no headword for. See Thallobionta