[wip] Tasks

[Wu-Urbanek]

• clean orthography profile
• language mapping
• concept mapping review
• wait for the real bib file
• update cldfs
• update test.py
• ask for review

[LinguList]

The concept list can be added to concepticon, this will be better. Wait, once I merged the data, so you can then check it and prepare a PR for concepticon.

[LinguList]

@MacyL, if the concept list can be interpreted as starostin, you can add starostin as the normal concept list, and you can make one exception for the one divergence. This is better than adding it to concepticon.

[Wu-Urbanek]

I would prefer not to add to concepticon, because it's basically the same concept list. I calculated the differences between the two, and there are about 5 concepts - 8 concepts (add the ones you mentioned in the PR).
My solution will be to hard coded a "dictionary" to fix these variations, it's a simple solution I can think off and it will not take too much time.
Another solution (I am not sure if lexibank can do this) is to have a etc/concepts.tsv with only these different concepts. And add something like conceptlist: ["Starostin-1991-100", etc/concepts.tsv] to the metadata?

[LinguList]

you didn't read my follow-up reply. Essentially, I proposed the same.  What you should do, however, is add the concepts with ID to the `concepts` variable, which we use as lookup. That's where this needs to happen.