lexicore with 80 words

[LinguList]

@xrotwang and @FredericBlum, the fix is even smaller. We just need to say >= 80 words are enough to be included in LexiCore. Then we should be fine. Grollemundbantu has 100 concepts, it would also loose most of the data otherwise. So with grollemundbantu, we can then make a new version 2.1 and use that as the basis for the paper.

[xrotwang]

sounds good

[FredericBlum]

I have provided the doi in #85 . With the updated svg-image for the core-subsets, we can go ahead with re-running and releasing this.

[LinguList]

Okay!

[LinguList]

Ah, important, @FredericBlum, please delete hsiuhmongmien from the list (for now). It seems to be ignored again. I do not know why, let us better ignore it. GrollemundBantu should be fine enough,a nd we have quite a few hmong mien languages anyway.

[LinguList]

I know why: it seems, there is just < 80 words per variety.

[FredericBlum]

I have now re-run the cldfbench converison, removing hsiuhmongmien and adding grollemundbantu, as well as adapting the threshold in the image and text. @LinguList I think having a second person looking over the changes would be important, in case I have missed something? The Core-set numbers (including ClicsCore) have changed quite a bit, so we hould double check that everything is going as we planned.

[LinguList]

Looks fine and numbers make sense!