geographic interfaces to OLAC data

[Jeff Good]

Hello,

I passed this note along to Michael Cysouw who I knew had been  
thinking about such things. Here's his response:

> The basic search that I would like to see should combine  
> geographical and genealogical relations, like:
>
> - give me all information available about the nearest 50 languages  
> to this one language
> - give me all information available about the 10 genealogically  
> most closely related languages
> - give me all information available , ordered by genealogical  
> relationship, and within that ordered to geographical relationship
>
> Visual output is always nice, but for me not really essential.
>
> Ideally, we would like to have geographical information prepared  
> for linguistic querying, but I still haven't found a geographer  
> that would like to help me with providing such information.  
> Basically, we would need two distance-measures for two point on the  
> globe: A) how much time does it take to walk from point X to Y?  
> Measure B) how much time does it take to go by boat from point X to  
> Y? Then we could ask questions like:
>
> - give me all information available for the downriver languages,  
> ordered by distance along the river.
> - give me all information for nearest coastal languages, measured  
> by distance along the coast.
>
> etcetera

I don't know if OLAC will have genealogical data, too--but if/when it  
does, this is something to think about.

The issue of how to measure distance between languages is an  
interesting one that I don't think there's much consensus on.  
Obviously, one could start with a very simple metric, but, in the  
long run, more complex ones might be nice, too. But, as Michael  
points out, that will require some outside help.

Jeff