[erlang-questions] A proposal for Unicode variable and atom names in Erlang.

[Loïc Hoguin]

On 10/22/2012 06:45 PM, Siraaj Khandkar wrote:
> My point with papers is that INRIA is a French national institution, so one
> would think it'd be very heavily biased towards using French everywhere. Yet it
> is not the case. Why? My understanding is that it is to widen the communication
> circle. Though incidental, English has become the lingua franca of computing
> (yes, a little pun is intended :))
>
> One of my interests is (or perhaps was) Archaeology, and in that field one is
> required to have a reading/writing proficiency in French or German in order to
> proceed with graduate studies, simply because (although incidental) those are
> the languages that a large body of important research has been published in.

Yes I completely agree for publications of computer (or any other) 
science, there should be a common language.

This doesn't change anything though. There will always (in my lifetime) 
be non-English speaking developers and if you want your language or tool 
to be used by them you have to accomodate.

This means Unicode in the source (variables are more optional but atoms, 
strings and binary strings pretty much required), translated 
documentation, translated tutorials and commercial support for each 
language.

> Though Fred is right to point out that this doesn't seem to have hurt other
> languages, I mostly side with Yurii's argument that there are other, much more
> important issues/features to tackle, amongst others: frames and native data
> types for pmods (I don't actually use those in Erlang, but I do enjoy a related
> feature in OCaml (though it's benefits in that context are different...)).

This change will take a few years to be completed, not because it's a 
lot of work, but because its support has to be phased into the language. 
I wouldn't think it's going to slow down anything significantly.

-- 
Loïc Hoguin
Erlang Cowboy
Nine Nines
http://ninenines.eu