Re: [Obo-format] OBO file character encoding

[Hilmar L. <hl...@du...>]

On Feb 7, 2008, at 11:49 AM, Chris Mungall wrote:

> I think the recommendation would be to avoid non-ascii where possible
> - most downstream consumers of obo files will react unpredictably


I would venture to suggest that meanwhile we live in an age in which  
i) most programming languages and libraries support different  
character encodings perfectly fine (for example, supporting a non- 
ASCII character encoding in Java is simply a matter of passing in an  
additional argument to the file reader constructor), and ii) in  
science we're collaborating globally. Needing to tell collaborators  
how they should specify their native-language names to fit the ASCII  
limitation doesn't feel that good, frankly.

Also, frankly, I would hate to have to entertain an argument of OWL/ 
RDF/XML vs OBO on the basis of character encoding support - my take  
is that that argument should be unfounded.

Maybe there is more involved than just putting an 'encoding' tag into  
the header, but it sounds unlikely that it's difficult to accommodate?

	-hilmar
-- 
===========================================================
: Hilmar Lapp  -:-  Durham, NC  -:- hlapp at duke dot edu :
===========================================================