Re: the viability of translators, and Guile itself

["Ian Bicking" <bickiia@earlham.edu>]

> But, back to translators: if they're such a great idea, how come they
> don't exist?  This is a very embarrassing question, especially since
> I'm advertising them as a feature on the Guile web page.

You cannot expect that by building the infrastructure for translation 
that people will write translators.  There has to be a will to build 
translators (and it has to be _willed_ -- this isn't what people do to 
scratch their itch).  Translators must be *part* of Guile, not 
something that Guile enables.  This doesn't mean they need to be 
integrated into the interpreter, just that they need to be part of the 
Guile pages, Guile CVS, and all that. 

Also, translators are targetted at people who don't care about 
Scheme, Guile, or extension languages in general.  They are 
meant to enable people to extend their favorite app without being 
forced to learn another language.  Before a translator becomes 
generally useful in this context it will have to be fairly polished -- 
easy to install, available to all applications, easy to specify in 
scripts, etc.  Otherwise the difficulty of using the translator will 
negate its own purpose.

Polishing the translator is probably as hard as writing it in the first 
place and involves a lot of skills that have nothing to do with 
language implementation.  In the likely case where the person who 
starts the project isn't up to finishing it, will someone else bring the 
translator to completion, or at least fill in the gaps?


Right now the Guile community has a number of concerns, few of 
which seem to be very related to translation -- particularly among 
the core developers.  As evidence, all the other replies to this 
thread don't have anything to do with translation.  [the same thing 
that happened the last time someone brought up translation -- 
people started talking performance]  Is there a will for this goal?  
Not a curiosity -- which isn't enough to carry something through to 
completion.  Not an obligation -- which created the prototype of 
CTAX, but wasn't based on an intention or desire to make a useful 
translation.  Something more is needed.


--
Ian Bicking <bickiia@earlham.edu>