[r6rs-discuss] JAR statement

[Alan Watson]

> I propose that R5RS be the baseline

I voted against R6RS. However, I recognize that there is a lot of good  
in the R6RS. I now use portable R6RS Scheme almost daily because for  
the overwhelming majority of my work, it is mor useful than portable  
R5RS Scheme. I more or less use R6RS Scheme as R5RS Scheme with a  
portable library system (and R6RS records instead of SRFI-9 records).  
I prefer to write in portable code, because the best Schemes for  
development are often not the best Schemes for production runs. Some  
of my projects require years of CPU time, so this is an important  
consideration for me. So, that's my bias.

We all have a finite amount of time to devote to Scheme  
standardization. Would it be better to use this time to "fix" the R6RS  
or to "rebuild" starting from the R5RS? My judgement is that fixing  
the R6RS would be quicker. Therefore, I am inclined to vote against  
any candidate for the steering committee who wanted to base the next  
Scheme standard on R5RS Scheme rather than R6RS Scheme.

Regards,

Alan
-- 
Alan Watson
http://www.alan-watson.org/