proper tail calls

Anton van Straaten anton at appsolutions.com
Thu Jan 24 12:12:59 PST 2008


Chris Pine wrote:
> Ahhh, I see.  Maybe that part wasn't communicated on-list, but was 
> discussed in the trac ticket (or in the phone meeting? can't remember). 

I'm just an observer, and have only seen the one ticket (#323) and the 
discussion here.  The ticket proposes a required annotation to activate 
PTC.  On this list, Neil argued against implicit PTC if it was going to 
compromise debug traces (a very valid concern, IMO), and later asked why 
some required syntactic overhead for tail calls would be such a bad 
thing.  I was responding to that question.

>   It was agreed that implementations would always be free to implement 
> PTC, and where pages depend on it from one browser, others will have to 
> support it there, too.  So, practically speaking, I assumed that it was 
> an assertion, not a requirement, and that we would all implement it 
> where we could.

You know that the minute there's some PTC around, people will implement 
crazy CPS-based programs and then complain about how browser X won't run 
them.  So y'all are going to have to get cracking.  ;-P

Anton




More information about the Es4-discuss mailing list