Email list hosting service & mailing list manager

Re: Initial comments William D Clinger (12 Aug 2008 06:24 UTC)
Re: Initial comments Alan Watson (12 Aug 2008 12:21 UTC)

Re: Initial comments Alan Watson 12 Aug 2008 12:20 UTC

> I do indeed intend for SRFI 99 to be useful in module-less
> R5RS systems.

And there I was thinking that lcd was just a standard procedure,
rather than a guiding principle. :-)

> Indeed, the record system of SRFI 99 has already proved
> its worth in Larceny, where records and record type
> descriptors are first class objects that are routinely
> passed between R5RS, ERR5RS, and R6RS code without the
> inconvenience of marshalling, renaming, or coercions.
> That objective convenience should not be sacrificed for
> subjective aesthetics.

I agree that avoiding marshalling in this case is very convenient, but
I don't see what that has to do with the names of the procedures.
Perhaps I am being dense.

While you may disagree with my desire for common names between SRFI-99
and the R6RS, it is a mischaracterization to say that my request stems
from subjective aesthetic concerns rather than (subjective) concerns
about clarity. However, if you want this SRFI  to work with module-
less R5RS systems, then I can see why you might want to insist on new
names for your procedures.

Thinking more about my concerns, I am worried that this SRFI is
justified because there is a discord between the procedural and
syntactic records systems defined by the R6RS. However, instead of
simply defining a new syntactic record that is compatible with the
R6RS procedural system, you have rewritten both from scratch. What is
the justification for that?

In other words, putting on my lazy hat, I can understand why I should
learn a new syntactic system (because I want one that works with the
procedural system), but why on earth should I learn a new procedural

I think you either need to explicitly justify the new procedural
system or use the R6RS procedural system for your new syntactic system.


Alan Watson