[r6rs-discuss] R5RS is not a baseline

[Michael Sperber]

William D Clinger <will at ccs.neu.edu> writes:

> How then did we end up with an R6RS in which *all*
> R5RS programs don't work?

A combination of two things, I'd say:

Many R5RS programs are very easily ported to R6RS - in my experience,
much more easily than trying to find the places where a pre-R5RS program
assumes that #f is null.

As you may remember, we debated specifying a "top level" similarly to
R5RS's top level repeatedly amonst the editors, which (I assume) would
have made many R5RS programs run straightforwardly.  As you may
remember, many of us felt that a top level would be nice (even though
some us didn't consider it nearly as important as you did), but we also
felt that it would be a significant effort - no one stepped up to the
plate to write it up in time for inclusion.  You yourself declined on
several occasions, if I remember correctly.

This, by the way, is an aspect that hasn't been very prominent in this
discussion: Good standards don't just materialize out of thin air by
virtue of people wishing for or even agreeing on something.  Somebody
has to actually do the work to write things down, figure out the
technical implications, and resolve the conflicts.  So you could say
that we ended up in the situation you complain about because of laziness
- being the project editor and sharing the responsibility for R6RS until
the very end, I guess that means *my* laziness.  (Your minutes of an
R6RS editors' phone conference actually show the sentence "Let's let
Mike chew on that." on this topic.)

You could argue that we shouldn't have presented R6RS for ratification
before all those "few issues" people complained about had been worked
out to everyone's satisfaction.  In theory, this may be a good idea, but
in practice, competent software projects have shifted to time-based
releases for the very same reason that we eventually committed to a firm
deadline for R6RS and (almost, at least) stuck to it.  Otherwise, we'd
still be working on it.

-- 
Cheers =8-} Mike
Friede, Völkerverständigung und überhaupt blabla