[r6rs-discuss] Implementors' intentions concerning R6RS

From: Ryan Newton <rrnewton>
Date: Sat, 27 Oct 2007 13:06:05 -0400

By the way, I should comment further that I see Marc's Snow project
as a win-win proposition. I don't think it needs to be in a zero-sum
competition with R6RS. Snow is a (relatively rich) least common
denominator that works across a fairly broad range of Scheme systems
-- including those that will run R6RS.

All Snow libraries will be widely portable, and R6RS libraries will
portable across a more limited (but important) set of implementations.

Well, great! This is still a very good place to be relative to the
historical status quo.

-Ryan

On Oct 27, 2007, at 12:57 PM, Ryan Newton wrote:

>
>> A standard which is largely unimplemented is a useless standard,
>> except
>> insofar as it is a list of suggested features. However, the Scheme
>> community already has and had a mechanism for making lists of
>> suggested
>> features.
>
> This mechanism has failed. If it were an outstanding success then
> that would make for a different story. But given the failure thus-
> far, I think the argument for the standard being useless becomes an
> argument for despair -- i.e. Scheme is dead, has failed, etc.
>
> I personally would rather call this a success. Running my code on
> PLT, Larceny, and Chez is much better than what I have now. In
> particular, that at least gives you access to PLT when you need
> libraries or debugging, and Larceny or Chez when you need performance.
>
> I sympathize with Jeffrey Siskind's argument that many Scheme
> implementations should be experimental playgrounds. But in my
> opinion, these implementations should be tailored for individual
> applications, used by a small number of people, and shouldn't be
> affected one way or another by a standard being ratified.
>
> -Ryan
>
Received on Sat Oct 27 2007 - 13:06:05 UTC