At 09:50 AM 5/4/00 , Tom Lord wrote:

>         I still don't like the current practice of allowing START and END
>         parameters to almost every procedures,
>
>I haven't reread srfi-13 lately, but remember coming to the same conclusion.
>I think that srfi-13 is valuable even if the parameters remain, but also
>predict that they will eventually cause an alternative, somewhat incompatible
>SRFI to be created.
>
>That's a good thing.  SRFI-13 is best existing practice and works today.
>No SRFI is the last word on its subject.

I have to put a word in in favor of START and END parameters.  I want the
power that they give me, and I use them all the time in other string
packages.  Shared substrings are an interesting idea, but they have so many
performance consequences, both good and bad, that I don't want to be forced
to use them when START and END parameters are all I need.

Olin has put in a lot of work on this SRFI and has revised it to accomodate
lots of people's comments.  If we try to make SRFIs into "diamond-like
jewels" like the Scheme standard, there will never be agreement.  I would
love to see the SRFI finalized so that I can expect to see it in a few
Scheme implementations.  If an alternative is needed, someone can write
another SRFI, as you suggest.