[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Some thoughts about Sather

From: Keith Hopper
Subject: Re: Some thoughts about Sather
Date: Sun, 10 Dec 2000 18:23:51 +1300
User-agent: Pluto/2.02b (RISC-OS/4.02) POPstar/2.02

In article
   Kipton M Barros <address@hidden> wrote:

> It took me a while to figure out what you meant by this.  Is the comment
> syntax "(* ... *)" not currently in the ICSI spec? 

     No!  It is usable in module files so far as I am aware - but, to be
honest, I like the Ada-style commenting!

> That sounds like something to look forward to.  Are you also considering
> adding support for simultaneous unpacking of tuples into multiple named
> variables?

     There is no point in having a 'half-baked' constructor primitive. 
There are one or two quirks in AVAL at the moment - buried so deep it would
be a major rewrite just to clear something that no programmer in their
(normally?) right mind will see - and no I am NOT going to give it away if
you hafven't found it for yourself!  A constructor should work at all times
- on the fly if you will.  Please remember that I am so far talking about
specification - not implementation - I honestly don't care -how- something
gets implemented - just that the implementation is correct!

 - - C - h - a - n - g - e - - - o - f - - - s - u - b - j - e - c - t - - 

     Anyway!  I need to diverge into another matter entirely.  My promised
next version of documentation by the end of November has fallen by the
way-side.  There is, however, what I believe is a good reason for this. 
The Numeric sub-section of the Required Library was reasonably well broken
and I thought that I had managed to straighten a few things out.  However,
when trying to look at the specification of it I discovered that all
numeric classes were very much inconsistent with all other numeric classes
- and with mathematics as I remember it.

     I have divided the classes into some sixteen mathematical concepts -
and made abstractions out of these - and then re-packaged different
combinations of abstractions to model the classes of number domains in the
library.  I am therefore looking, not at anyone in particular, for a kind
person who is rather cleverer and more knowledgeable about mathematics than
I am. I'm a retread electronic engineer by default - so I know most
engineering maths pretty well (I think!) - but not the higher regions of
numbers and number theory.

     Anyone who feels able to help so that I can get something on the web
before Christmas will be most welcome - I just have two files which need
reviewing carefully before I go into 'print' in the specification - all of
the abstract classes.  Help! Please?

                ...and a very Merry Christmas to all of you cold people
down there in the Northern Hemisphere.


City Desk
Waikato University
[PGP key available if desired]

reply via email to

[Prev in Thread] Current Thread [Next in Thread]