Re: syntax heresy

[Boyko Bantchev <boykobb@...>]

On 8/13/05, Jamie Webb <j@jmawebb.cjb.net> wrote:
> ..............

I respectfully disagree with most of your statements.
Your opinion on mathematical notation is unusual and
is lacking any supporting evidence.  APL, J and K (now Q)
are far from exctinction ... etc. etc.  I have a lot to say,
but I don't see much sense to do so, and it's getting
too off-topic.  So I'd better stop here.

> You're an academic, right? The great thing about being an
> academic is that if someone else doesn't understand your
> work, you are free to believe that they're just not smart
> enough, or should try harder.  In the real world ...

I happen to be living a little in both `worlds'.  On the academic
part, I love and value other things than being arrogant.

Rici Lake <lua@ricilake.net> wrote:

> I remember APL from my youth with a certain fond nostalgia.
> It had the advantage of not starting with an arbitrary character
> set, but rather designing a character set which more or less fit
> the semantics. I don't remember ever having much trouble
> remembering what a given symbol did, because most of them
> were well designed visually. Reading J code in ascii, on the
> other hand, drives me crazy.

I believe many people, including me, still love the APL character
set.  It was (is) charming, in a way.  Moving to ascii was a
compromise, and although the J operators were no less
thoughtfully designed (by the same person -- KEI), they are
probably harder to read.  In the presence of a more
comprehensive character set, such as a widely accepted and
universally supported Unicode, the shift to ascii would probably
not take place.  Btw, the APL character set is now really
present in Unicode, so nothing is lost yet. :)

> In any event, APL did not load down your memory with
> syntactic binding rules. One rule fit all :)

The syntax of J and K is almost that simple as well.
No precedence levels, no left/right associativity.

> Anyway, those who are fond of languages with a lot of 
> symbols (and really complicated syntactic precedence
> rules) might be interested in the work Sun is apparently
> doing developing "Fortress" ...

I didn't know of this language -- thanks for the link.
Let us hope they did not create another monster like
`everything needs a class', `you should public-static-void-main
before you do anything', `you need class instances of three
different classes in order to read a line' etc.
Although, `really complicated syntactic precedence rules'
already sounds discouraging.  :)

Regards,
    Boyko