1. Re: Jiri and his 'Hints'
Norm Goundry wrote:
>Jiri,
>
>[ Compiler Wars, Hints Part III ]
>
<What is it exactly that you are trying so hard not to say? <snip>
I think Rob has already answered your question. I do not really
understand why he also said 'it might not be possible to offer the
WATCOM graphics library', but I wish him the best of luck with the
rest of it. I am really glad he no longer feeds us the line that it's
too difficult, perhaps even impossible, and/or that we would not gain
anything much from it anyway. In the end I think it was just Mark
Brown who still believed him.
Talking about Mark, he writes:
>I would like to hear from anyone who has actually HAD to use a
>compiled language for a project just because Euphoria was too slow. I
>haven't actually heard anyone give a specific example of a project
>that was too slow when done with Euphoria and forced them into C.
How many examples do you need (ignoring the gratuitous C remark)? Just
a couple of days ago I posted a little illustration. The pascal
original, compiled, runs five to ten times faster and takes only about
8 to 15 kbytes of the precious disk space on my machine, depending on
the compiler used. And I am not talking about some miracles, not even
dirty tricks like in line assembly, just plain old pascal - 30 year
old technology. And its not just graphics either. Let me give you just
one more example. One of my hobbies are compression algorithms. I
would love to explore some of them with Euphoria, but life is just too
short for that! Ok, one more. In fact I think I already told you some
time ago how much I love writing small utilities, filters and such, in
Euphoria. It's real pleasure, but only until someone else wants it
too, and you have to give him or her 200 kbyte whale with a half page
of source code...
Mark continues:
>Unless speed is a REAL problem then I would rather see Robert using
>his time to add features to the language.
If you want features, Mark, try something else. Xbasic, for example.
It is a nice language, full of features, and I am not kidding. And
with a very active list, full of enthusiastic followers. Or you could
try C++, or Ada, features galore.
It saddens me to see Robert wasting his time on 'features', on
intricacies of database design, and similar details. All these things
can be done, and will be done, if really required, by us, the users.
He should concentrate on the big picture, do a bit of real planning,
as Everett urged him ad nauseum. Unfortunately as long as Robert keeps
all his cards so close to his chest, we cannot help him much with the
really vital elements, I mean vital to Euphoria survival, most of them
well overdue already: namespace solution, *real* compilers, better
'real world' interface, built-in assembler, etc.
I realize we all have different priorities. Rob is still a free man,
he can do as he pleases. I respect that, but I really wish he could
find a different mechanism how to exploit his brainchild. Seeing what
has happened to so many 'opened source' projects over the last few
years, Python, Lua, Clean, Smalleiffel, just to name a few, how much
they blossomed under the combined attention of so many good and
willing programmers around the globe, I wonder what sort of explosion
we would have witnessed, given Euphoria's potential...
Gosh, looking at the size of this post and irreverent tone of it, I am
really behaving just like sadly departed Everett. (Btw I really mourn
his loss. On many occasions he stretched my patience too, but I still
believe he was the best think that happened to us in a long while.
Without him the list has gone back to sleep (and I agree, let's
pretend, for everybody's sake, 'Mike the Spike' never happened). The
things are again the way we like really them, don't we: not much more
than David's clinic for a handful of Window$ challenged individuals.)
It's getting late, 3.15 in the morning. Good night. jiri
