Re: [colorforth] DOES> How is colorForth different from other Forths?

[Mark Slicker <maslicke@xxxxxxxxxxx>]

On Wed, 21 Jan 2004, Roman Pavlyuk (personal) wrote:

> Bon jour, Frederic,
>
> >> Wait! Wait! let me answer to that.
> >> Of course nobody has ever thought of these problems because Mickeysoft
> >> solves the problems it creates.
>
> Well, I'd not like to start a "M$ vs progressive mankind" flamewar here.
> But actually MS does not produce very bad quality products, _sometimes_
> better than competitors' ones (there are examples, but let's not start a
> discussion here please).
>
> What I don't like it that analytical and technical information is being
> hidden, and there's lots of sales and marketing junk available, but almost
> no real imformation. Almost everyone talks about their products, but there
> are almost no independent analysis and reviews. Maybe it's because of US
> legal system when it's better to keep mouth your shut :)
>
> Some topics: comparison of Win32 vs POSIX APIs;

Having strugled with a simple named pipe for four days at work, I don't
have anything good to say about the Win32 API. Its ugly, overly complex,
not fully specified. There is simply no way to have any confidence that my
code, dependent on Win32, will work in a consistent manner.

> DirectX vs (what
> alternatives exist there?);

OpenGL, which while being a poor abstraction of PC hardware, is not
impossible to work with.

> PDF vs PostScript vs DPS vs GDI; CORBA vs its
> absense; exceptions -- why they are good (not good). Well, in addition to
> comparison i'd like to read a series of books with a subtitle "Opinions of
> different people on how to do things right"  -- different (but good, at
> least for their sincere proponents) approaches to do things. With
> explanations why their way is better. Many people just do not have real
> knowledge of how things work and should work (an example -- for many
> programmers knowledge of multithreading is limited to "synchronized"
> keyword)
>
> It's actually state of the art and education system question where lots of
> infromation does not exist in "for dummies" form :) Also, much information
> exists in descriptive form, w/o analysis, what is maybe good for gurus, but
> not for inexperienced reader (like me). Well, and opinions about what we
> need all these technologies for? To increase GNP?

I think so. I think that is the only reason the U.S. government did make
any effort to dismantle the Microsoft monopoly. It is not simply a matter
of national competition, internationaly they want Microsoft to succeed.

The Operating System is the most fundemental piece of software. In
hands of one company, one organization, it is a tool of control.

>
> Also, because of lack of clear understanding of ColorForth message (let me
> use this marketing term here), we waste lots of time for discussion of
> unimportant details, etc.
>
> I would like to try to use it in real world (commercial) applications
> (currently -- a distributed control system, 100+ nodes to be more precise),
> but there should be real reasons not to use C++ and TCP/IP and files --
> though it's so bloated, PC/104 format card with Cyrix/VIA 600MHz processor
> is 20 or 30 bucks more expensive than one with AMD 133, and performance
> really differs 5-20 times. So, Forth could be really useful when there would
> be proof that it's _much_ smaller and yields much cleaner code, and allows
> using much more powerful coding templates (though Chuck hates this word, I
> mean "template" as a way of communicating knowledge from mnaster to
> aprentice. This does not replace thinking, though!), _real_ proof that code
> quality will be higher, showing that CF's approach is closer to satisfying,
> say, NASA or IEC safety standards than C and ANS Forth, whatever.
>

Good luck! I say that to Howerd as well. If the work is judged on quality
only, you may have chance.

I don't know that it is possible to do Chuck Moore style work and get
paid, unless you are Chuck Moore or are completely independent.

If there was a job posting that said "come up radical new ways of
programming computer systems", I would surely apply.

Mark

> [...]
> > My 2eurocents :)
> >
> >> ( Just out of curiosity, where are you living?)
>
> 70 km to the east from current EU border -- L'viv, Ukraine. Though I like
> travelling :)
>
> Btw, I tried to write a MachineForth-like assembler for JPB's ForthCORE
> (http://jpb.forth.free.fr) when I had some free time. (Don't think it's a
> complete offtopic here since it's a F21 in FPGA, and F21 is somewhat related
> to cf)
>
>
> >> Amicalement,
> >> Frederic
>
> BR,
> Roman

---------------------------------------------------------------------
To unsubscribe, e-mail: colorforth-unsubscribe@xxxxxxxxxxxxxxxxxx
For additional commands, e-mail: colorforth-help@xxxxxxxxxxxxxxxxxx
Main web page - http://www.colorforth.com