home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.lang.forth
- Path: sparky!uunet!starnine!mikeh
- From: mikeh@starnine.com (Mike Haas)
- Subject: Re: Free Forth
- Message-ID: <BuAow0.F8q@starnine.com>
- Sender: mikeh@starnine.com (Mike Haas)
- Date: Wed, 9 Sep 1992 05:07:11 GMT
- References: <1992Aug14.173059.20181@Informatik.TU-Muenchen.DE> <BtCvML.Lzu@starnine.com> <1992Sep5.151245.18215@mintaka.lcs.mit.edu>
- Organization: StarNine Technologies, Inc.
- Lines: 68
-
- In article <1992Sep5.151245.18215@mintaka.lcs.mit.edu> mikc@hal.gnu.ai.mit.edu (Mike Coughlin) writes:
- >In article <BtCvML.Lzu@starnine.com> mikeh@starnine.com (Mike Haas) writes:
- >>
- >>The trend is more & more memory in the machine, so what do I care
- >>if a Forth takes up 100k or 500k if I have Megs to work with?
- >>
- >
- > A small Forth fits in a human mind better than a big one.
-
- The only parts that need to "fit in your mind" are the parts you will
- use in programs. Identifying these words is the job of documentation,
- and HELP programs, not VLIST. The VLIST of a typical dictionary
- has always been a hodge-podge of babble.
-
- I want those hundreds of K to give me major functionality...
- Good, comprehensive, fast floating point (that uses hardware if it exists),
- Graphic & sound interfaces, Ability to call the OS with names instead
- of hex numbers, Smart optimizing compilers, Ability to create
- standalone small programs that run in the native mode of the machine
- (launched from shell, or icon double-clicked), performance profilers,
- source-finders, Ability to interface with custom platform libraries--
- code written in another language!
-
- And to tell me how to use these things, I want good documentation. THAT
- is why I think the job of producing Forth implementation should be left
- to vendors. PD versions do not document or support their products as
- well as commercial vendors (in the great majority of cases).
-
- >You can
- >do more with a system that you understand very well than you can with
- >a big system that you don't understand.
-
- Just because it's big, doesn't make it difficult. Indeed, some of
- that extra code should actually GIVE you something, like perhaps
- an online HELP facility, or an integrated text editor, etc.
-
-
- >Forth works very well in a
- >small space since it uses a few good ideas put together in a clever
- >way.
-
- Works fine for imbedded systems, but falls flat on it's face when
- you run on top of an Operating System (unless, of course, you live
- in the past and don't WANT to take advantage of it).
-
-
- >It is a much better way to do things than filling up all
- >available memory with more code than a single person can comprehend.
-
- Balderdash. If there ever was a language that SHOULD provide clean
- INTERACTIVE interfaces to complex Operating Systems and their facilitites,
- it's Forth. As long as it provides useful development tools, the more
- the better. (Don't you hate re-inventing the wheel?).
-
- It's a shame there is no standard for Forth to interface with graphics
- & sound, I/O requests, object files, etc. If ans forth is truly
- portable in practice, that would be a valuable stepping stone to
- such capabilities.
-
- Forth's strong points are it's interactivity, simple parser,
- extensibility, simplicity. It's too bad that it is so poor
- at interfacing with existing development technology and at
- "reaching out" into the worlds of the operating systems which
- now form the basis for future computing.
-
- We have to stop thinking along the lines of "oohh... the
- dictionary is getting so large!". (no porn jokes)
-
-