Subject:
|
Re: FW: Something else is needed, I think...
|
Newsgroups:
|
lugnet.robotics
|
Date:
|
Mon, 3 May 1999 23:58:38 GMT
|
Original-From:
|
John A. Tamplin <jat@liveonthenet.*AntiSpam*com>
|
Viewed:
|
1085 times
|
| |
|
|
On Mon, 3 May 1999, Ralph Hempel wrote:
> The FORTH bytecode producer is an UMBILICAL forth, and then you need one for the PC, for the MAC,
> for linux boxes, you get the picture. If the RCX interprets the code, then it HAS to work.
>
> Eventually we could make a headerless FORTH image that is way smaller, but let's walk before
> we run...
The problem is he doesn't understand how forth works (no offense intended).
Basically, all you save is the actual parser -- there is no "compiler".
A typical forth implementation (and I have no idea if this relates to pbForth
at all) translates forth code like this:
0 getvar 1 getvar exch
into a sequence of codes like:
pushint(0) call(getvar) pushint(1) call(getvar) call(exch)
(This is oversimplification, typically the resulting code is more compact
and just has addresses into a word table for the functions that are called).
Thus, you save only the code that parses ASCII text and looks up words. You
still have to have all the code that defines the words in the word table.
> I guess I've focussed so much on the host platform being as simple as a dumb
> terminal that I lost sight of what users actually want. Even the original RCX
> and legOS software need the PC to transmit the firmware to the RCX. You don't NEED
> to use a dumb terminal - it's the minimum tool for the job. If you want to write and
> debug and then use a GUI - that's cool. I just packaged the minimum set to get
> folks using pbFORTH - then they might know what the GUI or higher level interface should do.
> Sort of a develop the tool with the user approach, not hand the user a tool that no
> user has ever tested approach.
Personally, forth is a bit low-level for me. One thing that might help
people use it would be a library mechanism that would keep word definitions
on the host and then download them so you don't have to keep duplicating
them. It could be as simple as a #include mechanism.
John A. Tamplin Traveller Information Services
jat@LiveOnTheNet.COM 2104 West Ferry Way
256/705-7007 - FAX 256/705-7100 Huntsville, AL 35801
--
Did you check the web site first?: http://www.crynwr.com/lego-robotics
|
|
Message has 1 Reply:
 | | RE: FW: Something else is needed, I think...
|
| (...) I hope I'm not the "he" in this sentence - otherwise it's a pure accident that pbFORTH works :-) You have got the basic idea here, John. It gets interesting when you can factor repeated code snippets into words and then use them in a bigger (...) (26 years ago, 4-May-99, to lugnet.robotics)
|
Message is in Reply To:
| | FW: Something else is needed, I think...
|
| -----Original Message----- From: Ralph Hempel [mailto:rhempel@bmts.com] Sent: Monday, May 03, 1999 7:07 PM To: Mark Tarrabain Subject: RE: Something else is needed, I think... (...) The FORTH bytecode producer is an UMBILICAL forth, and then you (...) (26 years ago, 3-May-99, to lugnet.robotics)
|
32 Messages in This Thread:
  
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
This Message and its Replies on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
Active threads in Robotics
|
|
|
|
|
|
