Home

Thinking Forth & Starting Forth

Leo Brodie's books Thinking Forth and Starting Forth are now available for download and the first is back in print. This material had been hard to find for a long time!

By Luke Gorrie at 2007-06-29 05:43 | Misc Books | other blogs | 32498 reads

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.

Thinking Forth is a must read

...even if you don't like Forth. Be amazed at the term "[re]factoring" being used back in the 1980s :)

By James Hague at Fri, 2007-06-29 13:27 | login or register to post comments

Forth? In ... 2007?

I love Forth. I had to say that so that the title doesn't inspire a summary execution here at LtU.
I love Forth. Being a back-facing Lisp, it requires no parens. And that means Forth knows I am not a parser. That also makes Forth greater*.

But Thinking Forth is back in print in 2007? I wonder why. It doesn't seem like there is any Forth-ward exodus anywhere. Or is it more of a song of resistance?

*Yes.

By revence27 at Fri, 2007-06-29 16:19 | login or register to post comments

Forth still has uses

The company I now work for hired me a year ago to integrate a Forth interpreter into a library used by cell phone handset manufacturers for on-device performance data collection. With a bit of time and a lot of hard work, this particular Forth interpreter may become the most widely distributed scripting environment on the planet. It won't have any user visible components, and you couldn't write code for it, but in a couple of years you'll have a difficult time buying a cell phone without it integrated.

By Trey Boudreau at Fri, 2007-06-29 18:45 | login or register to post comments

Forth Forth Forth Forth

You could download those books months (maybe years?) ago, but either one of them being back in print is new. I like Revence27 characterization of Forth as a back-facing Lisp. But I'm not qualified to validate it. One of you who is care to expand or detract? Trey, the phone thing is just too cool.

By klausnrooster at Sat, 2007-06-30 03:03 | login or register to post comments

I agree

I only have minimal experience with Forth, but I agree with that characterization; it's a backwards Lisp without parenthesis. You also tend to see similarities between the languages, like how CREATE and DOES> in Forth are often used like Lisp macros.

Also, for anyone seeking more Forth resources, I found this tutorial to be excellent.

By Jimmy Miller at Sun, 2007-07-01 22:06 | login or register to post comments

That's a very shallow view of Forth

Compared to Lisp, Forth has a low-level approach
For instance Forth has direct acces to memory vs GC in Lisp. Higgin's Cat or Joy deserve much more this description of a Backwards Lisp.

By astrobe at Tue, 2007-07-03 16:14 | login or register to post comments

The big difference between

The big difference between Forth and Lisp in my view is lack of lambda expressions. However there is some recent work towards this end:
by Lynas and Stoddart: http://dec.bournemouth.ac.uk/forth/euro/ef06/lynas-stoddart06.pdf.

Thanks for mentioning me, but for the record it is "Diggins" rather than "Higgins".
:-)

By cdiggins at Tue, 2007-07-03 19:08 | login or register to post comments

Memory management with GC

I wonder who spends more time thinking about memory management, Lisp or Forth programmers? Reducing heap-allocation is a major theme amongst (Common) Lisp programmers.

By Luke Gorrie at Tue, 2007-07-03 19:40 | login or register to post comments

minimal memory management

Forth programmers obviously.
But concerns are different, because Forth programmers include the processor, the running program in the picture: what spends more time (and space) dealing with memory management, Lisp or Forth programs?
Again the concerns are different, because the applications and the supporting hardware are different.

By astrobe at Wed, 2007-07-04 12:06 | login or register to post comments

Many Forths Are High-Level

All this seems to assume when one says `Forth' he/she means GForth or Forth Inc. or the like. Nope.
You can write a Forth for the JVM, for example, and it won't have to bother with memory-management.

Lisp is much an idea, rather than the implementations. Same for Forth.

By revence27 at Wed, 2007-07-04 16:15 | login or register to post comments

you're mixing things

I don't think the ANS Forth specification allows a system to use dynamic memory allocation for ALLOT for instance.
It is true that a Forth system may provide some extension that may rely on automatic memory management of the hosting software stack (the OS, the actual implementation language), but it would certainly not be portable.

Memory management is a minor issue in Forth anyway. Thanks to MARKER or FORGET one may allocate static data for a job and release it very easily. Dynamic memory allocation isn't a Forth taboo, but static memory allocation is easier to deal with, more secure, and cheaper; this is quite valuable in embedded software.

I totally agree that Forth is much and idea rather than an implementation, and the idea of many Forth programmers about of Forth is not compatible with GC.

By astrobe at Thu, 2007-07-05 11:27 | login or register to post comments

Not that simple

Thanks to MARKER or FORGET one may allocate static data for a job and release it very easily

This is only useful if you're deallocating regions in the reverse order to that in which you allocated them. It seems to me that the kinds of applications for which Forth is well suited are partly determined by this fact.

By sigfpe at Fri, 2007-07-06 16:20 | login or register to post comments

It can be that simple

This is only useful if you're deallocating regions in the reverse order to that in which you allocated them.

True. That means that if one wants to use this, one just has to take care of that. As the main use of it is to compile and execute some part of the application when required (for instance, a database report), this is not a big constrain.
And that's a particular feature one may use, not the only way to manage memory. One can also use regular static storage, write a heap allocator if required, or even write a GC if mandatory.
But I know it's hard for outsiders to see how one may use and combine the unique set of features Forth offers, some of them being even rare among other languages, to do what one has to do.

It seems to me that the kinds of applications for which Forth is well suited are partly determined by this fact.

Well, I don't think Forth would have been used on space missions or in air trafic control systems, among other things, if that is true.

By astrobe at Mon, 2007-07-09 16:39 | login or register to post comments

Well, I don't think Forth

Well, I don't think Forth would have been used on space missions or in air trafic control systems, among other things, if that is true.

Funny, they're exactly the sort of apps I'd expect to fit in neatly.

The comparisons to lisp seem not inappropriate - I'm not sure I can say the same for your condescending comments about people who don't code in forth.

By Philippa Cowderoy at Tue, 2007-07-10 22:41 | login or register to post comments

Funny, they're exactly the

Funny, they're exactly the sort of apps I'd expect to fit in neatly.

I should have also mentionned CAD software.

The comparisons to lisp seem not inappropriate - I'm not sure I can say the same for your condescending comments about people who don't code in forth.

Wrong thread. BTW I made a comments about the comments of people who don't code in Forth. I may admit I spelled it bad, but I think it's appropriate not tell people to abstain from negative comments on languages they obviously didn't really use.

By astrobe at Thu, 2007-07-12 11:08 | login or register to post comments

Funny, they're exactly the

Funny, they're exactly the sort of apps I'd expect to fit in neatly.

I should have also mentionned CAD software.

That's more interesting, but for historical reasons not necessarily so surprising. Forth's nearest natural competitor has often been assembler.

The comparisons to lisp seem not inappropriate - I'm not sure I can say the same for your condescending comments about people who don't code in forth.

Wrong thread.

Right thread - the "we can build anything on top" approach is something that forth has very much in common with lisp.

BTW I made a comments about the comments of people who don't code in Forth. I may admit I spelled it bad, but I think it's appropriate not tell people to abstain from negative comments on languages they obviously didn't really use.

I think the comment in question may well have been appropriate and needed addressing directly rather than with an argument from authority. There are garbage collected concatenative languages that typically operate at a higher level and it seems likely that they would be more suited for applications where GC is near-ubiquitous anyway, no?

By Philippa Cowderoy at Fri, 2007-07-13 17:22 | login or register to post comments

Ah, Forth...

Forth was the first computer language I ever fell in love with, though it took me a long time to understand why. In my view, Forth's easy concatenation of functions to form new functions foreshadows the "point-free" approach to functional programming that is quite popular in the Haskell world. However, I always found writing anything useful in Forth to be a pain, mainly because of the incredibly low-level orientation of the language. Still, it's good to see these books available (though I have paper copies myself), and programming language buffs would do well to read them closely. One quibble: as far as I can see, Starting Forth is readable online but is not downloadable (well, at least not unless you use wget).

By Michael Vanier at Fri, 2007-07-06 00:15 | login or register to post comments