Re: A Brief Look at History

Jonah Thomas wrote:
> m-coughlin <m-coughlin@[EMAIL PROTECTED]
> wrote:
> 
>>    I noticed where there was some programming in Forth being
>> done at the Harvard Smithsonian Astrophysical Observatory, down
>> the street from where I lived, across town from where John Doty
>> worked. When I learned enough about Forth to maybe get a job at
>> the Astrophysical Observatory, I discovered they had fired the
>> Forth programmers, trashed their code, and were writing
>> astronomical imaging programs in some other language instead. I
>> investigated further and was told the programmers could not read
>> their own code and this was only because they were using Forth.
>> I knew that this is not true and it was because of poor
>> commenting style and poor management. I convinced no one. I have
>> not been able to convince any Forth programmer that Forth's
>> commenting style needs to be improved either. 

Funny, SAO is still using Forth, for some space projects.  They're good 
customers of FORTH, Inc., and we hear from them from time to time.  I 
posted a press release from them a year or so ago.

> I have heard this same exact story with the same wording from a large
> handful of different places, usually from the late 1980's but sometimes
> the early 1990's. An NRAO shop in suburban Maryland. A privately-owned
> cor****ation in Boston. A privately-owned company in Oregon. A large
> publicly-owned cor****ation again in Boston. "We fired the Forth
> programmers and quit using Forth because they couldn't read their own
> code."

NRAO never had a "shop" in Maryland AFAIK.  NASA, maybe?  But I know of 
several groups in NASA that are still using Forth.

We do, sadly, hear of companies that drop it.  In most cases, it's 
because management felt it wasn't "mainstream" and feared being unable 
to hire programmers cheap. We mostly hear from the programmers, of 
course, who are distraught (and sometimes quit).  At least *they* didn't 
think they couldn't read their own code!

> I don't know now whether there was some problem with Forth commenting or
> whether it was just the custom -- what managers told each other to say
> when they fired the Forth programmers. How did the managers test whether
> the Forth programmers could read their own code? Do they use the same
> tests to decide whether individual C++ programmers can read their own
> code?

I think  lot of these stories are either apocryphal or excuses made up 
by management who felt uncomfortable with a minority language.

> I did a google search on  can't|couldn't read his|their own code.
> 
> C/C++
> http://forums.civfanatics.com/archive/index.php/t-166431.html
> 
> Java IDEs
> http://forum.java.sun.com/thread.jspa?threadID=158369&messageID=467108
> 
> various, from comp.lang.misc
>
http://groups.google.com/group/comp.lang.misc/browse_thread/thread/984759db724234d1/4007325902c53b0e%234007325902c53b0e
> ----
>>>                        Scheme seems to share much of its syntax with
>>> Lisp, which tends to be a "write only" language.
>> Hardly true of Scheme/Lisp.  The only language that might be truly
>> said to possess this characteristic is APL, and only because of its
>> infernal character set.  (Because of this, APL is the one language
>> I've encountered in my life that I've never _wanted_ to master.)
>> Substituting each APL operator with a descriptive name would certainly
>> go a long way towards making it less of a "write-only" language.
> 
> Actually, FORTH is also an extremely write-only language.  I also find
> that Lisp is somewhat write-only when complex programs are programmed
> in a funcational style.
> 
> I suspect that languages are write-only more because they lack names for
> local variables or intermediate results that because they have unusual
> characters representing functions.  I mean, who would want to write
> out "iota" or "rho" every time you wanted an array? :-) :-) 
> ----

Interesting.

....>
> I'll bet if you find a hundred people who've never used Forth but who've
> heard of it, 98 of them will have heard it's unreadable. Probably 5 or
> 10 of them will have heard the story about the Forth programmers who
> couldn't read their own code. This says nothing about how hard it is to
> read Forth code, what commenting styles would help, or how many Forth
> programmers have trouble reading their own code.

Yes, that's the classic response from people who've heard of Forth but 
don't know anything about it.  A lot of this, unfortunately, originated 
in the 80's, when there were many articles in Byte, Dr Dobbs, and other 
places about Forth.  They all included code, and in virtually all cases 
the magazines "edited" the code to "fix" things like removing spaces 
before "punctuation" such as commas, periods, semicolons, etc., "fixing" 
spelling, and other things that rendered the code utterly gibberish.  I 
wrote some of those articles, and screamed loudly, but the mags said 
there was nothing they could do, it was their art dept (code was 
considered "illustration").  Folks can't be blamed for concluding that 
stuff was unreadable, and unfortunately that's the only Forth a lot of 
folks have seen.

>> I have
>> not been able to convince any Forth programmer that Forth's
>> commenting style needs to be improved either. 
> 
> Many Forth programmers have experimented with new commenting styles.
> Where are you continuing to see Forth code that's badly commented? How
> many lines of poorly-commented Forth code have you seen that were
> written since you failed to convince us to comment better?

Some programmers don't like to write comments in code regardless of 
language.  They shouldn't be tolerated: we've spoken of management in 
other contexts, and one of the responsibilities of good management is to 
be sure the team is professional about writing clear, well-commented 
code.  And one of the responsibilities of professional programmers is to 
write that way.

FORTH, Inc. is often called in to help ****t old apps written in Forth 
years ago to new platforms.  So, we get a lot of experience reading code 
written by random other programmers in other Forths over the years. 
Some of it's quite good, some quite awful.  We can always read it, even 
when the effort is painful and the comments deficient.  The likelihood 
that the programmers "couldn't read their own code" is vani****ngly small.

I think that's folklore.

Cheers,
Elizabeth

-- 
==================================================
Elizabeth D. Rather   (US & Canada)   800-55-FORTH
FORTH Inc.                         +1 310-491-3356
5155 W. Rosecrans Ave. #1018  Fax: +1 310-978-9454
Hawthorne, CA 90250
http://www.forth.com

"Forth-based products and Services for real-time
applications since 1973."
==================================================