From: david.brown@hesbynett.no

On 28/10/2025 20:14, Janis Papanagnou wrote:
> On 28.10.2025 15:59, David Brown wrote:
>> On 28/10/2025 03:00, Janis Papanagnou wrote:
>>> On 27.10.2025 21:39, Michael S wrote:
>>>>>
>>>>> [ snip Lua statements ]
>>
>>>> Algol 68 is a great source of inspiration for designers of
>>>> programming languages.
>>>
>>> Obviously.
>>>
>>>> Useful programming language it is not.
>>>
>>> I have to read that as valuation of its usefulness for you.
>>> (Otherwise, if you're speaking generally, you'd be just wrong.)
>>>
>>
>> The uselessness of Algol 68 as a programming language in the modern
>> world is demonstrated by the almost total non-existence of serious tools
>> and, more importantly, real-world code in the language.
>
> Obviously you are mixing the terms usefulness and dissemination
> (its actual use). Please accept that I'm differentiating here.
>
> There's quite some [historic] languages that were very useful but
> couldn't disseminate. (For another prominent example cf. Simula,
> that invented not only the object oriented principles with classes
> and inheritance, was a paragon for quite some OO-languages later,
> and it made a lot more technical and design inventions, some even
> now still unprecedented.) It's a pathological historic phenomenon
> that programming languages from the non-US American locations had
> inherent problems to disseminate especially back these days!
>
> Reasons for dissemination of a language are multifold; back then
> (but to a degree also today) they were often determined by political
> and marketing factors... (you can read about that in various historic
> documents and also in later ruminations about computing history)

I can certainly agree that some languages, including Algol, Algol 68 and
Simula, have had very significant influence on the programming world and
other programming languages, despite limited usage.  I was interpreting
"useful programming language" as meaning "a language useful for writing
programs" - and neither Algol 68 nor Simula are sensible choices for
writing code today.  Neither of them were ever appropriate choices for
many programming tasks (Algol and its derivatives was used a lot more
than Algol 68).  The lack of significant usage of these languages beyond
a few niche cases is evidence (but not proof) that they were never
particularly useful as programming languages.

>
>> It certainly /was/ a useful programming language, long ago,
>
> ...as you seem to basically agree to here. (At least as far as you
> couple usefulness with dissemination.)

I do couple these, yes.  I agree with you that there are many reasons
for the popularity of languages other than technical suitability, but
many of these add up to the general "usefulness" of the language.  When
choosing the language to use for a particular task, the availability of
programmers familiar with the language, the availability of tools,
libraries, and existing code, can be just as important as the language's
efficiency, expressibility, or any technical benefits.  Consider Bart's
language - if we believe him at face value, it is the fastest, clearest,
most logical, most powerful, and generally best programming language
ever conceived.  But for almost every programmer on the planet, it is
completely useless.

Similarly, Algol 68 may have been the technically best language of its
age, and highly influential on other languages, and yet still not a
useful programming language.  It could also have been a useful
programming language in its day, and no longer be a useful programming
language.

>
>> but it has not been
>> seriously used outside of historical hobby interest for half a century.
>
> (Make that four decades. It's been used in the mid 1980's. - Later
> I didn't follow it anymore, so I cannot tell about the 1990's.)
>
> (I also disagree in your valuation "hobby interest"; for "hobbies"
> there were easier accessible languages used, not systems that were
> back these days mainly available on mainframes only.)

I did not suggest that it is now, or ever has been, an appropriate
language for hobby programmers - I don't know the language enough to
judge.  I suggested that anyone programming in Algol 68 today is likely
to be doing so as a hobby or for historical interest.  (There may be the
occasional professional maintaining ancient Algol code for ancient
mainframes that are still in use.)

>
> As far as you mean in programming software systems, that may be true;
> I cannot tell that I'd have an oversight who did use it. I've read
> about various applications, though; amongst them that it's even been
> used as a systems programming language (where I was astonished about).
>

My understanding - which may well be flawed - is that Algol 60 and many
non-standard variants were used quite widely at the time.  Algol 68, on
the other hand, never took off outside.

>> And unlike other ancient languages (like Cobol or Fortran) there is no
>> code of relevance today written in the language.
>
> Probably right. (That would certainly be also my guess.)
>
>> Original Algol was
>> mostly used in research, while Algol 68 was mostly not used at all.  As
>> C.A.R. Hoare said, "As a tool for the reliable creation of sophisticated
>> programs, the language was a failure".
>
> I don't know the context of his statement. If you know the language
> you might admit that reliable software is exactly one strong property
> of that language. (Per se already, but especially so if compared to
> languages like "C", the language discussed in this newsgroup, with an
> extremely large dissemination and also impact.)
>

I don't know the context either.

>>
>> I'm sure there are /some/ people who have or will write real code in
>> Algol 68 in modern times
>
> The point was that the language per se was and is useful. But its
> actual usage for developing software systems seems to have been of
> little and more so it's currently of no importance, without doubt.
>
>> (the folks behind the new gcc Algol 68
>> front-end want to be able to write code in the language),
>
> There's more than the gcc folks. (I've heard, that gcc has taken some
> substantial code from Genie, an Algol 68 "compiler-interpreter" that
> is still maintained. BTW; I'm for example using that one, not gcc's.)
>
>> but it is very much a niche language.
>
> It's _functionally_ a general purpose language, not a niche language
> (in the sense of "special purpose language"). Its dissemination makes
> it to a "niche language", that's true. It's in practice just a dead

[continued in next message]

--- SoupGate-Win32 v1.05
 * Origin: you cannot sedate... all the things you hate (1:229/2)