On 14.02.2024 15:29, David Brown wrote:

On 14/02/2024 13:26, Janis Papanagnou wrote:

On 14.02.2024 09:46, David Brown wrote:

I hadn't done much with C++ before then. At that time, most of my
professional programming was in assembly (on small microcontrollers) and >>> Pascal, with only some C.

I'm surprised that Pascal was in professional use back then.[*]
Pascal had always been derided as "academic" or "toy" language.

In the DOS/Windows world, Pascal was a major language for a long time - starting with Turbo Pascal, then Borland Pascal for DOS then Windows,
then Delphi. Delphi's descendants are still in heavy use for Windows development. (And dotnet came from Delphi and related projects.)

Yes, I'm aware of the "popularity" (as Michael called it) of especially
the Turbo and Borland Pascal dialects [on DOS]. Being platform specific
(for a "personal computer" platform) and being dialects they have never
really been an option for our usage; popularity in general - and surely
in DOS contexts! - was never a criterion for us. (Using standards was,
though!) Yet, beyond popularity on DOS, I haven't heared much about it.
Other languages of the Pascal languages branch, like Ada, certainly had
their professional application areas (in aeronautics, for example)! But
Pascal?

Janis

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On 15/02/2024 07:54, Janis Papanagnou wrote:

On 14.02.2024 15:29, David Brown wrote:

On 14/02/2024 13:26, Janis Papanagnou wrote:

On 14.02.2024 09:46, David Brown wrote:

I hadn't done much with C++ before then. At that time, most of my
professional programming was in assembly (on small microcontrollers) and >>>> Pascal, with only some C.

I'm surprised that Pascal was in professional use back then.[*]
Pascal had always been derided as "academic" or "toy" language.

In the DOS/Windows world, Pascal was a major language for a long time -
starting with Turbo Pascal, then Borland Pascal for DOS then Windows,
then Delphi. Delphi's descendants are still in heavy use for Windows
development. (And dotnet came from Delphi and related projects.)

Yes, I'm aware of the "popularity" (as Michael called it) of especially
the Turbo and Borland Pascal dialects [on DOS]. Being platform specific
(for a "personal computer" platform) and being dialects they have never really been an option for our usage; popularity in general - and surely
in DOS contexts! - was never a criterion for us. (Using standards was, though!) Yet, beyond popularity on DOS, I haven't heared much about it.
Other languages of the Pascal languages branch, like Ada, certainly had
their professional application areas (in aeronautics, for example)! But Pascal?

I agree that standard Pascal was not (AFAIK) hugely significant as a development language outside of academia until the Turbo Pascal version
and its descendants in DOS/Windows. In particular, standard Pascal was
quite limited for "real" programs, so I believe that most people used
extended versions of various kinds. UCSD "p-System" Pascal was the most popular version until Turbo Pascal.

Perhaps the most significant Pascal program is TeX, which was written in
a combined documentation/coding language "Web" using Pascal as the
programming language.

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On 15.02.2024 14:34, David Brown wrote:

Perhaps the most significant Pascal program is TeX, which was written in
a combined documentation/coding language "Web" using Pascal as the programming language.

Oh, that's interesting! (I've never heard much about the TeX internals.)

Janis

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On 16/02/2024 03:49, Janis Papanagnou wrote:

On 15.02.2024 14:34, David Brown wrote:

Perhaps the most significant Pascal program is TeX, which was written in
a combined documentation/coding language "Web" using Pascal as the
programming language.

Oh, that's interesting! (I've never heard much about the TeX internals.)

It also has an interesting versioning scheme - it is currently at
version 3.141592653.

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On 2024-02-16, David Brown <david.brown@hesbynett.no> wrote:

On 16/02/2024 03:49, Janis Papanagnou wrote:

On 15.02.2024 14:34, David Brown wrote:

Perhaps the most significant Pascal program is TeX, which was written in >>> a combined documentation/coding language "Web" using Pascal as the
programming language.

Oh, that's interesting! (I've never heard much about the TeX internals.)

It also has an interesting versioning scheme - it is currently at
version 3.141592653.

It's Knuth's Literate Versioning. You can tell at a glance that
someone's TeX version is outdated when you see 3.14159265.

I'm surprised he didn't factor in rounding. I.e. this progression
of version numbers:

3
3.1
3.14
3.142 <- rounded up
3.1416 <- ditto
3.14159
3.141593
3.1415927 <-

--
TXR Programming Language: http://nongnu.org/txr
Cygnal: Cygwin Native Application Library: http://kylheku.com/cygnal
Mastodon: @Kazinator@mstdn.ca

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

David Brown <david.brown@hesbynett.no> wrote:

On 16/02/2024 03:49, Janis Papanagnou wrote:

On 15.02.2024 14:34, David Brown wrote:

Perhaps the most significant Pascal program is TeX, which was written in >>> a combined documentation/coding language "Web" using Pascal as the
programming language.

Oh, that's interesting! (I've never heard much about the TeX internals.)

It also has an interesting versioning scheme - it is currently at
version 3.141592653.

And Metafont is version 2.7182818.

G

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On 16/02/2024 17:38, Kaz Kylheku wrote:

On 2024-02-16, David Brown <david.brown@hesbynett.no> wrote:

On 16/02/2024 03:49, Janis Papanagnou wrote:

On 15.02.2024 14:34, David Brown wrote:

Perhaps the most significant Pascal program is TeX, which was written in >>>> a combined documentation/coding language "Web" using Pascal as the
programming language.

Oh, that's interesting! (I've never heard much about the TeX internals.) >>>

It also has an interesting versioning scheme - it is currently at
version 3.141592653.

It's Knuth's Literate Versioning. You can tell at a glance that
someone's TeX version is outdated when you see 3.14159265.

I'm surprised he didn't factor in rounding. I.e. this progression
of version numbers:

3
3.1
3.14
3.142 <- rounded up
3.1416 <- ditto
3.14159
3.141593
3.1415927 <-

Knuth did not release the main versions of TeX and metafont until he was confident that they were close to bug-free according to his
specifications for the programs (documented in his books). There were
plenty of test versions, of course. But rather than using traditional versioning, he used a scheme to mirror his view that the programs did
what he wanted them to do, but could get gradually closer to
"perfection" without actually reaching it.

So the final form of TeX was released at version 3.0, and has gained a
new digit of pi for each released fix - thus it has had 8 more releases,
each with smaller and smaller changes compared to earlier versions.
Similarly, metafont is on version 2.71828182.

Genuine bug reports were rewarded with payments that doubled with each
new bug (up to a limit - Knuth was not stupid!), as new bugs would be
harder and harder to find. (People rarely cashed these cheques - they
were often framed as certificates of merit.)

Rounding the version numbers would be a silly idea, and lose much of the
point (which requires a monotonically increasing sequence).

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On Fri, 16 Feb 2024 12:23:41 -0800, Chris M. Thomasson wrote:

It would be funny to use the convergents of infinite continued fractions
as a version scheme. ;^)

Particularly since they’re quite slow to converge.

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)