Hi, On Mon, Jul 11, 2022 at 7:04 AM Liam Proven <lpro...@gmail.com> wrote: > > On Fri, 8 Jul 2022 at 21:35, Rugxulo <rugx...@gmail.com> wrote: > > > > Everybody and their brother made Pascal derivatives: Ada, Modula-2, > > Modula-3, etc. While Dr. Wirth was not directly involved, there was > > also a newer "Extended" Pascal standard in 1988 (ISO 10206) that also > > had modules. But even Wirth kept going and started developing Oberon > > in 1986. "Standard" Modula-2 (N.B. GNU GM2) came in 1996 (ISO 10514). > > (So it was too many competing languages, honestly.) > > I think this is a misrepresentation and unfair. > > Pascal was just one point in a continuous series of work by Prof. > Wirth and his associates. > > As it happens, the computer industry seized on some of these, made > multiple 3rd party compilers, many with proprietary extensions. So you > say "Dr Wirth was not involved" as a global statement which is not > true.
Sorry, I meant specifically that he wasn't involved in either Extended Pascal or Modula-3 (although he attended one of their meetings). He wasn't directly involved in any standardization, AFAIK. His personal compilers that he wrote were for original Pascal, PIM Modula-2, and Oberon (original, -07). > Prof. Wirth's work goes: > > (Prehistory of Pascal: he was involved in the committee to modernise > ALGOL-60. His proposals were rejected, in favour of more complex ones > from Adriaan van Wijngaarden, whose revised language became ALGOL-68 > and which more or less sank ALGOL as a language.) Yes, Algol W. There's a third-party Linux transpiler for it. (He also wrote Euler and PL/360.) > Wirth took his proposals and made his own language, which he renamed Pascal. > > That got widely used and adopted. They literally gave away P4 (subset) sources to help propagate the Pascal language. > Wirth refined and worked on the language further and created Modula. > > (So in a way, Modula was Pascal 2.0.) > > Modula was not a success. He quickly moved on and made Modula-2. That > did quite well for a while. It was significant on DOS PCs in the early > days; JPS Topspeed Modula-2 was the fastest native-code compiler on > the PC and did well for a while. Acorn attempted to build the OS for > its new CPU, the ARM, in Modula-2. Yes, Modula-2 was a true successor to Pascal (with minor differences and many refinements). JPI was a spinoff of Borland. > So, Modula-2 can be seen as Pascal 3.0. > > Then others took Modula-2 and extended it, to make Modula-3, but that > was nothing to do with Wirth. Modula-3 was actually the cleaned-up successor to Modula-2+ [sic]. It was mostly developed at DEC. > Next Wirth built Oberon (1987). > > Oberon is sort of Pascal 4. > > Oberon is still around and still in use so it's arguably proved to be > a survivor. Oberon-07 is Wirth's latest dialect. There are several third-party compilers for it. > Then it gets really complicated. > > One line goes Oberon -> Object Oberon -> Oberon 2 -> Oberon/L -> > (renamed to) Component Pascal I'm not aware of any "Object Oberon" implementations. (Or did that inspire "Active Oberon"?) > Another line of development was: > Oberon -> Oberon-07 > > Another line of development was: > > Oberon -> Active Oberon -> Zonnon > > Wirth was involved with several of these, as he refined and > reconsidered his ideas. I believe Zonnon was mostly Jurg Gutknecht's work. > You also said: > > (So it was too many competing languages, honestly.) > > Also not really fair. 1982 was "standard" Pascal. 1984 was Byte's big issue on Modula-2. And then Oberon was created in 1986. That's a lot of overlap and confusing (especially in light of Ada, C, and others). > I mean, arguably, yes, but there are also dozens of variants of C. > > There's original C, K&R C, Plan 9 C, ANSI C, C 99, C11, C17 and soon C23. C17 is just minor fixes and clarifications to C11. Plan 9's differences are minor, IIRC. > All are C. All are different. Code from one may not work in others. > > And of course there is Limbo, Go, C++, C#, D, and myriad variants. > > All are forms of C with extensions and occasionally removals or > refinements or deprecations. GCC does have a D frontend nowadays, but I wouldn't call that "C with extensions". In fact, Walter had a sort of in-between from C to D called "betterC". > Some of them are hugely popular and widely-used, e.g. C++, but are > nothing to do with C's original designers. C++ is meant to be as compatible as possible with C. So it is possible to compile code with both. As you know, originally it was a transpiler into C code. > Some are directly from C's designers but are obscure, such as Plan 9 C > and Limbo. > > Some are directly from C's designers but also are big successes, such as Go. C was mostly Dennis Ritchie's work. Go is very different (garbage collection) but also inspired by Oberon. It's meant to be a better, safer systems language than C. > It's not possible to say which is more "real" or legitimate. There is > no single coherent version numbering system. Different compilers have > nearly-but-not-completely overlapping subsets, e.g. Intel C and GCC > and Clang. > > It's complicated. If you exclude ones from from the AT&T team as not > really being C then you exclude many of the most important variants. > > It's arguably too many language but nobody says that of it. I'm all for variety, but a single person really doesn't want to wrestle with classic Pascal, Extended Pascal, Ada, Modula-2, Modula-3, Oberon, C, and C++. Wirth, in particular, mandated only one language on each of his systems. So Lilith intentionally didn't have so much fragmentation. In fact, that's why Ada was born, so that they could unify and simplify to only one language in-house instead of hundreds. > Well, the Pascal line is the same. > > There are at least half a dozen generations. Given the ones that have > been adopted outside Wirth's institutions and used in many countries, > there are things that we could call Pascal, Pascal 3 (Modula-2), > Pascal 4 (Oberon), and several different successors to Pascal 4. Turing is another Pascal-inspired language. > There are also non-Wirth variants that had some adoption, including > Modula-2+, Modula-3,Turbo Pascal, Object Pascal, Delphi, Kylix and > FreePascal. All legit, all sold and were widely-used at some points in > time. Delphi has been around for 25+ years, but I wouldn't even lump that together as a single language. FPC devs used to say "Delphi 7" was the standard, but a lot has changed since then (2003?). I respect Delphi's work, but I dislike a constantly moving target. Just to be clear, if it compiles in Delphi but not FPC, that's annoying. > You can't say that there are too many Pascal variants unless you also > say that there are too many C variants. XPL0 is also inspired by Pascal. Actually, they (e.g. Ritchie) specifically didn't want subsets or optional features in "standard C". Of course, that ship has sailed. They all have different purposes and implementations. I'm not complaining (much). But Wirth himself doesn't use anything except Oberon-07 anymore. _______________________________________________ Freedos-user mailing list Freedos-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/freedos-user
