Mon, 19 Feb 2001 13:30:47 -0500
Not to keep the digression go much more but I sometimes get concerned
that some important lessons are getting lost. The Pascal effort was
significant in may ways. The first X3 meeting had over 90 people show
up for it. It quickly became one of the largest X3 standards
committees. There was a big political battle between IEEE and X3 over
who should "run" the committee. There actually were to chairs for a
We had establish the concept of an approved extension, which level 1 was
a good example, where the committee would approve an aproach for an
extension with the understand that this would be the starting point of
the next revision of the standard. We were trying to get the standard
out so that there would not be as many veriation as were developing.
The one that was most troubling to me was UCSD Pascal. We fully
intended to have a revised standard that would support level 1 and
more. Most of the people on the committee were using Pascal in industry
and had to use extension to get the job done. The goal was to get the
core language standardized quickly. Something the just never happens
Another interest part of the standard is the definition of I/O. It is
very well done. I always found Pascal I/O easy to understand and use.
PS for something Ada related. How many people know that the POSIX Ada
ballot was the largest ballot group in IEEE history. I don't know if it
still is. It was many times larger than the POSIX C ballot group.
Michael Feldman wrote:
> [said Steve]
> > Having been in the thick of this I would like to make a minor
> > correction. The US was against level 1 because it was an extension of
> > the original language specification in the Pascal Users Manual. The
> > original scope was to be ONLY the language defined there. I voted
> > against the proposal for that reason. At the time I was representing
> > the duPont company and not a coimpiler company. The major argument I
> > had against the idea was that you could not use the feature without
> > doing something outside the language or just coping the code into your
> > program. There was not library facility, preprocessor or separate
> > compilation facility. A more complete solution needed to be done.
> > This is just my opinion.
> > Steve Schwarm
> Wow, thanks for the clarification! I never heard that one before.
> As Steve implies, neither Level 1 nor Level 0
> described a language that was useful for more than small programs.
> Pascal was designed for teaching purposes, not to be industrially
> useful, and the ISO standard was, in that sense, a joke because
> any Pascal usefulness came from proprietary extensions anyway.
> There was some discussion of an extended Pascal standard, but I
> don't think this ever came to fruition. That being the case,
> the best approximation we've seen to a really grown-up Pascal
> standard is Ada. If memory serves, Red, Blue, and Yellow were genuine
> extended Pascals. Green was not, but OTOH it's a pretty close