I wish to add to this discussion, that another ancestor of
TP and Delphi is the original ObjectPascal from Project Macintosh.
As I read in the book Object-Oriented Programming for the Macintosh
by K.J. Schmucker (1986) "the syntax for Object Pascal was jointly designed
by the Apple Clascal team and Niklaus Wirth, the designer of Pascal,
who was invited to Apples's Cupertino headquarters specifically for
this project". It seems to me that the similarities with TP are obvious:
Unit, interface, implementation, object desing and implementation since
version 5, Graph unit design ...
For me is a mistery why Apple abandoned such a beautiful project,
why the academic computer community forgot this and the UCSD projects and
why Wirth does not know get the credit he deserves for his work, including
the idea of the p-code software development strategy now revived -without
a word of recognition- as java and bytecode.
It was like leaving a straigth and pleasant way for taking a tortuous one...
>
> Peter Gerwinski wrote:
> >
> > Hello!
> >
> > According to David Rauschenbach:
> > > [...]
> > > In my mind, Pascal is the *perfect* language to see a Java bytecode
> > > implementation.
> >
> > Something *very* similar to this already existed many
> > years ago: UCSD Pascal. I consider this the direct
> > ancestor of Borland Pascal. (Somebody knows some details?)
>
> Indeed, UCSD seems to be the ancestor of Turbo Pascal (a lot
> of similarities). AFAIK, UCSD Pascal produced portable
> p-code. Certainly, the Cabot implementation of it does. The
> p-code can be run on any operating system (that is supported
> by Cabot environment?). The Cabot UCSD system has its own
> file system, etc., and the compiled programs run in a
> virtual environment which is complete with its own file
> system (with long filenames, etc.). The environment itself
> is enclosed in a file which appears to the host operating
> system as just a single large file. The p-code binary is
> very small, because all the facilities that it needs are
> inside the environment. The idea is that, if a port of the
> environment exists for your operating system, then any
> p-code program written for any platform will run on your
> operating system as well.
>
> Now, I am writing all this from (long term) memory. I
> reviewed the Cabot UCSD Pascal compiler for the Pascal
> Magazine (ancestor of the Delphi Magazine) a few years ago,
> and I am writing from what I remember. I might have got some
> facts mixed up, but I think I remember the general idea.
>
> Best regards, The Chief