Fortunately not everyone has fully financialized their identities like Dougster has, and there is reason to learn more about old languages and operating systems, like COBOL, and to be good in Assembly.
SJG
Well assembly will help with those GPUs and crypto-currency mining. As for COBOL, PASCAL, and DOS? Time to get your psycho mind into the 21st century, @SJG!
No wonder SJG has a challenge dealing with people, only a tech centered geek would know what he is referring to.
Sometimes I wish I knew more about tech, but then you see a guy like SJG and are thankful for unanswered wishes.
Much to be learned from old computer languages and environments, especially when one is more interested in embedded and real time, than in the bloated and slug slow MS-WINDOWS.
One understand better when they apprehend the chronology of development of the key ideas.
SJG
Serious Shooting In Cleveland
[view link]
I'll have the PASCAL book within days:
[view link]
I'll have the 1995 edition. It is originally a 1974 book, and decades back when I was making lots of use of PASCAL, I remember now that I had the 1981 edition.
I always liked PASCAL. It serves a need and it works. But I was never a PASCAL proponent like I would later be with C.
C just seemed to speak directly to me right from the start.
Now I want to refresh myself as to all the details on PASCAL, and also learn about some archaic computer languages I have never used, because I need to be able to make lots of strategic decisions and I need to have vision of what is not now but could be.
PASCAL has what I have called 'training wheels', which C does not. C is more terse and makes tighter object code. But this is not always that important.
SJG
Jeff Beck ft/ Rosie Oddie
[view link]
Above book, I will have 1995 edition. But there is a 2012 4th edition, which it were available just to see what any of them have to say about any future for PASCAL.
[view link]
Some of my renewed interest in PASCAL is because some question the safety of C when it is a large project, and especially vehicular.
Where it is real time and about control of external devices, sometimes blinding fast computer speed does not matter, but safety does.
I have been told that ADA is PASCAL like, but I do not personally know this, or about the future of ADA.
SJG
FreeDOS, and PASCAL, and lots of other old computer things are likely superior when it comes to real time control, as opposed to things which are all user interactive.
SJG
Extreme Bikini's
[view link]
click on picture to see all of them
Jeff Beck ft/ Rosie Oddie
[view link]
I agree with both of the above. Also, there is no way that beginners could be able to learn such.
I love the object oriented paragdim. I've always used it everywhere I possibly can. But does it create more problems?
C is extremely terse, and with the its of compounding expressions. You have to be absolutely certain of the order of precidence. C is a langauge for showing off. Awesome results, but zero margin for error.
I will have the pascal book in hand today.
Is there still a need for PASCAL, or for ideas and approaches derived from it.
Was it a few decades back, when they crashed a glass cockpit A-300, because of a very small error in a mountain of programming written by dozens of people? Would PASCAL have made that less likely? ADA?
I do not know the answers to these questions, but I want to start finding out.
There is also a great need for interpretive langauges today.
PASCAL was like for people coming up from BASIC, and maybe FORTRAN too. But then C was for the hotshots.
And this again is why I see the old books as being just as important as the new ones, as you need to see how the thinking evolved.
SJG
Stones - Monkey Man
[view link]
The book I have says 4th edition, 1991.
I remember it well, though from when I used it back in the 80's. The book and the language are most impressive, so well thought out.
I remember around about 1980 a full spread add in Scientific American, where they showed a guy who explained that he was an Optical Designer. By this they meant complex lenses, and they showed some, not fiber optics.
He explained that he used his computer to write his own design software. And he was with his Apple II, and he used PASCAL.
This exemplified what I believe in. This was the idea of microcomputers, computers that you could buy. Is it realistic though? And people should be able to write their own programs to help them do what they do.
There is a mathematical approach called Matrix Optics. It gets you away from the thin lens approximation. You can consider lens stacks made with glass of various indices of refraction, and ground to fit right together.
But you are still operating under the ray optics instead of wave optics assumption, and you are still operating under the paraxial assumption. To get beyond these there is no recourse except to simulations.
And computer literacy should mean the ability to write programs to solve problems. This was the original idea.
But far too often now, even engineers are hired just to be simulation program operators. They use simulations to establish a division of labor, and as a busy box. That's another subject.
Anyway, it says in the book that during the 70's commercial use exceeded academic use. Interesting, but I would still think it was mainly scientists and engineers using it.
Today, is there a continuing need for PASCAL?
SJG
Beyonce, I like this better than the music video version because they aren't digitally processing her voice. And you actually can see her better in this live version.
[view link]
In the above book, it says that "Extended Pascal" was being defined just as the 1991 books was going to press.
[view link]
Interesting, but have to wait and see.
I am going to be looking at two more books which are PASCAL only.
Object-oriented programming in Pascal : a graphical approach / D. Brookshire Conner ( 1995 )
Advanced Turbo Pascal version 4 : now includes Borland's Turbo Pascal Database Toolbox and Turbo Pascal Graphix Toolbox / Herbert Schildt ( 1988 ), becasue I think data base connection is very important, and Borland had bought Ashton Tate and its DBASE.
FWIW, I never went along with Borland, never used their turbo Pascal or turbo C. Stuff looked like it was for students and hobbyists. The price was right, and they deserve praise for this. But I always had Microsoft and lots of other tools, and that seemed more professional oriented.
Today, I guess it is GNU.
I find more recent books about Fortran and Cobol, than I do about Pascal. I do think PASCAL would be good for beginners, but there are probably other newer things for that too.
I think most programming will want to go to Object-Oriented approaches, as the people get to be more and more sophisticated.
Again, reading the old books helps one see how the ideas developed.
And they say that the development of Fortran was being led by Pascal.
SJG
Finishing up with the above PASCAL book. To me, PASCAL was very well done, very functional. When I first started to learn C, I was surprised at the ways in which they deviated from PASCAL. I was never a proponent of PASCAL, like I would later become a proponent of C.
I feel that pointer to function, as it is with C, is extremely important. Not actually read this though, not yet.
If you write more and more sophisticated programs, they will have more and more complex flow control mechanisms, and this will involve user input, command line input, and even data and direction files. I feel that pointer to function gives you better ways of making programs which work this way. You are always to an extent, making your own variety of interpretive language.
Also I have worked with DOS based relational database systems which are extremely slow. If you want to "PROJECT" on two tables with some related tables, say one table has 10 records and the other has 100, the machine will make 1000 hard disc accesses. This becomes impossibly slow once you have enough data, taking beyond 24 hours.
What I did not do, but could have, was just to use C/C++ to write my own database app which would read those tables once, into core memory via C 'Structures' and arrays and linked lists, and then do all the processing and reporting from this C/C++ program. Not that difficult, but it would work GREAT. Pointer to function would have been greatly desired.
So this i why I want to read such stuff now, starting with the old books.
Also, I failed above to mention the language Perl.
SJG
[view link]
20 comments