The day computers died

[oldtobor]

They say that to understand the history of something, you can't be too close to the events. So I am referring to something that happened about 20 years ago, namely when pascal and turbo pascal started to become popular and started to substitute basic as the programming language for home/PC computers. That is the day computers died. Pascal was the next "big thing" and cool and also "harder" so it started to pick up. Basic slowly started to become out of style. But the question I asked then and I still ask today is "why say an old thing in a new way ?" I also guess that is when programmers started running out of ideas. When you run out of new ideas invent a new programming lanuguage.

 

I started programming microprocessors in summer 1978 using the SCAMP National semiconductors development kit in machine language (aka load accumulator with 0 was C4, 00). Then I learned basic and used it until about 1988. I then went down the road of learning all the sequence of languages: turbo pascal; turbo C; unix systems ksh; awk; perl; javascript; visual basic; java; c++; delphi; php; python.

 

After all these years I have come to the conclusion that we would have been all better off just sticking to basic and maybe improving it's compiler and creating a possibly huge library of reusable programs and modules. If all those man years used for just "java" woud have been used to really program, we would be light years ahead of where we are now.

 

Do you want an example ? Try to write this simple basic program in Java:

 

FOR i = 1 TO 5

a$ = ""

WHILE a$ = "": a$ = INKEY$: WEND

PRINT a$

IF a$ = "w" THEN SOUND 150, 2

IF a$ = "z" THEN SOUND 550, 2

NEXT

 

Hey don't forget we are in the year 2005 and we have all the great Object oriented java crap available!

[nameta9]

I mostly agree. I think Java and other object oriented hype was mostly market hype and mostly a social phenomena. Maybe if all that effort went to improving compilers and libraries who knows ?

 

This reminds me of the thread of science slowing down or going actually backwards. Anyways it is an interesting point of view to ponder. Of course the subject matter is quite complex but it is refreshing.

 

see

 

http://www.scienceforums.net/forums/showthread.php?t=10344

[Sayonara]

There's nothing wrong with hyping object-orientation itself; the point of computer science is to narrow the gap between the real world and the model that is being represented via metaphor. Object orientation provides a much better system of metaphor construction than non-OO languages can.

 

Although some problems are just as easily solved with old school programming as they are in an OO language, OOP is more advanced than Basic-style linear programming. Whether we like Java or not, or think it's over-hyped, is beside the point, surely?

[oldtobor]

Yeah, I know. Java is here to stay. Too many schools and companies have embraced a bad technology, bloated, complicated and not much fun. In the future I hope computer scientists will really try to simplify things!

 

Why say an old thing in a new way? Basic could have done everything with a top notch compiler and some really good extensions like "regular expressions" (something that is hard to use in java) and some good array manipulating things as perl has. Perl also has a built in sort and grep, which could have been added to basic. The point is all the thousands of students at schools that struggled to come to grips with java between 1998 and 2004 could have been programming and thinking about improving code than getting all confused with abstract fluff concepts that java forces you to deal with. Hell even assembly would have been more productive in the end. Anyways that is the way it is.......

[nameta9]

There are alot of critics of java on the web. I think the real problem is missed expectations. It was supposed to be this great new thing and it ended up being a clunker. Yes they use it in industry and schools but I am not at all found of it. Why so many complications ? why so many abstract concepts ? Really the only problem with software is organizing it well and documenting it well. Lets stop inventing new languages and code clearer and better.

 

People say the BASIC GOTO is wrong but if you consider exactly how the assembler code works under the hood, it is all GOTOs and actually jumps on conditions like JNZ (jump if 0) or JNE ( jump if not equal) etc. The real problem is badly organized and non documented code.

[oldtobor]

They say that to understand the history of something, you can't be too close to the events. So I am referring to something that happened about 8 years ago, namely when apple HYPERCARD was no longer supported or ported. That is the day computers died.

 

It is a general thing with the concept of "progress" that we tend to think of things "improving". Nothing can be so untrue. We have things that CHANGE alot and many things do improve, many do not and many stay the same. In 1987 Apple put out Hypercard that is still years ahead of anything out there today. They forgot about it, things change, sometimes we gain sometimes we lose.

[xa]

I'm starting to sense a really kick-arse language possibly being designed

OOP is useful, it makes many tasks that mimic life like events much easier. I'm too young to have been alive back then. Hell I was born after C++, which is my main language.

Perhaps you should invest some time into Lex and Yacc and remake BASIC with these powerful qualities.

In my humble but probably stupid opinion all interpreted code is pretty bad, but useful. Python is a great example, amazingly useful but slow. Java is even worse because it's intermediate code.

[Jacques]

I thaught htat Basic evolved to Visual Basic.

Also don't speak of compiler when speaking of Basic. It's an interpreter that run the code.

I like interpreted language for developpement and prototyping, but when you need performance you need a compiled program.

[radiohead]

Java is crap. I think that all we need is one versatile language like basic, or C. If we had stuck with one language, we could have found ways to use it for everything. We could have found ways to use C for web graphics if Java hadn't come along. We could have used C to make incredibly dynamic webpages istead of writing PHP or ASP. IMO

[xa]

Well you can do many things with C. You ever heard of CGI C? It's a compiled C form of CGI instead of perl, haven't tried it but I'm pretty sure it's needlessly complicated. I personally prefer C++ as an uber 1337 versatile language, because classes are one of the most useful things ever. In my personal opinion Java is only accepted because it's machine independant and can be used on any machine nearly.

[oldtobor]

If anything they could have kept the successive languages simple. Why do you need to learn an entire object architecture to program in Java ? They could have extended BASIC in a slight way to include GUI builders etc. Like when you click a button on a form it could have been written as :

 

if button1 = clicked then

print "I am stupid basic"

end if

 

and similar things. I just think there is way too much fluff in modern OO. Anyways new languages will come out constantly and it is also good to some extent, maybe the best deal is to create alot of very simple domain specific languages, so you program fast and well in given domains with very simple languages (like games or modeling software etc.).

[radiohead]

Well, some programming languages, like BF (or Brain****) are meant for very small programs. That is teh only reason I can see a new language coming out. To fulfill what others can't. C can do everything that C++ can. Assembly can do anything C can, and so on. But C can't make as small of programs as BF can.