Exactly. that was my point in my earlier posts. I really didnt understand the useless moral lectures from darkman or whatever he/she called himself/herself.

I have left him/her to wallow in his/her own ignorance.

Bossman, I agree with your entire post! My current place of work is an excellent example of this exactly what you describe, and I do not deny that at the present, mainframe/legacy skills are a definite plus.

However, I *do* disagree with the 'its not going anywhere' bit at the bottom. It *is* going away. Its just a matter of time.

Finally, lets keep my points of view in context:

Fortran and cobol obsolete?
My answer is no. Because though fortran has given birth to a legion of imperative children, COBOL remains cobol, and as such does not have an obvious modern succesor.

Is it a good thing to keep things in these languages around?

NO. Even though it is profitable to maintain such kludges, that would be like a doctor saying polio is a great thing cos it keeps him in business. We may get paid to maintain legacy crap, but it remains crap unless it is portable. For Gods sake, the C language is prehistoric too, but it is viewed as modern simply because it is portable, unlike dinosaurs like cobol and fortran. Cockroach versus Apatosaur. Who survived till the modern age, eh?


And as for clocky, if you would rather descend to the level of senseless name-calling, understand that I have written you off as a person of any intelligence whatsoever, and thus would be glad to engage you in a bout of mudslinging insults. bring it on.

Bossman, thanx a lot for ur contributions. I’m now convinced that people and companies still make use of COBOL and FORTRAN. But why is it that a lot of guys think or believe they are old-fashioned despite the upgrade versions that are as good as the newer languages especially in the area of OO programming? Is it because they are difficult to program with? I can see a lot of guys here are condemning them. Well. I’m into VB programming, but still like the techniques of QBASIC, FORTRAN, PASCAL and COBOL languages. At least learning and knowing them is helping me in my VB programming, so I don’t think it’s a waste learning how to use them. I still write a lot of custom statistical and numerical analysis programs, and I find FORTRAN very suitable to do that. I will still welcome your opinion if you have an alternative programming language to suggest for me.

Every language has its pros and cons. You definitely do not want to do any kind of RAD or design User interaction interfaces with COBOL, when there are other languages such as VB, Java, etc. that are well suited for that. OTOH, you do not want to do any kind of data processing that involves millions of records with any other language but COBOL. That's why at a lot of big IT shops, the COBOL applications are still there to handle the back-end stuff, but all the front end is done in other languages.

For example at my previous client, all the user interaction was handles using Servlet and JSP, but the data itself and processing is still done on the mainframe using COBOL. IBM has gone to great lengths to make sure the Java & COBOL interaction using websphere works well.

Bossman, well said. But I wonder why other guys are still arguing about FORTRAN and COBOL still being used by big companies; may be you will need to give them names of these companies for them to believe. But as for me I believe its still being used, else no body or company would be talking about it, not to talk of using it. Some schools in America and Europe still teach students these languages, its not only in Nigeria as some one had earlier said.

Yeah, its true, fortran/cobol are still being used. Languages that were popular in the past, are still in existence everywhere.

However, the question is, is it wise to start new projects using those languages? I believe not- more because they are nonportable than that they are old. I would give the same argument against VB version 6 (not .net, which is portable and hardware independent) because Microsoft has dropped support, effectively making it a legacy language.

The uses of fortran/cobol today is mainly in maintaining legacy systems, which means old systems that are still in use. As these systems are retired, the need for these languages will fade, but for now the skills are very much in demand. However it is wise to mention that the implementations of these languages that you will be supporting are usually upon proprietary, rare, and expensive operating systems- simply being good at watfor77 won't get you a multibillion dollar job at a multinational!

DAKMANZERO, I think you’ll have to agree with what clocky and bossman have said. I have a question also to ask: What makes FORTRAN and COBOL not to be portable? What do you mean by portability of a programming language? It may be different from the what I know.

Portability means sufficiently distinct from the hardware to enable the language maintain its independence.

C was designed with portability in mind, and most major languages since then have adopted it.

It means that your code isn;t tied to a specific implementation of the language. A portable language can theoretically never be obsolete, because it will always be compilable/runnable on the latest hardware with minimal changes.

Nonportable languages like cobol and fortran however are dependent on the hardware/ os they were designed for-hence the problem. When the implementing company ceases support, you are screwed, and moving to better hardware is difficult.

Thats why fortran and cobol skills are best used in maintaining existing installations that are too big/complex to upgrade trivially, or making 'glue' layers like clocky and his team were doing to enable interoperability with more modern systems

However to implement a NEW system with these languages is unwise at best. You don't want to be tied to a particular implementation.

Dakmanzero, I disagree with you. OO FORTRAN and OO COBOL are portable; at least to some extent.

really?

That's interesting. Could u give a link or two? I'd like to see that.

It would really be impressive if they can achieve such a feat without modifying the languages to such an extent that legacy code would be incompatible! If they did break compatibility however, then these languages should be regarded as modern languages, similar to their legacy cousins in name only.

Remember, all imperative languages are structurally derived from fortran. A noncompatible OO iteration of fortramn is no different from any other modern imperative in this regard.

Look 4 dis book, its on the internet. OBJECT ORIENTED PROGRAMMING VIA FORTRAN 90/95 BY Ed Akin, Rice university, Houston

Could u kindly excerpt the section on portability for the benefit of readers of this thread? Thanks!


An example of a project that suffers due to fortran's lack of portability is the Cambridge Crystallography Subroutine Library- there are other examples, too. The problems are mainly due to the fact that fortran comes from a time when portability was not a priority.

http://www.ill.fr/dif/ccsl/mk4man/c2node1.html

A quote:

Portability
There will always be problems trying to make the FORTRAN portable.

The areas which are particularly troublesome in portable programs are:

1)
input/output
2)
use of graphical output
3)
what the consequences of an error are

but there are other areas, which tend only to become apparent when the system is actually implemented on a particular machine.

We are aware of the problems of portability, but nevertheless we insist that CCSL must be portable. Some of the FORTRAN used may seem pedestrian as a result.

Dakmanzero, FORTRAN was developed when portability was not a priority; yes, agreed. But don’t forget that the language has undergone series of revisions and reviews, and portability was one of the areas that was touched or looked into. As I’m talking right now Fortran 90/95 and Fortran 2003 are portable versions of FORTRAN.

No, they are not.

You are thinking of the F programming language, which has been stripped of practically everything that makes fortran fortran.

However, remember I did NOT say that fortran is obsolete. Not by a long shot. Only that it isn't very portable. It wasn't designed to be.

fortran 2003 was designed to address many issues that you can see here:
http://www.kcl.ac.uk/kis/support/cit/fortran/f2003_new_features.html

portability is not one of them. fortran 2003 is an amazing modern language designed for use with scientific and mathematical programming due to its extensive and rich maths libraries- but it has a narrow problem domain and does NOT solve the problem of nonportable legacy code holding back progress in other areas.

Dakmanzero, you don’t need to discourage people. What you feel is the best for you could be the worst thing ever to some other persons. A lot of guys still develop applications with older languages like FORTRAN, COBOL, PASCAL and QBASIC; and they do it well; they love what they do cos that is what they want.

I still write qbasic programs :-)

once I was stranded at Gateway hotel with no link, no software, no nothing and I had to quickly code up a replacement for a logon script gone bad for a conference that involved 50plus laptops and autoconfiguring wifi connections. All I had was qbasic! And it served me well, lololololol! the details will just be boring, though,

Nah, no hatred to people using fortran, cobol, etc.

Not discouraging anyone. In fact, if you people actually read my posts rather than just jumping to conclusions you'll see the many arguments FOR legacy languages I give, and you wont be responding to my posts attempting to discredit me with points I have already raised!!!!!

So, for the 2,000th time I will repeat my position:

USING LEGACY LANGUAGES IS NOT BAD!!!!!!! IT IS THE NOTION THAT "REPLACING OLD NONPORTABLE SOFTWARE WRITTEN IN LEGACY LANGUAGES WITH MODERN LANGUAGES IS LAME AND IMPRACTICAL" THAT I DISAGREE WITH!

Aaaaaaaaaaaaaaaaaaaaaaaaaargh.  Listen for once.

Ishmael - It seems as though you already had a fixed stance b4 posting this thread!

Not discouraging anyone. In fact, if you people actually read my posts rather than just jumping to conclusions you'll see the many arguments FOR legacy languages I give, and you wont be responding to my posts attempting to discredit me with points I have already raised!!!!!

Exactly!!!

Darkman has raised the most objective, learned and conclusive facts about this question, to me, any further talk wud be like dribbling back to your Box-18! (aka Gege s'eyin).

Some schools in America and Europe still teach students these languages, its not only in Nigeria as some one had earlier said.

Good you said "Some"! Fact is in virtually every top class school, focus is mainly on C/C++ and some other "modern languages". Cobol and fortran are just discussed in History Comp. Sci. classes! Well, as I said, the original question,

FORTRAN And COBOL Languages: Are They Obsolete?

is already answered in Brief:

USING LEGACY LANGUAGES IS NOT BAD!!!!!!! IT IS THE NOTION THAT "REPLACING OLD NONPORTABLE SOFTWARE WRITTEN IN LEGACY LANGUAGES WITH MODERN LANGUAGES IS LAME AND IMPRACTICAL" THAT I DISAGREE WITH!

Also, Ishmael, lets be Objective here:

Dakmanzero, you don’t need to discourage people. What you feel is the best for you could be the worst thing ever to some other persons.

HELL, some people even still prefer to watch live soccer matches on DSTV on their 1980s Box Black and White TVs! (e.g Me.).

I hear you. Every body has what he/she likes.

SBUCAREER, thanx men. I understand and agree with ur points.

Hey sbucareer, I like your customized courier/fixed-width font posting approach. Makes sure that your posts always stand out! And insightful posts they are.

However, your last post has some misleading assertions: You said a big software company used cobol to write SAP and ABAP. That isn't entirely correct.

In fact, SAP is the name of the company, and they used their OWN language, a FOURTH-GEN language called ABAP, which is by the way a modern language that supports objects and advanced abstraction, to create the SAP R2 and R3 enterprse resource planning packages.

In fact, at first the SAP ERP packages were not written in ABAP. ABAP was initially intended as a data manipulation language to be used in conjunction with R2. Over time, however,


THE ENTIRE SYSTEM WAS REWRITTEN IN ABAP!!!! WHICH IS EXACTLY THE G&!%^"%"^"^^"^ POINT I HAVE BEEN MAKING SINCE MORNING!

TO REWRITE YOUR ANCIENT PROGRAMS IN NEW LANGUAGES IS NOT LAME OR IMPRACTICAL! SEE, SAP DID IT!!! JESUS CHRIST!

However NEITHER IS IT COMPULSORY! of course cobol code in existence will do just fine. If not, it wouldnt be in existence, ne? However as time and technology moves on, the code will need to be rewritten because that old code is not portable. Starting new projects in nonportable languages is asking for trouble. ABAP is obviously a portable language as it was originally designed to run on top of a hardware-independent platform ie SAP ERP.

dakmanzero, ok na you win. you can argue no be small. na u sabi programming pass; we agree. No body has been able to say anything on this issue that pleases you. Why is it so? Is it that other guys are not well learned or skilled in programming like you?

Actually, @ishmael, the purpose of this discussion is not to prove who knows programming 'pass' anyone else. That's for children and students. Once we graduate, we realise that everyone has their strong points, and even the world's most respected programmers can be wrong.

When we discuss in this way, it benefits us all because we get to see other points of view and if they expose flaws in our OWN point of view, it is to our benefit. If however we do not agree, then we argue the point in order to get the other guy to convince (and therefore help) us.

Also in a public forum, there are others that will tune in just to listen in and probably learn. For the sake of such people it is necessary to argue a point to completion- this was made known to me by nferyn and KAG, two really sharp guys who are known to thrash out issues. Simply giving up and going away, or doggedly sticking to one point of view without considering what others say, are detrimental habits.

The reason why I have been repeating myself is that due to the responses I recieve it is obvious people have not been reading my posts. or have been misunderstanding my point of view. Hence the need to clarify myself.

Its not a matter of winning or whatever. There are coders on this forum that I have learned to pay attention to on certain issues. There's no master of all knowledge.

_{Also, I resent it when people attempt to make me look stupid. That may be a weakness on my part, but for me its the same in real life- if you hit me, I hit back. shikena.}

dakmanzero, i hope u,re not offended by my words? i dont mean to make u look stupid in any way. infact if there is any body that has made this forum interesting and lively, then it's you. I must say that i ve really come to understand a lot of things arguing with u guys on this forum. Please dont stay away, i,m enjoying ur contributions. Cheers.

When we discuss in this way, it benefits us all because we get to see other points of view and if they expose flaws in our OWN point of view, it is to our benefit. If however we do not agree, then we argue the point in order to get the other guy to convince (and therefore help) us.

True, Great Talk Darkman!

Please don't stay away, i,m enjoying your contributions. Cheers.

O jebi! (lol ) Yeah Darkman, dont go oh!

Nah, I wasnt referring to you when I made that comment, no offense dude!

Sometimes I couldnt imagine that I was being percieved the way I was (It looked like I was claiming legacy languages are useless) when in fact I am usually the guy that comes up with all sorts of reasons why we shouldnt just jump to the latest and greatest language just because it's new! I greatly enjoyed my fortran classes, and indeed fortran played a large part in my final-year project at school (Numerical analysis), although I do hate cobol, lololololol!!!! but thats personal!

I have however been made to understand that portability is an issue with these old languages, and this was what I was trying to bring up. Unfortunately it looked as if I was condemning old languages and calling them obsolete, or attacking people that use them. Rest assured that I never did! In fact I was the first to give an actual example of someone who makes a truckload of cash on legacy languages alone! Another of my bosses left to work in Austrailia 2 years back- he got the job because the companmy had trouble finding local talent with the legacy skills he posesses. His salary there is insane! Plus they handled his visa, housing, etc.

I hope my position is clearer now.

Dakmanzero, i got ur points. How comes u hated COBOL and enjoyed FORTRAN? Its like we had the same minds in school then. I only did COBOL cause it was a compulsory course for me; but i never liked it, although i wrote some COBOL programs then but i never liked the english-like structure of the language.

didn't like cobol because of aaaaaaall the (needless) typing, and also because the language couldnt be used for much else other than business report generation (which is dry as dust)

Enjoyed fortran because it is closer to what got me interested in programming in the first place! as a matter of fact I enjoy all imperative languages, haven't encountered one I don't enjoy yet!

Dakmanzero, i enjoyed programming with QBASIC, FORTRAN and PASCAL cos of their mathematical nature and ease to code, unlike COBOL where u write essays. It was one of the worst courses i did in my school days. Its difficult to compile a COBOL program without it flagging errors. there must be an error,and i use to wonder why.

After all the (good) talk on Fortran/COBOL, could anyone please give a link to a free compiler, free resources, etc on FORTRAN (and COBOL wud be appreciated too). Most especially from the "proponents" of these two.

Dakmanzero and food4tot, you both lost me to a land were variables are no longer declare to be any of the premative type. I sat dawn for thousands of years trying to found my way of this redundancy but i could not. You both make my eyes dim and lost focus but one day, ninja go Find my IP address. With my elementry knowledge of programming, i think i could speak in a public room like you people do but for years, i was just trying to read and comprehend your posting. If i should use years in doing that, then you should know how long it will take me to learn programming and splash code the way you guys just did.
but i think there is a solution to this, if you can both carry me along, i mean chating my part in programming, lashing my booty if i don't take instructions, i think this will make me live my procedural way of programming and stat structuring my codes.
I just want you both to help me out of this blind world of mine. i have introductory knowledge of visual basic, java, COBOL, fortran, QBASIC and i want to learn c++, c# assembling language and likings. I want to be goosd at this, i need Una help.