Networks Generation

I love these misconceptions about COBOL. COBOL was a great language and still is but...

am a fairly recent graduate, and guess what I got my first job as a COBOL
developer. I have been a COBOL developer for three years since graduating.
I can truly say I wish I stayed away from COBOL and so should all you recent
graduates.

It really doesn’t matter whether or not there are 70 billion lines of COBOL
and all that other stuff; if you can’t get a job in this area (trust me I’ve
been looking). All you have to do is type COBOL in a job search (and there
we have it – the last time I got 100 hits) – “COBOL IS DEAD” if you want to
get a job. When you narrow down your search to say London and you only get
about 12 hits – so if you want to work as a software developer stay away
from COBOL.

It really doesn’t matter if some developer says “COBOL is not dead, I’ve
recently got a job in COBOL”. These people probably been programming in
COBOL for a life time (that’s like 25 years) so you are competing with these
guys for those handful of jobs.

It really doesn’t matter if one says “there will be plenty of jobs when
these old people retire and a company will give you loads of money if you
have COBOL experience”. They may give people with COBOL skills loads of
money to fix a bug but how often do these bugs cause faults (remember if its
working don’t touch it). So you maybe sitting unemployed for years before
anything goes wrong, waiting for your one off big contract payment (sounds
like too much of a risky investment).

Anyway I’ve been looking for work out their and finding it very hard (almost
impossible) to find a job. I am a First Class graduate and have a Masters
and even with these academics three years of doing COBOL has crippled my
chances. I wish I had not done COBOL and stayed with the skills (what you
should be learning at university) that have jobs. I have got 3 years COBOL
and someone has 3 years of Java or C or C# etc.. so I am out of the
competition when looking for jobs. Don’t listen to those that say it don’t
matter what the language is (because it does), all you have to do is look at
the job specs (they generally mention the programming language because it
really does matter).

I am now hoping to build on those skills at university and then will keep
trying to apply for jobs. I am also hoping to do some courses to refresh my
knowledge in areas I have neglected over the last three years (which from my
experience you don’t get from COBOL and my opinion is this language should
not be taught at university). Certainly its IT and you have too keep up
with recently skills set (that’s RECENT!! skills set), but why get into the
position I am in and many (I hope not too many) recent graduates may be in.

I know this is very negative and may not read well, but COBOL has made it
very difficult for me to forward my career and I really don’t want this to
happen to any other graduate.

Basically “COBOL is truly dead” if you look at it from the perspective of
your career. Go out there, do the research and see the truth for yourself !!

Oh and forgot - you don't have to be a pundit to do a job search.

Unless of course you've been developing in COBOL, working in the same company, for so so long that..

"The use of COBOL cripples the mind; its teaching should, therefore, be regarded as a criminal offence. - Edsger Wybe Dijkstra

If you are a COBOL developer good luck to you - and I bet you've been working in the same company for so long. If you want a change I here CWJobs are advertising 7 jobs citing the word COBOL in London(Not sure if they are the same mind).

Don't even think about wasting good education time to learn COBOL at university, its funny but some have started to teach it). But the people who like COBOL say it a easy language to pick up anyway (And I agree I never even heard of COBOL but I got a job as a COBOL developer - and now I am a Tester :-( ). Hopefully not for too long :-)

Completely Obsolete Business Oriented Language eh, Peter?

On a practical level, you are dead right.

Running a cold job search doesn’t immediately reveal a wealth of COBOL jobs at the end of the rainbow.

The point though is that many large companies still use mainframe technology.

Although some are moving to more modern platforms, which itself requires the ability to translate legacy apps coded in COBOL, some are choosing to stick with mainframe and COBOL because it is so reliable.

Those companies that stick with the mainframe should ensure that they have the skills to support it tomorrow – something that isn’t always immediately obvious with the brains in HR who write the job adverts today…

This blog is titled “Cobol programmers back in demand” when they are not – if COBOL was in such demand I think the jobs would be a lot easer to find.

Also you’re saying that a person which works in HR does not have a clue about the technology which is used by the company. I think you are putting these down and/or you are trying so hard to find some hope that COBOL is not dead, but the truth is that “COBOL IS DEAD” from the point of career prospects. Just imagine that you went for an interview for a SQL administrator job and they ask you questions on ISAM datasets (I think that person working in HR would be sacked). If they wanted a person with COBOL they do put COBOL as HR would do with any other programming language.

COBOL is a great language as no doubt it is running so, many, many transactions, in the back end. It’s like that old piece of hardware that works and works and never fails you. So if it works don’t replace it (and why take the risk if it has been doing what it supposed to for the last 30 years).

I am not saying that COBOL is obsolete it still will be running somewhere for years to come. But when something brakes sometimes its cheaper to totally replace it. How many years will we try to keep the mainframes going by simple adding/patching/modifying the code to get it to interface with the new?

Also many companies move away from the mainframe and migrate to Windows/Unix/Linux to run the COBOL. Where they are using different compilers and even have to modify the COBOL to get it to work (hence they can’t move away from the vendor which probably supplied the development tools, runtime etc). But to the point if you are using a different compiler/runtime what’s to say that it going to run as it has been doing for the last 20 years?

Everything dies sometime, but COBOL has left a legacy and was a great language, after all there are over 80 billion lines of it.

I used to be a COBOL developer (my first job in IT) and I did this for three years but the development experience was not good enough to get another job as a developer. Why did I leave you may ask, did not get pay in-line with my peers. After this I had to take a job as a tester to get exposure to technologies that are in demand, the longer I was doing COBOL the bigger the skills gap would have become.

Maybe I was not smart enough or just bone lazy. I leave you with those 7 COBOL jobs on CWJOBS for you to decide.

I love computerweekly articles – nothing comes specific to mind at the moment.

Thanks for the reply Anonymous – hope I don’t sound to negative :-)

http://legacymainframe.wordpress.com/2008/04/27/legacy-programming-languages-pay-range/

According to this post - COBOL skills went down in demand and went down in salary in the second half of 2007.

Sorry to hear that a recent graduate (Peter) chose the wrong technology (COBOL) to get into software development in his first years after graduating.

Peter although you lost touch with the skills that are in demand in the software industry, I do feel for you that you are now forced to do a job (tester) you really don't want to do.

My advice would be to get some certifications and maybe try to get involved in open source projects. This will allow you to get back as a software developer faster.

Also thanks for your post this has really open my eyes that "COBOL is dead" and companies (which are really big) which sell COBOL services are simply pushing misconceptions about it.

"My advice would be to get some certifications and maybe try to get involved in open source projects."

Some advice on the types of certification worth getting would be helpful.

Very interested that you mentioned open source projects – what would you say are the key technologies worth learning from a developer standpoint?

Hi Peter here.

I made a big mistake doing COBOL this was the real issue. It’s not about predicting which skills will be in demand, but learning transferable skills. COBOL does have some transferable skills but they are not enough.

For example, if I had learnt about relational data bases i.e. SQL these skills are not just in demand but transferable. But I got experience with ISAM which really does match up.

For example, if I had learnt Java or C++ these skills are also not just in demand but also transferable. But I got experience with COBOL. COBOL has a totally different syntax/program design compared to other languages. Even the basic constructs are totally different, assignments, declarations to name a few. If you write a ‘for’ loop construct in C is basically the same in Java, C#, C… Also other areas which you can’t get from COBOL are OO (I do know you can get OO COBOL today but most/or all of its not written this way) .

I am a developer I think a certification will allow me to understand what a language can do (learn the constructs and syntax). Also doesn’t matter the specific type of open source project, as trying to debug/code will allow me to gain experience on a development project with reasonable size code base.

Another point in the choice of certification can also depend on the subject area of development you want to work in. For example that’s up to the individual, for example one may be happy doing the low level of it and not interested in the higher level design.

But got to go – may continue this at another time.

Just wanted to point out yet another misconception about COBOL.

See http://www.itweek.co.uk/itweek/news/2215565/salary-boost-workers

This report is also invalid and an unfair comparison. Although its not saying that COBOL is not obsolete directly, it however implies it.

So you must be thinking wants my reasoning behind this "COBOL skills earn good money", I must be taking total rubbish.

Well the simple reason is that COBOL has been around for a long time. An when a person works in a company for along time its obvious that there salary would increase.

The real way to obtain stats in this matter is to include the number of years experience for a given salary. Not sure how you would do this but maybe the people that do these stats can come up with an idea (maybe?).

There are so many people out there just trying not to believe that COBOL is obsolete in respect to real career opportunities.

So will this be my last post on this blog? Any other questions you have for me about the "real misconceptions about COBOL"?

Or is there anyone else out there who really believes COBOL is a good career move (please read full post as your questions may have already been answered)?

Also, is there anyone old enough to have been forced into technologies just to stay in the industry and compete.

In the 80's is was COBOL and VAX/VMS. Jobs started to peter out early 90's.

In the 90's it was client-server, most typically UNIX and C.

In the late 90's-2000's (and now) virtually anything non-microsoft was percieved as obsolete.

I went through all these phases. I can tell you the best years was JSP/COBOL and UNIX and C. A programming job was both fun and challenging.

Nowadays, everything is pretty much run by M$ if you want a job, and now you have to have a slew of bloatware technology skills to stay competitive. The rules of the game for job seeking is simply what you can memorise from M$ monopolistic technologies just to get a job a hardly do any programming, just wax lyrical about how wonderful bill gates is.

It 2009, and I'm definatley changing careers for sure, the glory days of well and truly over...

I learned COBOL at university in 1989 - even then it was aging, so we quickly moved onto Object Orientated. As a software developer it was much more useful to learn C, C++ and now I mainly use ASP.NET AJAX. If anybody tell you to learn COBOL or tries to teach you it at University I recommend you make a formal request for change in course content. COBOL was useful in the 70s/80s - it has no place in modern business systems.