Slashdot Mirror
Battle of the Ages; Stereotypes Collide
JCOTTON writes "A CIO.com article By Phil Murphy explains that "The hype around the shortage of qualified legacy technologists grows each day. Pundits would have us believe that 1.5 million COBOL programmers will suddenly disappear one day, leaving any company with legacy technology in dire straits. The truth is that there are far more programmers with legacy skills looking for work than there are jobs for them, as evidenced by organizations like Legacy Reserves, which functions as a training and job matching service for unemployed or underemployed programmers wishing to modernize their skills."
This article explains many of the issues facing "the upper half" of Information Technology workers."
And here I thought there was going to be a great need for VB6 and that I would be viable for the next 20 years on that alone... Time to learn the new language of the month, I suppose.
I started programming after the days of COBOL. I wonder what everyone else will be using when I'm desperately looking for a Java job...
I just searched Google for Learn Cobol and only got 417k results. Not that popular a subject anymore I suppose.
- Michael T. Babcock (Yes, I blog)
In South Korea, only old people code in COBOL...
"I just searched Google for Learn Cobol and only got 417k results. Not that popular a subject anymore I suppose."
Neither is RPG, which I learned with COBOL, FORTRAN, PL/1 and PASCAL. Ah, the classics.
If they are unwilling to adapt, so be it. I spend a lot of time keeping up on the latest trends to ensure that I am always current. If a bunch of geezers are unwilling to do the same, why should they be given the preferential "upper half" connotation?
They have been predicting the demise of programmers since the invention of COBOL in the 60s. It was supposed to turn ordinary business users into programmers thanks to its easy, English-like syntax. We're still waiting. Now this writer is talking about running out of programmers capable of maintaining code that was presumably easy to write and maintain?
"Pundits would have us believe that 1.5 million COBOL programmers will suddenly disappear one day, leaving any company with legacy technology in dire straights. "
The funny thing is once the galaxan empire realizes we have 1.5 million COBAL programmers, they will invade and snatch them all up to maintain their computer overlords, who's code is in great need of modernization to support the new hardware archatectures that they built for themselves, and since they were originally programmed in COBOL it will be easier to just keep them written in COBOL...
Battle of the Ages -- Stereotypes Collide
Time for a Reality Check
Moderate this comment
Negative: Offtopic Flamebait Troll Redundant
Positive: Insightful Interesting Informative Funny
Nothing to see here
Actually, no they're not. At least not around here. I called a couple weeks ago. They're actually cutting jobs from the drivethrough. Pretty sad when McDonalds is laying off workers.
Perl is probably one of the more versatile languages out there, and IMHO it's the most useful for novices and beginners alike. (Note I did not say easiest, but with that said Perl is pretty easy once you get the hang of it). 0.02
People still using COBOL can migrate to COBOL.net. Fujitsu implemented a this abmonination.
Of course it's a complete coincidence that when the story mentions COBOL, the /. fortune cookie I get says "VMS must die.".
Where's the Kaboom?
There's supposed to be an Earth-shattering Kaboom.
Thus sayeth IT technical college.
God spoke to me.
to run Cobol.
Let the anecdotes begin! My friend's uncle has left the field of programmer/analyst (is that a valid title in english? I'm in Quebec and in french that's how it's called.) after 25 years.
The thing that most people forget in this type of story is that human beings change with age. They become higher level thinker, new interests develop and get more impatient with their old interests as they get older. The fact that there are a lot of older programmers no longer in the field means that maybe the guy left the field on his own to start up a furniture shop. Like my friend's uncle....
and you can fuck off with your shit site (oops i just added it to rbls)
iam suprised you can afford the internet, i guess those child prostitues do earn you money
it's a fscking advertisement.
Not that there aren't a few good soundbites in it, but come on, a consultant defending consultants isn't news.
now that the geniuses with their MBAs have figured out that overseas outsourcing is an even bigger disaster than domestic outsourcing was. ("But how can that be! It's CHEAPER!") I'm hearing from recruiters again. IT is such a huge force multiplier that it's stupid to do anything that will jeopardize its effectiveness. Labor cost is only one variable in the multivariable problem, kids.
Sure, the PHBs will whine about the need for cheap H1-Bs that they can abuse, but I don't see Congress being all that sympathetic at the moment, or at the very least they're too fragmented on the issue of immigration in general to get anything done.
Boom times are here again! Well, no, but at this point somewhat better than average middle class employment will do.
Hmmm, but we still make most of the CPU's and Operating Systems the world uses. Hell, even all the programming languages are English based. You guys overseas don't know about the "Secret" virus program we have to disable all systems in the world running outside of the US in the event of doomsday. Besides, we have our music, movies, and clothes in EVERY corner of the world. Culture is one way to conquer the world. Note this message is all in jest and not flamebait. :)
I dunno, I never have liked to tie myself to one language or another. Maybe it's the CS major, but I find that all languages have things in common, and that I can quickly become proficient in each.
Sure I have my favorite languages, but I treat each language I come across equally; hell, I tolerated and become proficient in Scheme of all things. This way, if the flavor of the day goes away, I can simply pick up a book on the new flavor, figure out how it does business, and get to work.
Good principles and techniques transcend language boundaries.
You can't defeat physics.
Sally Struthers is going to be on TV asking for money for aged COBOL weenies, and I learned PL1 when I heard that it was going to replace COBOL and Fortran. So, think of poor me -- almost forty years dealing with people who didn't know that COBOL was inferior, and all I've got to look forward to is 40 years having a hard time getting charity because I've got a disease that doesn't have a Sally Struthers, Mary Tyler Moore, or Jerry Lewis. I may have to start drinking and get depressed so that Jason Robards and Terry Bradshaw will be on my side.
I'm -still- trying to find a job with my Turtle Logo skills.....
Anyone need a 6502/6510 assembly language programmer? I'm a little rusty, but if I can just find my old book by Compute!'s Gazette, I'll be ready to go!
The CB App. What's your 20?
and fast cars, easy women, big paycheques... The university told me so! Never mind that my friend with a master's in EE is answering telephones! I gotta go give all my money to the YOUUUU-KNEEEEE-VEEEER-SEEE-TEEEE!!!!
So I can be broke and tired during the best part of my life, AND be a good obedient slave!
Tabulations was hard in cobol...
Also for old cobol program there was no COMPUTE statement, you had to do something like:
ADD A TO B GIVING C.
later it was
COMPUTE C=A+B.
easier
Of course this assumes common sense is used in implementing a new technology. It all comes down to upper management. A few will have a clue (usually those with an IT background) but the rest will make up for it with a moronic decision to hire Ernst and Young to to replace 50 million lines of custom legacy COBOL code in 6 months with another COBOL application written for the insurance industry in Australia.
Say goodbye to 70 million dollars and 200 legacy programmers.
Sounds like the Rapture to me.
I'm goin' to hell for that. But if you make me program in COBOL again, I'm taking you with me, rapture or not.
I know I will get flamed by some out-of-work programmers out there,
.
but...
There are too many companies that refuse to move out of the computing Bronze-Age; bite the bullet and upgrade.
The town that I work in (Blue-collar auto-industry) is filled with tool & die shops. Typical scenerio: The owner left the assembly line of Ford/GM/whatever 20 years ago and created his own company. He bought a DOS app to run his business on a 286-server/workstation, and he is surprised and insulted to find out that XP won't run on it.
I have seen shops that Net revenue >$10 million/year, and they depend on a app written in BASIC!!!! as their life-blood.
Holy shit people, it might be time to upgrade!
There is a reason we don't (all) still use Horse & buggys. There is still a market for companies to make horse-shoes and buggy whips, (and I bet that company has a monopoly) but there are valid reasons to upgrade.
There will always be a need for Legacy-based skills, but for the love of $deity don't hold onto old tech that you think "Well it used to be good enough!"
"The price good men pay for indifference to public affairs is to be ruled by evil men." ~Plato (427-347 BC)
"Unemployed PL1 Programmers are Unemployed Too"
Apparently the redundancy department is still employed.
at old technologies, not that the programmers "vanished".
Apparently labor markets are among the least efficient: supply and demand seem wildly uncoordinated...its a market even more influenced by psychological factors than the stock market!
I am NOT showing this article to my boss. I have a job turning old Ada programs into C++ and if I don't puke to death reading the code first, the difficulty that management percieves in finding less inexpensive college hires who know [or want to learn] this suddenly old language will keep ME employed until I retire.
SLASHDOT: news for people who can't concentrate on work or have no life at all and got tired of yelling back at the TV.
When they say about shortage, don't get it wrong. Quite probably they are after the people willing to work for a daily bowl of ramen or similar (aka "replaceable slaves"). Why do you think you are always asked whether "could we do it in VB?".
After all - you _are_ geek, you should enjoy cleaning dust out of office PCs! (Or coding VB which IMO is an equivalent).
"Straits", man. Not "straights".
Wow, will my 18 months of p-system pascal will suddenly become valuable in the workplace? How about my year as a Sage Retrieve (COBOL but nastier) programmer? I think not.
You think being a COBOL programmer is tuff? Try being a FrontPage dependent HTML developer left over from the dot-com days...
"I dunno, I never have liked to tie myself to one language or another. Maybe it's the CS major, but I find that all languages have things in common, and that I can quickly become proficient in each."
Yes they have things in common, but they also have different ways of approaching the same problem. For example the imperative way of programming, verses say the functional way, or the procedural way. Sometimes the barrier isn't the language, so much as it is "the way of thinking" that goes with it.
Unfortunately, most employers are looking for production work experience with those languages.
That's what causes the classic "chicken-and-egg" problem where there are many skilled technical people out there who could actually do the work required but who aren't seen as being qualified because of their lack of formal experience with a specific technical tool.
Mainframe/UNIX Bit Twiddler and long time Windows/Linux Hobbyist.
The Theorem Theorem: If If, Then Then.
I see, daily, an annoying point where IT users are OVER-trained in one technology set, which blinds them to more efficient and effective resolutions to company computer service and infrastructure.
My business concentrates on Mac OS X systems used in a publishing environment. They work much like their Windows counterparts and could even be integrated with the larger domain for more efficiency. But when I speak of this to others they look at me with confusion and, maybe, heresy?
These people act as if Macs are toys or inferior in some way. Of course, this is far from the case, but their training has changed how they see technology. This really isn't the old Mac/PC debate. (Apple lost the first war. But they still found an important place in today's computing world.)
No computer technology is perfect, of course. But the mistaken ubiquity that IT is Microsoft and Microsoft is IT makes all other non-MS technicial initiatives and products harder to sell in concept or through a store.
Vos teneo officium eram periculosus ut vos recipero is.
I guess I'll be talking to those guys at Legacy Reserves, because I heard that Java is the new COBOL...
8 of 13 people found this answer helpful. Did you?
In the same way modern coders learn "modern" langauges like Visual C#.net Java so you they can go and learn older langauge. I had to learn business basic when I joined a company it was a bit different having to use the most unforgiving editors and limited syntax but clearly within the reach of most programmers.
Also while there are probably less than 100 throughbred basic programmers in the UK if we really want software developed there are massive software houses in India full coders with cobol basic and a variety of other languague skills.
Odd thing was the count at google.ca for 'learn cobol' w/o the quotes was 417k, but for google.com it was 224k. Now, about five minutes later, google.ca is at 224k for the same query.
google.az still at 417k.
hmmph
However, when one is looking for work, it seems that one is usually labelled as a "specialist" in whatever technical platform and language used in the last position.
A person with both good knowledge of C and good knowledge of COBOL is usually seen as being a "COBOL programmer" if their last work experience was mainly writing COBOL code.
It sounds silly, I know, but that's what I've seen (and what many others I know have also seen) in the current job market.
Mainframe/UNIX Bit Twiddler and long time Windows/Linux Hobbyist.
The Theorem Theorem: If If, Then Then.
VB6 is dead. Start cranking out .NUT and C# on the double, drone!
Seriously, I think I remember reading that MS said that end-of-life for VB6 is coming up in 2006 or so, but can't find the article where I read that. If it exists, it's likely buried deep within MS's site.
The best advice I could give to someone who's been buried under a pile of MS technology for most of his/her education/career would be to go out and pick up some non-MS languages. That way, if Redmond (or its language of the month) disappears tomorrow, there's a chance that you'll still be employable, and you'll gain a perspective on programming that you might not otherwise have. That's just my opinion, though, and I'm sure there are thousands of MS flamers who would say that once you've gone down that path, you're damaged goods anyway. Take this sort of rambling in either direction with a tumbler of salt.
"Linux doesn't exist. Everyone knows Linux is an unlicensed version of Unix"- Kieren O'Shaughnessy
I've dealt with these guys. They are satisfied with what they're getting out of that type of system and will keep it till the power surge blows it away. Hell, I wouldn't be suprised if some of them checked Ebay for replacement hardware. I'm sure they know were they put those 5.25 floppies.
The simple truth is that interstellar distances will not fit into the human imagination
- Douglas Adams
When I worked for a major airline, the flight planning system I supported and helped enhance was written in Fortran and running on a Unisys 2200 mainframe (which is an older architecture but also a fairly reliable and *modern* platform in terms of its actual hardware).
Fortran was (and is) a perfect language for the type of problem being solved, since a lot of it actually does involve semi-complex calculations.
The mainframe platform is also ideal, as the system is designed as a centralized software app running on a large-scale server and being used by folks all over the world on remote terminals (be they "green screens" or web clients).
Sometimes the older languages and platforms in use really *are* a good fit. Or is it change for changes sake that you're asking for?
Mainframe/UNIX Bit Twiddler and long time Windows/Linux Hobbyist.
The Theorem Theorem: If If, Then Then.
There's the build, the lead-in, and the product. Sounds like he's selling Amentra to me. It's advertising YAY!!!!!!
There are better reasons to get rid of legacy code then lack of programmers.
Any semi decent programmer can learn a new(or old computer language) with out too much diffuclty.
A few months ago a bunch of programmer friends of mine were sent to a COBOL course(for some it was a refresher course) because the company had legacy code to be maintained. COBOL is still a needed skill as are many other aging technolgies but even if there is no fear in skilled personnal vanishing, A company with a bit of foresight will still strive to be rid of legacy code.
The fact is most computer systems become obsolete with in 5 years of making. Patching and repatching ancient code produces complex ugly unmaintanable code with no clear theme behind it, this is obviously bad. And this is the main reason we should get rid of legacy code.
Me
http://www.bls.gov/search/ooh.asp?ct=OOH
Pick just about any job and in the listing you'll find something like this:
Employment of XXXXX is expected to grow about as fast as the average for all occupations over the 2002-12 period. However, job opportunities are expected to be very good because a large number of XXXXX are expected to retire in the coming decade, creating many job openings.
Does anyone truly believe this? No. The only group of people that typically exploit this figure is someone trying to sell you something.
Now, that would be a pain in the ass.
Straits, dammit.
Victims of 9/11: <3000. Traffic in the US: >30,000/y
Read up on your history of programming languages, of Grace Hopper's writings on COBOL, and if you can find them (very difficult), contemporary advertisements/reviews of COBOL for the time - you will find that indeed, it was marketed as a "simple, english-like" language for business people. At the time, it was very simple - compared to custom assembler for each mainframe (which was almost always different between machines even from the same manufacturer, like IBM), COBOL was a breeze!
Reason is the Path to God - Anon
A carpenter is not a hammer-er, or a saw-er, or a drill-er. He is expected to be able to quickly learn and use any of those tools, as needed for the project. A new project can use a new tool (language, os, whatever) as needed for the application. When an old program needs maintenence, it may require some re-learning of the old tool, but that should not be difficult.
I suspect the harder problem is preserving the old development systems and tools. If the compiler (or some other tool) hasn't been used in several years, there is a good chance that it won't work. Or, that we can't find it at all because it didn't get loaded onto the new host before the old host was scrapped. Or, that the old hard-copy manuals (how to use the tools) have rotted and/or been discarded in the trash.
...but rather the database and transaction (or batch) environment that the COBOL itself runs in.
:-) Although he might survive the shock if he's been exposed to REXX...
An IBM CICS programmer familiar with DB2 would have a tough time coming into a Unisys A-series shop that uses COMS and DMSII, not to mention the culture shock when his JCL-conditioned mind runs into a job control language like WFL.
Mainframe/UNIX Bit Twiddler and long time Windows/Linux Hobbyist.
The Theorem Theorem: If If, Then Then.
...where can I use my extensive knowledge of FORTRAN 77?
I for one welcome our galaxan computer overlords!
... had in mind for this post. I tried to use "Score: -1, Troll" as my subject, but when I hit preview I get this: "Cat got your tongue? (something important seems to be missing from your comment ... like the body or the subject!)". So, I had to change the subject and now the joke is ruined.
So, for what it's worth, here's my original <strike>troll</strike> posting:
Ugh, COBOL. Bleah.
Oh well, at least it's not Java.
In the course of every project, it will become necessary to shoot the scientists and begin production.
I just don't have the exact combination of all 28 disparate skills every job seems to require. Though I do lack that one essential skill, lying. "Why yes, I am a certified Cisco network engineer, certified Oracle dba, and I know how to write device drivers for both Linux and Microsoft windows. Isn't everyone?"
- 1 COBOLonians 4:16-17
I'm goin' to hell for that.
I don't know, if God has any sense of humor, you probably assured yourself a mansion in heaven for that piece of work. I'm of the opinion that those who take the Holy Bible too seriously are breaking the 1st Commandment, elevating a written work to "false god" status.
Those who complain about affect & effect on
You're right. It doesn't let you submit anything that starts with "Score:" or "score:", probably, in fact, to prevent this from happening on a more regular basis.
Having been a consultant over the last 6 years, and an employee before that, I say that there are systematic problems in the corporations with their decision making, and mandates.
When it is policy to mutilate a system, and cause it not to function properly, of course the IT staff aren't able to perform their function! Enter the consultants, who recommend the same things that employees complain about in daily life, and voila -- the consultants look like heroes.
I've been on both sides, and it makes me want to no have anything to do with corporations -- except that they pay.
For those of us who are consultants and looking for our next contract, I have setup Recruiter-Rater in order to find and rate technical recruiters. Read some, post some.
Zhrodague.net - I do projects and stuff too.
Yes - even McD is outsourcing the drive-through.
try { do() || do_not(); } catch (JediException err) { yoda(err); }
We're already having problems finding decent applicants for our job posting (*) who know enough Perl. Which is frightning, because we do pretty standard web application development stuff. Most of this kind of work is now done in PHP, but few people want to learn from there and get proficient in Perl.
:-)
What a shame. Perl was my start into this business and it served me well. And that's just 9 years ago.
Yes, we did post an English summary of the job on jobs.perl.org, but all of the applicants coming from there did not even read the basic requirements.
(If you are interested: Please don't apply unless you consider yourself fluent in German. Oh, and yes, we did hire an American citizen recently. He speaks German.
------------------
You may like my a cappella music
One must be cautious when they hear the word "labor shortage". Lobbying organizations such as ITAA are paid millions of dollars from large tech companies to lobby congress and the papers about the doom and gloom of tech labor shortages. This is to justify more visa workers and offshoring. In other words, the "cheap labor lobby".
I am not saying that this is necessarily what the article's author has heard, but it would not surprise me. Organizations like ITAA are shrewd and tenacious. They recently managed to influence many small-city newspapers to publish articles about the dangers of tech labor shortages by quoting companies who allegedly will go under unless they import Indians or move to India. Their leader, Harris Miller, lobbied for more agricultural migrants (fruit pickers) from Mexico in his previous job, according to some sources.
The excuse is the same for tech as it was for agriculture: "Americans don't want fruit-picking jobs". At $3-per-hour, who would? They want to do to tech what they did for agriculture. Different career, same plan.
They should be on the same "geek enemy list" as SCO.
Table-ized A.I.
I don't believe in an upcoming shortage of legacy programmers, and if I did I would consider it a cause for celebration.
Anyone that takes 10 seconds to think about how IT journalism works can figure out the reason for "Expected Looming Skills Shortage Threatens $YOUR_HAPPINESS"
As a journalist, you want to interview the CIO's - the decision makers - to find out what they're going to buy in the future, what their biggest concerns are, etc.
You're less likely to interview the system administrators "on the ground" that really know what's going on.
The CIO's see that their highly-skilled people are overworked. They have limited budgets. They can pay people only so much. They want a future supply of the highest possible quality people for the lowest possible price they can pay. Simple. Easy. It's what drives them, along with the fear that the next virus or worm will compromise the crown jewels of the precious customer databases and the their latest presentation to the CEO about forming a Strategic Vision for IT-Empowered Passionate Agile Business with a Process Management Team of Stakeholders.
"Provided by the management for your protection."
found at top of page linked from article:
"Welcome to Legacy Reserves, the largest U.S. databank of Legacy Professionals over the age of 35"
I think that is a new low in setting the threshold for being "over the hill". This means I was old 20 years ago...god, somebody see if I still have a pulse!
SLASHDOT: news for people who can't concentrate on work or have no life at all and got tired of yelling back at the TV.
Money for Nothin'? ;)
There is a shortage...of workers willing to work 60 hour weeks for $15,000 / year. Truly, there is a shortage in that context. Also, by not being able to find "qualified" workers the backsides of incompetent middle-managers is kept nicely covered :).
When I worked for a major airline, the flight planning system I supported and helped enhance was written in Fortran and running on a Unisys 2200 mainframe (which is an older architecture but also a fairly reliable and *modern* platform in terms of its actual hardware).
But this isn't a mainframe, with its reliable architecture and a source of parts, it's a 286 running a basic app that no longer runs on new hardware - one power surge and they're screwed.
"We returned the General to El Salvador, or maybe Guatemala, it's difficult to tell from 10,000 feet"
Firms are populated by some of the dumbest blindest self serving retards on the planet. There is no shortage of labot. Let me repeat that - THERE IS NO shortage of labor.
What there is a shortage of is 45 year olds with 20 years of experience in a 5 year old technology and willing to relocate halfway across the country for a 50% pay cut on a contract basis for six months.
The COBOL jocks who are still around are not in it for altruistic purposes. They are in it to make a killing. Don't think so? Ok - more than half of the country's COBOL, etc. hands were fired in the mid 90's with 'consolidation' and 'modernization'. Then those same idiots who fired everyone freaked out when they simply couldn't answer their own auditors questions about Y2K.
It was magic. All the middle aged guys who got fired coming back to work and literally pulling a rate number out of their ass. $100/hr sound ok to you? $125? Good cause that's what it's to cost you.
Well here we are 4 years into a capital investment recession in IT and guess what? Those same old Mainframes are still around and COBOL and CICS and JCL are still running on them. Because that work NEVER got done ten years ago. It was too expensive and was crowded out by Y2K.
So second generation executards call in the oldtimers again, this time to 'fix' the mainframe problem because the leases are coming due and the CFO is absofuckinglutely convinced that and ICC capital lease iis more expensive than junking everything and starting over.
Hey I've heard this Opera before. It was called "Client Server Computer".
But make no mistake about it my fellow greyheads. They have about as much respect for you and your skills as they have for the beaker that collects bull semen. What you have to do is rape them on the contract.
And in 3 or 4 years and the progress is excruciatingly slow and they suddenly come back from Gartner executive retreats with the new found knowledge that mainframe is new paradigm they must strategize, optimize and leveragize they'll drop all the migration efforts and put their money back into mainframe system development.
Trust me, IBM would not continue to invest all that money in MVS and z/OS Large Systems if they thought there was a limited future in it.
Every couple of years there is the same old new revolution in commercial IT. It's part of the scenery like starving African children with automatic weapons. Sell them more weapons.
company with legacy technology in dire straights
What, will all the company board members become homophobic, and sit around saying "this newfangled java is teh gay"?
Or perhaps you meant straits, not straights.
My Karma: ran over your Dogma
StrawberryFrog
The "upper half"? huh..
At the Victoria Linux Enthusiasts recently we posted some old code in "dead" languages. Fairly interesting code posted, and worth a look.
Click here or here.
OH COME ON!!! All of these posts, and not ONE of you thought to correct his grammar / spelling??? dire straits n : a state of extreme distress [syn: desperate straits] http://dictionary.reference.com/search?r=2&q=dire% 20straits
Who the hell is the editor on this website?!?!?!
"There is a reason Linux is free"
~me~
There are probably as many C++ programmers today as there will ever have to be.
of mixing up programming languages and API knowledge. Please read this http://www.joelonsoftware.com/articles/LordPalmers ton.html, where the difference is made clear. The problem is that:
* Learn a programming language is not enough. You need to be good at API programming -- and there are lots of APIs out there.
* You can get 90% of knowledge by reading a book about a certain language or API. But the last 10% of knowledge you get only after several years of practice, and that is the knowledge you need to be a good programmer and a good debugger.
So you can't just use refurbished programmers. Experience is essential!
ptex
The hype around the shortage of qualified legacy technologists grows each day. Pundits would have us believe that 1.5 million COBOL programmers will suddenly disappear one day, leaving any company with legacy technology in dire straights. The truth is that there are far more programmers with legacy skills looking for work than there are jobs for them
Did you think outsourcing would stop with application programmers? There are lots of weasels making huge money selling the merits of outsourcing to decision makers. FUD like 'shortage of legacy programmers' is just one of a thousand punchlines being used by these guys to sell another contract to thier counterparts in india.
The trend is catching up to the dyed-in-wool technocrats? Ask me if I care, I don't. You guys should have been more vocal and more supportive of the rest of us while our jobs were being shipped oversees. Your turn fellas.
The government has a defect: it's potentially democratic. Corporations have no defect: they're pure tyrannies. -Chomsky
Linux users don't like Macs either.
:)
The proprietary hardware, lack of open source software (even worse than on Windows), lawsuit-happy Apple, and dumbed down straight jacket like Mac OS operating system (before they used a BSD core) are reasons.
How do you get a Windows user and a Linux user to stop fighting?
Say you think Macs are better, then run like hell.
Just because it CAN be done, doesn't mean it should!
"Repeat after me "The is no such thing a silver bullet". That statment is still as true and wise today as when Brooks said it."
Turn the tables, and have them describe their jobs that way.
... but seeing as how the VP is such a VIP, shouldn't we keep the PC on the QT? 'Cause if it leaks to the VC he could end up MIA, and then we'd all be put on KP.
... it's a 286 running a basic app that no longer runs on new hardware ...
I'm confused about why a basic app (either something written in BASIC language or meaning 'essential - low level') won't run on new hardware.
If the Windows PC is run with the 'cmd' program, it makes a DOS screen. Doing Alt-Enter makes the system look and feel like 286-era DOS, only running 30 times faster.
There might be a problem with legacy ISA bus expansion cards. This bus is being abandoned on all 21st century motherboards.
An approach to dealing with this problem is to use inexpensive microcontrollers to emulate early PCs. An Atmel AVR Mega that sells for about $8 can provide about 25 MIPS of 16-bit processing power. A major disadvantage is this approach is that microcontrollers rarely have address and data busses.
Anyway, I don't understand why legacy PC applications fail due to hardware reasons.
I've got some bad news for you.
You: I think I'll go for a walk.
LR: You're not fooling anyone, you know...
You: I feel happy. I feel happy.
*THUMP*
Your handle seems oddly appropriate to the topic, in a "that would be funny if so many PHBs didn't actually believe it" sort of way.
</tongue in cheek>
"Linux doesn't exist. Everyone knows Linux is an unlicensed version of Unix"- Kieren O'Shaughnessy
Lua rocks.
Not all of the programming languages are English-based. Don't forget Var'aq!
Well, here's the solution to said business-user's conundrum.
1) Get a modern PC.
2) Get VirtualPC.
3) Install DOS 5.0 or whatever on said VirtualPC.
4) Continue to run program within VirtualPC session until upgraded version can be produced, tested and used successfully by user community. If not possible/feasible to do so, user can continue to use old DOS app ad infinitum.
In the end, this isn't about technology; it's about solutions. And sometimes, the best solution also happens to be the easiest solution.
As far as I know, most of those aren't acronyms anymore. There ain't no other way to express 'em.
Mainframe/UNIX Bit Twiddler and long time Windows/Linux Hobbyist.
The Theorem Theorem: If If, Then Then.
I'm confused about why a basic app (either something written in BASIC language or meaning 'essential - low level') won't run on new hardware.
Most likely, it won't run on WinXP.
If the Windows PC is run with the 'cmd' program, it makes a DOS screen. Doing Alt-Enter makes the system look and feel like 286-era DOS, only running 30 times faster.
Nope. It resembles DOS, but it's different. Eventually, stuff specific to DOS may be dropped. Anyway, it's nice to have some support for your accounting software, don't you think?
"We returned the General to El Salvador, or maybe Guatemala, it's difficult to tell from 10,000 feet"
Besides, I've seen online stores selling *palettes* of 486-class boxes for almost nothing that would probably still run that older application, and one of those would give him spare hardware for the next several hundred years. :-)
Sounds to me like he's got a sane idea.
Mainframe/UNIX Bit Twiddler and long time Windows/Linux Hobbyist.
The Theorem Theorem: If If, Then Then.
For a perfect example of the need for legacy programmers. There are several major software vendors who supply medical information systems to the healthcare community who rely heavily on folks skilled in (M)UMPS.
If you're looking for a language less sexy than Cobol, look no further than M.
No battles to the death are recalled. Mumpsman can hit to attack and cause brainsmashing.
To implement the software on modern gear would require a tremendous amount of time just sorting out what everone does and why. It's a much larger problem than just sitting down and hacking it out, even if you have the original source and want to blindly follow the last guy's design.
And then sometimes they just can't match the performance of the old system. IBM's been trying to do away with their RETAIN system since I first started working for them back in the mid '90's. At the time they thought they'd go to a Lotus Notes app on their 486 servers. After all, the 486 was designed to give you the same performance on your desktop as a mainframe, right? Sure, for a single user! They never could figure out how to match RETAIN's performance. To this very day they're still maintaining it. I don't think anyone understands it anymore, really. It's millions of lines of mainframe assembler code from what I hear. It's like this ancient evil that lurks under the surface of the apparently peaceful company, just waiting to consume the souls of young programmers. With Tentacles.
I'm trying to teach myself to set people on fire with my mind... Is it hot in here?
They equate "old technology" with someone shoving a stack of paper cards into a steam-powered box the size of a warehouse.
/. were even born. :-)
They don't understand that some people were writing multi-activity programs (that's multi-threaded for you more modern weenies) before most of the folks who read
You don't tend to see issues like "buffer overruns" in most mainframe environments, and there's a reason for that: that type of thing was engineered out years ago. Decades ago.
Mainframe/UNIX Bit Twiddler and long time Windows/Linux Hobbyist.
The Theorem Theorem: If If, Then Then.
As I said in a response to another post, management doesn't seem to want "real" CS people. They want codemonkeys trained in the language or buzzword of the month, or are looking for an excuse to send the work overseas.
This is the situation we are in.
Management: We need people who know COBOL.
Retired COBOL Guru: I've retired, and am really tired working for you corporate schmucks, but I'll be willing to sign on as a consultant.
Management: Too expensive!
Younger CS Person: I don't know COBOL, but I'm a computer scientist who understands the theory behind languages. I am confident that I can learn COBOL quickly, and any other language you might need me to learn in the future.
Management: We don't want to spend extra money in training, plus real CS people are still too expensive.
Outsourcing Company: It just so happens that I have a bunch of teenagers in India who will work for cheap. I can train them in COBOL cheaply. Let me remind you that you also get a tax break from the Government for hiring Indians.
Management: India? Sounds good! We can increase executive pay with the money we save. Where do we sign?
"You spoony bard!" -Tellah
it confirms that you were killed by Stephen King and Bill gates in an apartment in Maine sometime around 1996
Sometimes the older languages and platforms in use really *are* a good fit.
True enough, as far as it goes. There are times, I think you'll agree, where an upgrade to a modern system isn't just prudent but necessary.
As an example, I've eliminated a mainframe system from a client. I don't recall the specifics of the hardware; only that the sole source of parts for the hardware involved ebay and paying a 2am visit to someone's garage for a disk controller. Moving them to an Oracle DB with a web front end does preserve a mainframe like paradigm, but also alows the use of supported hardware.
In another example, we're looking to replace our mainframe payroll system with something modern. Our customers are screaming for better integration with other applications; the time it takes to get data out of a mainframe system makes the data irrelevant by the time it's deleivered ( 1 - 2 weeks ).
Mainframes have their place, and will likely continue to do so. However, that place seems to be shrinking.
A Human Right
I have friends and family that are very happy with Mac OS X, and my limited exposure to it has been positive. That said Mac users are generally the worst OS bigots. They will tell you that every problem you've ever had on your PC can be solved with a Mac. And they will tell you this at every opportunity.
They conveniently ignore the never-ending stream of complaints about stuff on our corporate intranet that doesn't work right on OS X. It should be noted that we already support Mozilla/Firefox, so it's not like we just leave everyone without IE twisting in the wind. And they also forget the hundreds of dollars they fork over to AppleCare to fix stuff.
I suspect that the high cost of Mac ownership gives many users the impression that they are a member of the computing elite. Newsflash: it's lonely at the top, and IT cares more about supporting the majority of the users. No IT department would choose to multiply their support headaches with another platform to support unless the payoff was enormous.
I call myself a C++ programmer not because I am incapable of using another language, but because I would not want to. If someone asks me to write code in another language, I just tell them to give the job to someone else. It's just not worth the hassle. In C++ I have a friendly environment of many years of code upon which I can build; it would take a very long time to duplicate it in another language, so I just don't do it. C++ can do everything anyway, so I don't expect to ever need to.
Well, that will take my local robotic McDonald's down to only two employees- one manager and one stockboy/server.
Now all they need to do is automate the freezer-to-kitchen-robot connection and the kitchen-robot-to-customer connection, and they'll be able to operate it with only a manager.....
SJW: a person who perceives an injustice, and while correcting it, commits a greater injustice.
I wouldn't consider FrontPage users to be programmers.
Here's more information if you're interested: http://www.jovial.hill.af.mil/
I ended up spending some time reviewing it and figuring out how the functions worked, but it was frustrating.
Here's how it works:
1) Get shiny new Exec.
2) Shiny new Exec wants to use tool A
3) Tool A gets implemented about 85%
4) Shiny new Exec has power struggle with old dinosaur
5) Dinosaur wins
6) Shiny new Exec leaves
7) Dinosaur forces co-existence of legacy tool
and new tool - hilarity ensues.
8) Repeat from 1
Throw in 6-7 Rush Emergency LIFE OR DEATH Marketing nightmare projects ("We sent the mailing out and it has a URL on it.... should we have talked to you first?")
The only problem management has with IT is this - 99.99% of the shiny-suited class is too technologically illiterate to make their own toast in the morning and yet they believe they can guide a technical design process.
No skills. No vision. No clues.
But it's a RUSH.
That's the problem.
I have both in my cube. The PC is routinely down for patches/virus cleanups/etc/etc/etc.
The Mac just works.
Got a love that.
d
I have a friend who is a mainframe COBOL programmer -- knows MVS, DB2, CICS, etc. and hasn't been able to work in his field for over 4 years. And it's not because of a lack of skill -- he's good at it. It's just that the jobs aren't there.
Chip H.
From the article:
Based on my experience in several companies, projects get dragged out for months, mired in meetings to decide what to do. Too many of those meetings leave technical people out, and so the management ends up making highly mis-informed decisions. They might take 10 months to make a decision, then demand the doomed project be completed within two weeks.
But this reminds me of a project in one of my earlier employers (involving what we today call legacy systems). A committee of 6 top executives spent about 6 months trying to decide whether or not it was worth the effort to try to integrate accounting data between two unlike mainframe systems. Eventually, they decided it was worth doing. The project was assigned down to my manager, who then assigned it to me. It was a top priority project, meaning everything else I was doing would be handed off to other people, and I would be taken off the rotating on-call cycle for the 3 months they expected the project to take. When my manager told me what it is they wanted done, it was hard to keep myself from laughing. It got the assignment during the 10 AM meeting once a week. I started working on it right after lunch the same day. 45 lines of assembly language code patching, and it was done in under 3 hours. ALL DONE! I came in at 5 AM the next morning to take the system offline and actually test it. It worked perfectly. All accounting data was transferred with all batch jobs and came out on the reports exactly as intended. So I just took it easy for the next couple days, then finally went to my manager and told him it was all done. His response to me: "you should have held onto it for at least another couple weeks or so". We both had a 10 minute laugh fest, thinking about the executive committee and their meetings.
now we need to go OSS in diesel cars
...and that is the root of the problem.
Pick your poison:
1) Qualified individuals exist, but the job is listed at $35,000
2) Qualified individuals exist, and the pay is acceptable, but the company won't pay relocation.
3) The qualified individual is established in the community he/she lives in, and does not want to move. The company does not want/can't use a teleworker.
4) The company doesn't want to pay anything
5) The HR dept doesn't want to pay anything
6) The hiring manager doesn't value the position at a pay scale that is reasonable
7) The pay is bad
8) The pay is horrible
9) They only want you part time so they don't have to pay benefits
10) They would like you to pay them for the opportunity to work at their illustrious company
http://shit.slashdot.org/article.pl?sid=04/12/14/1 441255
Some companies run help wanted ads to make it look like they are hiring. It helps their reputation and standing as a corporation. I guess the reasoning is that if it looks like they are hiring, it looks like they are doing a lot of business, when in reality they may be doing poorly.
eat shiat and bark at the moon
i received my degree in computer science and cobol was a mere little elective.. tucked away deep at the bottom after all the C++/OOP/misc data structures/VB/Java coursework. no one that i graduated with took cobol. it was offered as a mere elective if you didnt have anything else to take and as a supplement elective for non-computer science majors. doesnt it stand to reason that after a certain period of time there will be next to non-existant cobol and other legacy programmers out there? i have checked other universities..most of them either offer C++ or Java as the first language and 95% of them or thereabouts offered C++ as the main language(the language that will take you through your degree, i.e. data structures, AI, programming languages, etc etc).
The authors of Perl and TIBET(tm) are all over 40. The authors of Rails are largely under 30 but they aren't stuck on Java's marching moronism.
Seastead this.
There's a basic industry-wide degradation in lower-level programming understanding. I've heard the lament from many retired coders who say that programs these days are just so -clunky-! They're used to working in assembler language, and got frustrated by the amount of processing-time used up by in-house software generated by higher-level-language "experts" who didn't understand the resource implications of their code.
I've even heard some go so far as to say that the escalating inefficiency of code (in CPU terms, NOT coding terms) will eventually have a serious impact on the relative acceleration of computing power.
But then, what do I know? I'm sat here coding in Java... *ducks behind cover*
Meta will eat itself
Mainframes: good at tasks that require massive I/O.
Smarter people set up data feeds from the mainframe OLTP system to backend RDBMS, either for off-line querying and reporting, or for further consolidation into data warehouses/Hyperion/etc.
"Mind your own business": if techs are fired, companies say "tough", and suddenly we have to care about them not having skilled workers...?
Just because I'm ranting doesn't mean it is not true....
Hugo
Let me ask you. Is that the UAL unitmatic and apollo systems ? I know they run on Unisys key. The reason I ask is cause my fiance was a UAL flight attendant and I used to use unimatic to check her schedule. I was actually very impressed with the system... the user interface is almost like a text based hypercard or web type dealy. And whoever wrote the online tutorial had a fantastic sense of humor :)
what you still use german in germany? ;)
Next thing your gonna tell me you still use perl
now seriously, if you realy want someone, give them a chance of learning perl and german.
you will not find someone who will be efective first day of arival anyway he/she needs to get acustomed to your legacy code aswel.
anyway post a wanted note on www.vdab.be
german is our 3de languish so perhaps you'll find someone usefull.
and uhm no the ducky didnt help, it would be way better if they ducky were to be deformed after i try to release some stress on it!
It would also be possible to run the application in DOSBox. This would allow them to keep their existing application and run it on newer hardware if/when the old hardware finally dies.
now seriously, if you realy want someone, give them a chance of learning perl and german. you will not find someone who will be efective first day of arival anyway he/she needs to get acustomed to your legacy code as well.
It may be news to you, but an applicant should bring _some_ of the skills that the hiring company needs with him. If the company has to train the employee first the programming language and the company code and the first language used by the company and its clients, they spend months training before the employee is effective.
Some day, when our minds are deprecated, legacy third party integration will be but a fond memory!
To blog is sublime
I meet plenty of people who like my work, from which I draw examples to show at my LUGs and SIGs (which in turn seems to draw in more contracts). Even when my audience is not technically oriented, it allows them to get a good idea as to what PHP/ASP/JSP could be used for. If you start negotiating a contract, your potential contact don't care what back end you use, so long as it works.
This is how I learned PHP. I used to be strictly an ASP/MS-SQL programmer. When an old friend wanted to pay me to do a football pool website, I told myself I'd teach myself by doing it in PHP and MySQL. It was a fair learning curve, but more importantly, contingent on my getting paid. It wasn't an exemplary use of PHP, but it fulfilled the original spec, and it was at least a start.
My next PHP contract allowed me freedom to choose whatever backend I wanted to use, once again. This time, it was better planned, and I had a much sharper grasp of the language. All the client cared about is what it looked like in the end.
After about five such contracts, I could now put my Publicly-Accessible PHP accomplishments on my resume, which actually look professional.
In essence, I consider this Chicken-and-Egg Problem to be evolutionary: create a tiny chicken/egg which then hatches/lays a bigger chicken/egg which then hatches/lays an even bigger chicken/egg until you have one which lands you a really nice contract/job.
Solomon Chang
"Twice half-assed makes an ass whole." --Solomon K. Chang
Every prof. and admin. in th CS dept. of NIU ought to see this. The entire dept. and every prof justifies their employment on the idea that teaching CS students COBOL and other mainframe tech. is worthwhile. Over half of all the classes there use the mainframe in some way, usually either COBOL or S/360 ASM. As if anybody writes mainframe ASM anymore.
Wow... I'm used to dealing with mainframe apps that also dump data to a satellite relational database in real timem, and even Unisys has CIFS integration now in their mainframe lines (well, on the 2200 side of life, anyway).
Given that kind of turnaround, I can see why folks are wanting changes, but I don't thing the mainframe itself is the issue. Maybe the software designer...
Mainframe/UNIX Bit Twiddler and long time Windows/Linux Hobbyist.
The Theorem Theorem: If If, Then Then.
What do you think I'm making my living writing code in? Yet again? :-)
Mainframe/UNIX Bit Twiddler and long time Windows/Linux Hobbyist.
The Theorem Theorem: If If, Then Then.
That way I can be dead twice over! :-) And with OpenWatcom available for OS/2, it should be easy to do! Look out, world! :-)
Mainframe/UNIX Bit Twiddler and long time Windows/Linux Hobbyist.
The Theorem Theorem: If If, Then Then.
In an economic environment with rapid job turnover the legacy systems might be one of the few places institutional intellegence is being maintained.
I'd have to assume a Basic application written for the 286 Dos system dropped support long long ago. By 1991 286 computers were already getting harder to buy new.
ofcourse your right, thats why i sugested looking in a country close to germany.
i can only say this about dutch for certain, but it's close to german. and a lot of us had atleast some german in highschool. So i guess it wouuld be a refresh course rather than learning a whole new languish.
i guess the same can be said about other neighboring countries like denmark norway and sweden.
anyway if the aplicant is a CS he brings his skils in to the company.
But yeah it's a road only to take if the position realy cant be filled anyother way
"All we do is get the analyst to draw a some use case and sequence diagrams and were all sweet"
Management has been bouncing from silver bullet to silver bullet ever since they had to take off a shoe to count the number of operational computers. Absolutely all of it can be traced back to an irrational belief that somehow Taylor's ideas can be applied to software development as if it were an assembly line turning out identical widgets.
Note that this has very little to do with professional best practices and programming paradigms such as functional programming, OOP, structured programming, etc. While management also jumps on those as part of their silver bullet, they spring from within computer science and engineering and persist beyond the flavor of the month attempt to proceduralize programming. Management adopts these ideas in much the same way that a chimpanzee might watch a master carpenter hammering a nail, then pick up the hammer and proceed to hammer a nail, then a bananna, the wall, the door, a window, etc.
The most useful observation I have yet seen is that some developers are ten times more productive than others. That is, each of that subset can replace a ten man department.
Management wants to bottle whatever that subset has so they can take anyone with a pulse and turn them into a super developer at a much lower rate of pay (expand supply and cost goes down).
The problem for management is that all of their attempted proceduralization is a very effective tool for dragging the productivity of that subset down to the average level.
The heart of the problem is that coding (aka code monkey) may be procedural, but development is a gestalt. Evidence suggests that management is procedural. This means the twain shall never meet except, perhaps, at a level of 'meta-management'.
Whenever you find a shop that is using the various management drivin 'development paradigms' AND productivity isn't down the toilet, you'll find developers who ignore management and just do their thing, then have junior team members play fill in the blanks with the management 'tools' in order to justify what they did after the fact or possibly to create a total fiction that LOOKS sort of like what they did as long as you don't try to understand it.
Please note that I'm not talking about 'cowboy hacking' that produces volumes of spaghetti that just happens to work so long as nobody sneezes. I'm not saying that code reviews, design for testing, or anything like that is wrong (quite the contrary). I'm not saying pair programming or extreme programming (or any of the many X programming) is wrong. All of that can be perfectly valid until management attempts to proceduralize it or even worse, lock it into a (probably unrealistic) schedule on a chart.
I find it interesting the way management seems to always want X years experiance with Y (sometimes X is greater than the age of Y) for development, and demands to specify the tools used, but never expects to tell physical plant people what tool to use for their jobs. The latter is probably fortunate since we don't really want carpenters banging nails with a pipe wrench.
</RANT>
Now that I have all of that out of my system, the summary is don't worry, the shops that ACTUALLY drink the Kool-Aid will die early of extreme unproductivity. The remainder will fake it until the next flavor of the month comes out, then they'll fake that one.
Some shops still need the special features which a mainframe provides (mainly things like recoverability and very high reliability, but some don't, and are still using a mainframe environment mainly due to inertia.
If you folks are the latter case, then more power to you. There probably *are* better solutions out there.
I find it strange that you have mainframers who don't like working with relational systems. Each shop is different, I guess, but every mainframe shop I've ever worked at has had its own mainframe-resident relational database group as well as applications using various other types of data storage structures.
A relational database is just one tool of many, and it's very good at solving certain types of problems. It does so at a cost, however, and the performance trade-offs can be important when running a high-end system.
Hopefully it won't be an issue in your case...
Mainframe/UNIX Bit Twiddler and long time Windows/Linux Hobbyist.
The Theorem Theorem: If If, Then Then.
I've done significant work in a dozen languages over my 15-year career. That's one of the benefits of working on multiple mainframe/server platforms and also being a PC and Mac hobbyist on the side.
I wonder if you might've tossed away some gems there...
Mainframe/UNIX Bit Twiddler and long time Windows/Linux Hobbyist.
The Theorem Theorem: If If, Then Then.
My point was that one only has so much room on a 1- or 2- page CV, and that it's quite possible that there are multiple major projects that the person simply wasn't able to list given space considerations. I know I certainly couldn't.
It seems that many (most?) HR qualification processes these days are largely based on buzzword matching, anyway, and a list of languages, tools/environments, and other similar things might be the only way to bring that knowledge to the employer's attention outside of the cover letter (since it's almost impossible to talk to an actual human when searching for work, much less get an actual interview).
For what it's worth, I've had the strangest items on my resume trigger fairly solid job interviews, so I tend to list a number of things on my own CV which might be obscure but which I actually do know fairly well.
Mainframe/UNIX Bit Twiddler and long time Windows/Linux Hobbyist.
The Theorem Theorem: If If, Then Then.
I was only including programming languages from Earth... Besides, I bet the Ferengi have a better programming language than the Klingons. Probably like COBOL...
First time - after six rejects. Seventh is the charm! If I had known that my submission would have been posted, I would have checked back more often. I would have had some comments on replies too. OK, cookie crumbles. I will check back regarding my next submission.