Leadership

The top 10 IT skills on the way to extinction


Training, and retraining are a constant part of working in IT. You simply can't rest on your skill set and stop learning or the technology world will pass you by with little remorse. With that in mind, Mary Brandel over at Computerworld has put together a list of the top 10 skill sets that are on their way out in IT:

  1. Cobol
  2. Nonrelational DBMS
  3. Non-IP networks
  4. cc:Mail
  5. ColdFusion
  6. C programming
  7. PowerBuilder
  8. Certified NetWare Engineers
  9. PC network administrators
  10. OS/2

If you want to hear Mary's reasoning, you can read the original article.

She's right, for the most part, although I don't know that PowerBuilder was ever quite as hot as she believes it was, and I certainly don't believe that OS/2 was ever a primary skill set for many IT pros.

I would replace those two with "PBX installation" (still out there but destined to be almost completely replaced with VoIP and managed through IT) and "Printer maintenance" (also still out there but printers have gotten easier to manage and most companies just don't print nearly as much as they used to).

What IT skills do think are headed toward extinction? Join the discussion.

About

Jason Hiner is the Global Editor in Chief of TechRepublic and Global Long Form Editor of ZDNet. He is an award-winning journalist who writes about the people, products, and ideas that are revolutionizing the ways we live and work in the 21st century.

223 comments
CraigCameron
CraigCameron

The bad thing about IT is nothing ever really becomes fully extinct.

nkjaykay
nkjaykay

I disagree. PB 11 is now a .NET development tool.

cf
cf

Fnck you man! ColdFusion will live forever... silly ass hat .net is for kids

happy.hermit
happy.hermit

I think she's right in the context of mainstream roles and training. Unfortunately (or otherwise) legacy systems will always exist that will require specific expertise in tired technologies (e.g. C programming). Look at the banking and insurance industries and some of their systems!

bleyers
bleyers

I have been in the IT industry for more than 30 years now and I am constantly amused by statements about COBOL becoming extinct. Most corporation???s today still run application programs that are built on 1960 architecture which hover between assembler and COBOL. I seriously doubt that COBOL will retire in my life time. I have managed applications that run off the mainframe and servers and would put my money on the mainframe coded applications any day of the week. The sleepless nights, marathon conference calls on bridge-lines while trouble shooting server problems wins it hands down??????

renowayne2006
renowayne2006

Isn't Cold Fusion is gettting ready to release coldfusion 8? I don't think it is "on the way out" at all. If FACT, (which Mary's article suggests very little of), I don't believe that to be true at all....

cvestal
cvestal

I replied to the original article by Mary at computerworld in a very civil tone a month ago, but now regret that it was not more ferocious and pointed. Both the original author and Jason need to actually do REAL research and come up with REAL content, versus just inventing/recirculating a half-baked list that is either extremely obvious, or gratuitously controversial. I do not pretend to know the motivations behind the propagation of this heap of trash, but it seems very apparent that neither show any professional determination to actually do their jobs and RESEARCH instead of passing down "gossip". Is this what techrepublic.com has sunk to? If so, I expect I will cancel my subscription to the newsletter and encourage every other developer that I know to do the same. Most developers I know do not subscribe to gossip columns... There should be a public apology by the original author and those that continue to distribute it. Do your job. Stop being lazy! Maybe write a followup about all the technologies on the list that are in actuality THRIVING - like ColdFusion. ColdFusion 8 will be released very soon... Chris Vestal

krolrules
krolrules

Since I'm only a number 9 in your list, and really don't have much todo with COBAL or cc:Mail or any others listed, all I can say is... as long as there are PC's or workstations in the world, there will be users to "screw them up," and admins to fix them!

khalid1690
khalid1690

I am little bit of agree with the Author..... He totally left the hot demanding skill set which are rare i.e. Information Sysmetms Security and IT Project Managment...... These two skills are very necessary, mandatory and highly paid.....

bcswartz
bcswartz

Um, so you basically repackaged the original, dubiously researched article and didn't really add much in the way of new insights. Not exactly impressed.

esequeira1
esequeira1

I do not want to beleive it but its true. The above were already out but needed somebody to pen it

nick
nick

What about DOS Batch File Scripting? Now that was an IT skill worth having

r.rangachar
r.rangachar

I don't personally agree completely...I think C is still widely used as well as needed! Otherwise though, the cc:mail is also not necessarily required on this list either. it's always good to have that option..right??

Brandon_Forest
Brandon_Forest

IMHO: DotNet will eventually crowd out Java in the next 10 years. Everywhere I go, Enterprise scale applications done in Java are being re-engineered in DotNet. I think that Java is doomed for the same reasons as C. It's to unweildy to create applications in a timely manner.

ChEng
ChEng

"... most companies just don???t print nearly as much as they used to" Printers have gotten better, overall, but I have seen throughout the 4 companies I work with that printing has increased. Significantly. Availability of information on the Internet, coupled with the better printers and quick printing has given the paper blob a new lease on life. Hopefully my experience is an anomaly, but I don't think so in the SMB world.

Tony Hopkinson
Tony Hopkinson

.Net is for kids, cookie cutter web application building is for whom ? Infants ? If you'd have said C or Pascal or assembler, or something you might have had a point...

apotheon
apotheon

Have you tried responding with a little maturity? (edit: and maybe some proper capitalization and punctuation while you're at it)

mattschmatt
mattschmatt

While I agree that ColdFusion (CF) is not as popular as other server-side scripting languages like ASP, PHP, and JSP, but it's NOT on the verge of extinction. I think the authors probably ran queries against job databases and whichever ones were appeared the least made the list. I'm a CTO and have been using ColdFusion since 1999. It's funny how smug IT people will rip me and others for using it, but it works and works well for me. I've developed major web applications for doctors in CF and have had no problems as long as you learn how to properly configure and scale CF. It integrates nicely with Java and is also open to work with just about any other language/platform. I can build a simple routine in a few lines of code that would take 5 times as many with .NET or PHP. Version 8 is coming and I'm anxiously awaiting to utilize a few new features like the PDF tags.

oarf
oarf

Tell everybody you know that these skills are on the way out. That way, those of us with those skills can only go up in value! Also, in that previous list of What Makes a Good Programmer, add the ability to learn new programming languages....

GoodOh
GoodOh

Read number 9 again and place the emphasis on the middle word and it might strike you as having a different meaning than it seemed the first time you read it. I think you may be responding to a point other than the one being made.

apotheon
apotheon

Wow, what a waste of time. It's kinda like taking Unix shell scripting and sawing off all its limbs.

GoodOh
GoodOh

C is not dying (the evidence suggested for its decline makes more sense to me as evidence that the market is saturated by the existing developers not that the language itself is dying or likely to be replaced soon - not sure if that's true but it's an argument worth discussing). While so many drivers and games and other work continues to be done in C (and the existing systems need maintaining) there will be a call for it C and its derivatives (C#, C++, Java, etc). are still the core of most university education in programming which has to tell you something about the regard it is held in as a 'foundation' language. It also explains why the market is saturated and why people need more than 'just' C to rise above the pack and get the good jobs. While .NET might well continue to rise in importance relative to Java I really don't believe that it will 'push Java out' while Java is cross-platform and .NET is Windows only. MS dominates but to suggest that the vast majority of developers will simply cut themselves off from the small but real proportion of the world running OSes other than Windows in the next decade or so seems highly unlikely to me. I also don't believe it's highly likely that large numbers of public funded universtities will abandon C (etc) and change to directly teaching .NET etc as a foundation of their courses. Imagine a degree based around .NET and MS abandons it in the student's second year? Even if a student never uses C (etc) in the real world its longevity and shared history (not to mention the complete absence of any cost to the student for IDEs, compilers, etc) mean its foundation role is stronger than many give it credit for. However I've been horribly wrong in the past so... [2nd EDIT - Remove 1st edit inserted in error because I can't follow a thread properly. Time for sleep it appears.]

mark
mark

Java is a language (sort of): .NET is a proprietary marketing exercise

apotheon
apotheon

It helps that the printers used by businesses these days are pretty much all Postscript laser printers with built-in TCP/IP print server adapters. When buying a printer, anything less is basically just a waste of money -- which is why I have a big ol' HP 4050N. In fact, I got the thing for about $35, used, from the overstock warehouse at a local university -- with that kind of accessibility of superior hardware (superior, that is, to the crap you can get at something like CompUSA) for half the price you'd pay for the consumer garbage most people buy for home use, there's [b]no reason[/b] to settle for junk. In any case . . . quality printers are more available and make network printer setup and maintenance much easier than it used to be. It's no wonder printing is so common in the office. The flipside of this, of course, is that dedicated printer service professionals are pretty much limited to field tech work for printer manufacturers, while on-site maintenance is usually the purview of the general IT staff -- so if you're a printer repairman, you should be learning other skills or preparing for retirement, unless you work for Xerox or HP.

Tony Hopkinson
Tony Hopkinson

It will work for a bit with some business types, but if they start believing the environment is dead, start re-skilling.

yyyc186
yyyc186

Did he bother to put "natively speak English" in that list?

hillman.d
hillman.d

C# code can be compiled and run in GNU/Linux, Windows and FreeBSD using Mono. There is also a module that can be plugged into Apache web server to run ASP.NET. The Web Forms model is very nice and it smoothes over the differences between Web applications and desktop nicely. Performance is not bad, either. Couple that with Generics and other nice things and it's not that bad. I have actually uprooted a medium-sized ASP.NET Web application from Windows to GNU/Linux with no problems. I think the community should help Mono as much as possible. The way Microsoft keeps changing things and hooking .NET more into Windows will make it much harder to be cross-platform.

apotheon
apotheon

The .NET framework and it's Common Language Interface provide some nice benefits for certain types of development. Unfortunately, it has been seriously overhyped and overmarketed, and I suspect that in a few years many businesses that have made stupid decisions like migrating enterprise back-end software infrastructures from Java to C#.NET will find themselves in dire straits. Many of them, of course, will be entirely unable to associate the problem with their choice to migrate everything to a different, less well-suited technology, but that doesn't change the fact that .NET fever will surely be a big part of the problem in many cases. The .NET framework suffers some significant shortcomings as a general-purpose development toolset. Among these is long-running server software performance. Another is the simple fact that the .NET framework is, in general, very poorly suited to most innovative development -- it is for the most part sort of a "drag and drop" programming tool. Doing anything significantly new goes against the grain of the benefits of the .NET framework. Don't misunderstand me: it's great for certain types of CRUD programming, and that's a big chunk of what people need in the software industry. It's just not the panacea everyone seems to think it is, and a lot of corporations that have bet heavily on .NET without fully understanding the limits of the technology are going to be hurting in a few years because of it, when the limitations of the technology become more keenly felt.

rdrainer
rdrainer

Headhunters are telling me that there is a current demand for combined Java and COBOL skills. I haven't had a chance to go with it, but headhunters never lie, right?

NickNielsen
NickNielsen

[i]...if you're a printer repairman, you should be learning other skills or preparing for retirement, unless you work for Xerox or HP.[/i] Not necessarily so. I see a booming market for printer support at my level. Although the big four (Canon, HP, Xerox, and Lexmark) do the majority of the work themselves, many SMBs have found local contractors who can do the work faster, with the same (or better) quality, and without the hours on the help line ("Now, plug the network cable back in and restart the printer..."). I do some side work repairing laser printers myself. It consists mostly of training users, cleaning the printers, and replacing worn rollers, drums, and fusers, but it could become full-time very easily were that my choice.

apotheon
apotheon

I intended that to mean that it was a waste of time to develop such a broken implementation of the idea of shell scripting in the first place. Microsoft should have had its developers working on something useful -- like either a proper shell or porting a proper shell to DOS -- in the first place.

ElGoliath
ElGoliath

"DOS batch files . . . ? Wow, what a waste of time. It's kinda like taking Unix shell scripting and sawing off all its limbs." I take that to mean you don't think it is a handy tool. Handy tools don't tend to be a waste of time imo, but you are free to have your own opinion-

ElGoliath
ElGoliath

But I guess I am speaking from a quick fix point of view. So far writing small batch files is the fastest way I have found to do recurring mundane non critical tasks (scheduled tasks). If I'm using a batch file, it's probably because I want to be done in under 2 minutes so I can move on to something else. It's not the most robust option, but it works very well at what it does not to mention that unlike cmd, bat is compatible with win 9x which unfortunately is still necessary these days.

Tony Hopkinson
Tony Hopkinson

have more than dos batch files. On top of that you could use one of several scripting environments under windows. There's not that much more support for scripting in other OS's in terms of volume but on largish admin scripts or ones where you want to interact with the OS, the benefits are extraordinary. Remember Choice.com ...

apotheon
apotheon

"[i]It's fine and dandy that other OS' have better options, but for the time being it's a windows world, and knowing how to write batch files is a very handy tool to have under your belt.[/i]" I don't think anyone meant to suggest that it wasn't a handy tool when working on MS Windows systems. I just pointed out that working with DOS batch files is a bit like using a proper shell with its limbs sawed off, and Tony Hopkinson provided a little explanation of my meaning when you expressed confusion.

ElGoliath
ElGoliath

But that helps me when I'm working on a Windows OS how? It's fine and dandy that other OS' have better options, but for the time being it's a windows world, and knowing how to write batch files is a very handy tool to have under your belt.

Tony Hopkinson
Tony Hopkinson

available in similar options in other OS' it's next to useless.

apotheon
apotheon

"[i]C# code can be compiled and run in GNU/Linux, Windows and FreeBSD using Mono.[/i]" Unfortunately, most of the libraries that put C# on equal footing with other, similarly useful tools are not portable across platforms. "[i]I think the community should help Mono as much as possible. The way Microsoft keeps changing things and hooking .NET more into Windows will make it much harder to be cross-platform.[/i]" I agree. The more people exposed to Mono as an alternative to .NET, the more people will get the opportunity to consider moving to better platforms.

apotheon
apotheon

. . . but it's not what I was talking about. There's a big difference between a client application and enterprise-level back-end server software. Much of what these overmotivated .NET fanboy managers are replacing with C# on the CLR is software running behind the scenes written in Java that handles high-load server software, with the equivalent of your GIS/Google mashup application running on top of it. "[i]None of this was done using drag-n-drop.[/i]" I hope you don't think I literally meant "drag and drop". I referred to the fact that most .NET programming consists of writing glue code to piece together .NET library calls, slapping an interface on the front, and doing something interesting with the way it's assembled rather than writing the underlying code itself. That sort of assembly of preconstructed pieces is where the .NET framework's strengths are realized, because of the massive availability of library functionality. Where it does [b]not[/b] excel is in long-runing server process performance under high load, rapid (and quality) development of entirely new libraries and software that doesn't leverage many pre-existing libraries, and coding anywhere near the metal. "[i]I've used Java, PHP and .NET. They are all great tools.[/i]" None of them are, in my opinion, "great". They all have strengths that make them almost indispensable, but none of those strengths really make the "great" in my estimation. PHP is probably the worst of the lot, and at the same time the most commonly useful because of its few (but important) strengths. "Great tools" are things like C, Objective C, Objective Caml, Common Lisp, Ruby, et cetera. They, too, have their strengths and weaknesses -- but in general, their weaknesses are not central to the design of the tool, and their strengths often [b]are[/b] central to the design of the tool. I guess, by way of analogy, it's like the definition of a great man. Thomas Jefferson, for all his failings, was a great man -- thanks to exceptional qualities that were intrinsic to the character of the man himself. Bill Gates is just materially successful and filthy rich.

hillman.d
hillman.d

I am no Microsoft supporter. However, I have got to give .NET its props. We just finished an application that pulls data from an Oracle Geographical Information System (GIS) database, performs calculations, combines it with data from Google using Web Services and allows users to dynamically get data about traffic patterns, construction projects, etc in real time on their PDA. This application was also designed using proper O-O and component techniques. Javascript, CSS and other resources were properly packaged and the whole thing can be moved from a GNU/Linux box to a Windows box with no problems. There is also a GTK-based Desktop version that uses the same components that the Web version uses. None of this was done using drag-n-drop. I've used Java, PHP and .NET. They are all great tools.

Locrian_Lyric
Locrian_Lyric

They want people who can interface with the mainframe, and troubleshoot if something doesn't work quite right. I know a few shops that are using the newer technologies as front ends to mainframe.

apotheon
apotheon

We're looking at another round of companies who want to migrate away from COBOL. They're looking for people who know how to translate COBOL code into language_foo code.

Locrian_Lyric
Locrian_Lyric

COBOL + whatever is a hot ticket right now. I've seen: COBOL + VB, COBOL + Java, COBOL + SQL Server, COBOL + MS Office (all).

NickNielsen
NickNielsen

I had a call once where "nobody on site can print." When I got there, I found all six networked printers displaying "Load Paper" and the OS/2 print server locked up because the print queue had reached its maximum size of 4095 jobs. :0 I rebooted the server, purged the print queue, and took the site manager around to each printer to show him why his people couldn't print. With each printer, he got redder and redder. When I walked out the door, he was calling everybody on site to the conference room: "I want every single life-form in this branch in the conference room NOW!" If I hadn't had a dinner date, I might have hung around to listen... :^0 Edit: gramer and speeling

apotheon
apotheon

I think you might have missed the references to home users and people who provide support to them.

ssampier
ssampier

Seems rather foolish to me. The printer is cheap, but the ink is priced like gold.

apotheon
apotheon

That was meant to be covered by choice B.

Big Ole Jack
Big Ole Jack

Bringing the piece of crap back to Staples, OfficeMax, etc for a replacement or a new model, as covered by the "extended warranty" plan that some people buy. I know a guy who goes through 6 printers in a year because he exchanges them like worn out gloves. He simply takes advantage of the extended warranty plan and gets a new printer every few months.

apotheon
apotheon

Keep in mind that when home users have problems with their crappy, low-quality inkjet printers, they: A) throw them away, B) send them back to the manufacturer, or C) come to the PC support tech. Other than at the manufacturer, there isn't really any room for a dedicated printer repair professional in there. edit: If the printer repair trade is going to survive, at this point it's not going to be in the home support market.

apotheon
apotheon

I still foresee some shrinkage in the printer repairman trade, at least outside of the major manufacturers like Xerox. On the other hand, based on the points you bring up, I think I'll retract my earlier suggestion that such a trade is dying. You've convinced me.

Big Ole Jack
Big Ole Jack

many inexperienced users are scared to death of opening up the case of their PC and poking around the motherboard and internal expansion cards. It's these folks who keep the PC repair shops in business and folks like me in business. Also, the bulk of my business consists of repairing PCs' that have been trashed by viruses, spyware, and crapware. Users can be warned and told numerous times to be careful on the web and to download with caution, but advice falls on death ears. It's like scolding an obese glutton of a person when they stuff their face with food, knowing it's not good behavior, but they will continue to do it because it has turned into a compulsive disorder.

NickNielsen
NickNielsen

Repairs also include board replacement and rebuilds. Note also that the user is not expected to change anything more than paper and toner on many of the higher-capability printers. And I ask you, given what you know about the technical abilities of most office workers, would you trust them loose in a printer with a screwdriver? Neither do their bosses... ;)

apotheon
apotheon

I guess I've overestimated the intelligence of humans again. It seems absurd to outsource printer "repair" where such "repair" consists basically of following the instructions in the manual for replacing rollers, drums, and fusers. It's not exactly rocket science.

Editor's Picks