Slashdot Mirror
Your Tech Skills Have a Two Year Half-Life
itwbennett writes "Eric Bloom, an IT leadership coach and former CIO, has answered that eternal question 'does working on old software hurt your professional marketability' with a somewhat surprising 'no.' But, Bloom adds, 'a techie's skill set from a marketability perspective has a two year half-life. That is to say, that the exact set of skills you have today will only be half as marketable two years from now.'"
Depends really on how specific your skills are.
Knowing, for example Java or .NET programming languages won't decline in value that fast. Perhaps specialising in certain specific products will- and certainly the development environment will.
On a non-programming side- knowing the basics of computer hardware doesn't decline in value that fast. Perhaps specialising in certain models does.
"That's the way to do it" - Punch
Suppose I know some amount (X) of C now (Just out of college)
Will that be less valuable after having 2 years experience in the field?
That is to say, that the exact set of skills you have today will only be half as marketable two years from now.
But taken together with Moore's law, you can market them twice as fast. So everything works out the same.
As a general rule I don't even list things on my CV (resume) that I have less than two years experience in, these days...
I'm willing to accept this is the case for startups wanting the latest buzzword filled technology, but a LOT of places are happy at a much slower pace.
Take this with a grain of salt, because every two years another set of software is old in absolute IT terms.
Uh, this article has a catchy subject but very little content and nothing to back up its claim. C'mon slashdot! Go to reddit for stuff like this...
.. an IT leadership coach ... uh-huh. Veiled message is "take my course, buy my book". I'm still employed using skills I learnt in 1980. Eric Bloom can get the hell off my lawn.
"The greatest lesson in life is to know that even fools are right sometimes" - Winston Churchill
I still program in Java which I've been doing since 1998. I also sometimes program in Python which I've been doing since 1997. Obviously some things about those languages have changed, but many things haven't.
OO languages are fairly similar to what they were 10 years ago. As is OO design, etc. There have been large changes to frameworks etc, but there is a significant "core skill set" which transfers over.
In my case, my skills have not become become less marketable at all over the last two years. Recently I spent two years out of work (voluntarily), and when I returned to the job market I had no problem whatsoever finding a job.
I think the half-life of skills is more like 15 years.
Eric Bloom, an IT leadership coach and former CIO
Doesn't seem like his coaching help himself any.
That said, I've been coding QA software in some VB-Form language since 1994. My pay during that time has only increased. This is the first year that I've had to do anything in a C-form language.
The unfortunate fact of the matter is that a lot of new technologies are horse puckey. C++ was an actual improvement over C. The .net platform, for all its many faults, has actually increased my productivity, but much of the rest, Windows Presentation Foundation, Python, Ocaml, Ruby, Silverlight, et. al are nifty, but nobody *needs* them. Frankly, if the world standardized on Java tomorrow, and we just used extensions thereof for different platforms and purposes, we could all concentrate on getting useful work done and quit dicking around with learning the latest obscure and allegedly more elegant syntax. The best language and syntax isn't the most logically consistent one, it's the one you know. In productivity terms, human factors trump formal systems elegance every time.
Please do not read this sig. Thank you.
This certainly fits my experience. I'm "over 39" and have specific tech skills that date back to the early 80s. Those are worthless. I continued doing highly technical work and staying current into the late 90s, when I went back to school to build up some of my non-technical skills. Not such a good idea as it sounded. I emerged from school several years later with just enough still-marketable skills to land a tech job that offered little opportunity to further advance my skills, then got laid off from that, took a retail job as a life raft.... and now my "freshest" marketable tech skills are a dozen years old, and close to worthless. I guess it's time to get out the paintbrushes and see if I can swing a new career as an artist; at least the half-life on those skills isn't as short.
http://alternatives.rzero.com/
Even COBOL refuses to die. C, C++ and it's variants are still everywhere (Objective C for Apple's iPhone App Store) decades later. Java has outlasted the fads of Ruby and Rails. HTML has been around ... well ... since the Internet. Javascript continues to be the #1 web scripting language.
So no, your skills don't have a half-life of "X" number of years.
It depends on the specific skills and industry specialization. Among (many) other things, I've been intermittently doing embedded C code for 2+ decades. If the half-life rule applied here, then my embedded C coding skills would be roughly 1/1000th as marketable today as they were 20 years ago. Embedded C is still used in the defense and avionics industries (among others)... there's still fair demand for it (though admittedly not the sort of demand there was 10 or 20 years ago).
You are not having the point. He is making comparison with the nuclear half-decay rate, which says in general the same, that your skill will have be decreased half in 2 years, and then another half after 2 more years, etc.... Here you could calculate when your skill would become worthless, if you have some math skill of course (which in general has a bigger half-decay period)
I can play pong just as well as I did in 1984! I took my Atari 2600 out of the basement, fired it up and hit an all time high score again. Time to hit the Wall Street Occupy protest to complain about evil CIOs and how their greed is destroying my reputation as a highly qualified gamer from the past.
For software engineering, I could agree with him that languages, IDEs, paradigms, etc., are still evolving very quickly. For all I know, they will be evolving at that speed in perpetuity.
On the other hand, I don't think this is true for all "techies." The tools for electrical design, for instance, haven't changed much since the introduction of 2D CAD tools for PCB layout in the 1980s. If you've been soldering, prototyping, debugging, and laying circuits out for the last 20 years, chances are pretty good that your skillset is still market competitive with people who've just been trained. If you've been out of work for the last two years, I doubt that you are any less good at doing those things than a fresh college grad. Perhaps more so, since there are many finer points of electrical engineering that only can be learned via experience. You might still have difficulty getting a job in the current environment, but it won't be because your skills are out of date.
And although new versions of SolidWorks (for instance) come out every year, the tools for 3D mechanical modeling haven't changed since the introduction of parametric modeling over twenty years ago. If you were designing manufacturable parts five years ago, and were able to produce quality detailed drawings and discuss designs with other people, you should be able to jump right back in and do that today: you haven't missed anything substantial. The biggest change, perhaps, has been the introduction of rapid prototyping machines. You can approach them differently than you would traditional manufacturing, but that's hardly necessary to start making use of them.
...but twice as profitable?!?
You might be the only one left who understands RPG III, laughing on your bed of C-notes, dressed in clothes made only of diamond, flying around in your hover-car made of pure aerogel, and renaming nation states with a click of a mouse!!! LOL!
Well if the "leadership coach" says my current skills are only good for two years I guess it's time to ditch Linux, Apache, PHP and MySQL.....
Perhaps that's why he's a "former" CIO...
My skill set includes excellent problem solving ability, communicate effectively and work hard.
That skill set seems to only get more valuable as I get older.
"a techie's skill set from a marketability perspective has a two year half-life"
Well, a marketie's skill set from a technical perspective has a zero year half-life.
My unix "Tech Skills" are still quite marketable after more than a decade. Sure, specifics like futzing with IRIX software streams might not be useful any more, but a good 80% or more is still standard.
I do not know why this is in the front page, and I do not know why the educated crowd of Slashdot listens to BS from the CIO/CEO/CXO of the day and his new genius theory to quantify things he should not, mainly because he does not understand what technology is about. These guys should be in marketing. There are new technologies and old technologies, and jobs for all of them if you are good and know the right people. If you are very good at Fortran or Cobol you can get a job. If you excel at Java or C you can get a job. None of these are new technologies by far, and the skills are highly portable from one to the other. The basic knowledge you need is always sort of the same, a mix or common sense, knowledge of the basics (algorithms, data structures, and a brief background on the problem domain you are working on), and some minimum social skills.
What if your tech skills are troubleshooting and adopting new technologies rapidly? What if your tech skills really have nothing to do with a particular piece of software? This former CIO wouldn't understand that by being a CIO he knows less than the people he had working for him. Which to me makes his opinion on how long someone's skills are "valid" invalid.
In Maine, the technology is 10-15 years out of date, so even with 4 year old skills you'll be like a person from the distant future.
CIO more than 2 years ago , so what is his half life relevant CIO experience .
Time for everyone to learn agile development and working methods :-)
Tinfoil hats, everyone, please.
Has anyone noticed how often you see expressions like "erosion of skills" in the popular press? In a "Who Moved My Cheese?" sort of way, it won't belong before every schmoe quietly accepts the idea that, if you're unemployed for one @#$%ing day, YOUR SKILLS WILL HAVE ERODED. ALL OF THEM. In IT it will be worse. "I don't care if you're a Java expert. Because you were using it 2.5 years ago, you're only 50% of a Java expert. Accordingly, I will pay you only 50% of the going rate. Anyway, those skills can be acquired in 21 days, so why should I think much of your expertise?"
...or is it most of the poster here, who bash him?
Did he say that C/Java/whatever decays, or only the worth of people who use this skills?
I can imagine an interpretation of his statement, which would make sense. In my youth I coded in BASIC, Forth, PASCAL... This was somewhen in the middle ages. I was ok for that time, but today those skills are decayed to nothingness. I made some money in JAVA projects. Only a few years ago. Since then I didn't use JAVA at all. I image it would be much more difficult for me to find a JAVA project now.
So the marketability skills certainly do decay, but usually only if they are not constantly used.
Soft skills like playing well with others, selling ideas and products, listening, etc. and "non-HIGH-tech" technical skills like driving, using a pen and paper, typing, etc. probably have far longer half-lives.
Knowledge is how to play a game, intelligence is how to win, wisdom is knowing what game to play.
The primary skill at my embedded software positions for the last 20 years has been C. It's what I use most of the time and what I'm quizzed about in interviews. However, I continually need to pick up new skills to supplement my toolkit.
With exactly zero evidence to back it up. The faster we ignore this entire story, the better.
AccountKiller
Windows 7 may go just as long.
Some industrial systems still have windows 9.X, ISA cards and other older stuff.
2 years is to quick and lot's of places may do long testing time of new OS's / software any ways before roll outs.
Totally flawed analogy. The figure might hold true for latest fashion in development technology, but its insane to think that fortran skills for instance will be half as marketable 2.5 years from now. They will probably have declined by a few percent but difference in value of 40 year old tech versus 41 year old tech is negligable. Its more like the value of a technology falls by 100/(2+years-since-hot) percent every year.
http://rareformnewmedia.com/
If you are trying to market yourself with buzzword technologies and languages, then yes, your marketability decreases over time. On the other hand, if you are marketing yourself for less trendy technical work, maybe not. There are still a lot of COBOL and FORTRAN programmers out there, and they command some pretty competitive salaries. There are a lot of systems that were installed over decade ago that work just fine and just need people to support and maintain them, along with occasionally adding an interface for some newer system.
Palm trees and 8
This may have been less the case for someone that knew the ins and outs of Xp a few years ago, since the platform has been around for 10 years (almost to the day). But Xp is being phased out in many business markets and many say that an operating system will never have such dominance again, I could definitely see a 2 year half life sounding appropriate. Couple that with ever changing technologies and software and this sounds right on the money. Further illustrating this is how Microsoft and other vendors offering certifications (Cisco is another example) are now putting an expiration on their certs. You may get your CCNA or MCP now, but if you don't take another test for 3 years (I think that's the average now) and your cert expires.
The article is mostly about IT in the sense of database/SQL skills. When all you have to sell is the ability to code in some vendor's API, and when new versions keep appearing at some regular intervals, you need to keep running to keep your place, like in a treadmill. But there are many jobs where the coding skills are essential/necessary but not sufficient. In scientific application development (CAD developers like AutoCAD, Ansoft, Ansys, Fluent, Cadence, Mentor etc) the marketability could improve with experience, if you could demonstrate that you other skills have benefited by experience. I am very sure the analysts, architects and other higher level workers in IT will see their value and marketability improve with experience and demonstrable successes. But if all you have skill are the ability to program in Oracle version XYZ, your marketability will be tied to that version of that software.
sed -e 's/Chuck Norris/Rajnikant/g' joke > fact
Every single statement referenced the "software vendor". Every software vendor's goal is to lock you in to not thinking and just buying your way out of any problem. Saying you have technology skills because you know some software from some vendor is like saying you can play guitar since you've got such high scores on the XBOX/PS3/Wii for Rock Band. Even if you know something from that "software vendor" inside & out, you don't know shit unless you understand the fundamentals under the hood of what the toolset is doing. That's why so many Windows "Administrators" are idiots - unlike the harsh world of *NIX, they don't (think they) need to understand what's going on under the surface. Just point & drool.
*** Sigs are a stupid waste of bandwidth.
This is a reflection of a serious problem in the area of hiring decent techie folks. There's a difference between a "marketable" skill and a "useable" skill. A marketable skill gets you hired by people who are clueless about what makes a good techie (hardware or software) and only know buzzwords, whereas a useable skill is what the people who you're going to work with and for HOPE you have. Sometimes skills overlap between marketable and useable, but my own observation is the larger the company doing the hiring the less overlap there is.
im earning my living from php/mysql/html/css for the last 6 years. and im earning even more today. and having to turn down potential new clients.
Read radical news here
I suspect that the Bloom is referring 'tech' skills in a general sense. Most IT people are not programmers, and thus consume rather than create software products. If you have 'skills' using Office version X, it will probably not be as valuable in two years when a new and improved product, Office X+1 comes out.
.Net have been around awhile and are not going to go away.
Obviously, if you think of IT as just programers, what he is saying makes no reals sense, since staples like C, Java, and
HA! I just wasted some of your bandwidth with a frivolous sig!
The bastards told me I'd only ever have to learn electrical engineering once, or does he mean "tech skills" for certain specific values of tech.
As a programmer, I can say that programming itself, that is, *how* to write code, in terms of methodology -- is a skill that will never leave you once you have acquired it (so long as you keep using it).
Almost any programmer worth their salt can learn a new language in a few weeks, if not days. Granted it may take more time to develop understanding of any idioms or warts the language may have, but you can learn that stuff on the fly, unless you're writing HA/mission critical code, in which case, there'd better be a review process, and it's reasonable to expect that someone on the team will be an expert in the technology being used.
So I'd say unless you've given up programming entirely and have moved on to a different career, your skills are still valuable, and will stay reasonably "fresh" even if you're writing code in a 30-year-old language (as the article says), as long as you actually think while you write code, and aren't just a copy/paste/munge wizard, not that there's anything wrong with that, for certain kinds of things.
This of course doesn't even consider the (imho) much more valuable part of being a software developer: being able to converse with non-technical people, in whatever human language you use, and then translate that into some sort of actionable programming work. That's often more than half the battle. Then of course there is testing, testing and testing.
The article isn't completely wrong, but (like much of the "IT industry") I think it missed the point of what skills are actually important to doing software development. Knowing how to use a specific bit of kit is pretty far down on the list, I think, for any reasonably competent programmer/technologist.
I treat anything with the word "marketability" in it with suspicion.
I've been using Unix & SQL for over 20 years now and still make good money doing so.
I think the best understanding of this is when you dont use the skill we talk about. For example you program with Java and then decide to do graphics design. In two years, assuming you didn't forget anything of your Java programming, your skills will be twice less relevant.
I have nothing against ongoing training and learning what is good from the latest fashions in the market. But I think your "marketability" (how attractive you are to potential bosses/hiring staff) depends on how much your skills are tied to any specific tool/software, and how you reflect that into your CV. IMHO, you should focus on providing good highlights of your previous job roles, and clear descriptions of your skills more than your obsession with the latest tools& versions. Things like personal projects, philanthropic and academic achievements (papers, posters, projects, not your grades, unless you are fresh out of high-school) are also surprisingly good eye catchers for hiring staff/potential bosses. Unless your potential boss demands a ultra-super-duper expert in any piece of software, they will simply look for the most outstanding CV from the bunch, in terms of clearly understandable experience and roles, and displaying 'the right' attitude (team worker, self driven, self teaching, etc).
I just left a job doing objective-c and iOS stuff to work on a windows application that is all c++ and opengl.
I work at an extremely profitable semiconductor fab making the latest chips. Many of our systems run DOS. Dozens of them. We make billions (not exaggerated) of dollars with DOS variants and IBM mainframe/3270 terminal based systems, that frankly, work very well. I have a stack of floppys on my desk that actually get used. Support for this old stuff is expensive but its barely a blip compared to the rest of our costs.
but how many years until HL2:E3?
.....i know devs that spent ages learning html5 (six months in one case) - only to realize after a few short months that the hype was fading and no one actually wanted their skills at all.
cheers for that w3c - xhtml3 anyone?
OMG, so my skill of using a keyboard then has put me into a complete untenable position, career wise. I've been using a keyboard for over 28 years now. What do you people use today, do you talk into your mouse?
You can't handle the truth.
*My* most valuable tech skills are trouble shooting and adopting technologies rapidly. Remember that House episode where he diagnosed the lady in Antartica? I do the tech equivalent, but without a video link and I have to get the diagnosis right on the first try. And I adopttechnologies so rapidly that I am correcting vendor engineers knowledge of their own product before the SOW is worked out.
I recently had to look for a new position after mgmt botched a customer relationship and I can tell you those skills are very valuable but not marketable. I couldn't get past the buzzword bingo to talk to mgrs who didn't already know me. Because you can't assess those skills easily in an interview the way you can product knowledge.
That's what the author means about the skills not being marketable.
your skills have a 2 year half-life...from a MARKETING perspective.
Let's take it in that order.
I've been reinventing myself every 5 years (roughly). I'll ignore my first 5 year gig (Fortran..sigh), and jump to C. I stopped doing C code (mostly) around 2002. Jump forward 2 years and my C skills are about half as marketable as before. Jump another 2 years and they're 1/4.
Doesn't mean I won't get paid what I'm worth or that jobs aren't out there. Rather, it's harder to find the next gig.
Would you hire someone who hasn't done C for 4 years? That answer should be "maybe".
I think he's a little aggressive. That half-life might be 3-4 years....but other than that, it's fairly accurate.
I design and configure GUI automated testing systems. The particulars change. The principles don't. I've started to design server and virtual machine environments to control the systems more precisely and easily. Think virtual machines are going away? Doubtful. I'm sure the particulars of these too will change, but the principles won't.
Please do not read this sig. Thank you.
It jobs need apprenticeships not 4-5 years in a class room. IT is a trade and CS is the high level stuff.
The fact I can troubleshoot classic MacOS 7.6.1 up through 9.2.2 and a number of old-world PPC related hardware issues over the phone without being anywhere near the machine in question is hardly Buzzword Compliant in this day and age.
The fact that I learned basic troubleshooting out of self defense in that environment, however, gave me a great baseline for dealing with hardware and basic software issues in the general sense. While any classic MacOS-related "certifications" may be long useless, the fact that I got that knowledge in the field with plenty of practice instead of out of a book or classroom lecture provided long-term benefits that no class or HR-friendly tickybox ever could.
The fact that hard-won knowledgebase went from being Current to Niche to Hobbyist over the course of a couple of years is one of the major reasons I've stopped giving a shit about staying "current" on hardware and software. It's a moving target, and I have much better things to do with my time - namely using the production software everything else is there to support.
Back in the "olden" days, I grew up programming basic, assembly, and later in "C" on Vic20 and C64.
Now work for a major chip manufacturer, and some of my duties involves developing bootloaders, firmware, apps, and cryptographic libraries for SoC (System on Chip) and smart card platforms. Those old programming skills allow me to develop on platforms with extreme memory, storage, and performance limitations. I noticed that this is a challenge for those young wipper-snappers that only developed on platforms with giga-bytes of RAM & storage, multi-core processors, etc., and never had to learn to be efficient with limited resources.
Programmers should be thought of as artists. Just as a skilled oil painter can easily switch to water paintings, it should be just as easy for a skilled programmer to switch to different programming languages and platforms.
Being current in some area of software today is about knowing its defects. You can read the manual about how it's supposed to work, but knowing what actually works is essential to high productivity.
I've been thinking about this recently in connection with Mozilla Jetpack, which is a library for making add-ons for Firefox, etc. There are two websites, a blog, a forum, a Google group, a development committee, two completely different sets of development tools, "hack sessions", and an IRC channel. There's even an "app store" of sorts with a review process. With all that, you might think it actually works.
No such luck. Only four Jetpack add-ons have made it through the approval process, and they're all rather simple. I've been converting over code from a working Greasemonkey script, and I've spent 80% of my time dealing with bugs in Jetpack. I've filed two bug reports so far, neither of which has generated any useful maintenance activity from the developers.
That's the kind of information you need to be current on to be effective. Knowing that A works and B sucks is important, and that info does degrade rapidly.
The problem with programing is that the supposed good ones (They aren't especially good, just narcissistic and lonely.) are claiming that you need to spend your nights and weekends (all your free time) learning new languages, learning new skills, new technologies. Basically sit in front of a damn screen all the time letting your body and soul deteriorate... No thank you!!
Continuing education is a good thing, but keep it in balance. I know for some people the balance is different than it is for me, But I also know far to many people who horribly depressed and do not realize it and feel the need to spend all their time in front of screens and think less of people who don't spend their free time learning ocaml, perl, ruby on rails, or whatever the flavor of the month is.
What this world needs is more well rounded people in the figurative sense not in the literal sense.
People say do what you love. I love to do different things with my free time than what i do during the work day. Besides, strong social skills (which I admit I need to improve) are just as important as strong technical skills in career advancement anyway.
Im sorry if I offended anyone. Im just sick and tired of programmer/IT culture and working on my career path to get away from it.
Maybe for the world of IT tech support. but not for anywhere else.
Over the past 10 years I personally have observed....
Embedded systems experience has only went up in value.
C programming for embedded systems and critical systems has skyrocketed.
automotive computer programming has skyrocketed.
digital video experience has went up in value.
SCADA systems experience has went up in value.
AV integration and smart buliding experience has skyrocketed in the past 10 years.
What has went down in value. Windows support, Office support, C programming for desktop apps.
What have I seen crater in value? IT Management skills.
...instead of CS. Ohms law has awesome resale value.
Based on my experience, two years is a bit of an exaggeration. Just look at Windows XP. The software is ten years old and still in wide spread use. The real danger is missing out on major trends. In my own career, I almost missed out on virtualization. My employer did not have any plans on virtualizing and that is where the industry went. If I had stayed put, my career would have been dead.
On the other hand, my knowledge of IT allowed me to make a move into a better position with a company that did not have its head stuck in its ass. By knowledge, I mean platform and technology agnostic understandings about how IT should work. Things like, you need to have a good data protection strategy. You need to have a security architecture in place. You need a DR plan. I would add some soft "skills" to that, like being able to translate business requirements into technical solutions, and being able to guide management to a consensus position on what processes are needed, and which technologies can support those processes.
Unless he stated "on average" (didn't read the fine article).
If you work with:
- M$ software, two years is realistic, as people going to work with Windows 8 now realize (btw, sorry, but you have been warned and mocked the ones who warned you... so bear with it).
- extinct platforms (mostly killed either by Windows or by Linux IMHO), this also happens and all of a sudden, which is really sad because some of them are extremely cool (e.g. Amiga, BeOS etc.).
- free software -- specially the most famous and most commonly Unix-based/related -- then this is BS. You know things that last for decades.
In 2009, I started to use Vim at work (I already used it at home since 2000, I guess), but I first used a program like it in 1983 (if memory serves me well) which had more or less the same commands, including macros and regular expressions... under CP/M!
I don't need to talk about C or Unix itself. Do you think "ping" will be around in 4 years? "less" was created but "more" still exists. I could hire some 70-year old Unix expert and he still could be useful.
Even what I learned with a PDP-11 helped me with CP/M, and then with DOS, and later yet with Linux.
A 2 year half-life isn't bad, if the remaining 2 years recedes at a geometrically decreasing rate. /nerd
Well, considering that the average bachelor's degree is 3 year and the increasingly common masters is 4 years, almost half our degree will be useless by the time we are finished. So why am I studying again?
You're a ways ahead of me but what I learned between 2004-2010 for Windows "help-desk" stuff is still good for some other 3-4 years. I purposely stayed away from the harder volatile server side stuff, because I like Durable Knowledge. So I'm backup Helpdesk and a "line" accounting administrator.
My first Journal Entry ever, in 8 years! http://slashdot.org/journal/365947/aphelion-scifi-fantasy-horror-poetry-webzine
i just was hired for skills which i acquired 8-15 years ago (but maintained them) which are unrelated to my PhD title. So i dont exactly know which IT skills he is talking about, but the only skill of mine which lost value strongly is actually my perl coding skill. Or is he talking about the clickedmins which don't find the control panel and need training if the windows version changes.
I can relate to this. I design embedded systems where memory and clock cycles are always in short supply - my latest project was about 4K of C and assembly. In that environment, it's all about squeezing the maximum performance out of each dollar. Oddly enough, the skills and low-level system design stuff that I learned as a teenager in the mid-1980s now makes me wickedly competitive when it comes to developing consumer electronics.
2 and a half years is co-incidentally the same amount of time it takes to train a fresh college grad to be halfway useful.
IMO that's a better to think about it than how fast it takes for you to degrade.
If tech skills' marketability have a half-life of two years, then 22 years ago we were 2^11 times as marketable as we are right now, which happens to be about the same as the number of work hours in a year. Hey everybody, remember 1989 when you made your current yearly pay, every hour? That was so fucking awesome! I must have partied pretty hard back then, because I have no memory of what I did with all that money.
As copyright owner of this comment, I authorize everyone to defeat any technological measure which limits access to it.
Some people learn only the cookbook level of tech skills. They know just barely enough syntax, buzz words, and key words to hack their way through a project. For such people, even the 2 year half life is optimistic. I once met a guy who knew how to configure TCP/IP networks, but only using MS tools. He didn't know how routing or DHCP configurations worked, but he knew how to set it all up on Control Panel.
Others learn the true spirit of what technology is all about. They know not just syntax, they understand why it works the way it does and what can be built at the outer limits of creativity. These people can get a lot more than a 2 year half life, and they can even parlay a current skill into new tech skills, using the old skill as a base.
I've been changing jobs every 2 years in the last 6 years (average), my professional lifetime.
I'm positive, don't belive me look at my karma
I've got two years' experience now. In two years, I'll have four, so I'm covered!
Now I just need to figure out how to get 4 years' experience in the two years after that...
If I have been able to see further than others, it is because I bought a pair of binoculars.
The converse is true, that you don't understand large and highly-parallel applications until you develop them, as well.
Brian Fundakowski Feldman
UNIX, C, Bash, Perl and SQL have around since before the dinosaurs and are still in high demand - the demand is just getting higher actually. There could be a problem in Microsoft land, but I don't do Windows.
My reasoning for being hired all these years was BeOS knowledge on my resume.
I will agree that the keywords you use, and the specific subsets of the technology you have on your resume change about every two to three years. But categories don't change. Supersets change, but rarely. You just have to be creative in what you call yourself, and use a lot of rich keyword terms in your resume. You're ability to pack a resume full of useful keywords is actually more important than your ability to do the job, in my opinion.
This signature has Super Cow Powers
The actual skills have a marketable life only as long as those skills are actually sought after. Unfortunately, this explains why there are still those among us who program in COBOL, which is not so much a language as it is a festering carbuncle held together with bubble gum and duct tape.
However, the real goldmine is information harvesting and data mining, which is pretty much irrelevant to a specific programming language, and is ultimately a true test of near-indefinite skill.
The article seems to make the presumption that technical skill is based solely on the knowledge and use of specific software applications (and the hardware they run on). As most people here will intimately understand, this is not the case. Application specific knowledge is _one tiny part_ of your average engineer's technical skill arsenal and to suggest that her / his entire skill-base is eroded by 50% as a result of the relentless marketing of new applications is, frankly, insulting.
The two-year half-life is exactly true but very dispiriting. When I think back to the huge amount of time money and effort to turn myself into a Novell Netware Ninja, and for what? You can't just give up and stop learning but it's still annoying. Brain surgery students have it easy by comparison...there will always be brains that need fixing. Nobody is going to suddenly show up with brain 2.0 either.
I compensate by deliberately spending time learning non-IT skills such as sport and foreign languages where your hard work stays relevant forever.
Your parents weren't as stupid as you thought.
- I've got bad karma because I won't parrot everyone else's opinion
I suggest value has multiplied rather than decreased
Of course. Resumes can only be so long, and the scanner looking for keywords is going to be looking for the "latest" keywords. I could fill my resume with my extensive experience in SNOBOL, FORTH, and Z80 ASM, but that would displace something newer. So the article's author (pretty short article and fluffy, by the way) estimate the half life at 2 years. From this and the knowledge of release cycles and resume lengths, you could probably derive some (inaccurate) model.
Still have the disks and everything.
Even Half-life 2 is like 4 years old.
No, I didn't RTFA.
Be seeing you...
Granted, professional and experienced ass, and the use of a halflife as a measurement is appropriate and to-the-point, but still it is a number drawn straight out of someone's ass.
For a non-trolling point: it really depends on what your skills are on- knowing Java, C/C++ and your way around Linux ALWAYS pays off.
The three laws of thermodynamics:(1) You can't win. (2) You can't break even. (3) You can't even quit.
This is marketspeek at its worst.
If someone starts to speak from a company, and has a label C*O then it's time to take any such crap and throw it into the trash. Comments like these shouldn't even be taken with a 40'-container-sized grain of salt, just throw it into the trash.
Your skills are only valuable, it seems, if you are willing to work a 6 month contract in the mountains of minnesota somewhere. Ok so i am not really being honest but those sort of things happen all the time to me. That and they want people that can manage some MFC work along with Rational Rose and every other IBM'ish legacy system that would require me to be born 10 years earlier than i was. Then of course you have the HR people call you up and try to verify how many years of experience you have with visual studio and no matter how much you explain to them that its just an IDE and you've done eclipse, netbeans and everything else, they just dont seem to get it.
I know what a variable is. I know what a loop is. Those haven't really changed in the last 20 years.
Coder's Stone: The programming language quick ref for iPad
This is why my resume will only list technologies that are less than two years old. But of course most employers want substantial experience in any job candidate.
So if anyone is hiring, I have 5 years of experience in Windows 8 administration.
who prays for Satan? Who in 18 centuries has had the humanity to pray for the 1 sinner that needed it most? ~Mark Twain
Relatively speaking if you are grounded in solid principles of design and execution then the language you choose is purely based on the elasticity of the environment you're working in. Knowing the fundamentals, and being able to communicate those across many different levels is far more valuable than being a top-notch coder in a specific language. In any large organization you will never be able to create a system by yourself in the timeframes demanded; however, if you can adapt and provide a solid understanding about the "why" and communicate that they you become much more valuable. Languages are the whip-end of and core understanding, design principles, thinking and digging, are all on the handle end...they are the most important...what language you use will vary with time.