Leadership investigate

Tech skills: Is it getting harder to keep up?

Professional skills and experience are hard won from education, training and time in the industry. But it's amazing how many people get by despite a fundamental lack of knowledge.

Written in Philadelphia and despatched to TechRepublic at 30Mbps over an open wi-fi hub in my Pittsburgh hotel later the same day.

Almost everything I learned at college and university has been used at some time during my professional career - and I still lean on that seminal education when faced with new and challenging problems today. But by degree, the speed of change has seen many technologies and techniques sidelined by progress during my years in industry.

To probe the rate of change I recently asked an engineering class for a show of hands on a series of topics to get a feel for the knowledge evaporation rate. Out of a class of about 100 mature students the count went like this:

  • Who has seen a thermionic tube? = 5
  • Does anyone know how they work? =0
  • Has anyone know how a cathode ray tube works? = 1
  • Does anyone know how a transistor works? = 1
  • Who knows how a laser works? = 3
  • Who knows how an LED works? = 2
  • Who knows how an LED display works? = 3
  • Who understands Maxwell's equations? = 0
  • Who knows how an antenna works? = 0
  • Has anyone heard of the radar range equation? = 0
  • Who has heard about the Schrödinger equation? = 11
  • Who knows what a compiler is? = 15

I won't go on as I'm sure you get the idea. The big question is: does this lack of fundamental knowledge matter? Perhaps not. So long as someone somewhere does understand, the tech world will keep on spinning. But should the last one with knowledge die, we could quickly be in trouble.

For many of us, keeping abreast or ahead of the game is now an accelerating challenge driven by technologies that span every sector and aspect of companies and society.

We can no longer read all the R&D publications or attend all the conferences and courses to get a filtered and distilled view of progress.

Putting all these issues into some quantified context using the best practice I have come across involves the concept of knowledge half-life. The calculation methods are varied and hardly comprehensive, or indeed fully justified, but it is all we have as a guide to the challenge we now face.

The simplest technique is to reference the citation rates of scientific, technology and engineering publications. On this basis, I have to put together the following graphic for a broad selection of disciplines.

Image: Peter Cochrane/TechRepublic

The most interesting observation to make here is that the medical and marine biology students are out of date before they can even graduate, while the physicists have about 11 years of grace.

What does this tell us? The education system as it stands is no longer doing its job and can't possibly work as we move forward and the situation gets worse.

It is obvious that we have to move on, and it all has to be online and available anywhere, anytime, and in a form fit for purpose. But perhaps the biggest leap will be delegating the role of tutor to some disembodied entity - some machine - able to rapidly access, filter and format what is required well, or just, in time.

In addition, individuals will have to assume a greater responsibility for their own course of study. They will have to choose what they follow as wholly prescriptive education paths fall by the wayside. Moreover, education will be full time from cradle to grave for those in the fastest-moving sectors.

In many respects the world is demanding that students grow up and mature far faster than ever before. They will have to assume greater responsibility and achieve greater authority earlier than any previous generation, and they have to do it in concert with a world of machines and escalating complexity.

Will our young people be able to rise to this change and challenge? We'll soon see, but I certainly intend running with them and giving it go.

About

Peter Cochrane is an engineer, scientist, entrepreneur, futurist and consultant. He is the former CTO and head of research at BT, with a career in telecoms and IT spanning more than 40 years.

37 comments
DaveDopp
DaveDopp

In my opinion, any technology-dependent society will always lead to specialization in technology knowledge. The key to successful application of the broader technology is management skills. Thus, a project manager doesn't necessarily have to have intimate knowledge of every detailed aspect of the technology used in a project. What he does need is the ability to assemble and manage the diverse technical skill sets represented by the project team members. Years ago I was the project manager of a team designing and developing software to service mortgage loans. My only previous exposure to programming languages was a one semester course in Fortran and my self-taught skills in Lotus. My task was to keep the programmers focused on delivering a product that was user-friendly (and a most difficut task THAT was!) The final product was elegant, efficient and very easy to use - from an end user's perspective. My point is that even though individuals may be unable to keep up to date in the broad fields, generalists with even a passing knowledge of the technology will blend the specialty fields into useful end products.

mkogrady
mkogrady

I agree that this topic gives me a gut reaction. I have been in the field of IT since 1986 and have not stopped taking classes the entire time. I have talked to new students at my former college on their career training choice and informed them that IT is NOT for the squemish. It's a tough job to be sure, but also rewarding both professionally and financially. As for those professional students out there (in reference to Mr. Newton) I have met a couple that at least academically speaking are much smarter than myself, but they're so focused on theory that they could not possibly deliver a project yet alone an ongoing service to their respective employer and internal company customers. One was a real JERK and I truly wanted to see him dance before his peers and managers to read their reaction, but alas it never happened and I am pretty sure he is still a research student living in a test tube. The skills to be successful are to be Technology Driven, Organized and Customer Focused. You need a bit of Bean Counting and Public Speaking too if you want to climb the company ladder to be an "official" manager and make the bigger-coin. On that note, most of my managers have some basic IT skills, but for the most part are MBA trained. That means they know how to spell Windows, but sure as heck don't clean them (some are well qualified in IT though - so I'm not discounting their contribution). In conclusion - you need to stay current as best you can - PERIOD. There's no rest for the wicked and not much hope to climb the ladder to greener pastures. However, once you do gain experience, there is along career ahead of you since any company that goes to great lengths to hire, train (continually) and integrate you into their culture would be foolish to not take care of you. Enjoy the ride while you can.

krishimprov
krishimprov

I think the old knowledge is getting transferred to machines. So if the machines or the automation to create those machines are lost, real loss of old knowledge can happen. Not sure how much is it of real problem. The problem could happen if something like this happens in medicine industry where knowledge gets completely transferred to machines and doctors lose some of the fundamental knowledge

Smedley54
Smedley54

As the knowledge base expands, professionals in that field must specialize. In that sense, keeping up is no more or less difficult than it has always been provided we are selective. The days of knowing everything about IT were over by 1985 and it's time to accept that and adapt. Basic education has changed very little: Freshman still need binary and introductory programming courses. It does take longer to teach just the basics that every IT professional needs so there's less room in the degree plan for specialization, so choices get driven into either OJT or graduate school. Students entering do know more, so introductory level courses are more advanced than in the past, but we're still pressed to get it all in. Students don't know the things you think they should? Talk to professors and department heads and find out why before going curmudgeonly.

lehnerus2000
lehnerus2000

I say "Yes". The amount of information is increasing every year. How much of it is useful information is debatable. Sir Issac Newton could have (conceivably) mastered every scientific discipline that existed in his age. IMO, it is impossible to do that now. Of course (in theory) it is easier for us to access information.

Tony Hopkinson
Tony Hopkinson

It's no harder than it ever was. There's simply a lower percentage of people who have any interest / need in doing so. I'm having a really hard time with those questions, the audience and the conclusion as well. In fact I'd suggest that the people involved in the study stopped keeping up themselves a long time ago...

Michael Kassner
Michael Kassner

Could you please describe how you determined the set points on the graph?

Tony Hopkinson
Tony Hopkinson

It's not popular. First if you don't claim expertise, you must be crap. Out performing those who do claim expertise really irritates them. It makes you far harder to replace. So stop it. :)

peter
peter

Nicely stated. My management solution was a diversity of race. colour, creed, profession, education, employment history and experience. That won the day for a while, but is now being out flanked by the tidal waver of progress. Machines like IBM Watson will no doubt help plus networking - but I suspect we will need still more in the near future. BTW I have always enjoyed the ride no matter how tough it got :-)

Smedley54
Smedley54

This is justified. Good development (my opinion) iteratively transcribes skills and methods into code so it can become more widely available. People naturally, then, move onto other things and either forget or never bother to learn those skills. It is liberating, because with a calculator or spreadsheet almost anyone can do complex calculations, but it also makes us dependent on our tools. But we are dependent on our tools for civilization. Somebody, somewhere, knows how to capture their special knowledge in an artifact and share it with the rest of us, and we're all better off. Someone made every article of clothing I'm wearing, and it wasn't me, but there's no risk that we, as a civilization, will forget how to sew although I am concerned there's nobody left in the USA that sews.

peter
peter

New paper has a half life of about 50 years, whilst old paper is about 100years, whilst velum is of the order 500 years. Right now we have no means of securing all our knowledge in electronic or hard form. Libraries burn down, machines crash. The best we can do is multiple copies in multiple formats and locations. BUT paper/print is much less than 0.1% of the total and we cannot afford the atoms to change that ....

Tony Hopkinson
Tony Hopkinson

machines hold data. Humans turn it in to information and some take it in and become knowledgeable.

peter
peter

I am a professor :-) Within the limited space of a blog I cannot squeeze everything in! I am merely trying to lay out a major challenge we face and one that says we have to change education, training and the way we work. And for electronics it was over way before 1985....

Tony Hopkinson
Tony Hopkinson

If he had spent all his time mastering every discipline of his age, he would have had no time to exercise his genius. Even then, you had no chance of keeping up with everything at once. That was one of my problems with the questions. Unless all your engineers had done theoretical physics, asking them what Maxwell's equations are will find a couple of people who like me remembered a paragraph in Brian Greene's The Elegant Universe...

peter
peter

When I came into industry it was possible to read all R&D reports/papers in electronics/telecom. It was also possible for one man to understand an network from one end to another. None of this is now true. We have 'experts' in narrower and narrower fields of everything - liked 'doped optical fibre' for example. BUT at the same time the world is getting more complex, more non-linear, and more chaotic and we need people with a good overview and problem solving skills. PhDs in narrower slivers really don't help a lot unless they are in their particular niche - but that does not last long ! Knowing everything about nothing is the limiting case for the trajectory we have set ourselves - and it won't work in the short, let alone the long term! Not only is it getting harder - it is accelerating - and our only help is a symbiotic relationship with our machines that do have the ability to span all the knowledge in every field. The point is that education and training are not learning and neither are the professions - we have to change fast. If you were engaged in any new and challenging field you would feel the full force of all this - believe me!

peter
peter

I used citation indexes for publications as the prime indicator of continued worth. I also took previous examples computed by many others. The big problem is that there is no standard algorithm for the calculation of half life....but at least if you adopt one there is at least some constancy. Looking across the work of numerous authors on the topic I would say the error bars on my graph are around +/- 10% I tried to take a middle road - others are far more severe in their predictions abd selection criteria.

CharlieSpencer
CharlieSpencer

Too many companies are still plagued by undocumented institutional knowledge.

peter
peter

And that is increasingly the case...IBM Watson is just the start...

Smedley54
Smedley54

IT was easily divided into mainframes and mini-computers and then along came Apple, Microsoft and IBM with their silly little boxes, and Novel with that dumb networking thing... And there went the neighborhood. I should have checked your bio! You know, then, that there's only so much room in the lecture, their attention span, and the homework load you can lay on before the Dean drops by for a chat. Sometimes, even the priority items lack enough priority to make it into the semester. Our degree plan is designed in conjunction with local industry so (a) graduates can find a job and (b) we can demonstrate value to the community and pick up business support. So a second semester of COBOL (yup, still a big deal) edged out another term of AI. Crap. But onward.

peter
peter

How very true.....but in my experience it is increasingly everything!

peter
peter

But he did - and that is why he impacted so many fields including astronomy and economics.....but his advantage was that the knowledge of the time was thin, sparse, and narrow....far easier than today...

lehnerus2000
lehnerus2000

Some of those questions did seem to be a bit obscure. :) I think that a similar result would be obtained by asking similarly obscure questions in any class (e.g. literature). I still think that it was easier to be a "Renaissance man" in Newton's time than it is today. I often think that the only thing that held back the great geniuses of the past, was [b]easy access[/b] to the relevant information and high quality materials for experimentation.

Tony Hopkinson
Tony Hopkinson

IT is much more than tech. You entire premise is cockeyed. I'm knocking together network process, I don't care whether the fibre is doped or not, do I? I don't care about the exact chemical composition in the substrate of the chip. I don't care that the electrons move the other way.... Nobody ever knew a network from one end to the other at the level of detail in the multiple disciplines you are talking about. I've been doing this stuff a few years as well. I'm just as ignorant now as I was then and achieving that much comprehension is as hard now as it was then.

Michael Kassner
Michael Kassner

That doesn't help. It's still what you say. Can you provide links?

Tony Hopkinson
Tony Hopkinson

You're a knowledgeable guy. Is AI predicted to happen before or after the paperless office?

Tony Hopkinson
Tony Hopkinson

Even if you credit Herr Liebnitz with the math, he conjured it out of nothing. In my opinion that's way harder, than say bashing an atom and measuring how far the bits went. Seems to me you are seeing trees and not forests.

Tony Hopkinson
Tony Hopkinson

Working at different levels of abstraction and keeping the levels separate is a key part of programming. We are stood on the shoulders of giants. There are been a lot of them though, so the effort required to clamber that high is exhausting.

peter
peter

Correct...and it seems impossible for all of us to be good at everything....and to some extent team working cures the problem...but someone/something has to have an overall view. I don't think that this is wholly negative or depressing by the way, quite the reverse, it is a considerable challenge we need to rise to...

peter
peter

Spot on 2 :-)

Tony Hopkinson
Tony Hopkinson

To be renaisance man your comprehension has to be both broad and deep, in the past it was narrower and shallower. Not sure it's been possible to be a renaissance man since the information age began. Certainly most of the engineers I've worked with were crap at programming, and while I know the basics, I wouldn't have a prayer at touching chip design, or the solid state physics than underlies it. Even if you managed to get to genius level in more than one area, is at the same time doable any more.

Tony Hopkinson
Tony Hopkinson

and another attempted insult. Some of us get challenged everyday, because we don't believe we already know it all... Of course circuit designers knew how components they wanted to use worked! What has that got to do with knowing an entire network?. Another question would be, did they need to know how they worked, or did they only need to know their characteristics. Your initial argument was based around asking a bunch of unidentified specialists, a set of general knowledge questions. You show me the bunch of engineers back in the 60s or 70s that could have done bettter. Classic case of piss poor science, you had a conclusion, and geared your "experiment" to prove it... Scientific method rule one, No control = unprovable conclusion.

peter
peter

You are plain wrong! Even in the 1960s people in telcos understood in great detail how their transmission and switching systems worked. And circuit designers understood how valves, transistors and other components worked too. How do I know? I was there! No wonder you get challenged everyday!!