This is an interesting topic, but if we look a the way that (for example) China has built enormous expertise in a very short time I don't think it is a problem with losing experience or 'institutional knowledge'. Ideally as you/we (ok I) get towards retirement age we will have been passing on some aspects of our experience to the next generation. (and not just by being the grouch in the corner telling war stories of how we saved the day or invented the hyperwheel). But there is a problem - both here in Europe and also (I think) in the United States, and that problem is that engineering students leaving school or college don't seem to have such a well targeted enducation as they did in my/your/our day.
Actually, this should not be a surprise to any of us! The field of electronics is orders of magnitude bigger than it was when I took my first degree in the early 70's. In almost any field microprocessors, display technology, memory architectures, microwave electronics, assembly technologies, etc etc etc developments over the past few years has been greater than in the previous few years. We have accept the need to mould the next generation to become the experts of the future, its not got any easier - and it never will, its just different and it always will be.
I'm sure in ten, twenty thiry and forty years from now there will be successive 'old grouches' (sory - Old timers!) who will complain that they have all the knowledge. But there will always be a few industry super-gurus and there will always be generations of new engineers that will revere their knowledge and their writings (or their blogs) they way we revere Jim and Bob for theirs.
What is lost with the passing of guys like Bob Pease and Jim Williams is the "elegant solution" to a problem. These guys were masters of their art and could squeeze almost perfect performance from imperfect components with good design. These days, the solution tends to be to just chuck some more bits / megabytes / bandwidth / processing power at a problem, which doesn't always seem to work, especially with analog.
But as pointed out, no-one is indispensible, and in a couple of years time most will have forgotten these names, except maybe a few of us old-timers....
Being close to retirement age, i think about these things a lot. I've been in the same company most of my career, and I don't really see my replacement anywhere. But it probably really doesn't matter. They don't know how the pyramids in Egypt or Machu Picchu in Peru were built, but they have found ways to build even more technologically advanced buildings without that knowledge. I worry more about when man has finally managed to destroy the earth to the point that civilization collapses and they need to reinvent society from scratch, the only problem being there isn't anywhere to plug in their computer.
I agree with the comments that nobody is indispensable, and that fresh-out engineers tend to be hard-working, eager to learn, and the immediacy of their education may bring advantages of new tricks or techniques with which the old guys may not be familiar.
I am not quite certain what "institutional knowledge" is, but I do know what intellectual property is -- and I'm not talking about the company's patent portfolio, I'm talking about the IP that enables derivative products or enables the next generation product to have higher performance and lower cost compared to the current generation product.
The kind of IP I'm talking about isn't a patent, written in a bunch of legalese -- it's more often associated with files on a network drive, schematics and source code and layouts and that sort of stuff that we engineers recognize as a design database.
And part of that IP database is in the minds of the engineers who developed it, who last tweaked it and solved the latest bug -- those who touched it, wrestled with it, and helped make something of value from it.
When those guys go, without first having brought in new blood to learn the IP, to take part in the tweaking and the wrestling, then the company is left with nothing but a bunch of files on a network drive.
How many times in your career have you had to reverse-engineer something that was developed by your own company? Something for which the answers to all your questions about why something was done one way instead of another way were simply "sorry, the guys who did this don't work here anymore"?
THAT is what companies lose when a generation of engineers retires without helping bring up a new generation to take their places. The company loses its IP, and its ability to quickly generate competitive new products.
Literally, this has been happening for 10s of thousands of years, with homo sapiens.
Leaving rants about management aside, I have come to notice that new people do step up to the plate, when this is necessary. Remember that it only takes one person with extra vision, among many workers, to keep things running smoothly. Maybe things will be done differently from what the Old Master had been doing previously, but that doesn't mean worse.
Another point, I think, is that even if Company X was doing amazing things while the Old Master was working there, and even if this Company X folds after this guy retires, there are other companies in competition that will take the lead.
Surely, we have all witnessed this change over the years, and how new guy sometimes does turn out to be fantastic? Surely, this will happen again.
While I may not believe that any individual is indispensable (management included), I find the trend of reducing and further reducing staff while maintaining or even increasing the workload to be very distressing and even more damaging than the loss of any one person.
At some point in the "right sizing" process, even the most brilliant of engineer can do little more than go into survival mode. Then there is no time for mentoring, little time for designing in quality and little opportunity to let creativity really flow.
the sad truth is the belief by a company's business leadership that no one is indispensible, except for themselves, of course.
The examples of where a company starts to perform much better after "right sizing" out experienced (read expensive) personnel are very few if at all. Going back to Brian's analogy, the docks rarely just collapse through rot, these guys are cutting the support posts with chainsaws (how's that for a metaphore).