Slashdot Mirror
Tech Geezers vs. Young Bloods
Lam1969 writes "Robert Mitchell talks about how technology is dividing him from younger generations: "The technologies I've watched grow have shaped an entire culture of which I am not a part." Adds Dinosaur: "Ask them [members of the younger generation] HOW the things work, and they have no idea. They are really riding on the backs of the 'old folks' like us that built the goodies they enjoy.""
Abe: I used to be with it, but then they changed what "it" was. Now, what I'm with isn't it, and what's "it" seems weird and scary to me. (Episode: 3F21 Homerpalooza)
It's only going to get worse as the pace of change continues to accelerate. In ten years a few engineers will be designing new classes of electronics based on quantum principles. Or totally new types of devices based on photons or magnetic spin vs. electron charge. Ten years later, that will be passé and maybe we'll be doing something with neutrinos. Who knows how things will work 30 years from now. It will all be magic by then.
The NSA: The only part of the US government that actually listens.
I feel kind of odd watching flamewars about who is tougher and more hardcore, C++ or some other language group, and I think to myself, "maybe they should have to actually deal with assembly, logic, and bits for real before they start talking hardcore. I remember when we were putting together kits out of catalogs with hex pads and light up bulbs and calling it computing.
Oh well. I think all this excitement has gotten to me. I'm going to go take a nap now. Where's my cane?
If my grammar and spelling are off, I am [distracted/tired/careless] (take your pick)
Take for example a small group - technical support folks. Since I started doing technical support, things have changed. Back when I first started, most people DID understand the underlying mechanics of what was going on. They COULD do things command-line and know precisely what to expect to receive back. They also often had knowledge of a wide range of systems and levels of technology from the front end, to the server, to everything in an entire corporate network. Today, technical support folks know how to click mouse buttons and change graphical settings without having any clue as to what exactly is happening to the system or why. Furthermore, they're specialized down to the point of knowing only a few systems, instead of the broader range.
I agree technology has changed how people use it. I agree that the masses have technology in ways they never could have back then. I agree that most people who use it don't and shouldn't need to know how the underlying systems work. I also agree that there are people who SHOULD understand more about the systems they work with and don't. I sum it up to DOT-COM frankly when floods of people came into the tech world and lingered too long knowing too little.
"As far as I'm concerned, I prefer silent vice to ostentatious virtue." ~A. Einstein
Conversely, and this relates to the parent post, when population numbers decline inventions are sometimes lost. He sites examples of societies that had acquired and then subsequently lost, writing, the wheel, and other technologies.
Test 1 2 3 4
I don't know anything about diverging fieldds and such, but it's simply a sign that the technologies have become commodities. At the early part of a technologies lifecycle only the early adopters and geeks get into it and have to know the nuts and bolts of how it works because you have to make it and maintain it. Then some others come in and you have to maintain it for them. Then the tech gets matured to the point that it becomes a commodity and they still need many of the originl geeks and adopters to maintain it, but they don't do nearly as much down in the guts of the tech anymore. Very few tech geeks nowadays could tell you all the parts that go into a working steam turbine electrical generating system, but they can sure plug in a gadget and use the electricity. This allows the next generation to focus their efforts on the next new technology, which will eventually become the next commodity, ad infinitum.
"The world of the future will be an evermore demanding struggle against the limitations of our intellegence, not a comfortable hyammock in which we can lie down to be waited upon by our robot slaves."
-Norbert Weiner (1894-1964)
Each suceeding generation begins a couple steps ahead of the old. That shift in point of origin allows the younger generations to view the old's accomplishments as the beginning of something more, while the old can only see the tremendous effort it required.
Pessimists.net - as if life wasn't depressing enough.
I am often put into the role of teaching others how things work. I am 29 years old and have no CS background (I am entirely self-taught). I talk to most techies and they have no idea how things work behind the scenes. I am not talking "this IM client sends the message to the server which sends it to the other IM client." I am talking an in-depth understanding of how things like TCP, IP, and UDP work. They generally have no clue. I actually had one student who had several years of IT experience tell me that he thought UDP and ICMP were the same thing...
;-)
How did I understand how these things worked? I started by reading the oldest documentation I could find. Part of the problem is that computer professionals have become very good at confusing eachother (using the OSI model to discuss TCP/IP for example) and the other part is that the document writers in general don't understand what they are writing about. Then I could go and read newer documentation and have some sense of what it is worth. Good documentation in this industry is a rare thing.
Maybe it helped that both of my grandparents on my mom's side were writing programs before I was even in diapers
LedgerSMB: Open source Accounting/ERP
The only people I can think of who wouldn't are a few of the people I know who have learned the technology trade, not grew up with a passion for the machine. (Note when I say friends above i mean the latter. I make friends with similar people.)
So yeah, I think anyone who has a real interest in computation studies knows with some interest how circuits are arranged, how Turing machines work, is at least afraid of the y-combinator, and knows that language fights are dumb. :) I think once the pay starts decreasing again, then things won't be taken for granted *quite* as much.
One mild caveat to all of this, however. Managing complexity means abstracting. As we continue to add complexity there's a point at which some people just won't want to understand how a machine works inside. They blackbox it and move on. Hopefully they'll still get a top-level from it.
People who quote themselves bug the crap out of me -- Me.
I'll bet the people who maintain or design and build the Tesla turbines know how they work. That's what I mean by divergence versus commoditization.
With commoditization, as you describe it, the common fear is that all the knowledge will one day be lost because no one has to use it anymore. You see this in a lot of B Sci-Fi movies set in the distant future, often leading to religious-based uprising (religion being the clear enemy of science, what?)
Whereas, with technological divergence, you end up with the breakup of a field into two, like "computers" into "hardware" and "software." There are plenty of electrical engineering students who know what NPN and PNP mean, and haven't a clue about, say, the pros and cons of classes versus structs re functional programming and modularization.
Thus I say it is a point of divergence, because the field has broken into component fields. Commoditization is a realistic fear, which was certainly described somewhat in TFA, but I think it is somewhat narrow-minded at this point. Most people don't know a carburetor from a transaxle, or what ring and tell have to do with traditional land-line telephones. That doesn't mean that knowledge is lost.
I, being a believer in meritocracy, ignore the actions of the end-users, who know not what they do.
When you're afraid to download music illegally in your own home, then the terrorists have won!
Exactly, this is a basic economic problem relating to specialization.
People don't need to understand everything about every mundane detail in life to be able to be a functional and productive member of society, and indeed we shouldn't strive for this. Honestly, I don't know how to change my own oil in my car, but I doubt that the dude at Jiffy Lube knows anything about software development. We all have our own absolute and comparative advantages in life. For me, and for society, it is better for me to take my car in and get my oil changed rather than me taking up more time doing it myself. This allows me to save time which I can spend doing things which I have an advantage in (like developing software).
It's a basic "Jack of all trades, master of none" issue. The guy at Jiffy Lube could spend all his time learning to program, but his time is better spent serving society with something that he has a comparative advantage in. I could spend much of my time learning about raising crops for food and changing my oil, but it is more efficient to allow others who are better at this to specialize while I work within my own area of expertise.
In summation, the opportunity cost of changing my own oil is higher than taking it to Jiffy Lube. Simple economics...
I have always been of the beleif that the seperation or hardware and software has actually been holding the computer industry back. I am one of the few "young bloods" that actually have a good understanding of the hardware (most of the people I work with don't even know what a register is, let alone how to use it). This lack of knowledge by newer software engineer has caused many programs to be much slower and more memory intensive than they need to.
A classic example of this shows up in Java alot, where large amounts of data are loaded into memory (undoubtedly swap) for manipulation later. Those of use with good understanding of the underlying system realizes it's a pretty big waste to read from disk only to have it swap right back to it and instead maintain references to portions of the data that we need to rereference. This is a basic example, but a fairly solid one. That fact that many modern languages do not even allow you access to the underlying registers, operations and other processing structures (excpet through other languages like using JNI in Java) makes true optimization pointless.
Most modern software is highly pessimized, using layers of abstraction for development convenience rather than optimized for performance (include system requirements). The idea that you can always through more hardware at a problem has lead to software bloat and this unneeded pessimization. The only bright side is that there are atleast a few people out there still considering the hardware and how it can be used by the software, leading to things such as the use of GPUs for general vector processing and not just graphics. If we loose the few of us that do have knowledge of both ends of a system we will run into a stagnation of technology. When I began programing on commodore machines many years ago I had to learn how to use the processing power of each sub component and not just hope that a pre existing library knew what it was doing and could some how optimize for what I am trying to do exactly.
I would like to see more people try and teach the values of understand what your system can actually do and not allow the inner workings to behiden by unneeded abstraction.
Although I agree with some of what you say, I have to disagree in places. (disclaimer: I taught TCP/IP and OSI at a technical college and have been a programmer of mobile and more classic network-enabled software for about 12 years now).
Prototyping is fantastic. But sometimes people just never bother to finish a job. TCP/IP seems to be one example. How many systems have ported original TCP/IP stacks? Why is it that I see the same unimplemented methods in stack after stack? Someone had enough wit to realize they'd be handy, but the guts of a TCP/IP stack are no trivial matter. And the protocol went out and became ubiquitous long before it was complete. And now, bits of it never will be.
You can damn OSI for being slow off the mark, and that's typical of standards bodies. But for all you say about TCP/IP, I've also written an OSI prototype over TCP/IP as a proof of concept and a goodly portion of the services can be easily delivered (the parts that map well together). And the OSI semantics are probably more intelligible than the TCP/IP ones. (That's an opinion, YMMV).
The OSI model, on the other hand, is a perfectly good *model* for understanding the role of a tiered networking stack. Why is this so useful? Sure people abuse it in the real world and many apps span several layers of the stack, etc. But the conceptual idea of encapsulation of function and also the conceptual ideas of what the layer's functions should be is a good start. This lets you look at real world divergences and then realize where they might be good, bad and what the tradeoffs might be. If you never had the reference model, you'd have a harder time quantifying these differences between real world implementations and understanding why they might be good or bad.
I ran across one instance of this not long ago where someone had taken a shortcut in a networking stack and not exposed some lower level service primitives. Sure, as long as all you wanted to do was the basic subset of tasks as imagined by their developers using their higher layer interfaces, you were okay. But if you wanted to do something a bit different, you didn't have access to some key lower level primitives. This is a case where the developers didn't think beyond their own application and they didn't obviously have much of a concept of a tiered set of functions. And lo and behold, a less useful result.
OSI isn't the holy grail, but it is an instructive learning tool. All standards are produced in some ivory tower and where the rubber meets the road things are different. Yet at the same time, those standards and those theoretical models have great value, especially as individual implementations come and go (TCP/IP has got a lot of traction and has had a long life with no end in sight, but the same cannot be said of many other technologies and even TCP/IP may one day see the a twilight of its days).
To blindly say the OSI model must be killed because TCP/IP got out there and did some things is like throwing the baby out with the bathwater.
-- Mal: "Well they tell you: never hit a man with a closed fist. But it is, on occasion, hilarious."