I think this is a really good point, but I think there are some difference between specialization in traditional engineering and specialization in computer engineering. Please, correct me if I'm wrong as I don't have as solid a grounding in traditional engineering as I do the computer side.
It seems to me that in computer engineering, the specific skillset is less important than the general grasp of concepts in the more general field. IE, if I am a network engineer and I specialize in ATM network, I probably also have a grounding in a number of related technologies. If ATM suddenly becomes entirely obsolete, I am more than likely to have an ability to fall back on one of these other skills to develop. I think the sub-diciplines of computer engineering are less spread out than traditional engineering, which are perhaps less inter-relateable.
A counterexample would be speclization in something like Microsoft products, or only learning a single web language like ColdFusion. This is in fact a dead end because it doesn't require a common background with the majority of the current computer engineering field (I can say this after recently having dealt with hoards of Microsoft trained NT people who simply do not understand the basics of things like... well, anything below the Application layer, whereas UNIX sysadmins *have* to, by the nature of the system). In fact, if that one thing died out, the specialist might be screwed, depending on their personal adaptability.
You know, just last night I was commenting to my SO that the guys in sales are always really hot - good looking, clean cut guys who dress well, all at least 6 feet tall...
Are you one of those people who believes in plug and play as a way of life? If M$ couldn't get an operating system to have even *vaugely* reliable plug and play (where there are really not that many standards, all told) what on EARTH makes you think that anything as diverse and nonstandardized as the Internet will *ever* achieve that level of plug-and-play? Funny, but PC consultants are still in hoards years after the introduction of such concepts into the PC world.
The closest we have for plug and play to the end user for networks are things like DHCP, which still require quite a bit of setup and administration on the back end - and that's just *host* addressing.
File this along with the genetic algorithims writing code. Fantasy, this decade at least.
The fact is, regardless of ideals or licenses, people end up making money off their craft, in one way or another. Read the book Hackers by Steven Levy which tracks, among other things, the path that homegrown software took from being exchanged for free to being sold, marketed, and profited from. The idealism disappeared about the same time the greed kicked in, and that will continue to happen. It's a cycle, something which is apparent to me at least after reading that book. I highly reccomend it.
A lot of companies on the west coast already know this - it's how a good number of us got where we are, by practical experience and knoweldge, not by a degree. In management, of course, you still need a degree to get anywhere, but then... most techies I know were never interested in that anyway.
After watching my SO interview these last few weeks, I've noticed that there are a couple of kinds of companies, and some get it more than others. There are companies who have the hiring manager, who is technical, do an interview. Often these are the best kinds, as he knows specifically what he's looking for skill-wise and isn't going to give a crap about certifications or degrees. Then there is the Director Level interview. In these you're frequently doomed if you can't quote chapters from TCP/IP Illustrated V1/II verbatum, and have a college degree, because people at this level rarely understand the skills required for the job well enough to base a decision on it.
So what does this amount to? There are companies which have a lot of really bright, skilled people without degrees and are doing fine. There are companies who have lots of bright skilled people with degrees and are doing fine. It's just a matter of finding the former, and laughing at the latter for their limited thinking.
I've put a lot of thought to these topics lately, so here are a couple of thoughts...
Suits:
Well, on the West Coast that is most definitely true. Almost every kind of office has transitioned to "business casual," even the German-owned insurance company my father is a director at, which previously required ties of even the technical people. However, many of my friends have noted that on the East coast, the traditional dress has for the most part remained, and in fact that they have to dress up more when visiting East Coast offices. It's a culture difference for the most part. On the East Coast you have many companies which are older than every Slashdot reader - on the West Coast, there's been a boom, and culture changes with a boom like that. When was the last time you went into a tech company and they didn't have free sodas? Were you surprised?
IT shortage:
Here's what it boils down to. As a network engineer I can only speak as to the network portion, but examine the field. It's only been around for 30 years. Before 30 years ago, *there were no network engineers* (not counting telco engineers, who have laregly stayed with their telecom networks, and not transitioned to IP, though I have no doubt that will change...). We are second generation at best. We are one of the youngest industries around. So it only makes sense that there is a lack of trained professionals, or even of mechanisms for generating trained professionals. Secondly, examine the fact that there has been a higher growth rate for networks and the internet than probably almost any industry to date (perhaps excluding broadcast media such as TV, which grew almost as fast, I think... I don't have the facts to back this up, but it's not really important). Again, this means you have far too few people to run far too many networks.
Will this be true in 40 years? No, it will be a fully formed industry, with channels for training, set career paths, and A lot of very senior people. That's part of the key - in 20 years, I'm going to be damn senior as network people go. That's because I choose to learn constantly and develop. I'm not concerned.
Also, the fact that networks are for the most part a closed kind of community means there won't be a massive influx of people until schools start teaching it, and even then, we all know that schools often produce people with no practical technical knowledge at all. Most kids can't reasonably set up a routed network in their apartment or dorm. This makes the learning curve a bit more difficult to mount, as compared to system administration.
I think the sysadmin world is a little bit different. The main issue being that anyone can boot up a free UNIX on their home computer, or NT for that matter, and administrate it. They can learn enough from that to admin a small company's boxes, and maybe a large company's. There *will* be a glut in the market for sysadmins, because it's so easy to learn. However, as with everything, specialization is the key to preserving prosperity. Sendmail gods will still be sendmail gods. Security wizards will still be security wizards. But if you bill yourself as a UNIX system administrator, you might want to consider adding a specific specialization to your resume - you're a dime a dozen, from my observation.
Is Sun really a friend of Linux? Is SGI? Of course not. The closest you have to support in business is a vauge alliance towards a common goal.
I'm sure one or two of you out there have jobs, and for those jobs you receive paychecks. Those paychecks are dependant on your company making money. In fact, your company *exists* to make money - a concept abhorant to all of you, I know, I know.
Now, here's my point. You see Linux and the open source movement in general to be basically an altruistic effort, for the good of the community. You draw more from Stallman than you'd probably like to admit. If you think that companies are altruistic, and are going to support Linux because it's a good right thing to do, you are simply uneducated in the ways of business, and for that matter, the real world.
Businesses exist to make money. Let me say it again.
Businesses exist to make money.
And to make money, they will do whatever they think will get them the next slice of pie. So enjoy your hobby, and get off your hobby-horse - Sun, SGI, and Microsoft aren't going to bow to your bedroom experimenter's whims. They will simply use it as they see fit. None of this should come as any surprise, and frankly, I can't figure out why it even counts as news.
Slashdot Mirror
I think this is a really good point, but I think there are some difference between specialization in traditional engineering and specialization in computer engineering. Please, correct me if I'm wrong as I don't have as solid a grounding in traditional engineering as I do the computer side.
It seems to me that in computer engineering, the specific skillset is less important than the general grasp of concepts in the more general field. IE, if I am a network engineer and I specialize in ATM network, I probably also have a grounding in a number of related technologies. If ATM suddenly becomes entirely obsolete, I am more than likely to have an ability to fall back on one of these other skills to develop. I think the sub-diciplines of computer engineering are less spread out than traditional engineering, which are perhaps less inter-relateable.
A counterexample would be speclization in something like Microsoft products, or only learning a single web language like ColdFusion. This is in fact a dead end because it doesn't require a common background with the majority of the current computer engineering field (I can say this after recently having dealt with hoards of Microsoft trained NT people who simply do not understand the basics of things like... well, anything below the Application layer, whereas UNIX sysadmins *have* to, by the nature of the system). In fact, if that one thing died out, the specialist might be screwed, depending on their personal adaptability.
You know, just last night I was commenting to my SO that the guys in sales are always really hot - good looking, clean cut guys who dress well, all at least 6 feet tall...
But who the HELL would date a sales guy? Ugh.
Are you one of those people who believes in plug and play as a way of life? If M$ couldn't get an operating system to have even *vaugely* reliable plug and play (where there are really not that many standards, all told) what on EARTH makes you think that anything as diverse and nonstandardized as the Internet will *ever* achieve that level of plug-and-play? Funny, but PC consultants are still in hoards years after the introduction of such concepts into the PC world.
The closest we have for plug and play to the end user for networks are things like DHCP, which still require quite a bit of setup and administration on the back end - and that's just *host* addressing.
File this along with the genetic algorithims writing code. Fantasy, this decade at least.
The fact is, regardless of ideals or licenses, people end up making money off their craft, in one way or another. Read the book Hackers by Steven Levy which tracks, among other things, the path that homegrown software took from being exchanged for free to being sold, marketed, and profited from. The idealism disappeared about the same time the greed kicked in, and that will continue to happen. It's a cycle, something which is apparent to me at least after reading that book. I highly reccomend it.
A lot of companies on the west coast already know this - it's how a good number of us got where we are, by practical experience and knoweldge, not by a degree. In management, of course, you still need a degree to get anywhere, but then... most techies I know were never interested in that anyway.
After watching my SO interview these last few weeks, I've noticed that there are a couple of kinds of companies, and some get it more than others. There are companies who have the hiring manager, who is technical, do an interview. Often these are the best kinds, as he knows specifically what he's looking for skill-wise and isn't going to give a crap about certifications or degrees. Then there is the Director Level interview. In these you're frequently doomed if you can't quote chapters from TCP/IP Illustrated V1/II verbatum, and have a college degree, because people at this level rarely understand the skills required for the job well enough to base a decision on it.
So what does this amount to? There are companies which have a lot of really bright, skilled people without degrees and are doing fine. There are companies who have lots of bright skilled people with degrees and are doing fine. It's just a matter of finding the former, and laughing at the latter for their limited thinking.
I've put a lot of thought to these topics lately, so here are a couple of thoughts...
/dev/null
Suits:
Well, on the West Coast that is most definitely true. Almost every kind of office has transitioned to "business casual," even the German-owned insurance company my father is a director at, which previously required ties of even the technical people. However, many of my friends have noted that on the East coast, the traditional dress has for the most part remained, and in fact that they have to dress up more when visiting East Coast offices. It's a culture difference for the most part. On the East Coast you have many companies which are older than every Slashdot reader - on the West Coast, there's been a boom, and culture changes with a boom like that. When was the last time you went into a tech company and they didn't have free sodas? Were you surprised?
IT shortage:
Here's what it boils down to. As a network engineer I can only speak as to the network portion, but examine the field. It's only been around for 30 years. Before 30 years ago, *there were no network engineers* (not counting telco engineers, who have laregly stayed with their telecom networks, and not transitioned to IP, though I have no doubt that will change...). We are second generation at best. We are one of the youngest industries around. So it only makes sense that there is a lack of trained professionals, or even of mechanisms for generating trained professionals. Secondly, examine the fact that there has been a higher growth rate for networks and the internet than probably almost any industry to date (perhaps excluding broadcast media such as TV, which grew almost as fast, I think... I don't have the facts to back this up, but it's not really important). Again, this means you have far too few people to run far too many networks.
Will this be true in 40 years? No, it will be a fully formed industry, with channels for training, set career paths, and A lot of very senior people. That's part of the key - in 20 years, I'm going to be damn senior as network people go. That's because I choose to learn constantly and develop. I'm not concerned.
Also, the fact that networks are for the most part a closed kind of community means there won't be a massive influx of people until schools start teaching it, and even then, we all know that schools often produce people with no practical technical knowledge at all. Most kids can't reasonably set up a routed network in their apartment or dorm. This makes the learning curve a bit more difficult to mount, as compared to system administration.
I think the sysadmin world is a little bit different. The main issue being that anyone can boot up a free UNIX on their home computer, or NT for that matter, and administrate it. They can learn enough from that to admin a small company's boxes, and maybe a large company's. There *will* be a glut in the market for sysadmins, because it's so easy to learn. However, as with everything, specialization is the key to preserving prosperity. Sendmail gods will still be sendmail gods. Security wizards will still be security wizards. But if you bill yourself as a UNIX system administrator, you might want to consider adding a specific specialization to your resume - you're a dime a dozen, from my observation.
flames to
email to grey@enigma.mips4.com
Is Sun really a friend of Linux? Is SGI? Of course not. The closest you have to support in business is a vauge alliance towards a common goal.
I'm sure one or two of you out there have jobs, and for those jobs you receive paychecks. Those paychecks are dependant on your company making money. In fact, your company *exists* to make money - a concept abhorant to all of you, I know, I know.
Now, here's my point. You see Linux and the open source movement in general to be basically an altruistic effort, for the good of the community. You draw more from Stallman than you'd probably like to admit. If you think that companies are altruistic, and are going to support Linux because it's a good right thing to do, you are simply uneducated in the ways of business, and for that matter, the real world.
Businesses exist to make money. Let me say it again.
Businesses exist to make money.
And to make money, they will do whatever they think will get them the next slice of pie. So enjoy your hobby, and get off your hobby-horse - Sun, SGI, and Microsoft aren't going to bow to your bedroom experimenter's whims. They will simply use it as they see fit. None of this should come as any surprise, and frankly, I can't figure out why it even counts as news.