"...application servers, XML, and Web services" are not core elements, they are implementation details. One of the great problems in modern curriculum design has long been the insistence by members of every discipline that the current state of the art of their niche must be included at the expense of basic understanding. Most scientists, engineers and other professionals forget that they did not learn most of these things until well after graduate school.
Furthermore, the overall importance of these details is not obvious until viewed from the perspective of history. One is reminded of Bill Gates' forward to the OS/2 Programmer's Guide:
"I believe OS/2 is destined to be the most important operating system, and possibly program, of all time."

The main cause of cancer is the accumulation of genetic mutations from a variety of causes. This leads back to applications development in the sense that most modern software suffers from a similar accumulation of defects, and many of the "features" of modern deployment environments are attempts to ameliorate the impacts of those accumulated flaws.

The community where I reside is presently undergoing a wave of indoor smoking bans. Scientific inaccuracies aside, the point of how some people insist on inflicting their own chosen lifestyles on others while at the same time attacking more minor flaws in others, is an age old human dilemma. There is a point to be taken in that most of the same people complaining about people smoking in public are the same ones spewing tons of exhaust from their excessively large SUVs while sitting in traffic commuting to and from their sprawling suburban homesteads.

Most people are inherently self-interested and will insist on disapproving others' lifestyle choices, even when the real effects on them are minor. At the same time, they will defend vehemently their own, regardless of the impacts.

Since all human behavior (or that of any other species) affects the environment, and indeed may be regarded as an innate component of "the environment," perhaps people should learn to be more tolerant.

Jack, I'm sorry you're a cigarette addict but using this forum to justify your habit is very inappropriate. Trying to back up your case with the gasoline = tobacco analogy is so laughable. Anyone in your position should stick to topics that you have knowledge in. Your tirade on this topic only serves to highlight your complete lack of knowledge of chemistry and combustion. Do a little surfing Jack and you will see very quickly that your comparison holds no water. Same goes for your XML is here forever argument. It may be the latest and greatest but like all things man made, it will be replaced.

I think I get the point that students will benefit from understanding web services and OSes the same way they benefit from understanding how electricity is delivered to their house. It's not necessary for day-to-day life, but it's interesting to the curious, and can be fun to learn about. Maybe that's where the author is coming from -- IT people tend to be curious about how things work.

I'm sure that many students who learn about web services will not care, the same way they don't care about catalytic converters and electricity. The ones who *do* find it interesting will probably also be interested in other technologies, because they're just curious people. In a few years, maybe they'll enter the IT field as developers.

An app server should be an extension of the OS. Almost every java app server leaks memory and needs frequent restart. When you redeploy or rename JNDI names, you need to restart the whole Websphere App Server.

Redeploy EARs in Websphere Studio? Restart the server.
The result is a lot of waiting. Whoever solves that is going to be the king of app servers and will wash away the "java is somewhat slow" forever.

As much as I like having a central place for applications, there seem to be a huge overhead or "housekeeping need" for an app server that very seldomly benefits me.

I want less and faster features!

Kind of reminds me of one of my favorite movie scenes from The Breakfast Club...

Brian Johnson: I'm a fu**ing idiot 'cause I can't make a lamp?
John Bender: No, you're a genius 'cause you can't make a lamp.
Brian Johnson: What do you know about trigonometry?
John Bender: I could care less about trigonometry.
Brian Johnson: Bender, did you know without trigonometry there would be no engineering?
John Bender: Without lamps there'd be no light.

Object oriented programming teaches us that when we understand the interface to an object, then we can use it without understanding details of the implementation. In daily life, we can flush a toilet without having to be able to explain the plumbing involved. It is very basic to a plumber, but please don't ask me for a detailed explaination. Sorry, don't care.

Sure, application servers, XML, and Web services are here to stay. I agree 100%. But how much most people need to be aware of them is where I disagree. To most people they are the digital equivalent of plumbing or the power grid. Sure, it would be nice if everyone understood all about them, but mostly people just need to know a few basic things like not to put your finger in the wall socket.

Another problem with the gasoline/cigarette analogy is that smoking cigarettes produces no worthwhile gain to society, whereas burning gasoline delivers fast transportation. So, if I could only choose one or the other, I'd choose the gasoline because it also delivers a great service to me (and the world).

Short answers -

XML and HTML will be around for a very, very long time.

We will see "web services" (no capital letters) pretty much everywhere, though "Web Services" - with capitals as a proper name for the standards - look to be a sure bet in the same way the ISO stack was clearly superior to TCP/IP (not).

Application servers in the model of WebSphere -- a mega-all-rolled-into-one package, along with servlets and J2EE, while all worthy first tries, are all likely to lose out to much lighter weight and more efficient models.

Just a little poking around on the web tells us that per gram cigarettes generate about ten times as much nasty stuff (probably nastier per gram too - but that is only a guess).

Given that more folks suck on cigarettes than tailpipes, that smokers are more likely in enclosed spaces, and cars are more likely in open air -- I think annoyance at the guy who lights up nearby is a bit more justified.

The comparison between burning gasoline and burning cigarettes seems a bit stretched. At that simplistic level, we may as well say that burning hydrogen is as bad as burning the same mass of gasoline. Or one pound of chlorine gas is the same as one pound of helium gas.

And should we wish to make the analogy clearer, most people throw wobblies when people smoke indoors. I suspect that they would also have a problem with people running a car indoors.

Getting to the substance of the article, about whether AppServers are a "fad", the substance of the argument seems to be that we've spent years installing them so therefore they aren't a fad. People spent years buying bell bottoms, but that didn't make them any less of a fad. The question you should be asking is whether AppServers deliver the value they promise at an acceptible cost, given the growing number of alternatives. You should be asking whether we will be looking at AppServers as an embarassing feature of our youth.

I think this article and the preceding article are highly irrelevant, and that probably puts me on the side of those who think that application servers, XML, and Web Services need to exist, but don't need to be understood.

If I need to talk to a business partner in Houston when I'm in LA, I don't want to have to know how the telephone switching network operates, or whether or not my connection is analog or digital, or what the difference is. I want to pick up the phone and punch some buttons and speak. Simple, stable, reliable, uneventful. In a word - BORING.

This is exactly what Application Server, XML, and Web Services are to non-tech people, and that's just fine. As you said, it only takes patience to use applications, but it takes intelligence and skills to write them. But most people aren't in the business of writing applications, and most of them don't need to be.

Mr. Paige was accurate when saying the STUDENTS were more comfortable "on computers." But that doesn't mean that they're more knowledgeable about HOW they work, or that teachers need to be so.

My niece is 12 and can surf her way around 3 or 4 sites and 3 or 4 IM conversations all while loading her MP3 player with tunes. She doesn't know what an ISP is, or understand anything about HTML, or client-server applications, or audio compression methods or file-transfer protocols.

She just knows that she likes music and chatting, and that a cord has to go from the computer to the wall and she can do what she needs. Then she's able to spend her time thinking about boys, movies, homework, or whatever else an 11-year would rather do.

One of the most powerful laws in economics is the Law of Comparative Advantage, which could be applied in this case to dictate that it's actually better for a person to spend LESS or NO time investigating app servers, xml, and web services, but to give that job over to other while they concentrate on the areas in which they have their own advantages.

Only us IT developers have to worry about all this and hopefully we can contribute to this global Information System and make it as boring as the phone network today.