The World Wide Computer, Monopolies and Control

Ian Lamont writes "Nick Carr has generated a lot of discussion following his recent comments about the IT department fading away, but there are several other points he is trying to make about the rise of utility computing. He believes that the Web has evolved into a massive, programmable computer (the "World Wide Computer") that essentially lets any person or organization customize it to meet their needs. This relates to another trend he sees — a shift toward centralization. Carr draws interesting parallels to the rise of electricity suppliers during the Industrial Revolution. He says in a book excerpt printed on his blog that while decentralized technologies — the PC, Internet, etc. — can empower individuals, institutions have proven to be quite skilled at reestablishing control. 'Even though the Internet still has no center, technically speaking, control can now be wielded, through software code, from anywhere. What's different, in comparison to the physical world, is that acts of control become harder to detect and those wielding control more difficult to discern.'"

Comments

big server farms, thin clients at home

[primadd]

I'm fearing for the days when all you have at home is a thin client to some virtual machine inside some big server farm. You buy CPU time, like in the old mainframe times, get billed by cycle.

No need for anti piracy features, you don't get to see the executables or source anyways, all tucked away from your prying eyes.

--
Bookmark me

Re:big server farms, thin clients at home

[MightyMartian]

There may be little you can't do, but there's a lot you can't do well, reliably or securely. As cool as Google Apps may be, you're essentially trusting your data integrity and security to an outside company.

The web, as it currently exists, is a really shitty software platform. Web 2.0, if it meaningfully exists at all, is built on some rather horrible hacks that break down the server-client wall, and for certain kinds of limited applications that's fine, but building substantial applications, like accounting and financial software, in AJAX would be an unbelievably difficult job, and a rather hard one to justify.

I think this guy is, as with his last great proclamation, overstating his case. Yes, in certain arenas, like home and small business email, apps like GMail certainly can play a role, but I can tell you right now that the business I am in, which deals with confidential information, will be waiting a long time before farming out this sort of thing.

Re:big server farms, thin clients at home

[PPH]

And then some enterprising guys, working in their own garage, will develop a machine that you can own, can program yourself and mantain complete control over.

It will be the 1970s all over again (except without disco).

Re:big server farms, thin clients at home

[kwerle]

I'm fearing for the days when all you have at home is a thin client to some virtual machine inside some big server farm. You buy CPU time, like in the old mainframe times, get billed by cycle.

Look around. There are no thin clients. The iphone is 100x more powerful than my first computer. The macbook air is 1000x more powerful than my first computer.

Imagine 21 years from now. Imagine computers 128x more powerful than they are today. That means that the iphone of 21 years from now will be 10x more powerful than "the lightest laptop available today."

You're talking about "thin clients". But a really powerful computer will be the size of a thick piece of paper.

Yeah, I'm dreaming - but how else do you expect to keep up!? In my professional career (say 18 years), computers have become 100x more powerful, and fit in an envelope.

The only reason for "thin clients" is because the client wants and agrees to be thin.

Re:big server farms, thin clients at home

[dodobh]

The browser is the software version of the thin client.

Re:yea

[kcbanner]

...and so it begins. Not on the frontiers of outer space, not launched from Mars during the night...but here, on Slashdot. They have found how to infiltrate our minds and compel us to respond, waste our mod points, and upset the balance of society itself.

World Wide Computer

[gringer]

Otherwise known as a botnet

world wide computer, eh?

[swschrad]

10 stop war
20 fix domestic problems
30 printf "Woo!"
40 goto 10

hmm, doesn't seem to be working. hairbrained theory, anyway.

it would probably take 80kb to do that in visual C.

All Control-G's are now Taco Bell

[NetSettler]

Internet still has no center, technically speaking, control can now be wielded, through software code, from anywhere. What's different, in comparison to the physical world, is that acts of control become harder to detect and those wielding control more difficult to discern.

Or from nowhere. The risk of a bad guy taking over is serious, but the risk that no one is at the helm is much more likely to lead us to death by Global Warming, for example.

You have to look no further than the US Congress to see a worked example. If you idealize every single member of Congress as intelligent, and I think a similar analogy can be made for people on the net or for companies on the net (where you still have to question intelligence sometimes, but let's not and say we did), it's pretty clear that the problem isn't just the sinister taking hold of someone with total power. It's also that it's easy to cause behavior that no one can take responsibility for, and that isn't in the best interest of individuals. The Internet is no different, but not because we didn't have examples of this before. Just because we didn't heed them.

To paraphrase Charlie Stross

[monopole]

The definition of a real utility computing environment is one where somebody can hold a coup d'etat in it and make it stick in the real world.

Ahem....

[GaryOlson]

"The tighter your grip, the more star systems will slip thru your fingers." Princess Leia of Alderaan

This guy obviously has no sense of history....real or fictional.

The IT cycle?

[jase001]

Isn't this just the IT cycle, everything gets centralized, short term costs are saved. 10 years later decentralized, and long term costs are saved Vs short term.

Re:affect on the backbone

[dbIII]

would be maintained by professionals, assuring they will be virus free

Oddly enough that currently defines the difference between the professional level operating systems (some of which are free) and a hobby system that was pushed into the workplace (which you have to pay for). The wide range of malware is currently a single platform problem and is almost all the fault of poor design of two applications - Internet Explorer and Outlook.

Frederic Brown's "Answer"

[Tancred]

Here's a classic sci-fi (extremely) short story on the topic of an immense computer. Frederic Brown's "Answer":

http://www.alteich.com/oldsite/answer.htm

The mainframe is back

[Duncan3]

So we're back to the point in the cycle where centralized mainframes you rent time on rule the world again. Can you guess what happens next? Privacy problems, reliability problems, outages, and we all go back to personal systems again.

Old is new again.

Ridiculous comparison

[Dun Malg]

Carr draws interesting parallels to the rise of electricity suppliers during the Industrial Revolution. Interesting comparisons? More like spurious comparisons. I read the linked interview and, as someone who has read quite a bit about the rise of industry and its relationship to the availability of power (basically, the history of power generation), I can say he's a typical unrealistic abstractionist. He handwaves away the fact that the purpose and nature of electric power generation and electronic communication are similar solely in topography by claiming that they are both "general purpose technology" and are analogous economically. Of course, his entire line of reasoning is balanced upon a precarious point of assumption which is highly questionable: that people will find off-site centralization easier than in-house. Really, it's the same old crap we've heard for years. How long had we been hearing about how "real soon now" thin clients will be all people will need? It's ludicrous. Just think about how much lower latency and greater reliability would be required before people would be willing to offload any significant percentage of their storage and computational needs. We're not there yet. We're not anywhere near there. I'd say you'd be lucky to get 2 nines of reliability out of such a system, much less the 4 or 5 nines you'd need to make it what this nutter predicts. Really, the parallel between remote IT service and electric power is nil. All power requires for reliability is a good run of copper wire and generator.

Half-joking. Half.

[Eco-Mono]

And where's Alderaan now, pray tell?

Re:Half-joking. Half.

[Anonymous Brave Guy]

It's where baby Leia is growing up. Even I know that, and I've only seen the first three films!

Why, is something bad going to happen in episode IV?

I agree, to a point.

[Vellmont]

There's some technologies that everyone wants, and there's a solution that'll fit 90% of the populace.

Examples would be hosted email, contact management, and calendaring. A central provider can just simply do a better job at providing all these things that an IT department does, and the requirements are all extremely generic. Users seem to want infinite amounts of email storage, and the ability to find an email at a moments notice. That's difficult to manage unless you want to dedicate someone to JUST knowing the email systems.

The thing I disagree with is that the IT department is going away. Simply not true. The difference with other utilities is that the IT department doesn't provide a single, simple resource like electricity. IT provides automation and tools that increase productivity, many of which are going to be way to specialized to centralize.

IT departments may evolve, like they've been evolving for the last 50 years. I've heard many years ago (before my time at least) there were people dedicated just to swap tapes around. We don't have that anymore of course.

Privacy Laws

[Roger W Moore]

As cool as Google Apps may be, you're essentially trusting your data integrity and security to an outside company.

Just to drive home your point further what can be even more important is that, as trustworthy as Google may be, they are subject to US law. This is a huge problem in places like Canada which have privacy laws since using, for example, GMail means that your organization can end up breaking Canadian law because the US government has free access to any data in your email which you may be legally responsible for protecting.

This is not Nicholas Carr's First Attack on IT

[caramuru]

Carr wrote the May 2003 Harvard Business Review's "IT Doesn't Matter." His argument (grossly simplified) was that IT is a "utility" and businesses should not invest in IT because IT cannot differentiate one firm from another. In a well known (to the business community, but apparently not to ./) rebuttal to Carr's article (Smith & Fingar's "IT Doesn't Matter, Business Processes Do", Meghan-Kiffer Press, 2003,) it is argued (again, grossly simplified) that IT is critical to optimizing business processes - the true source of enterprise value. A business that optimizes its processes differentiates itself (positively) from its competitors. In fact, Business Process Management Systems (BPMS) is a new layer on the enterprise software stack. For those of you coming from the SOA space, BPMS is the choreography layer.

Carr's current article's argument that IT functions should be taken over by functional units only perpetuates the silo thinking of most organizations. Budgeting IT resources on a departmental basis perpetuates islands of automation, redundant/conflicting rules, ridiculous internal interfaces., etc. Outsourcing some or all IT functions may be reasonable in some cases, but turning control of IT over to the various functional units in an organization is insane.

Re:This is not Nicholas Carr's First Attack on IT

[Tarwn]

I'm not sure the idea of splitting the IT responsibilities into other departments is insane (hold with me a moment). Consider the current situation of an IT department that is a separate department, usually with their own goals, budget, etc. This department is notable for not always getting new PCs as fast as they are wanted, for not implementing software changes immediately when requested, and for demanding additional money when deploying technologies like video conferencing so they can upgrade the internet connection or some other foolishness. Oh, and they always act like they are busy, but we all know the systems hardly have problems.

Unfortunately the suggested solution, of splitting IT into the surrounding departments, is going to look like a good idea to many director level people. It will (in their minds) ensure immediate service for new equipment, allow a higher level of control over the purchase of items they think are unrelated, and allow them to have changes made to software at a higher level of priority. To the outside manager or director, they generally only see what we are not supplying, not what we are. If we are good at our jobs, but have poor systems, they don't generally realize just how bad things are because we are keeping the system limping along. A lot of our expenditures are due to reasons they just don't understand. If we buy a 48 port managed switch with fibre but were rolled under one of these departments, it could very easily turn into a refurbed 48 port hub off ebay, since they both have lots of connections and thats all you really need.

What about change control? They don't see it. Time for testing? That will get reduced further. Developing in test environments? But those are good machines, they should be used for something important. Oh, and why do you need fancy development tools? Joe down the way made an application to do that in 45 minutes using MS Access, but it takes days in this fancy technology, we'll just use MS Access instead.

The whole idea of splitting IT up into several departments is like a startup company (non-tech) in reverse. Money will go to IT-related resources last, it will be in no one's interest to spend the time, resources, or money to ensure there is a strong infrastructure capable of growth, in house software development will be on-the-fly and likely based on technologies like MS Access. On top of that, larger initiatives like data warehousing, global data management will be left to whoever wants to pay for the whiz-bang consultant to come in and do it their way. Backups, email, directory services, all of this will end up on someone's plate who will forever be trying to drop it off on someone else.

I realize that the author of that article was likely thinking that IT resources would not need to deal with most of these things in the future, and for that I can assume he has not worked in an IT environment in quite a while. While technologies are available to streamline our jobs and allow us to grow the department(s) more slowly that in the past, splitting the department so that no one has these responsibilities is going to have one positive thing going for it: The consultants that come in to clean up the mess after the takeover are going to be set for a good long time.

Re:affect on the backbone

[TheThiefMaster]

That was true in the past, but nowadays malware is mostly spread by the good old "User wants free porn" method.

A.k.a social engineering.

I don't remember encountering any malware since at least before 2000 that could spread itself without relying on the user to infect their own machine. I've had several pieces of malware try to email or even msn file transfer themself to me from an infected pc though.