Intel Predicts Death Of WWW

LostCluster writes "Forbes is running a report saying that Intel's CTO claims that the WWW is 'running up on some architectural limitations' that will eventually cause its downfall. He's pushing a project called PlanetLab that has Princeton, Cambridge, Hewlett-Packard and AT&T on board, but Cisco is notably absent from that team."

Internet 2

[Icarus1919]

It's unfortunate that while a newer, faster version of the internet is in the works, it's supposedly going to be limited to use by scientists and other researchers. Perhaps the system with increased complexity that was previously reported on slashdot is the answer to our difficulties?

Corporate boardroom conversation

[nysus]

"Gentleman, the bad news is that there isn't much more money to be made selling Internet hardware. Profit levels in Internet-related hardware are down 300% from 5 years ago.

The good news? We've just landed a top notch PR firm to help sell our message that we must upgrade and overhaul the whole infrastructure. We'll be monitoring the impact of this message over the next several months. If successful, we expect to see profit levels soar again within 3 years."

Re-architecture

[el americano]

A worthless article to be sure, with no discussion of the web's architectural problems. (bad Slashdot) There is obviously more to the architectural problem than will be solved by IPv6, but allowing for IPv6 and higher capacity routers alone, I'm sure the web could go a long time with no other upgrade. I can only wonder how much money Intel will spend on convincing people that the web will die "sooner than you think." If it's anything like the $300 million they spent on telling people they have the best/only Wi-Fi solution, we'll be hearing this for a long time.

Re:Well...

[Alsee]

Last time Slashdot ran this story I did a little digging. I'm not sure of everything Intel has in mind, but part of their plan is to impose a Trusted Computing layer on top of the internet. "Security" "viruses" "authentication" blah blah blah. What it really amounts to is that if you do not "voluntarily" submit to Trusted Computing and turn over control of your computer you will be locked out.

Slashdot ran a story quite some time ago about Cisco Working to Block Viruses at the Router. [slashdot.org] The way they really work is they first scan that you are Trusted Computing Compliant, then they can scan exactly what software you are running, for example to ensure you are running the mandated firewall or anti-virus software or whatnot.

If you do not submit to Trusted Computing, or if you are not running the mandated software, then the router "quarantines" you until you come into compliance. In other words it denies you a network connection. Compliance is "voluntary", but you are blocked from the network until you comply.

-

The web sucks

[autopr0n]

I really think the future lays in some sort of system that distributes server load to clients. sort of like an interactive bittorent.

A lot of things that are done these days over the web are extremely simple and could be done on the client side, but can overwhelm a server when it needs to be done for thousands or millions of people. And bandwidth still isn't free.

Intel CEO Lyle Lanley, everyone!

[adoarns]

The Web was not created by companies like Intel. It wasn't created by companies at all, only in some cases co-opted by them.

When companies like Intel pitch these wide-ranging changes, it comes over like some seedy traveling salesman pitching a monorail.

If we want to make changes to the web, we will.

Re:Very Vague

[leifbk]

At least to me, they have not said what the problems are to begin with and further more have not said how they are going to address each one.

The problem, from a financial point of view, is of course that it isn't that easy to make money off the Internet as a lot of investors may have thought. TFA suggests as much when it's said that "the Internet will begin to collapse as millions of new computer users from developing nations begin to sign on." My guess is that most of those new users from developing nations hardly have the potential to generate profit remotely in proportion to their consumed bandwidth. So the Internet as a means to stockpile return on investment may well soon be a thing of the past.

And that probably sums up Forbes' interest in the case.

However, as long as the infrastructure of the 'net mainly consist of rather cheap hardware and essentially free software, I can't foresee the imminent death of what we really love about it: The free exchange of information around the globe. It's not the death of the Internet, then, it's rather a full turn of the circle back to Tim's vision. And good riddance to the money hoarders.
--
defenestrare necesse est

Re:That's not bad at all

[grahamlee]

The primary reason that Gopher's contents are so interesting is because it never caught on as a commercialisable medium - i.e. there are no ads, pop-ups etc. Unfortunately there is no Mother Gopher any more, nor is there a reliable VERONICA either, which means that you need to know where in gopherspace something is before you can look at it. [For instance: a NeXTSTEP gopherspace at sdf-eu [sdf-eu.org].]

Re:Well...

[Kaenneth]

Couldn't a Virus/Trojan Spoof the authentication?

It will never happen

[Cokelee]

The entire idea is asinine. To paraphrase who many simply call the inventor:

The Internet will never be controlled by a single company or a group of companies; Any company that tries to control it will fail.

-- Sir Tim Berners-Lee

I was there

[Groo Wanderer]

I was at the IDF Keynote, and I bounced some questions off Gelsinger afterwards. There are two problems with this that make it a non-starter in my opinion.

First, they are talking about layering another level of obfuscation on top of the net as a fix for the underlying problems. Rather than dealing with the problems, they ignore them, and make a shiny thing. Wow, that's architecture for you!

Next, with the innovation fom HP lately, and the fact that they are going commercial with it rather than open and standards based, it is doomed to be a niche idea at best. As one questioner afterwards pointed out, the internet was built on open ideas. This is looking to go the opposite way. NEXT!

Vint Cerf was cool though. They said there would be a special guest, but to my horror, they only meant Vint. No telletubbies in bondage gear this year. I can only hope for spring 2005 IDF....

I plan to rant about this on the Inq as soon as I recover from last week.

-Charlie

PlanetLab = mobile code. Again

[Animats]

PlanetLab is another "mobile code" scheme for running your stuff on other people's machines.

As an operational model, this fails. Either the system gets take over by hostile code, or there's some central adminstration that controls who runs what. (This last is the 3G cell phone services model. It's not working.)

This is one of those ideas, like "push technology" and "micropayments", which fail because the people who benefit are separate from those who absorb the costs. Only in a monopoly situation can that work.