NSF Ponders New And Improved Internet

diorcc wrote to mention a Wired article about a NSF Project that could completely rebuild the Internet as we know it. From the article: "The National Science Foundation is backing a major initiative that could lead to a completely new internet architecture, with built-in security measures and support for ubiquitous sensors and wireless communications devices, among other things. The Global Environment for Networking Investigations, or GENI, will include a research grant program to fund new architectures and an experimental facility, which has not yet been planned in detail."

Comments

Idea!

[Knight Thrasher]

Let's name it "Internet 2!"

Misleading..

[PDXNerd]

So in other words, this is just an experimental research facility with possible long-term finds that may impact the future direction of interneworking.

To rebuild the internet is insane. To slowly change the direction we are building it is more likely.

Re:Misleading..

[mfh]

To rebuild the internet is insane. To slowly change the direction we are building it is more likely.

I agree. It's about standards that companies should follow. Those that fail to follow the standards will lose relevance and compatability.

And yes, the article title was misleading. They won't be rebuilding the Internet any time soon.

Re:Misleading..

[DRobson]

It's about standards that companies should follow. Those that fail to follow the standards will lose relevance and compatability.

For some reason all I could think of after that was the phrase 'Internet Explorer and CSS Support'... (That said, I still mainly agree with the idea).

Comment removed

[account_deleted]

Comment removed based on user account deletion

NII2

[Doc Ruby]

Didn't we already give them hundreds of millions of dollars, and trust that they'd deliver the "New and Improved Internet" to us with Internet2? I know I2 is doing a lot of good for a bunch of universities, medical centers and corporations, all of which therefore are getting their N&INet (NII) to contribute to their hugely profitable enterprises, subsidized at taxpayer expense. Where is the delivery of I2 to the rest of us, who pay for it, who need it, who represent most of the American economy (foreigners are welcome to ride for free, as usual ;)? Why should we give them even more money, when they just got paid to learn they can get paid not to share it with us?

Re:NII2

[Doc Ruby]

It's true that the government's NGI is actually independent from Internet2, though they work closely together. The NSF funds organizations to connect to Internet2 with tax money. And the I2 is about 80% funded by universities, which are largely funded by public money, government and otherwise. Where's the return to the public?

I don't know why expecting public money to return products of its investment is "stretching it". We're buying R&D, we should get the R&D. Except where secrecy is important to, say, national security (tiny percentage of research), or the results would be premature to release, of course we should get access to what we bought. Why not?

If an org wants to keep its research products private, it should use only private money. Perhaps there's a case to be made for proportional return on proportional investment (eg. publishing 80% of I2), but that's surely balanced by 1> the critical enabling support of the public money; 2> the vast public research predecessors on which all this new research depends; and 3> the essential role of publishing research results anyway, to science, culture and business. Otherwise, siphoning off all the oxygen produced will leave the system stagnant, and the private systems will wilt and die also.

Re:NII2

[guet]

And the I2 is about 80% funded by universities, which are largely funded by public money, government and otherwise. Where's the return to the public?

The return to the public is in research and education (which is what universities do). Where else would you expect it? Serving inane comments on Slashdot quicker? Supporting the latest dot.com fad? When the industry is ready to embrace new standards (hint, this is not a rational or controllable process) they will come to the mass-market. Not before.

I'm afraid your free market ideology is blinding you to the benefits of public research and public funding. The 'all power to the poeple' line is very seductive, but research takes time, and the best research is not calibrated, and is not predictable. It does not obey the laws of the market and will never do so.

(foreigners are welcome to ride for free, as usual ;)
What a tired old canard. Where did the tags your writing is surrounded with come from? (hint, not the USA).

Re:NII2

[Doc Ruby]

Universities don't teach the return on their R&D. They productize it. Before you talk about economics, and university research's role in it, learn something about it first. Especially if you call my demand for better managed public funding for public research a "free market". It's a demand for a "free market" only in ideas: government subsidies aren't free marketing. Therefore your complaints are irrelevant. When the results are ready, they're currently privatized into university patents and thinktanks. They should, on the same schedule, be published and indeed taught the way you wrongly believe they are currently.

Then there's your ducktalk about HTML tags. I didn't say the US invented everything. But we did invent the Internet. And until an American invented the IMG tag, the Web wasn't useable by most people. So take a hint, and show some gratitude, instead of your jealous spite. We're not cranking out this tech for your thanks, but you could at least show some dignity when you accept our gifts.

Good News / Bad News

[sdpuppy]

Good News:

They're rebuilding the internet to make it more secure, eliminate spam, virus, spoofing and so on.

Bad News:

Initiative will use Microsoft programming techniques as its foundation.

:-) :-) {just joking}

GENI are Evil

[Bruha]

Sure they helped give us some nukes to kill a wraith ship but I still think they're bad.

Hell I didnt even know they had a internet.

Strategic Incrementalism

[ngr8]

There was an old McKinsey article that talked about "Strategic Incrementalism" back in the 80s. Idea was that with a clear vision, one could tweak the way to "good enough".

While there are intrinsically very ugly problems in client and server software right now, it seems that "Little Science" is displaced by "Big Science" (viz, NSF) in addressing incremental substantive improvements in security and availability for the Internet masses.

So, for example, as valuable as a *waving hands* non IP infrastructure blah blah might well be... there could be greater good achieved with work on secure computing environments, strong authentication, one time pad encryption methods and etc.

As a very dear friend of mine was fond of saying "if you want security, pull up your own shorts".

So, while big honkin backbone and new architectures are and will be very important, some think time at the "big level" regarding applications architecture and services would, likely, produce faster returns and shorter implementation times.

New and Improved

[k4_pacific]

Maybe I'm drifting off topic here, but how can this internet thing simultaneously be new and improved? If it's improved, it existed before. If it's new, it didn't.

Finally!

[bullitB]

The current version has clearly been a complete failure. Maybe if they start over from scratch, this Internet thing will actually become popular.

Trusted Network Connect

[tepples]

It could use IPv6, but "built-in security measures" makes me think of Trusted Network Connect. Imagine if you needed a Trusted Platform Module plus an approved, unmodified operating system plus an approved, unmodified dialer program that verifies the "integrity" of your machine just to get an IP address. Some analysts claim that most major cable and DSL ISPs are likely to require TNC by 2015.

Re:Trusted Network Connect

[griffjon]

Especially given the paranoia/security/centralized control mode we're in with the current regi^H^H^H^Hadministration. I wouldn't be surprised to see a new attempt to enforce key escrow, and for all the "trusted" computer to have "secure" backdoors into their crypto systems that "only" the govt can access with "a warrant"

(I also hear that there's a movement for a sarcasm tax per-double-quote in the house committee, so I'm tryin' to use 'em while they're free!)

This all being said, the concept of a mesh network and the work of the guys at DefCon WiFi Shoot-out might be very, very valuable sooner rather than later. Man, wouldn't that be fantastic? A geek-run national wifi mesh... It's be just like 1990s Internet again, until the FCC started raids...

this is easy to do...

[3seas]

simply track every transaction on the internet and allow law enforcement to invade and abuse it whever they will it...

Considering we can break anything we make, no matter what is done, it comes down to this.

giving access to personal and private information to other humans...

May as well just start installing gps tracking and personal data recording chips in all humans...
Then it really won't matter what internet or other future tech we make use of.

Of course included is a punishment system of shock therapy and AI second guessing what you do to stop you from doing anything on the list of things not to do..... A list created by a few faulty humans of course....

The point is, there is nothing we can build that we cannot break.

Making this whole "better internet" just a carrot to get the donkey to move...... in circles.

In related news...

[KillerBob]

In related news, industry analysts have examined the expected content of this "new & improved" web, and have decided to call it the "National Science Foundation Web", or "NSFW" for short. When asked for comment, an official replied "finally, the Internet will have a name that accurate reflects the majority of its content."