Who Will Fix the Internet? No One, Apparently

blackbearnh writes "It seems like everyone focuses on the latest and greatest killer Internet applications, but the underlying infrastructure that all of them run on is showing its age. That's the claim made by a recent article in the Christian Science Monitor. IPv4 is relatively ancient, and even stalled improvements like IPv6 aren't significant enough to matter, according to some researchers. With no one 'in charge' of the Internet, it's almost impossible to get any sweeping technical improvements made, especially since there's no financial incentive on the part of the ISPs and telecoms to invest in basic infrastructure. CalTech Professor John Doyle puts it this way: 'To the extent I've been working in this field for the last 10 years, I've been mostly working on band-aids. I'm really trying to get out of that business and try to help the people, the few people, who are really trying to think more fundamentally about what needs to be done.'"

Comments

Re:The Whole Point if the Internet...

[gbjbaanb]

I suppose IANA could start handing out IPv6 addresses only from now on, that'd shake the industry up quickly enough; and if ICANN announced that they would turn off IPv4 on its DNS roots, it'd have the same effect.

I Thought We'd Been Through This?

[segedunum]

IPv4 is an absolutely fundamental part of virtually every network in existence today, and given that networks are a fundamental prerequisite in the modern computing world and see very, very, very, very heavy usage every minue of the day no one is going to take any time out and start tinkering because people think networks and the internet are broken. There's no financial incentive for ISPs or any companies to invest in IPv6 yet and there won't be no matter who is 'in charge' of the internet to 'force' it to happen. You can't mandate anything in an open market, and I just find the possible motivation for that statement bizarre.

Basically, it'll start to happen when we really do run out of IP addresses and things get desperate and it will happen when someone comes up with a sane and straightforward guide for making IPv6 co-exist happily with existing IPv4 networks and making sure everyone knows about it. Until those things happen there is zero incentive to rip out and replace or tinker with something so fundamental. Band aids are the order of the day and have been in every piece of fundamental infrastructure since time imemorial. We must leave this 'rip out and replace' culture in computing far behind otherwise no one can ever take us seriously.

Re:Hmm

[chill]

The article author thinks IPv6 is just a band-aid, though he admits it would fix the address shortage. He is talking, vaguely, about an architectural upgrade but doesn't really say *what*. He only says "more research is needed", which I translate to "give me more funding".

Do you have any insight as to what he's talking about, other than "get off your ass on IPv6"?

Proactive...not

[Stenchwarrior]

There will be no proactive solution; this sort of thing will only be improved upon in increments as things break. John Doyle mentions "Band-Aids" but that's exactly how it needs to evolve....like any other living organism.

Re:Hmm

[Anonymous Coward]

I honestly don't even think IPv6 is needed. We just need recall some of those huge blocks of IP addresses that have been allocated for no good reason and implement NAT/proxies more widely.

Just about every single company uses firewalls nowadays anyway, there is absolutely no reason for them to have huge blocks of IP addresses like they currently do (they don't even use them!).

The only thing "on hold" is the USA, not IPv6

[FreeUser]

The Internet is improving everyday as better routers, faster servers, new better cables/antennas are deployed, the last mile connection options are also multiplying. IPv6 is put on hold as there is no real need for it at the moment.

IPv6 is NOT on hold. Most of Asia are already using IPv6. If you use Apple there's a good chance you're using IPv6 without even realising it. The EU is mandating moves to IPv6 in the coming years, and I imagine most countries are doing something similar.

The US may have its head in the sand, but that doesn't mean everyone else does.

Re:The only thing "on hold" is the USA, not IPv6

[tsotha]

The US doesn't have its "head in the sand". The US government and corporations are simply in the position of having large blocks of IPv4 addresses, so there's far less urgency. Of course China is using IPv6 - they came along too late to get many v4 addresses, and v6 already existed when they started building out infrastructure. The US market is relatively mature, as well, so you're not going to see the kind of demand growth you have in other places. We could last for decades on NATs.

Re:Hmm

[kalidasa]

Here are two examples of the kind of research he is talking about:

• Multiscale networking, robustness, and rigor, by John Doyle, a simplified discussion of his own research area
• Patterns in Network Architecture, a book by John Day on new paradigms for network architecture

Re:Hmm

[mellon]

The problem is that congestion control on the Internet is strictly based on the Van Jacobsen hacks to TCP/IP. These work pretty well, but they have problems. First, a lot of IP traffic is not TCP. Second, various IP protocols like Bittorrent actually game congestion control to get more than their fair share of the pipe, and there's really no way to prevent this (e.g., what Comcast tried isn't a good solution).

The belief that no-one is working on this is incorrect, however. There's some very good work being done in the IRTF (a research organization associated with the IETF). They did a really cool presentation on their work at the Stockholm IETF this month. There are really good people at various ISPs and running the backbones. It is not the case that it's all on autopilot and slowly decaying. E.g., check out Hurricane Electric. Comcast has a very good team.

The most hopeless thing I see on the Internet is the continued prevalence of operating systems that are highly vulnerable to attack due to poorly-thought-out security models. Apple is starting to do some interesting work on this - they recently hired the guy who did BitFrost for the OLPC project, for example. A big complaint about Bitfrost is that it's not necessarily all that useable, but if anyone can fix that, it's probably Apple. Would be nice if Microsoft weren't backsliding on this.