Friday, May 29, 2009

Is the Internet Broken? Do we need a new model?

One thing I find fascinating in discussions about the Internet filtering trial is the extent to which industry professionals seem prepared to wash their hands for responsibility for the experience of users.

After all, it doesn't seem like an unrealistic expectation that for people using the World Wide Web specifically, rather than the internet in general, shouldn't be able to be afforded the same protection as they get with movies or television of being advised of the nature of the content they are about to navigate to. That is, a person just "surfing" through a standard browser looking at html pages.

To make that happen would require develoment of standards that could lead a global debate on unified content standards. It could be conducted in a secular manner so that the classification scheme was against some objective criteria. But the industry response is it is eithr too hard because it is global or because it can be subverted.

But it is not the only place where the model of the intrnet is broken. Everyone seems to accept that IPv6 implementation is critical, but doesn't do much about it. At least IPv6 activists who want to Act Now have a place they can go.

Another interesting avenue is the The Pouzin Society whose purpose is "to provide a forum for developing viable solutions to the current Internet architecture crisis." The crisis itself is described in the following terms;

About 15 years ago, it became clear that IPv4 was reaching its limits, and the IETF responded by creating IPv6. In 2006 came the tacit admission that there continue to be fundamental scaling problems in the Internet routing architecture which would only be exacerbated by IPv6, and that Moore's Law could not save us this time. Several solutions were proposed, all based on revising IPv6 addressing using the concept of a locator/identifier split. Work has proceeded diligently, but a few months ago, it became clear that not only was this approach fatally flawed, but by implication, so was IP, or any variation of it. Academic efforts, beginning with NewArch and continuing with FIND and GENI are no closer to finding a solution than we were a decade ago.

A jumping off point for this work is John Day's Patterns of Network Architecture, and I think the society's name is in honour of Louis Pouzin whose main claim to fame seems to be criticism of the US centric model of Internet governance and architecture.

My policy question is whether we are thinking of any of this as we sail off to build the world of the Australian NBN.

3 comments:

HeathG said...

David,

On the content front.

This isn't a new idea and there are a few schemes like this around e.g. RSACi, PICS, ICRA - whereby content creators label their pages based on criteria defined by the scheme operator.

Out of the box(download) Internet Explorer comes with RSACi installed, but disabled. Once enabled you can vary the level (rating) you are prepared to accept, including blocking of all unrated content.

Some issues with this approach

1) There's currently not a single standard. This has implications for adoption both by content providers and end users.

2) It may be very difficult to get agreement on a single standard since you;re trying to get people to agree on what is essentially a moral standard, rather than a technical one.

3) Even if you do end up with a single agreed standard, getting it widely adopted by content providers will be problematic.

In tradtional media (e.g. TV/FILM/PRINT,) content creation & distribution is concentrated and primarily undertaken by professionals. This is not the case when it comes to the internet.

These are just the top three challenges I could think of briefly.

I'd also ask
- how does this solve the problem of misleading links caused by accidental or deliberate mislabelling?

- what is the marginal benefit of this approach over simply installing one of the multitude of existing filter programs?




DISC: Pesonal view only, don't reflect the views of my employer etc

Anonymous said...

heh, isnt the NBN a bitstream...one big Layer 2 Ethernet...no L3, no IP issues....somebdy elses problem perhaps!

David Havyatt said...

Anonymous will probably never come back - but that comment was extra-ordinary. I don't think I mentioned the NBN anywhere - and certainly my sources weren't talking about Australia.

Who cares about L2 vs L3 - I want anonymous to tell me who the "somebody else" is. In the final analysis "they" are "we".

The distorted values system of a GenX if I'm not mistaken.