Network World Senior Editor Carolyn Duffy Marsan sat down last week with the IETF's new chair, Harald Alvestrand, to discuss his plans for the group's future. Here's an excerpt from their conversation:
How will the IETF change under your leadership?
My job is to facilitate the change that has to happen. We have to get a better feel for managing our relationships with other standards organizations. We have to get protocol documents out on time, and we have to make sure that they're finalized. We have to do a better job of resolving conflicts and not letting them stew. But I see no need to change the basic structure of the IETF.
The one criticism about the IETF is that the group takes too long to develop standards that the market needs. Some say that might cause the IETF to become irrelevant.
Many companies come to us and would like us to standardize something they've developed. We look at it and find that it's broken here, it won't scale and it isn't secure. Being right takes some amount of time. Companies should know that it takes a serious investment in time to get the work done.
How will the IETF's mix of work change in the future?
It won't. We do everything from what runs on the wire to the applications. It's a matter of getting the Internet to work.
What do you consider to be the IETF's most significant work?
That's a difficult question. IPv6 [the next generation of the Internet's main communications protocol] is significant because it alters the possible scale of the Internet. But it is probably several years before IPv6 will be widely deployed. Two other big issues are mobility and security. Mobile communications are already changing the culture, and they're going to change the Internet. We have to consider what it means to have connectivity with security that allows all functions to reach you wherever you are.
You also co-chair the Domain Name Supporting Organization within the Internet Corporation for Assigned Names and Numbers. What's your view of ICANN?
The Internet needs certain things done that smack of regulation, like keeping track of domain names and adding new top-level domains. We at the IETF are definitely not good at regulation. However, ICANN is not good at it either. The biggest problem I have with ICANN is the lack of logic behind what it does. But the bad thing is that I don't know how ICANN could do its job any better.
How will the IETF's relationship with ICANN change under your leadership?
I would like to see at least the [Internet Assigned Numbers Authority] function being identified as a separate thing, something clearly controlled by the IETF and the IETF only.
What do you think of New. net, a start-up that recently announced plans to sell names using 20 top-level domains that were not approved by ICANN?
What I think of New.net is not fit for print. Identifiers on the Internet need to be unique and verifiable. When two people create the same name and pass it around to a third person, that third person is in deep trouble. Of the 20 domains that New.net is selling, 18 are already operational with other alternate domain name providers. These companies believe new [top-level domains] can be created by thinking of them first. That idea neither has forethought, nor fairness, nor quality control.
This story, "New head of IETF facing a full plate" was originally published by Network World.