May 23, 2006

Network Neutrality - There Are Other Shoes To Drop

It seems likely that through action or inaction our legal framework is going to permit telcos who provide IP packet carriage to impose differential prices.  That's sad, because it is likely that these price differences will be imposed not to cover actual cost differences in providing different classes of service but, rather, to hobble products that compete with those offered by the carrier.

Right now we seem focused on non-equal package carriage on the part of carriers, ISPs.  We've not noticed that there are others out there who are equally able to take their pound of flesh out of the internet.

For example, let's look at our old friend, Verisign, and .com.

Verisign makes no secret of the fact that it is a for-profit company.  And were I part of Verisign I'd be thinking to myself, "how can we make more money out of .com?"

One way would be to drive users to pay for "enhanced" name resolution services.

Here's how I would do it - and as far as I know there is nothing in Verisign's contract with ICANN to prevent this:

I'd begin by limiting my build-out of "public" (i.e. "free") resolution system of .com servers so that the typical free name query/response cycle is on the order of 100 to 200 milliseconds.  That may not sound like a long time but when we start to add up the name resolution cycles that most people generate we will quickly notice that this will make the internet appear sluggish to many users.

At the same time I would build out a "preferred customer" system of servers that would answer queries much more quickly.  Access to these would, of course, require a subscription fee.

It would, of course, require considerable finesse to establish this dual-class regime without causing too much of a shriek from the bulk of internet users.  But if Verisign is anything, it has demonstrated that it can consistently negotiate the pants off of both NTIA and ICANN.  And Verisign need not allow the "free" service level to erode in one obvious jump; it could simply be a slow fade as demand increases without a commensurate build-out of free servers.

And from a technical point of view there might need to be a bit of creative work to separate the first class queries from steerage, but I have full confidence that this hurdle could be overcome.

Posted by karl at May 23, 2006 2:24 AM