[57] in Commercialization & Privatization of the Internet
Re: competition [answer to the question]
daemon@ATHENA.MIT.EDU (Kent England)
Tue Oct 30 17:59:39 1990
Date: Tue, 30 Oct 90 17:43:20 est
From: kwe@buitb.bu.edu (Kent England)
To: kwe@buitb.bu.edu, schoff@psi.com
Cc: com-priv@psi.com
> From: "Martin Lee Schoffstall" <schoff@psi.com>
>
> Kent,
>
> I just wanted you to establish the facts, otherwise I would
> be inaccurate, you've given your normal honest answer.
You flatterer. But that's why I like you so much.
> I'll summarize
> in the most contractural manner, you have 20 contracts with
> academics and 36 contracts with commercial people to provide
> a service.
>
That's exactly the way I look at it. We have 56 contracts for
non-commercial internet service.
> >From the "Commercialization of the Internet" presentation at Interop90
> you asked the panel "when would the commercial networks interconnect"
> and the response from Rick and I was "when our customers ask for
> it, which they haven't".
>
How nice of you to notice it was me. You flatter me again. I
just asked the question to embarrass you, you know. Didn't work, did
it?
Interconnection of commercial internet service providers is an
issue of concern to me, since competition can get in the way of rich
interconnectivity when we have unregulated services. However, I found
it hard to accept that AlterNet and PSInet customers don't want to
communicate with each other, but if you and Rick both say so, it must
be true. Still, I don't know why they wouldn't embrace the service if
you two buddies got together and interconnected. Why not surprise
them and just do it? Spirit of co-operation and all that.
> Then I in my normal agressive manner said that the real problem
> was the commercial organizations participating in the R&E networks
> such as NEARNet which imposed restrictions as to use. Those
> are the people that some % of the 150 networks on PSINet want
> to get to right now and do commercial work with.
>
There may be 150 networks on PSInet that want to do business
with the 36 commercial organizations on NEARnet, but those 36
commercial folks on NEARnet understand that that is not within the
charter of a non-commercial network. They knew that when they signed
on. It's our dilemma, Marty. It takes time to work these things out.
> why don't you remove your restrictions or go commercial so
> that your customers and mine can be happy.
>
If I did that, you might refuse to exchange traffic with me
after you had gained sufficient market share in New England, like you
and Rick are doing to each other now. NEARnet is much bigger in New
England than PSInet at this time, so it would help you a great deal in
getting going in Boston if we interconnected now. I wouldn't mind
that at all, but would you disconnect from NEARnet after you had
gained market share in Boston, in order to gain competitive advantage
against NEARnet?
If so, I would much rather that NEARnet and PSInet continue to
be richly interconnected thru the NSFnet than not connected at all
because we become similar-sized competitors, and after we had been so
nice as to allow you to gain market share in our backyard and grow to
self-sufficiency in Boston. You are welcome to keep a POP in Boston,
just so long as you are losing money on it. :-) I'm not serious; I
would be glad if you made money in Boston. Let me know when you do.
Remember, you and I had made a deal to interconnect NYSERnet
and NEARnet when you were still with NYSERnet. NEARnet had the leased
line in and we had the router set-up and we had an agreement and
everything. And we waited and waited. The NYSERnet/NEARnet
interconnect was canceled by NYSERnet for no apparent reason after PSI
was formed.
So you see, I'm not convinced you really want to interconnect
to other competitive networks, except as it is to your advantage to
gain a foothold in a market dominated by another. Am I wrong to be
suspicious? But you do see the problems I anticipate when
similar-sized competitive commercial networks refuse to interoperate
because they do not perceive some competitive advantage accruing.
Same thing happened with X.25 services, so I think I am right to be
concerned. I advocate rich connectivity over competitive advantage
and I think you and Rick had better address this problem area if you
are going to convince people that commercial networking is better than
non-commercial networking.
--Kent
