[9673] in Commercialization & Privatization of the Internet
Re: Telecommunications Competition Act of Washington State
daemon@ATHENA.MIT.EDU (Karl Denninger)
Sun Jan 16 03:42:21 1994
From: karl@mcs.com (Karl Denninger)
To: ittai@ans.net (Ittai Hershman)
Date: Sun, 16 Jan 1994 02:40:25 -0600 (CST)
Cc: karl@mcs.com, com-priv@psi.com
In-Reply-To: <CMM.0.90.2.758666200.ittai@shemesh.ans.net> from "Ittai Hershman" at Jan 15, 94 03:36:40 pm
>
> MCSNet's T1 came online last evening, and with it FULL CIX connectivity
>
> Congratulations, Karl. I am not sure, however, what you mean by "FULL
> CIX connectivity"? The CIX, in so far as I have been told, provides
> equal services to all members in a neutral fashion.
Yep. TO MEMBERS. However, there are lots of "non-members", including one
here in town that is attached to ANS, that mooch off the CIX connectivity
without being members (ie: they're reselling connections and aren't members
of the CIX, yet route via it through your membership).
We are on exactly equal footing with the other members of the CIX as a
result of our membership. That other provider, and those who choose not to
join, cannot make that claim.
> I saw another posting of yours (on a.i.s. I think) in which you state
> you have FIX access. I am curious what that means since the FIX is an
> exchange point for Fed nets and has a AUP-like traffic restrictions.
>
> Or should I just ignore the marketing talk?
Those customers of ours (and ourselves, since we agree to abide by the AUP
where required) who choose to comply with the AUP can have access to the
FIX as well via our SprintLink connection. We submit NACRs just like
everyone else, and if approved the routing is enabled in the PRDB.
How is this different than for an ANS customer? Well, I guess you don't
have to submit a NACR, but other than that there is no real difference. You
are still supposed to comply with the AUP when you are talking to a R&E
customer, regardless of the provider you are connected to. That there is no
formal statement of complience required of <your> customers is more an
artifice of what I consider to be an unfair marketing advantage granted to
ANS by the federal government.
One, I might add, that I have consistently said should be remedied. Either
ditch the PRDB or apply it <equally> to all provider customers. Including
yours.
> you know, it is really impressive when
> we can get universal connectivity with one T1 line from a national phone
> carrier. Seems that the wonderful word - "competition" - has done what all
> the Government regulators, and the NSF, utterly failed to do --
> does it not?
>
> Or have I missed something?
>
> Well, maybe. The NSFNET Backbone Service supports over 15,000 networks.
> The number of networks available on the CIX router, is significantly
> smaller. And the number of those that are only available via the CIX
> is even smaller than that (1019 to be precise, according to my latest
> information).
So what? The full routing tables are what -- 16000+ networks? You claim to
support "over 15000", but how many of those are <direct> attachments? Not
nearly that many. Just like the CIX doesn't have 15000 <direct> attachments
-- but that's not the point, is it?
If you want to play numbers games add up the routes carried in the core BGP
tables and figure out who talks to who. You'll get a slanted view of that,
since you're looking from your perspective (and thus your "best routes" will
of necessity run through more ANS links than otherwise). We'll have a
different view of the world, since we're not on ANS. We see that most of
our routing goes through non-ANS circuits, and that very few sites are
unreachable without transiting your backbone.
Its terribly disingenous for you to say that "15000 routes reachable via
ANS". Well, of course! But not ONLY via ANS, and for those who aren't ANS
customers a large percentage of those don't have a "best route" via your
services.
Finally, since we are attached via SPRINT, much of the international traffic
never leaves <their> backbone! They have a large percentage of the
international links, and as a result those never see either the CIX or the
FIX. They just route and go.
> I think the bottom line here is that the commercial Internet became
> possible because the NSF primed the technology with the R&E Internet.
Yep. And, IMHO, botched the job. That the net worked out anyway is more of
a fluke and the result of hard work by those who wanted it to succeed than
the wisdom of the NSF in the way they did business.
> Now that Internet services are a competitive market, the CIX -- as a
> trade association of commercial Internet service providers -- is the
> next logical step. There is no need to promote one at the expense of
> the other.
You're right. And you start, Iatti, by not raising points which bear
relavence only from your point of view (the routing numbers you cite
above).
--
Karl Denninger (karl@MCS.COM) | MCSNet - First Interactive Internet and
Modem: [+1 312 248-0900] | Clarinet feed in Chicago. Send email to
Voice/FAX: [+1 312 248-8649] | "info@mcs.com" for more information.
