Paul and OX, The following was posted over the weekend on the "Last Mile" thread. The mention of Covad reminded me of it. The part relevant to us starts about halfway down (By the way, Frank Coluccio is one of the really excellent and helpful posters on SI, and knows the technical stuff). One phrase hit me, concerning "next-generation dsl". I wonder where Avidia fits in this evolution?
To: +WTC (3334 )
From: +Frank A. Coluccio
Saturday, Apr 10 1999 1:14PM ET
Reply # of 3364
Tim,
I enjoyed reading your most recent post, upstream from this one, as always. Thanks for
the enlightenment concerning the ILECs' potential in fulfilling advanced DSL services
through the use of deeper fiber builds to DSLAM deployments at serving area interfaces
(SAC/SAIs).
The potential for colo conflicts there is real. It reminds me of the colo issues of the early
cellular and pcs deployments, when all the dither was about competing carriers sharing
towers and actual cell sites. In the long term, it proved to be viable due to the cost
sharing attributes of such a model, especially for some of the smaller providers, where
power and air conditioning requirements were concerned, and where aesthetic
constraints existed in the environmental sense.
I would imagine that companies like AFCI and IFCI and others will benefit from this
both through increased sales [AFCI] of equipment and services [IFCI]. I was
wondering if the SBC AFCI field tests had anything to do with the DSL deep in the
distribution plant? Would you know? Or, I suppose the more appropriate question to
ask would be, 'are you at liberty to say?' I've not seen any detail on those tests. It would
be interesting to list the other companies whose products are not intrinsically DSL, but
who would benefit, if the ILECs proceeded along these lines. Anyone?
Would you say that some of the ILECs are actually at this time pushing for a full-fledged
deployment of the FSAN model, complete with passive optical networking (PON)
elements in some parts? Or have I been over-using this term lately?
Is FSAN only a 'straw man' model around which others merely point to, for reference
purposes? Or do you see it as a near-term interim- (or even a ten year end-) game for
the incumbents? Perhaps you can comment on the following as well. Thanks.
Thread,
I was asked in a PM to comment on Covad, Northpoint and Rhythms..
...comparing their business models, and what this all means in the presence of @Work
who is also using DSL.
My take is that the @xyz's who use DSL technologies are a deep dark closet issue
for at least one of the factions within @Home. The fact that they had to resort to DSL
in the first place is an embarrassment for some within this cable modem consortium,
and I would imagine that there were some religious civil wars fought over its initial
deployment.
The culturally induced animosity between the coax and twisted pair camps is not to be
underestimated. This can be witnessed by inferences picked up in the sarcasm,
innuendo, and digs almost every day on some of the cable lists and Use Groups.
Sometimes you have to read hard "between the lines" to pick it up, but it's not only still
there, it's growing as we speak, due to the current and more pronounced face-offs
between the two approaches, as each strives to reach the promise land of what each is
mis-labeling as "broadband."
Of course, who am I to step in and meddle with the evolution this trade press- and
web site- inspired language. In my book, sub-45 Mb/s [155 Mb/s?] speeds no longer
constitute "broadband," rather, they connote wideband. And at the actual speeds
being squeezed out of these pipes, I would tend to characterize them more as
narrowband, in many cases. This will increasingly be the case for many users over the
next couple of years. Mark my words. But such are the ways of hype, and I have
digressed once again...
I wouldn't consider myself very up to speed on the differences between Covad,
Northpoint and Rhythms, the three cos you cited. Ours took a slightly different twist,
targeting the independent CLECs a while back. We had performed some
more-than-rudimentary research and projections which led to a business model that
resembles all three in certain ways. This was about a little over two years ago. Of
course, at that time we were looking at ways in which to satisfy virtual LEC (VLEC)
requirements, mainly a means of supporting VoIP at Layers two and three by the
creation of a last mile alternative to using ILEC services.
There were no media gateway protocols at the time, just the emergence of an H.323
draft at the ITU-IMTC. SIP hadn't been conceived yet, much less anything that truly
resembled a means of arbitrating between "IP telephony" and "Internet Telephony,"
two distinctly different approaches to voice services on the Internet. But we at least, at
that time, identified a necessity to replace, or reshape, DNS to be more functionally
aligned with the capabilities of SS7 (or equivalent). This, too, is only now coming to the
fore in the IETF, close to three years later, and it doesn't appear that there is enough
consensus yet to allow meaningful predictions yet, as to when such a list of capabilities
will exist.
We found that it would be very tough to make a buck even at the physical layer, DSL
alone, at that time, and I still feel that right now, in fact, due to the chasm stages that
both DSL and VoIP (for lack of a more descriptive term) are still in, considering the
still-high price-points being faced by the service providers to get in at this time. Most
network elements are still only available in their discrete forms, i.e., in an unintegrated
manner.
But this is changing, in several ways, most identifiably in the larger platforms now being
produced by NT, LU and CSCO which take into account many of the next gen
requisites. But there is still not agreement for voice at some very fundamental levels.
For this reason, we may wind up with many islands of connectivity over the next five
years, which doesn't do much for ubiquity of reach.
Getting back to the service providers themselves, making a profit doesn't seem to
matter as much these days as it once did, as long as the promise exists of there still
being yet untold star bursts beyond the horizon. This is euphemistically termed
'Internutsy.'
And the V-Cs don't seem to mind, as long as the stock prices keep going up.
EBITDA? In some ways this metric is only an artificial barrier to entry for those who
still believe in projecting a point in time when financial viability can safely, or
predictably, be reached. Cross-over points, however, will rarely have an opportunity
to be realized. At least that is the hope. But like I say, cross-over doesn't seem to be
on the minds of many principals who we read about these days. These upstarts would
defer to other means of extricating themselves, mostly through exit strategies that
depend on takeouts and mergers before the test of time has had an opportunity to take
hold.
Some [glamour] players will realize their dreams, but many of them will learn the
meaning of an unfriendly shakeout, instead. Perhaps not the companies you cited,
although that remains to be seen, but they are only the tip of the iceberg, numerically
speaking.
There are over 6,000 ISPs out there, and many more alternative forms of SPs who will
all be looking at this model, where they haven't already. It was an easy thing to satisfy
when surfing via http was the only game in town. But voice and other forms of
integrated services will prove too much for many of these outfits. For those who are
able to adapt, it will no longer be a means of differentiation, rather it will become, at
some point, a must carry issue in order to survive.
I think that the three companies you cited are only in the very early stages of
deployment, and it's too early to formulate an opinion about them right now that would
speak to the longer term. The ILECs have yet to show their hands in force. When they
do, then we'll be able to see how these startups are able to compete and retain their
subscriber bases.
If the ILECs wait out only one more generation of DSL infrastructure, then the new
unit level economies will be heavily skewed in their favor, leaving some of the present
startups with some heavy baggage to carry on their books going forward. Both from
the standpoints of technological and financial burdens. It's an old trick in this industry to
allow the enemy to pave the roads of public awareness and build the bridges which
span behavioral habits, so that one's own troops could march through, unencumbered.
In this way, the spoilists allow others to take the proverbial arrows.
On the other hand, if the ILECs are able to get their protected DSL infrastructures
implemented on a structurally separated basis (regulatory speaking), which means
that their new deep fiber feeder plant would only be usable by them (relegating the old
twisted pair to the D/CLECs), then we may see a truly superior form of next
generation DSL, as outlined in the FSAN model, that would be on a par with even
sparsely-populated cable TV coaxial segments. Isn't this the kind of sanctuary that T
has been able to carve out with the FCC for it's new found cable TV environments,
such as that of ATHM and the packet cable initiatives it is about to embark on for
voice services over cable? If this were to come to pass for the ILECs, then they would
be at a great advantage both pricing- and feature- wise.
Regards, Frank Coluccio |
