Curtis,
Now that we have lunch settled on, let's get to the issues.
>>Now QWST vs LVLT---Both are great in what they do-- LVLT from what
I see, has an edge in data communications, and they will have a more
modern fiber infrastructure. Fiber already in the ground won't hack
it at higher transmission speeds, nor in the DWDM fancies. The edge
goes to the data guys in service differentiators cause they know what
it takes to "do data". It just ain't a TDM AddDropMux and a 5ESS.<<
I don't think that I can agree with you on many of the things you state, as the devil goes off to get a cup of coffee...
---
1. Qwest and LVLT are huge, cash-connected, and promising, but that doesn't make them great. Qwest is getting there, but their efficacy in the long term still needs to be proved in the face of growing competition. There is no Santa Clause, in effect. But I do hold out a lot of optimism for them, and am enjoying their current antics with the regulators and the incumbents.
2. LVLT has the potential in every conceivable way, but if they go full bore into the "intelligent network" regime, as implied by the core and fabric of your arguments on their behalf, as opposed to providing apertures to almost 'endless' supplies of clean, dumb bandwidth, they may be painting themselves into a corner, setting themselves up for a Pyrrhic victory, or worse.
The current arguments which center around transparency, translucency and opacity apply here. In brief, the argument states, for the sake of the uninitiated, that the more intelligence you place in the middle of the network, the more restrictive it becomes. Conversely, the less intelligence, the more open it is. We could do a whole thread on this one, alone.
Notwithstanding, top-heaviness in the protocol itself, from a network-centric perspective, can backfire in a couple of years if not done with extreme care on providing for forward-migration capabilities. A prospect that is at the very least, extremely capital intensive, and at the risk of premature write-offs in an age of 18 month obsolescence. That is, of course, unless these carriers wind up stalling progress to favor their 'own' depreciation cycles. Where have we seen this movie before?
3. True, it ain't "just" TDM and Add-drop multiplexing anymore. In the higher electronic layers, especially. But from an opto-electronic perspective, and more to the point, at a photonic layer perspective, it is increasingly a matter of FDM (brought up to date with the monikers WDM and DWDM) and Add-Drop capabilities (still called such in emerging optical cross connect fabrics) that characterize the state of the art, and going forward.
George Gilder's Fiber Sphere personified? No. But getting there. And most of these enabling "breakthroughs," in actuality, have their roots in ancient analog technologies, to boot. Starting with the first open-wire heterodyned two-channel carrier system along railroad rights of way, threatening the telegraph (which also used variants of this technique) during the turn of the century.
4. The switchhead/packethead syndrome may be coming to a close, as I've indicated in the VoIP thread, due to cross-pollenation, turnover in livestock, and the influences of the Internet that are inescapable by even the most staunch and devout Bell-head. Two-hour loop trouble-clearing times are managed by relatively mundane practices, and people who, for the most part, have on their agendas, primarily, families to feed. They are necessary and require people to do it, be they ISP or BOC employees. When the Fiber Barons, CLECs and ISPs finally gain share in the loop, they, too, will acquire the traits that have traditionally been associated with the switch-folk genre.
It just happens that there are still more of the former BOC types at this time. This should not be confused with what board room decisions are about, though, when it comes to current and future competition initiatives. I will give you this: There remains a great deal of legacy thinking through inheritance in the traditional carriers. But some of it has already spilled into the ISP realm. Case in point, the herding of MFS higher ups into LVLT. It would have to be this way.
And where did most of those MFS types come from? I know a couple of IEC and RBOC regional heads who could tell you the answer to that one.
Intrinsic packetheads, if I comply with the vernacular of this argument, don't have the heuristics to deal with the scale and breadth of global services administration that the 'Net portends supporting going forward. I don't mean this from an ideal point of view, rather, from a practical one.
5. Th 5E-ESS and DMS2xx/5xx Series of switches may not be elegant means of deploying IP services right now, although we are still forced to use them through a series of circuitous contortions, conversions and detours. And I certainly wouldn't begin designing a new platform at this stage that would architecturally resemble them.
But how long do you think it will be before they begin supporting IP in its native appearance, for both voice and data transmission? In some cases this will occur through hardware/DSP upgrades, some through adjunct bypasses (XLSW & SUMA), all supported by revs in the generics. It is being planned, prototyped and tested as we speak. Rest assured.
----
BTW, this month's Business Communications Review ( bcr.com )has two supplemental packets that are noteworthy.
One is on the evolution of WDM/DWDM Evolution by Ciena, and the other is Voice2000. I suspect you'll need to get your hands on the hard copy for these, although I haven't searched deeply enough to be sure.
Now, if I can only pull myself away from work and these bloody boards long enough to read them...
Best Regards, Frank Coluccio |
