re: bandwidth after death: paranoia on the 'net. Coffe, Tea, and NSPs
Ed,
Ever since your eloquent post on the merits of network intelligence, and the sustained need for more of it as we move forward, versus a sure-fire solution through the execution of more bandwidth which you characterize as a myth, I've come across, and been involved in, a half dozen similar discussions, at least. And dare I say, this wont be the last.
Heck. Even my fifteen year old is now attuned to the imminence of a higher speed line. Her friend has one, she sometimes reminds me. Her "friend," as it turns out, is setting himself up on the web, along with his older brother, with a coffee and tea import & export operation. That's another post, entirely. But what a lesson this now-seventeen year old kid taught me!
Getting back to the push pull discussion on bandwidth vs application creep, however, the whole topic is starting to wear me down a bit. Maybe it's time for a rest.
I'm beginning to see a level of... I don't want to call it futility, exactly... let's just say mundane superficiality and uselessness at this point, perhaps an impertinence about it all. (Don't worry Ed, this is only a passing phase I'm going through.) Permit me to demonstrate to you how this topic is now starting to take its toll on me:
While pondering some of the near- and longer- term vagaries of the 'net with a fellow poster here today, over the phone, against the backdrop of some recent outages and other calamities, it became apparent to us, at least to me it did, that the possibility exists that none of this stuff actually works. At all.
[The first time I felt this way about the legitimacy of what I was observing with my own two eyes was the first time I saw the Beatles on the Ed Sullivan show.]
What if, instead, we down deep inside merely trust that all of this stuff works? I talk about photonic liberation all the time, right? Yet, try and get two physicists who weren't joined at the hip at birth to agree on the nature of light, much less the photon.
We may even think that it works in the smaller and more gullable parts of our brains. We certainly will give the impression to others that we know that this stuff works. And we will actually come out and say to others that we know how it works.
We perpetuate this charade each and every day as we give the appearance of accepting all that is disseminated in the press and other market venues, simply to avoid being tagged a Luddite, alas, as one of those poor souls who does not yet "get it."
Much of this is true.
It's as though we've assigned "getting it" the status of being the overriding criterion, a prerequisite of the highest order, for all earthly entitlements going into the next millennium.
And a lot of this is true. If you don't get it, you wont have it.
So, what do I do in order to shake off these notions? I go back to the books and review some of the basics. That'll work. I turn to the diagrams and foot notes at the bottom of each page which "prove" to me, once again, that the architectures being depicted are for real, giving me new-found comfort in the fact that they actually do work. What else do I do, just to make sure?
I pull out my copy of The ISP's Survival Guide which talks about peering and settlement relationships... but only in such a way that the author doesn't pin himself down. Which I knew already, because I've read this book before. But the book is a year old.. which causes some stress.. is it still accurate?
I fast read through the most poignant sections of the latest Cook Report that I've annotated and underlined. I looked up a couple of book-marked locations on the web that I'd saved since first becoming 'net struck, only to find that they've already died and gone to cyber heaven. What do I deduce?
I deduce that contrary to some of the well heeled clichés I've learned to accept in other, normal situations, you cannot easily predict the future, or even understand the present, of the 'net by analyzing its past. All of my usual reference points which I've turned to in order to assure myself that I wasn't losing my mind were already, if not fast becoming, obsolete. I find that even some of next year's terabit routers, by a growing number of measures - depending on what camp you are in - are already obsolete.
The business models of emerging service providers of every color are about as predictable as the turbulence found in a hurricane. Startups, up and down Silicon Alley and across the nation, are planning next year's e_commerce cybershops, auction outlets and more, but they are stymied. They require new modeling and analytic software, we find - and faster platforms to run them on - that hasn't been coded yet, much less debugged, lest they employ legacy code that everyone else has been using. Brick and Mortar business software doesn't fit the 'net model, they soon learn. And local wiring to cash registers and LANs do not a TCP/IP long distance transactional and relational model make.
Using off the shelf software, as it turns out, isn't exactly conducive to the unique types of cyber solutions the VCs are demanding, either. But they settle for it, nevertheless, lest they lose their windows of opportunities.
So, the startups wind up fluffing it in their business plans. In so doing, they often create an aura that they possess the next new miracle on the horizon that will save the 'net from itself, or allow you to buy tomatoes without having to feel them first, for which they will IPO within record time frames, avoiding much of the usual due diligence that was necessary before. And when they do, they will IPO at a reasonably unreasonable price. [Credit for that last quip goes to someone else whose post I can't find. Sorry, whoever you are.]
Where does this all leave me as I struggle with this new found paranoia?
I profess to have some fairly accurate notions as to where we'll be in twelve months in certain types of solution spaces, maybe two to three years, but that's really stretching it. It's not as easy to predict these time frames, relatively near term, as it is to shotgun into the longer term, when the latter forgives specificity.
And I know what is no longer relevant, what is decomposing, and who might have gone to Heaven or Hell already, because their cemetery plots have been filled and the tombstones say who's inside.
Where does that leave us, though, with respect to today? Is it even worth contemplating where we are today? Today was already dead yesterday, or last month, or last year.. depending on how fast you can type up a business plan. What was that, you said? Did you say, "Send in the chimps?"
Is it worth contemplating at this point which will win out over the next phase of the Internet's evolution? The abundance of bandwidth or adding intelligence to POPs? POPs, by the way, which had no business being on the 'net in the first place, if the architects had done their job right all along?
I mean... adding new intelligence at this time. After all of the celebration that was afforded to David Eisenberg's superb treatment in the way of his stupid network encyclical?
-------
And what else do I do to shake off the bad vibes when they return? I turned to Boardwatch.com on the 'net for a dose of reality. But instead, here's what I found:
-------
I came across some verbiage in this month's Boardwatch.com which seemed to reinforce some of the paranoia I've spun above. I took it out of its original context, from Bill McCarthy's editorial titled "Theater of the Absurd:..." and pasted it below:
From: boardwatch.internet.com
"If you follow this stuff closely, your head ends up spinning so fiercely and becomes packed with so much conflicting information that all life seems to be a Federico Fellini film. “Realism is a bad word. In a sense everything is realistic. I see no line between the imaginary and the real,” said filmmaker and director Fellini, whose neo-realist film movement evolved beyond the strange....
..In these days of cultural relativism and the open forum of the Internet, everything seems to come down to trust in a cyberscape of transitions."
-----
Now, what was that you were saying about application creep constantly chasing, often overcoming, bandwidth supply?
Regards, Frank Coluccio
|
