Meltdown Myths


Psst! Want in on a secret? You know the Internet Meltdown "They" are predicting? Well, it's part of a big corporate conspiracy to control your brain! Who are "They" and why would they want to do something like that? Can't tell you here -- not secure -- I'll call you from a pay phone... Conspiracies aren't just fodder for USENET kooks and vacuous television shows, you know. They are sometimes real, though seldom are they as intentional and elaborate as is depicted by the aforementioned sources. Conspiracies are more often the result of people with power and money attempting to influence events to their advantage. Since the thought patterns of wealth aren't big on creativity (think about how little imagination a CEO has to exhibit to be called a "visionary"), these individually-hatched plots are often very similar in motivation and execution, and their external results can have the appearance of careful coordination and, yes, conspiracy.

What does that have to do with the Internet and its physical collapse? The death of the Internet has become a hot topic for columns such as this one for the past number of months, and awareness of the possibility is entering a larger and larger segment of Internet users. First mentioned by Bob Metcalfe, the idea attracted attention due to his credibility (an inventor of Ethernet after all!) and his habit of shouting his predictions to the rooftops at every opportunity. Metcalfe's dire warnings were initially considered to be alarmist, but as time went by, user frustrations about the "World Wide Wait" started to lend the meltdown theory increased acceptance. Even the "World Wide Wait" catch phrase has been adopted by the same mainstream press that beat the "Information Superhighway" to death about twelve months ago.

The theory goes that the rush of new users and new applications to the 'net have clogged all the routers and switches, and, TCP/IP being ill-suited to high traffic conditions, large portions of the network as a whole will fail catastrophically, causing "brownouts" and denying access to millions for long periods of time. This scenario misses two critical factors, thankfully, one technological and one having to do with open market economies. First, while TCP/IP doesn't tolerate high traffic, it does adapt well to adversity, finding routes around "damage." Second, where there is a surplus of demand and a scarcity of supply, companies rush to be first to satisfy it. There is certainly demand for the Internet, and both established companies and start-ups are indeed rushing pellmell to fill the gaps in supply. All the present "backbone" suppliers, Sprint, MCI, et al, are upgrading their connections as quickly as is possible. In addition, companies such as Digex, Best, and others are rapidly deploying their own nationwide networks and negotiating space at the Network Access Points, or "NAPs." Finally, there are serious research efforts to boost switching speeds to the terabit range, again from the established suppliers and some promising new dark horse contenders. All this activity is certainly enough to convince this observer that we will have all the bandwidth we currently lack and some left over to take us to the next bottleneck.

The doomsayers also conviently ignore the fact that the history of personal computing is really a history of bottlenecks. First the processors were too slow, then the data bus, then the video, then the network, then the processors again, etc. ad nauseum. And that's ignoring completely the perpetual game of leapfrog in which hardware and software are locked. So, why are the stories of gloom and doom taken seriously and not relegated to our mental kill files? News requires repeating to remain news, and the predictions have flown from first one mouth to a large chorus of voices. Media hacks hear the same theme from several sources and think in headlines, and the Death of the Internet ends up on the cover of a news weekly. The interesting question to ponder is why Metcalfe's grumpy rumblings spark sufficient interest for all those parties to take them up and make them their own.

The answer probably changes depending on which source and which version of the story you are analyzing, but one common thread persists across all involved parties. Repeating and amplifying the collapse myth is seen as a strategic move that may somehow shift events to benefit the party who helps it spread.

Some versions of the myth are easy to see through: take the telcos, for example, who focus on the local loop instead of the Internet proper as ground zero. Their version goes something like this: "Internet users' average call duration is twenty minutes, and our network is designed around a three minute call model, therefore all our quality of service is suffering." The potential gain? Telcos are actively trying to convince the FCC that ISPs have to pay local access fees to reduce their own contributions, and their peculiar version of the meltdown myth may sway opinions in Washington. Of course, there's another theory, too, slightly darker and more nefarious: RBOCs such as Southwestern Bell, Pac Bell and Atlantic Bell are all jumping into the Internet Service market with their own offerings, and most ISPs would fold if they suddenly had to pay local access fees... but I would never suggest such a thing! See what I mean about justified paranoia?

Of course, some companies would like a little worry about network capacity all the time. Router manufacturers, like car manufacturers, love to sell you a new router every eighteen months or so, and what better motivator than the fear that you might remembered as the loser with the old router that brought down the Internet? Too much hysteria, however, is likely to scare off both consumers and potential providers of network access, not to mention the inevitable mumblings about hardware failing to cut the mustard, so you won't hear Cisco or Bay supporting the myth much at all.

Then there is a giant software vendor whose business model depends on bloated applications that attach themselves to users like enormous ticks, making us all dependent on them apparently by their sheer girth. We won't name names, plenty of other observers do all the time, often more for the thrill they get from being naughty rather than constructive criticism, but you know of whom we speak. The founder and central nervous system of this company conceded to George Gilder in an interview published in Forbes ASAP that he is hoping that consumer Internet access settles in at the "middleband" level for quite some time. That would mean ISDN to the home, not ADSL or cable modems or anything faster. Broadband, high speed access would suddenly make the component model offered by Java, Open Doc and Castanet eminently reasonable, and would probably dig the grave for bloatware. A quick compare and contrast of the concept behind the NC vs. the NetPC is enough to prove that. Would you ever want to even consider installing Office 97 over a network? Any network?

Of course, whenever an entity is so large and so dominating a market, the temptation to spin wild conspiracies is overwhelming. The darker undercurrent of the Redmond giant's support of the meltdown myth is that, if the public network loses favor with the consumer market, someone could build a proprietary, private replacement and sell it as a superior replacement. That's what MSN was supposed to be, remember?

In fact, this is the sinister subtext that swims below the surface every time another industry type repeats the meltdown myth. It's silly to propose that large corporations are meeting secretly and plotting the overthrow of the Internet, but it's even sillier to forget that for most of these companies, anything that is that big and unownable is a threat and a menace. When the Internet took off in 1994, you could feel the lumbering giants look up and shudder as the shadow passed by overhead. Some have attempted to embrace it, others ignored it, and some have quietly analyzed the market, looked at their own goals, and cooked up strategies to engage it in battle. Everyone agrees that Sun's Java is a good thing, but Microsoft still forges ahead with its own "flavor" that will be incompatible with the very concept of cross-platform computing. It's unlikely it thinks it can "win," but it probably stands a good chance of slowing Java down sufficiently that it dies all by itself. Sun itself, along with Oracle, have shown with the NC that they see the opportunity to subvert the Internet to their own goals, namely, a return to essentially dumb terminals supported by lots of expensive hardware and software. They can find lots of support for this idea in the halls of corporate America, where the idea of the employee being in control of his or her own information never really got digested.

This is a dangerous time for the Internet, and for all of us. Ethereal and, for increasing numbers of us, ubiquitous, the Internet can too easily be seen as invincible. In our society, the shape of our lives is too often ruled by the almighty dollar, and everything that unfolds is shaped by greed. A huge public "property" that will extend its reach into every home in the country is the ultimate bait for the greedy mind. Every entrepeneur and executive with the slightest technical bent has thought, at one time or another, "if I only owned the Internet..." When the prize is that big, some of them are bound to try, in one way or another. What we must do is not blindly embrace or discount the myth of the meltdown. Instead, if we care about our collective ownership of the Internet, we must listen closely when we hear someone say "World Wide Wait," and ask ourselves where he heard the phrase, and what does he gain by its repetition and propogation?


Return to column index


© 1997 Sebastian Hassinger