WWW8 Notes - Predictions
07/10/2000
by Edward Piou
It's been over a year since WWW8 - the Eighth Annual World-Wide Web Conference - took place in Toronto, Canada. Several books from early web technologists (including Tim Berners-Lee himself) have come out since then, tracing their views of the development of the Web. In addition to looking at the past, though, many of the WWW8 attendees had an eye toward the future. It's worth looking now at the predictions made by two of the conference speakers and see the progress we've made.
Much of Tim Berners-Lee's keynote dealt with his vision of a "semantic web" - a web where both basic information, and machine-understandable information about using basic information, would be important. A lot of technological progress has been made, in developing both XML and the tools and protocols - including SOAP - that will make distributed, automatic data exchange a reality. But the semantic web still isn't here - the web is still about information, rather than meta-information exchange, and in the trenches, designers and programmers are still not working on a truly distributed medium.
Of course, distributed technologies like Gnutella - which make every computer a server, and use the network to great effect - may eventually provide a quicker and more robust route to inter-computer resource sharing than the technologies currently being developed in the more traditional, standards-based, top-down means.
Bob Metcalfe, founder of 3COM, is infamous for his predictions - and also infamous for the stunts he'll pull if they fail to come through. (In 1996, when the Internet did not suffer the major outages he predicted in 1995, he ate the paper on which he'd written his predictions - as promised.) In his keynote speech at WWW8, Metcalfe started out with some fairly tame, and fairly believable, predictions:
These are all visions which will take more than the year that has passed to come true, and not that controversial. To liven things up, though, Metcalfe made seven more specific predictions which can be judged now:
On November 8th, 1999, the stock market bubble will burst.
Wrong. The stock market bubble didn't burst on November 8th; it wasn't until the year 2000 that a tech meltdown occurred, and even so, the damage was somewhat selective, with many mediocre Net companies continuing to hang in there on the market. (Of course, nobody expected Metcalfe to get the date right, but he was rather specific...)
Y2K will be a nonevent
Correct. In the sense that he meant it - that we wouldn't have catastrophic computer or societal failures - Metcalfe was right; because of a combination of hard work across the industry, plus decisions to finally upgrade decades-old hardware and software, we survived.
"Pretty soon you're going to find drugstores serving cappucino"
Wrong. The theory was that because you'd be able to shop for anything online, real-world stores would need something better to draw in customers to their stores. As it turns out, being able to see and feel products have kept customers from abandoning bricks-and-mortar storefronts for Internet shopping. That, plus the problems online retailers have had with home delivery.
In 1999 ... the direct access internet will begin its deployment and into the millions; multimegabit, and always on.
Correct. DSL, once thought an also-ran, is making large strides; and there are plenty of customers who are happy with their cable modem access. There are millions of customers with always-on, megabit access to the Internet - but many millions more without.
The growth of the Internet will drop from annual doublings to less than 100% a year
Correct. Internet growth is still high, but growth is limited by how much people are willing to spend on computers and connections.
Open Source - "I predict it's going to fizzle."
Wrong. Open-source software is still going strong, from the perspective of both developer mindshare and money spent (on support, service, and - yes - software). Of course, closed-source software is still selling well too. Perhaps there's room for both.
"A gigalapse of the Internet will occur before the end of the year 2000."
Unknown. We've still got 5 months for a single event to cause a network outage costing 1 billion user hours. But enough has gone wrong to demonstrate the point of Metcalfe's prediction: there's a lot that can go wrong with the Internet - and does.
Edward Piou is an ahref.com producer and runs ep Productions, Inc., a development company based in the Washington, D.C. area.
This site copyright 1998-1999 ep Productions, Inc. Text of any articles is copyright of the author.