The Web didn't change much in 1999. I think more fundamental changes are in store for 2000.
For years, I have been saying that micropayments are needed for quality services and for sustainable business models on the Web (since advertising doesn't work). And for years I have been disappointed at the slow progress at getting micropayments implemented. In particular, it was sad to see MilliCent go to waste after Digital was sold to Compaq. They could have become the Federal Reserve Bank of the Internet economy but simply sat on it for years not knowing what to do.
I now finally believe that the first wave of micropayment services will hit in 2000. MilliCent may still have a chance, but there are several other solutions that are just as good - maybe even better.
Initially, micropayments will only be used for highly important content and services that specific users feel a strong need to get. In the beginning, users will be reluctant to download, install, and subscribe to a payment service. Especially since they will have to do so multiple times with competing payment services. It would have been much better for micropayments to have been integrated in the operating system: pre-installed single solution chosen by somebody else. But the way things currently are, users have to suffer. Which means that micropayments will not be used for casual browsing in 2000.
During 2000, micropayments will be restricted to vertical markets and value-added content. Much of this content may be so valuable that it is priced at several dollars instead of a true micro payment.
You cannot charge for, say, a movie review, if doing so requires the user to install special software. But once enough users have micropayment solutions in place, it will become normal to charge for good content, even when it is not mission-critical. I expect this to happen in 2002.
Despite my poor record of predicting mobile access for 1999, I think it may finally happen in 2000.
People who have wireless Internet access usually rave about it. For example, Ricochet users give it top review ratings because of its high quality and fast bandwidth (28.8 with 128 kbps coming Real Soon Now). Unfortunately, Ricochet is only available in Silicon Valley, Seattle, and Washington, DC: this may be good for somebody who does a lot of computer consulting, visits Microsoft, and lobbies the Federal government, but the coverage area surely doesn't satisfy the average business traveler's needs.
Ricochet has received a $600M cash infusion that allegedly will be used to roll out national coverage in mid-2000. This could be one of the main drivers of wireless Web access in 2000. Several telephone companies offer wireless Internet access, but they suffer from limited service areas, slow bandwidth, and oversold capacity - and worst of all, denial that open IP is the future.
Despite the problems, mobile Internet access will multiply in 2000. The biggest advance will be the launch of PDAs that have wireless modems built in as standard. The Palm Pilot VII was the first, but surely not the last Web Device.
One variant of mobile Web is the possibility for limited interaction on a cellular telephone. Of course a telephone will never be the same as a true PDA with a real screen, but they will still play an interesting role as a temporary solution in 2000.
One of the upcoming mobile Web services is the ability to buy from Amazon.com over a PCS telephone: an ideal service since books can be fully specified by an ISBN number which would be easy to enter on a telephone keypad. If this service takes off, it will kill many independent bookstores: you are out shopping and see a book you like. Instead of waiting in line to buy it and then having to lug the heavy package around the mall during the rest of your shopping expedition, you whip out your phone, type in the ISBN, and have the book delivered at a 30% discount.
In general, the Web makes it easy to buy products from the cheapest vendor. At the same time, websites that sell things are becoming less and less credible in terms of helping customers find what they want (as opposed to what's currently being promoted).
Solution: unbundling different levels of service. Advice is offered from one source and fulfillment is offered from another. Both get paid: The site that delivers the product will hopefully make a profit from the sale (I know: this is not the case right now). And the sites that offer different kinds of advice will make money from micropayments and affiliate programs.
I expect to see physical experience environments that charge a fee to allow people to browse, see physical products, and try them out. Though I don't think unbundled physical experiences will be common until 2002 or so.
The Internet infrastructure itself will keep running, but some individual websites will suffer Y2K problems. Too many sites run on software that has been patched together in a hurry without the oversight of a senior software architect. Sometimes, a Y2K bug will be lurking inside this low-quality software.
Example: Paramount's Star Trek website listed an airdate of January 1, 1900 for the episode "Fair Haven."