Belly of the Beast I was at a Microsoft presentation today where they spoke about their .NET strategy and XP. Here are my reactions:

1) Microsoft's new licensing scheme is designed to increase the cost of ownership for those whose upgrade cycle is longer than 3 years while making it cheaper for those with upgrade cycles of 2 years. Their argument is that since it costs them more money to support old software, they should pass those costs onto the appropriate users. Personally, I have never had any support from Microsoft for any Microsoft product, so I don't know what costs they're talking about. Their real motivation is to force people into a faster upgrade cycle, which is more lucrative for Microsoft. Microsoft's only competition on the desktop is old Microsoft products.

2) Passport authentication is still in flux. The presenter told me with a straight face that ultimately, each user may end up having their own personal authentication server on their desktop which would then serve authentication up to other devices. For 99.999% of the population, this means all you need is a text expander on your desktop to store your passwords. No one needs their cell phone to ask their desktop for permission to buy off Amazon's useless WAP site. I'll write about the Joy of Text Expansion soon. Personally, I doubt Redmond will give up the power centralized authentication gives them.

3) .NET really is a big deal for the company. The presenters honestly believed that Microsoft supports open standards because they use UDDI, XML and SOAP. This is disingenuous, because a centralized authentication server is essentially a closed system even if it runs on open protocols and platforms. And they are shifting all their apps to this base. Moreover, the Word XML formats will continue to be partially proprietary, just as Word document standards are now, so don't expect great openness there either.

4) Smart tags are here to stay. Although they may have taken them out of the browser (for now) the rest of Office XP is riddled with Smart Tags that default to Microsoft standard services (such as Expedia over MapQuest). Users will probably never change their Smart Tag defaults, so unless companies pay for the privilege, they're not going to have access to these services. This clearly should be an active anti-trust area.

5) The SSSCA did not have high priority within all parts of the organization (these folks had never heard of it).

6) The future of Microsoft is built on massive interoperability where applications can serve up information (possibly over a network) to each other. So Outlook will serve addresses to Word, Access will serve numbers to Excel, and PowerPoint will send data to Visio. Given that the transaction related parts of XML will be kept proprietary, this essentially makes Office even more monolithic than it is now. When Microsoft talks about interoperability, they only mean interoperability with other parts of Microsoft. By converting the PC to a server as well as a client, they've automatically created (and own) 250M low end servers. That's server market share that Apache can never dent. It also locks Java out of the peer-to-peer game, the one area on the desktop it actually had some relevance. This is probably partly why Java will not be included in XP.

It will become increasingly expensive to use Microsoft. It will also be hard to operate outside their network, unless you use truly open protocols. The applications will become thick clients that increasingly mediate between the user and their data. So expect more frustration. And since monocultures are toxic to a network, expect more viruses that are more pervasive than anything we've seen so far. Businesses are going to have to get MUCH smarter than the current crop of execs who chase after any gee-gaw convinced it will finally make all their IT investments pay-off. If I was a business seriously interested in productivity through technology, I would stay away from XP and .NET.
Link to this article