This is the first of three essays that speculate on Microsoft's Internet strategy, aka its ".Net initiative." Unlike other analyses, which are based on Microsoft's announcements, I take the approach of asking, "If I were Microsoft, what would be my strategy?"

This exercise reminds me of a novel from many years ago. Our nation's literary culture has not always been as strong as it is today, when we can enjoy such towering masterpieces as Who Moved My Cheese? In the 1960's, we had to settle for other works. One of the books that dominated the best-seller list in its time was Saul Bellow's Herzog.

In Herzog, the main character is in the habit of writing letters to remote figures, such as the President of the United States or God. For me, writing suggestions for Microsoft's strategy is an equally quixotic gesture. Hence the title for this miniseries, "Bellowing at Microsoft."

My observation is that there are three factors at work when Microsoft is successful.

1. Microsoft focuses on software. Microsoft has tried to expand into hardware with a few devices (such as WebTV), into the ISP market (with Microsoft Network) and into entertainment in various ways. However, it does best by sticking to its core business of software and letting other companies build hardware and provide other services.

2. Microsoft levers its installed base and takes advantage of network effects. Microsoft will have an advantage in marketing a product only if it is related to existing Microsoft products, so that Microsoft can take advantage of its large installed base.

3. Microsoft levers the work of other software developers. Microsoft does well when it can recruit third-party software developers to write programs for its platforms. When it succeeds in doing so, it achieves market domination. When it fails to do so (as in the handheld market), other companies are more successful.

With that as background, I believe that there are three problems that need to be solved on the Internet that Microsoft could profitably address. One is that of data preservation and security. Another is to facilitate formation of clubs as a means of paying for information goods. And a third is developing mailbox protection services. The rest of this essay will focus on the first issue, with the others deferred to subsequent essays.

Data Deposit Boxes

Garfinkel's argument is that the files that we store on our personal computers are vulnerable to at least two types of catastrophic events. One, of course, is a disk drive failure, due either to mechanical malfunction or a malicious virus. The second catastrophic event is a technology upgrade. You get a new computer and you leave important files either on your old computer or--worse--on older media that are not even supported by a new computer. For example, 5-inch floppy disks.

Garfinkel argues that we need a place where we can store precious computer files that will survive either a disk drive wipeout or a migration to new technology. Hence, the idea of a data deposit box.

I have argued in Personalization, Portability, and Ownership that our personal data should be stored for us on our behalf. This is in contrast to the conventional wisdom that the best way for us to obtain the benefits of personalization is for corporations to gather data about us.

I would like for my personal database to include my resume, my medical records, my financial records, my personal purchase history, and other information. In addition, I would like to use my personal database to store my email address book, emails and other files that I want to save, and other information that now sits precariously on my computer's hard drive. Having this information on the Net will make it easier for me to work with it when I travel as well as make its storage more reliable.

Naturally, I would like to have complete control over access to this data record. Again, the analogy with a safe deposit box is apt. Just as a bank will ensure that no one else can access my safe deposit box, I want my personal database to be something to which I control access. (That does not necessarily preclude other parties, such as insurance companies or credit card issuers, from having independent sources of information that they can use to protect themselves against fraud.)

Microsoft understands the value of a personal database. Reportedly, Microsoft plans to offer consumers the ability to maintain a personal database on their network.

Many commentators have reacted negatively to the idea of storing their data "with Microsoft." They fear that Microsoft will take advantage of the data in some way, regardless of what it states as a privacy policy.

The solution is for Microsoft to focus on developing the software infrastructure for creating, reading, updating, and deleting information in personal databases. However, the management of the data storage could be performed by banks. Banks could use their experience in managing physical safe deposit boxes and personal financial records to develop policies and procedures that provide consumers with reassurance that our "data deposit boxes" are secure.

The various parties and their responsibilities for maintaining data deposit boxes would be as follows.

1. Consumers.

   Consumers would decide when to deposit data. Consumers would authorize third parties to view data. Consumers would have the ability to create custom categories or folders for data, but data also would be cataloged in standard ways to facilitate authorized uses by third parties.

2. Microsoft.

   Microsoft would incorporate the function to deposit data into your data box into every program. Just as now you can click on "File, Save" from almost any program, you would have the ability to send information to your data deposit box.

   Microsoft would develop programs and program interfaces that would enable third parties to request access to data and to view authorized data. Microsoft also would develop programs for banks to use in setting up data deposit boxes.

   Microsoft would develop standard data "views." For example, there would be a standard way to extract personal contact information from a data deposit box (with permission of the owner of the data deposit box required).

3. Banks.

   Banks would provide permanence and security for data stored in consumer data boxes. Microsoft should allow--and encourage--banks to develop their own security mechanisms for consumer data boxes. The more diverse and multi-layered the security systems, the greater will be the cost-benefit ratio for hackers to try to break in to any particular system.

The conditions for a successful Microsoft product would be satisfied as follows.

Therefore, I predict that Microsoft will try to foster the use of data deposit boxes. My guess is that they will have better luck getting consumers to adopt the concept if Microsoft works in co-operation with banks.

Related Essays	Club Infrastructure	Email Protection