Summary: How a couple of simple inventions and smart gathering of data could really kick Microsoft in the pants.
Knocking The Fat-Cats Upside the Head
Simple inventions have a habit of triggering huge shifts in power and money. What may look like an ordinary TV remote control, for example, is actually a deadly machine gun to the eyes of the once triopolistic broadcasting networks (ABC, NBC, CBS). Before the remote control, the bulk of entertainment automatically dropped in quality to only what was good enough to stop you getting out of your chair to cross the room. Or then there's the bar-code and scanner. Since companies like Gillette knew when a TV commercial was going to air and send people out shopping, they could force the supermarkets into displaying their razor blades exactly how they liked it, and when they liked it, and how much to re-order. But give the supermarkets the power to gather incredible volumes of data on their shoppers, detailing not only when shoppers bought something, but also what else they bought, and that information can be used to hold Gillette to ransom instead.
Looking at these examples can help you understand the economic and power shifts that have occurred in the computer industry. The installation program removed the need for expensive consultants to set up a company's computer software, for example (up until bad programming brought it back). But it can also teach us how to Kill Microsoft.
Lets start analyzing the situation. Microsoft gets on 99% of all computers by pre-loading practices, much like how the broadcast networks get into every TV set through their affiliate stations that dot the country. The remote control - an easy way of changing channels with the twitch of a thumb muscle - killed the chances of the stations getting away with mediocre entertainment. When cable and satellite came and joined the remote control, the smaller stations could now reach a whole continent without the need of affiliates.
Parallels? Boot manager software lets us switch between operating systems very easily, just like the remote control let you change channels. The internet gives small software companies the power to distribute their product to the whole world cheaply, just like cable and satellite did for independent TV stations.
Boot Manager was probably the smartest feature that IBM could have put into OS/2. It was boot manager that gave millions the confidence to try OS/2, knowing it wouldn't be an all-or-nothing deal. If this software didn't exist, or PC hardware didn't allow multiple bootable partitions to exist, then OS/2 wouldn't have existed outside an ATM machine at all. Is it really a surprise that practically every operating system for the PC comes with some kind of utility like Boot Manager? All, that is, except for Windows?
Making it easy to switch between operating systems, easier than what Boot Managers do, is priority number one for anyone who wishes to knock Microsoft upside the head. Invent the equivalent of the remote control for computer operating systems, one that can let you jump from one OS to the other without requiring a shut-down. What we have right now is a bit like a remote control that has to stop the VCR from recording the show you want on channel 2 before you switch to channel 3.
Next, learn from FreeBSD when it comes to installing operating systems. IBM, take note. Make OS/2 freely downloadable and give it a "one floppy" installation program. The idea is this: You download a program that creates a special, bootable floppy disk. That floppy has enough OS/2 to boot, plus a minimal internet dialer and TCP/IP stack. After you've booted from the floppy, you connect to the internet and install the operating system via FTP. It works for FreeBSD so well that it's what got me to try it, and it can work for OS/2. You solve the problem of not being stocked on the retail outlet's shelves and help with the lack of pre-loading by making this installation as easy and painless as possible.
The next step -- learning data acquisition tricks from the supermarkets -- is made harder by the fact that Microsoft is already doing pretty good at it. Marketing tricks such as the creation of "Office Suites" (smart bundling) came about because someone was watching what software was being bought by who and when. Not only is Microsoft powerful enough to wrest sales data from the retail stores, they also take thousands of technical support questions per day from people who tell them 1) what they have installed, 2) when they bought it, and 3) what they use it for -- all as a matter of course in a normal tech support call. For Microsoft, it's was easy to see that those who bought a word processor also bought a spreadsheet and database.
It would seem that to compromise Microsoft's monopoly you should go into the third-party technical support business instead of trying to compete with them on the desktop. When Microsoft's best technical advice is to "delete and re-install," you'd think it wouldn't be so hard. Gather all the information about what people are doing with their computers and you've got MS by the throat, since if you sold that information to someone else like Corel or Be, they could make a better office suite or operating system with no trouble. It's just like how the supermarkets could sell their shopper statistics to Bic instead of Gillette.
That kind of technical support takes a lot of resources to supply, however, and it seems as if the one company already doing it -- IBM -- isn't using the data they're collecting.
Or are they?
IBM wants to be a solutions company, one that solves problems rather than ship boxes that create new ones at the same time. They've been suffocating OS/2's destiny, but they've also been taking the destiny of Windows out of Microsoft's hands. IBM ships Apache now, not IIS. It could be that IBM lost the desktop war against Windows, only to dangle Microsoft itself on the end of a string.
In the war against Microsoft, all the ammunition comes in ones and zeroes. Visit a repository of user-supplied munitions in our interactive forum and dump any steel-tipped rounds you have handy too.
|Copyright © 1999 - Falcon Networking||ISSN 1203-5696||February 16, 1999|