MC Press: Dr. Putnam, within the world of colleges and universities, there seems to be a degree of disdain for Microsoft. Why is that?
Dr. Jeff Putnam: I wouldn't call it disdain, really. More like distrust. Microsoft doesn't do a very good job of playing fair--not that playing fair is something that corporations necessarily do. Corporations have no soul to damn nor body to kick. So they're kind of immune to those things that people can do. Microsoft, having as much money as it does, is very immune.
I've talked to people who worked at software companies that Microsoft has basically destroyed. They wanted the business. And these are companies you know the names of, but because I was told these things in confidence, I can't say what these things were. There are several instances where there was a company that Microsoft has lied to, and then Microsoft has gone off and said, "Well gee, we're not competing with you," and then actually decided to compete with them and essentially kicked them out--destroyed the market--and the smaller companies are now gone.
But it's not just lying to them on a business level; it's lying to them on a technical level, too. For example, at one point Microsoft told this one company, "Well, here's the API that is going to be in the next release of Windows that will allow your product to work." The company counted on that to allow their product to work and to be able to ship immediately after the release of Windows. Microsoft took the API out of the operating system. No, they didn't take it out; they just renamed it. Which they then used in their competing Microsoft product. What this means is a single vendor, Microsoft in this case, can essentially lock everyone else out. And they've done that.
You may remember a spreadsheet package called Lotus Improv. Best spreadsheet ever. It was available on the Next system when Next was out. Absolutely the best spreadsheet I've ever used. But, it wasn't Excel, it wasn't Lotus 1-2-3, so basically what happened is the 1-2-3/Excel kind of model pretty much killed off Improv because nobody wanted to learn to use it. Even though it was 10 times more powerful and easier to use in the long run.
So, if you don't use a Microsoft system, if you want to use Linux or you want to use something else, you will rapidly discover how much of a grip Microsoft has that people don't see. So you try to use Linux on something, and you suddenly discover that you can't get drivers for something because Microsoft has bought up all the rights to the hardware specifications or the company won't release it because there's no market for anything but Microsoft products. I've got hardware in my house right now that I can't use because the only drivers that are available are for Windows. I paid good money for it.
Try to buy a computer without Windows on it. Even if you don't want Windows, you must pay the Microsoft tax. There's this thing about getting a rebate if you don't want to use Windows. Nobody will give you the rebate.
MC: A rebate for not using Windows?
JP: If you accept the Windows EULA [End User License Agreement], if you click through that, then Windows is loaded as your operating system. If you do not click on that EULA, if you do not want Windows as your operating system, then you're supposed to be able to go and get a rebate for the price of all this Windows software. Except that Microsoft wouldn't give it to you; the company that sold the computer to you wouldn't give it to you. If you bought it in a store, the store wouldn't give it to you. Each of these would just point to the next in line. And this is still the case.
And in the colleges and universities, we're bringing up our students to know only Microsoft. So basically, they have a stranglehold on the industry. What they have done is make the industry such that if you don't interoperate with Microsoft, if you don't go along with Microsoft, you might as well not exist. But, of course, they hold all the standards. So that, for example, to get into a .doc-formatted document, people have had to reverse-engineer the format in order to make things work and to get something else to write the format. And with the Digital Millennium Copyright Act, it gets even worse. The DMCA says that Microsoft could withdraw all of its software products from the marketplace and essentially make it illegal for you to read your own files with a non-Microsoft piece of software.
And the funny thing is, because of the way Microsoft has done this--actually, it's not that funny; really, it's kind of tragic--a lot of really cool advances in the industry have been lost, even on the Microsoft platform. When I started off, I didn't care about Microsoft one way or the other. It was only when I started using Linux that I began butting up against Microsoft, that I started to discover all these things. Basically, people know I'm not a big Microsoft fan, so they tell me all these stories.
MC: Is there anything good that can be said about Microsoft?
JP: Yes, Microsoft has done some good things in some ways. Some very good things. One of the things they have tended to do is build a software monoculture where everyone is running essentially the same platform. This can be a bad thing though. To use an analogy, consider the fir trees in the western United States. Fir trees are all Douglas fir. Pretty much all the fir trees are of the same species. So if a particular disease should hit the area, all of the trees will be affected. So a certain amount of diversity is a good thing.
However, the benefit of a platform grows dramatically, maybe exponentially, with the number of people that use it. That is, the benefit of 20 people using a piece of software is four times greater than the benefit of 10 people using it. Files can be shared, and so on.
MC: What is the impact of Microsoft policy on the business sector in general?
JP: Microsoft has made a practice of promoting and giving software away until it becomes a standard and charging you over and over for it. How many times have you bought Windows or Excel or Word? This has been very expensive for the business community in the long run. And you can't move from it. For IT organizations, this is a real pain and has probably cost 10 times what it should.
Further, IT departments shouldn't be putting machines on everyone's desk. Well, we should put machines on everyone's desk, but not ones with disks in them.
MC: Thin clients?
JP: Thin clients. Yeah, and I know the thin client thing has been around, and people thought, "Gee, that was such a bad idea," but it's not, really. It's the best idea. It's the way to do it. Because then you can have a single server, back things up off the server, control access, and so on.
MC: The old argument of central versus distributed processing?
JP: Yes, but the nice thing about thin clients is the processing isn't really centralized. The processing is distributed. The thing that's centralized is the software and the data. You know, how many people back up the data under their desks? How many people do that even within an IT department? Most of the desktop data is not backed up. Most IT departments don't even have a program for it.
And part of that is we've got this notion that desktop computers is what people want so that's what people get. But some of that is because people learn the basics of Excel and Word or whatever, and then they don't want to learn anything else because they don't perceive it as being germane to the problem, it's not really something they need to do, whatever the reasons. They want to stick with exactly that and not change it. And that's a really serious problem.
For example, Microsoft Word is the wrong solution for 90% of the problems it's used for. There's no metadata in it. There are few templates and none that can index the document content. There's no way to [programmatically] go into a business letter, for example, that is written in Word and identify things like who the letter is actually to. Typically, there may dates in the letter, but you can't tell when the letter was actually sent. There's no metadata. So what we've done is take three steps backward with something like Word. And now everyone has it, and they don't want to do anything else because they already know how to do this.
MC: How should documents be created then?
JP: Going to an XML Smart Editor kind of thing, where you actually fill in information according to a common set of metadata specifications, would be so much better. Then, if you file all these things--rather than file them on paper, file them electronically--if you file all these things, we can then search. For example, "find all the documents sent on the topic of submitting bids sent between January and June of that year." Now I'm willing to bet you can't do that in 90% of the IT organizations around.
Why? There's no metadata. There is nothing that says what this thing is. And yet that's exactly what you'd like to be able to do: take the data out of those documents and find out what you've bid. How many IT organizations have been successful in doing some sort of data mining over documents like that? Nobody does that.
MC: How does XML bear on this?
JP: XML is the technology to watch. It's been belittled by a lot of people, even here at this university, saying there's nothing to XML. In some ways they're right. There's almost nothing to XML. In other ways, it's an enabling technology that's going to change more--way more--than anyone can even imagine now.
There's a recent announcement on slashdot.org about the new Longhorn, and Longhorn is going to be using XML in a services-oriented architecture. XML is going to be the sub-strata. Imagine what you can do with that if you have data in Access; you can get it out into an Excel spreadsheet or into your word processor, whatever, and do it all in such a way that you haven't lost what the data actually represents but you can change the way it's displayed. Just because it's in XML doesn't mean you have to display it that way.
Then, what we can start doing is putting things like these business letters, marked up in XML appropriately, into some sort of database in the right way. And now suddenly that's part of your organization's data, accessible to a variety of applications and stored centrally.
Next month's column will continue the interview with Dr. Putnam and his thoughts about the future of our industry.