|PHOTOS © TERRY LALOS|
Can one company be the dominant OS, application, and development-tools provider for Linux? Well, if anyone can, Ottawa’s Corel Corp. is certainly a leading candidate for the “company most resembling the Microsoft of Linux award.” They sell applications, an OS, and until their acquisition of Borland/Inprise fell through, Corel was looking to provide a wide range of Linux developer tools. This April, Corel’s CEO Michael Cowpland invited Linux Magazine Executive Editor Robert McMillan to tour his company’s shiny gold headquarters and to talk Linux.
Linux Magazine: Corel Corp. is the first company from outside of the Linux space to really bet its business on Linux. How did Linux first come on your radar? It seemed for a while there that you were hoping Java would give you an alternative platform to Microsoft.
Cowpland: Well, the way that Java was originally advertised was a neat concept. But the Java virtual machines didn’t perform as advertised and still don’t.
Sun basically is several years behind in the development timetable and even now they’ve still got their claws in it from the point of view of licensing. Hopefully IBM will beat them down on that to make them more reasonable.
LM: At what point did you realize that Java was not going to be a suitable desktop platform for you?
MC: I think as we began to get the actual product converging, we realized it wasn’t really happening at the speed that we were used to. We saw that it wasn’t converging and Sun kept shifting their release dates. So we thought: “This isn’t the way to go. Let’s go to a thin client model.” It was just six months after we’d gotten into Java.
LM: Did you have a Linux evangelist within the company?
MC: We had a few. At that time Eid was the chief technology officer at Corel and Pat Beirne was our chief engineer. The two of them pretty much said that this was the technology to adopt.
I know Pat Beirne really well. We go back fifteen years. When he told me that the more he looked at Linux the better he liked it, that was the real turning point.
With Java, you would never hear that kind of observation. People would say, “Wow. This is pretty buggy stuff. Good idea, but I wish they’d make it work better.”
LM: So at what point did you decide to port products to Linux? When did you realize that WordPerfect, for example, would work well with Linux?
MC: We did that shortly after acquiring it [from Novell in 1996] because we had the Unix versions of WordPerfect. We ported the Unix code to Linux fairly early on, with WordPerfect 7. At that time it was priced pretty much like a Unix product. Then we made a bit of a bold move in December of 1998 — we decided to offer it free. And that was a great move because we’ve had two million downloads.
LM: That seems to represent a mind-shift. The traditional Unix market consists of high-end workstations with expensive software, and here you are releasing a product for free and getting two million downloads.
MC: Well, we had to get an education from Bob Young. He came out from Red Hat before their IPO, when they were only about thirty guys. He explained the Linux model. He said the more stuff they give away, the more money they make. That was the funny thing about WordPerfect 8. Even though we had the two million free downloads, we still sold a few million dollars worth of the boxed version. People were still paying for the manual and the CD-ROM.
LM: Even with the pricing, at the time that you released WordPerfect 8 for Linux, there was a perception that if you didn’t make whatever software product you had available for free, you could forget about it. But look at Loki Entertainment Software right now. They’ve been pretty successful selling games. I think that perception is going away.
MC: There’s a difference between the highway and a shopping center. You don’t want to pay a toll every time you drive down the highway, but you don’t mind buying stuff along the way. That’s the way we look at it and some things that are connected tightly to the OS you want to open source as well.
LM: How much of Corel’s revenue comes from its Linux products?
MC: About five percent. In the first two quarters [of fiscal 2000] we did about $5 million with Linux, which is a total of about five percent of $100 million in two quarters.
We expect that to be higher this quarter because we’ve got our office suite shipping. The other good thing is that we’ve brought forward the launch date of CorelDRAW and Photo-Paint to July from October.
LM: When did you first use Linux personally?
MC: I’ve been using it for over a year now and I really enjoy it. I find that it’s much better for browsing the Web. The apps are getting more and more complete.
LM: What percentage of your company is doing Linux development?
MC: In terms of getting involved with it, probably half, but they’re not exclusively working on Linux. Quite often they’ll be covering the product that’s going to be on both Linux and Windows.
Why Corel Linux?
LM: What made you decide that Corel needed to become a Linux distribution vendor?
MC: We didn’t find any [distribution] already out there that was easy enough or polished enough for the mass user. We think polish is important for a mass user. They expect a certain look and feel, and we know how to do that from our nine releases of WordPerfect. So that experience translates into a certain look that users feel comfortable with.
LM: Don’t you think that works against the ease of use as well? I consider myself a mass user. I’m an editor. It would be a lot easier from my perspective if there were only one Linux distribution.
MC: If it took you a day to get it going with an expert…most people just don’t have time to do that.
LM: But couldn’t you have partnered with Red Hat or Caldera on this?
MC: We initially tried, but it’s tough to get somebody on the same urgency of agenda as you are yourself. We just found that it wasn’t working. Encouraging other people to do the work wasn’t as effective as simply doing it ourselves.
LM: It depends on the scope of the problem, I guess. To a certain extent, if you really felt a feature needed to be there, couldn’t you develop it, open source it, and partner with Red Hat in some way?
I wonder about this because with all the different balls you have in the air right now, it’s easy to ask the question: “Just what is Corel?” I think of you guys as an applications vendor but now you’re involved in the operating system. You’re looking at developer tools; you’ve done hardware as well.
MC: If you look at all of the companies in the Linux market, which company looks like the one who can make Linux a reality on the desktop? It’s got to be us.
The other ones are kind of small and start-ups with a few dozen people. We’ve got about 2,000 people. We’ve got nine years’ experience. Before Corel, people said that Linux could not be a desktop product. We took it into the desktop market. It wasn’t there before.
We’re now the fastest-growing distribution out there because there are more desktops out there than servers. So while we’re making Linux work on the desktop, we’re really also making it complete. As a server operating system alone, it would never get to be that popular.
LM: People talk about the Linux standard and whether Linux will go the way of Unix and fracture. What are you doing to prevent that from happening?
MC: We basically work with the standard committees. I tend to agree with Bob Young that there’s no economic merit in fracturing Linux.
The reason Unix fractured is because everyone was using different chips and had different hardware to sell. But that’s not really what’s driving Linux now. It’s the common interest that’s driving it. There is not really any economic incentive for it to fracture.
LM: Where would the applications tie in? What if somebody else has a distribution that’s using a different office suite and it’s based on GNOME for example? This KDE/GNOME thing seems to have some possibility to divide Linux.
MC: I think that KDE and GNOME will probably tend to blend. That’s one of the beauties of open source: You tend to start out with multiple options and then gradually, because of the fact that it is open source, they tend to go into the best of the best combinations. Who needs to have separate projects when they have the same functionality?
LM: But we do. We have Qt, we have GTK+, we have Motif….
MC: But the fact is that you get people choosing whatever they want to use and then gradually they tend to evolve into one.
LM: Hopefully. This should have happened with Unix, but didn’t.
MC: That’s why open source licensing is so good, though. That way it should tend to evolve naturally.
LM: As a distribution vendor, are there things that you think you can do to prevent that from happening?
MC: We’re fortunate in the way that Unix did fracture so nobody won. Microsoft won because it won by volume — a lesson everybody learned.
We also found that, even with IBM’s might with OS/2 and its billions and billions of dollars, it still didn’t work. So even IBM has realized that Linux is a good thing to adopt.
LM: That sort of gets back to my initial question. I was asking why Corel wants to be a distribution vendor.
MC: One very important turning point there was when we decided to put all of our enhancements back into open source too. We could’ve said, “Let’s keep the parts we’re doing ourselves as an edge.”
That would have been a bad sign because that would be almost fracturing the top end of the distribution. But the fact that we didn’t do this is good because, being a commercial vendor, we could still see that it was better for us to follow the open source model. That means that we’re not going to rely on that little lever to gain an edge. We’ll just have to rely on having the applications speak for themselves.
Look! we did it
LM: What is your relationship with other ISV’s? Can you go to a vendor like an Intuit and encourage them to do a port of Quicken?
MC: We’re encouraging everyone.
LM: What can you actually do to make that happen?
MC: Wine really helps a lot. We basically enroll a lot of people and say, “Just try Wine out. Try your application. It may run almost out of the box and you just have to tweak it a bit.”
LM: You’re saying, “Look! We did it.”
MC: That’s right. The Office Suite’s turning a lot of heads because it’s really good and it’s not a minus one or a minus two version. This is surprising to Linux users because they’re used to everything being immature, you could say, compared to what would be available on Windows.
LM: But you sell a Linux distribution. Look at Red Hat: They have ISV programs with branding and certification and they’re encouraging people to port their applications to Red Hat Linux. I don’t see that a lot with Corel. Are you following that strategy?
MC: We’re focusing more on the mass distribution markets. Where most of our customers will be computer manufacturers that pre-install Linux.
LM: So you don’t see the need to create the same kind of ISV programs?
MC: Not really, no. That’s the good thing about Linux. Each company does what it’s best at, and each one has a different focus.
LM: Let’s talk about the stock market. Corel’s stock soared to about $40 last fall, but it’s trading in the $7 range right now. What were you thinking when your stock was that high?
MC: “Great. We’re getting the recognition that we deserve for our Linux activity.” Because compared to other Linux companies, it’s still a lot. So if people are recognizing that we’re not just in Windows, then that’s good.
LM: So you didn’t think those valuations were crazy?
MC: I didn’t think ours was. I thought some of the others were in the sense that we had much more going on in terms of real activity than almost anybody else. So it was gratifying to get that kind of recognition, because we do have the end-of-the-line reality to support it. A lot of the start-ups and so on don’t have much behind them other than the name.
LM: Do you worry about competing with any open source office suites?
MC: Not really, because I think it’s a time-to-market versus completeness thing. StarOffice is good for 80 percent of what you want to do but the last 20 percent is not all the way there. Sun’s record of nurturing software isn’t exactly fantastic.
So we’re very happy to have a robust office suite out there that’s got a great reputation so people know what to expect and how it’s going to work. In terms of all the free stuff, it’s a lot of fun to play with and so on but it’s never going to catch up.
LM: Were you happy when you heard that Sun bought StarOffice?
LM: You could move up the chain and get to office suites and say if the chips can be commoditized and the operating system can be commoditized, why can’t the office suite be commoditized?
MC: It can. The fact of the matter is people are very busy getting the OS up to bat, and on certain levels the OS will begin to fill in more of the stuff we might be doing in office suites. So a future version might take advantage of that. But I think that because the price Corel is used to getting for our software [around $100 for WordPerfect] is low enough that there’s always going to be some deluxe functionality that’s worth having. Even as the base level maybe goes higher.
LM: Do you think there will be componentized parts of the office suites eventually?
MC: It’s definitely a moving target. We’re looking towards much more componentized software, XML type of structuring, and back-end Java components. So the software is definitely evolving in that direction.
LM: Are you thinking of open sourcing any applications right now?
MC: Not really. I’d say the sub-applications like the file manager, for example. Those are basically keeping us busy on the open source side. The other ones we’re really spending a lot of energy on to get them over to Linux as soon as possible. We have a lot of work on our plate.
Robert McMillan is executive editor for Linux Magazine. He can be reached at email@example.com.