Building Borland back from the ground up

Scotts Valley, Calif.-based Borland Software Corp. has had a varied and colourful history in development circles.

From the introduction of its Turbo Pascal compiler in 1983, to its early contribution to the CORBA community, to its name change to Inprise and back again, the company has seen financial woes and has bounced back again. At a recent eBusiness Conference held in New York, Gail Balfour, ComputerWorld Canada assistant editor, had the opportunity to speak with Ted Shelton, Borland’s senior vice-president of business development, about the changes the company has endured, how it has fared so far, and what lies ahead.

CWC: Describe the impact that the Internet has had on the application server market, both on the individual developer and on enterprise development.

Shelton: The interesting thing we’ve seen with the Internet is the way in which those two markets have started to come together. Because as people are building applications for the Internet, the individual developer is actually facing problems for the first time that really only enterprise developers have had to face (in the past): having applications that are dealing with many, many simultaneous users – very large scale, distributed and having to be managed over the entire world as an operating space.

And actually in some ways, the problems of Internet applications are larger than the problems in the enterprise because in the enterprise I at least first of all know who my audience is – I know how many employees there are, I know when they use the system. The other thing is when I build an application that just runs inside my corporate firewall with only my employees using it, if it doesn’t work the only bad thing that happens is employees get mad at me.

CWC: How has this changed the role of development within the enterprise space?

Shelton: The interesting thing happening to enterprise managers is that now they are being asked to build applications that are going to be seen outside the firewall. And so they start having the kind of problems that were often seen with the individual developers where they don’t know how many users are going to be hitting the application, they don’t know when the peak times are going to be. And when it doesn’t work, instead of having somebody be mad at you that’s just another employee, you actually have your shareholders mad at you. So it’s a completely different problem that the IT manager has.

So for us, both those sets of customers start coming together and having the same set of problems. And we’re in a position to address both those customer bases because of the way we are bringing all of our product lines together. Now we are able to bring together both the individual developer and the corporate developer and manage these large-scale development projects where you have a lot of people in different locations who all have to work together.

CWC: Does this tie into Borland’s new “DSP” initiative?

Shelton: Exactly. That stands for development service provider – we’re complementing the idea of an application service provider and management service provider.

CWC: Has the Internet and the open-source movement created more opportunities for developers – do they have more freedom now than ever before?

Shelton: I think there have been a whole series of things that have happened that have helped the developer in that way. One of the things that we were actually the initiators in was component libraries, so that as a developer you would have a whole set of pre-built tools that you’d be able to add to your application. Our Delphi product was one of the first products to build a component architecture around the application. And we have today, still, one of the largest third-party communities of writers of components for application development.

The second step was the whole open-source movement. So now you’ve got not only the idea of “Gee, I can build my applications by pulling all these components together” but now I can even see the source code of these components because they’ve been built in a community where everybody’s giving back to that community. So the Internet is an enabling medium for allowing people to exchange those components and allowing them to participate in open-source development, but it really wasn’t the Internet itself that created the change. The change was already underway in the industry, and the Internet certainly accelerated everything that was done in that area.

CWC: How does the name change from Inprise back to Borland fit into the company’s overall strategy?

Shelton: There was a process that started in the middle of 1999 in which the company was looking for a partner that it would be complementary to. And of course there was the announcement last January (2000) around Corel. Subsequently that was re-evaluated as a bad proposition for both companies, so we broke off that decision. And in the course of going through the thinking about why were we looking for a partner to merge with, the board of directors made a decision that they really wanted to reinvest in building this company as a stand-alone business.

So when we started in May, one of the things we talked about was the Inprise name – we’d never gotten traction with it. Everybody knows Borland – we (decided we) should change it back to Borland. Companies that are for sale don’t go and change their names. But this company had four quarters of losing money and had not turned the corner on profitability, even though it was doing a good job of getting costs under control. And what we decided was that our shareholders, upon learning that we wanted to go and spend money on changing our name, would much prefer to hear that after we had shown that we could get back to profitability. So we said “Let’s get a couple of quarters of profitability under our belt here as we turn this company around, then let’s change the name.” And that’s exactly what we did. It’s a great relief that we are back to the Borland name.

CWC: To what do you attribute Borland’s recent return to profitability?

Shelton: I have to say that there has been nominal revenue growth, but a significant decline in expenses. And one of the things that we have done very aggressively is we’ve rationalized the business and reorganized the business into specific product units. Equally important, however, is revenue growth. But the thing that you don’t see when you just look at the numbers is that we’ve been keeping to a level of revenue – or slightly growing in revenue – during a time when we had no new product releases. So in fact, what that shows is that we’ve been bringing in new customers for our products – not simply bringing a revised product release to the same customers. That’s going to be very, very important in the next couple of quarters now that we are releasing new versions of our products.

Our Delphi and C++Builder product group has represented, in the past, as much as 50 per cent of the revenue of this company. We had no new product release during the year 2000 of our Delphi product, because all the energy of the development team has been focused on building a product for the Linux market which we call Kylix (see facing page). So all those developers were spending all their time rearchitecting, from the ground up, the Delphi environment to be cross-platform. So, with no new product releases, that means all of our existing customers didn’t buy anything from us in that category during the course of last year. So to be able to keep the revenue at the same level was actually a terrific testament to how we are growing our customer base.