
"We're not going to put our heads in the sand. We could never get the market reach that Eclipse can. But we have the ability to be one of the premier engines for Eclipse," Douglas said. "Certain aspects of developer tools will and should be free. Other aspects people will continue to pay for if they add demonstrable value."
CodeGear is also adapting by targeting emerging markets, such as the PHP and Ruby on Rails development worlds. But expanding the product line carries risks. Developers still haven't forgiven Borland for releasing and then discontinuing Kylix, a Linux version of its Delphi and C++ IDEs. Ben Smith, CodeGear's interim CEO for several months before Douglas' appointment, said CodeGear couldn't find a viable business model for Kylix.
Swindell said CodeGear won't repeat its past mistakes. "Kylix was a very successful product, just not at the same scale as Delphi," he said. "One of the things we learned as a business is that you have to develop these new markets very agilely. Don't take two years to develop a product; get it out to market and get customers on it," he explained. "Kylix was a very big, expensive technology. For us to do that kind of thing in the future, it needs to be much more agile and lightweight. When we build something like the Ruby on Rails IDE, it's a smaller team."
Douglas is bullish on the Ruby market and thinks it's ready to support a commercial IDE. "We think the big enabler is Rails. We think it's ready for prime time now," he said. "We see Ruby on Rails as an emerging platform as both a Web technology and a potential enterprise technology."
CodeGear's long-term future is still somewhat up in the air. Though Borland retained the group, it did so largely because it couldn't get the price it wanted from private equity groups interested in backing an independent CodeGear.
The rest of Borland is moving away from CodeGear, figuratively and literally: The application life-cycle management company is moving its headquarters to Austin, Texas, while CodeGear remains in Scotts Valley, Calif. CodeGear's executives are carefully noncommittal when asked about how long they expect to say in the Borland fold.
"Operationally, we're very different than Borland," Douglas said.
One way CodeGear is different: It's profitable. Borland's most recent quarterly report, for the three months ended March 31, broke out CodeGear's operations for the first time. Borland as a whole had a loss of $9.2 million on revenue of $71 million, yet CodeGear reported operating income of $1.9 million on sales of $14 million. Since CodeGear has only begun shipping its new products, its early profitability bodes well for the overall health of the business.
"As soon as Borland announced the spinoff [was] complete, we started receiving orders again. In fact, our component sales revenue is approaching what it was before the first spin-off announcement," Arcana's Southwell said. "That indicates to me that customer confidence in CodeGear and their mission is as high as ever."
