Borland [BORL] has, historically, been very closely identified with developers, having originated Rapid Application Development (RAD) for Windows and carved out a healthy market share in Java 2 Enterprise Edition (J2EE) with JBuilder. During the last two years, Borland has raised that debate by speaking about application lifecycle management (ALM) – combining requirements gathering, modeling and testing across development teams.
SDO is a higher level sell, founded on the principle that today’s software development processes are not working. Too many projects fail or fail to meet business requirements because the business managers and decision makers who specify or co-ordinate projects lack visibility into what’s going on inside projects.
Inadequate insight
Borland chief executive Dale Fuller told ComputerWire the reason for project failures is business people don’t have adequate insight into projects, yet make changes. When projects start running late or going over budget as a result of changes, the answer is to de-feature and software fails to meet business needs. SDO is designed to avoid this.
SDO is a roadmap to create real competitive advantage from software by aligning people, processes and technology together – that’s aligning things so that when business makes a change they can see how that change will impact the project – how they will impact time to delivery, quality and costs, Mr Fuller said.
Senior vice president and chief marketing officer Rick Jackson said while developers will remain influencers, Borland must add more classic enterprise class messaging and marketing to SDO. Who cares you have the right priorities? That’s going higher up [in the organization] – the VP of application development, the CIO.
For Mr Jackson, it could be seen as deja vu. Until recently senior vice president for worldwide marketing at BEA Systems [BEAS], he helped take BEA from a company selling the WebLogic application server to developers to an enterprise-class business pitch. That’s why I joined Borland, to take the marketing to that level, Mr Jackson said. He added Borland has been transforming its sales force to call on individuals higher-up in enterprise class customers.
Visibility and control essential
Driving SDO, according to Borland, are the pressures of increasingly distributed application development through outsourcing, the need to come to market quicker than ever before, reduced manpower and fewer financial resources, competing demands and the need to manage complex IT environments. All of these reduce the opportunity for cost or time overrun on a software development project.
SDO attempts to rectify the questions of visibility and control. Two of three technology concepts, named Hyperion and Prometheus, will provide a portal-based dashboard into projects and individual developers’ roles. An architect, logging into a project, for example will see all the tools, code, bugs and tasks they need to do their job. Underpinning both is the third concept, Themis, where elements of Borland’s ALM products will be boxed according to individual’s job requirements.
Borland, though, faces growing competitive pressure from the industry’s polar opposite software architecture leaders – Microsoft [MSFT] and IBM [IBM]. Borland, like many in ALM, has been compelled partly by the competition to move up the vision stack.
Microsoft on board
Microsoft is coming into the market in the first half of next year with Visual Studio 2005 Team System (VSTS). VSTS will, slowly, drive the value out of providing ALM point products, like unified modeling language tools such as Borland’s Together, for example. Microsoft will make its own Whitehorse modeling platform available to a wider developer audience, going beyond the traditional UML audience courted by Borland and even Borland’s ALM competitor IBM Rational.
At Borland’s BorCon conference in San Jose, California, this week, Microsoft’s VSTS general manager Rick LaPlante expressed solidarity with Borland and supported SDO, while Fuller supported Microsoft for building their strategy around ALM.
Borland’s ALM rival IBM Rational, meanwhile, is due to ship an update of its suite, codenamed Atlantic, by the end of 2004. Details on Atlantic are scarce, but it will be based on the open source Eclipse 3.0 framework. Free software such as tools based on Eclipse, have presented Borland with a business challenge that is pushing Borland to seek new ways to add value to the customer.
Borland, though, says it has architectural superiority over Eclipse and Rational. Senior vice president of software products Boz Elloy claimed Rational’s heavy reliance on the rational unified process (RUP) produced an inflexible methodology that required IBM Global Services to install.
Every CIO and customer is baffled [by RUP], Mr Elloy said. RUP as a methodology is important but it’s not enforceable, is flawed and not flexible.
Mr Fuller, meanwhile, believes the answer to Eclipse-based tooling is to add greater value through innovation and integration. You are going to see more and more of the [Borland] products having feature sets of the others in them, Mr Fuller said.
You are always going to have free – that’s good for people with lots of time on their hands and money… but if you are going to start to transitioning into making money, you want productive – you don’t have time [to integrate].
