the company released Mylyn M4, an open implementation of its model for SLI.
“It’s time we stopped looking at ALM [application life-cycle management] as something we get from one vendor, and look at it as an architecture,” said Mik Kersten, CEO of Tasktop.
There are macro forces at work—notably the cloud and mobility—pushing the pendulum back to being able to mix and match tools for specific needs. Requirements are different, development is different and testing is different, so a traditional ALM suite might not meet those needs.
Best-of-breed is no longer a choice; it’s a reality, according to Kersten, but he sees the No. 1 problem with best-of-breed for ALM as a lack of integration. RESTful APIs offer “so much hope, and a bunch of confusion. There’s a new wave of data integration with a new value chain attached to that. We’re creating an API economy.”
With a software integration architecture, developers could use Atlassian tools on demand via APIs, for example, or TFS and Git. In fact, this already is happening in the industry. But Kersten said a new data model is required “if we’re going to hook up the life cycle.”
His main point: “The social flow of information has to cross silos that weren’t meant to talk to each other.”
To that end, Tasktop is defining what it calls a collaborative life-cycle bus, something that is highly scalable and that allows comments to flow from ops to dev to the business side, connecting them with traceability and offering large-scale agility.
With that, as demonstrated in Mylyn M4, are patterns and taxonomies for integrating back-end systems and outside partners for a lean software life cycle. The open-source project is a community-facing effort that will put out intellectual property for users to use, review, comment on and get involved in developing out further. Kersten said Tasktop is moving the project to OASIS because, he said, Tasktop believes “it’s the core footprint needed for SLI.”
So it seems the APIs and tools are in place for the kind of loose coupling Kersten described. It’s the commonality of architecture and taxonomy, along with this collaborative bus, that he hopes will drive this effort toward best-of-breed tooling even further.
David Rubinstein is editor-in-chief of SD Times.