There is a nice multi-part series on mashups by IBM. They call them Situational Applications or SAs. The series, written by IBMers, Andy J. F. Bravery, Luba Cherbakov, and
Aroop Pandya covers the “applicability of Web-based situational applications, what to the enterprise, their relationship to SOA, and how you can use them to improve the current state of corporate IT.” Here is a summary of Part One: Changing computing in the enterprise.

The name situational applications comes from Clay Shirky’s essay titled “Situated Software” that describes software “designed for use by a specific social group, rather than for a generic set of ‘users.’” Clay argues that “most software built for large numbers of users or designed to last indefinitely fails at both goals anyway.” This is an excellent point. I have seen many times where software to support a particular work process within a particular company culture was dragged around by numerous consulting companies hoping to leverage the initial success into a practice. It never seems to work the same way.

However, while mashups are often developed as situational applications, some go beyond the situation and others do not. I would prefer that they use the more common term, mashups, that describes the somewhat unique process, than use the term, situational, that can be applied to many other software development efforts. But this is a minor point, you can do the translation when you read the articles and I will call them mashups in these two posts.

There is a nice summary of the factors that led to the rise of mashups. They compare mashups and SOA, noting the opposing views on their relationship and take a middle ground. “Our observations show that while SOA and SA have contrasting development life cycles and motivations, different usage patterns, and even dissimilar enabling technologies, they also have important aspects in common: separation and exposure of legacy applications’ functionality into reusable services, software as a service, and composition or assembly of a new solution from distributed reusable and remixable parts.”

Contrasting mashups and SOA, they cover development life cycles, usage patterns, and the social aspects. There is nice summary of the business value and the improvements to business solutions are worth listing: “creating applications that are better fit to LOB problems, satisfying short-term knowledge workers’ needs, addressing niche market (The Long Tail) requirements, combining tactical SA-based solutions with the overall IT portfolio, focusing on the end user and creating a rich user experience, enriching IT data portfolio with unique, hard-to-recreate data that’s created by individuals and small teams; they get richer and “cleansed” as more people use them.” They go to cover the OI aspects and adoption challenges. I found this s useful summary of the general mashup situation. My next post reviews what they said about mashups within IBM. Thanks to Tomoaki Sawada for pointing it out to me.