Interface: Transforming Information Management with Enterprise Mashups
Can you feel the thread fraying? All around us, the fabric of doing business is unraveling, then slowly being rewoven in new ways, with outsourcing, offshoring, repurposing and restoring. Everything's going mobile. And everyone is either working harder or trying harder to find work. Jobs are scarce, yet data is more abundant than ever... which means there's plenty of work to be done.
It's the classic conundrum, a scenario that forces clichés into action, because necessity really is the mother of invention. "If something must happen within an organization, it's happening," mused a savvy consultant some years ago. To wit: A historic transformation is happening right now. The tumblers are realigning at this very moment. The wire frame of information management is being redrawn.
When the dust settles, mashups will emerge as the new strategic view of enterprise data. There are at least three reasons for this: 1) the Web will increasingly dominate commercial activity; 2) agility will be the hallmark of successful companies over the next five years as the world recovers from still-reverberating financial shockwaves; and 3) interoperability will be paramount.
Mashup technology was designed with the Internet in mind. A mashup combines data or functionality from two or more external sources to create a new service. The term mashup implies easy, fast integration, frequently using open application programming interfaces and data sources to produce results. So, mashups live and breathe online and have implications much larger than just mashing geospatial data.
Mashups enable enterprise agility by decoupling the painful side of analytical application design from its end goal of delivering value to decision-makers. When designed properly, mashups allow a fluid mixing and matching of data sets that traditional business intelligence solutions would take much longer to produce. The result is value generated within days or hours instead of months or weeks.
Perhaps the ultimate benefit of mashup technology is interoperability. A mashup's goal is to create a simple means to easily combine a variety of data sets quickly, which is why open APIs and data sources are fundamental. This interoperability facilitates the sharing of data sets between organizations and their channel partners, customers and even prospects.
If the interoperability argument sounds familiar, that's because it is. For more than 20 years, many consulting firms and even software vendors have espoused the value of service-oriented architectures and composite applications. The line of reasoning is that fine-grained services can be combined quickly and easily to create coarse-grained, business-focused functionality.
Think of mashups as data visualization-focused extensions of this model. In fact, many consultants years ago questioned how data would be handled within an SOA, noting that special attention must be paid to its treatment and the role of metadata. In essence, the movement toward mashups represents the proverbial other shoe dropping: With mashup technology, the data now has its vehicle for delivery.
Still, many challenges abound. Browsers continue to crop up and change without notice, frustrating information consumers and managers alike. In late 2009, a consortium of companies collaborated to form the Open Mashup Alliance, which is promoting adoption of the Extensible Mashup Markup Language, or EMML, an XML markup language for creating enterprise mashups.
The OMA boasts several major backers and numerous startups. The list of companies currently associated with OMA includes: Adobe, Bank of America, Capgemini, Convertigo, DreamFace Interactive, Hinchcliffe & Company, HP, Intel, JackBe, Kapow Technologies, ProgrammableWeb, Synteractive, Xignite and Zinnia Systems. The official OMA Web site (http://openmashup.org) quotes industry analyst and consultant Dion Hinchcliffe, principal of Hinchcliffe & Company: "I've personally examined EMML and can attest that it's a clean, powerful design that includes potent capabilities such as declarative data transformation, advanced procedural logic, parallelism, metadata and much more."
According to OMA, a key value proposition of EMML is that as a domain-specific language, it can theoretically eliminate the need for complex, time-consuming and repeatable procedural programming logic to create enterprise mashups. As with all proposed standards, the proof will be in the usage. If OMA manages to foster a critical mass of adopters, then EMML will thrive.
But adoption of standards is not assured. Software companies tend to gravitate away from standards that are too transparent, simply because the closed-source software model has done so well for so long. Powerhouse players like Microsoft, IBM, Oracle and SAP don't make money by giving away their code, and with mobile application development on the rise, even HTML could be in trouble.
Then again, some ideas are too powerful to fade. Renowned philosopher Victor Hugo noted: "One resists the invasion of armies; one does not resist the invasion of ideas." Here's hoping that the Open Mashup Alliance will help keep the Internet transparent, agile and free.