Monday, August 20, 2007
JSR-170: If at first you don't succeed...
For a proposed standard to be successful, it needs the right mix of good design, timeliness, vendor backing, and simple luck. JSR-170 has a good chance at meeting all those requirements, which is why Xythos has been building a JSR-170 layer above the existing API. In 6.0, we finished the first phase, the read-only interfaces ("level 1," in JSR-170 parlance). We'll soon be finishing level 2, the complete read/write interface.
JSR-170 has a lot more momentum than its predecessors, such as ODMA, the now-obsolete effort to specify a document management API standard. ODMA came too early, pushing a standard when organizations viewed DM as a niche function. To make matters worse, IT departments successfully implemented DM systems only about half the time. The other half were, of course, highly expensive failures.
JSR-170 has emerged at just the right time. While not everyone knows the term "basic content services" (BCS), most people in the workforce today understand the need to manage documents better. On the stick side of the incentive question, IP protection, corporate risk, and compliance have enlarged the circle of people who need to think about their documents more carefully. Increases in productivity, the rich variety of web-based applications, and the Web 2.0 generation of collaborative applications have expanded the number of people who want to manage their documents.
Just as importantly, people in the technology biz, from software vendors to IT departments, have learned how and why to implement standards. For example, the SOA movement couldn't move anywhere, if there weren't standard interfaces among the different components in the larger
JSR-170 also benefits from a trail already blazed by another content management-ish standard, WebDAV. At some point, enough applications added WebDAV support to give the standard the critical mass of acceptance. Not just Apache, but Dreamweaver, had a lot to do with WebDAV's success.
Is JSR-170 a better standard than ODMA? A comparison of the two APIs might be illuminating, but it won't necessarily tell you why ODMA failed, and JSR-170 has a good chance of success. The best-designed standard has no chance of success, if the timing is wrong.
P.S. If you're interested in being a beta customer for our level two JSR-170 implementation, drop us a line at support-beta@xythos.com.