Site icon Big Men On Content

ECM 2.0 from EMC – I’ll wait for SP2

I was reading over the Open Text announcement of their Enterprise 2.0 strategy. I am reminded of the old line. “I’ve started working on my second million. The first didn’t turn out so well.” When exactly was 1.0 finished and just what is so different that it warrants incrementing the major version?

The announcement highlights a number of key points but the central idea for us is in the middle of the list

I first heard EMC talking up Enterprise 2.0 at EMC World last year but I haven’t seen as succinct a summary of their strategy as we find from Open Text.  When EMC talks 2.0, it’s much more about platform and frameworks than specific end user capability.  All this new world dawning talk is nice but If you already have Documentum in house, is this something to consider.

Word of Pie , Billy Cripe and Jed Cawthorne have all chimed in on the topic of ECM 2.0 in recent months and there is general consensus that the social app features most often associated with the 2.0 label are in fact natural extensions to the ECM tool set. Forums and Wiki’s and Blogs (Oh My) are just not that remarkable. RSS has been around for years now and there is nothing fundamentally unique about streaming video except the that corporate networks have finally caught up and can handle the load.

There is one 2.0 quality that I believe falls outside of “natural evolution” and is not mentioned. SOA. Service Oriented Architectures are a fundamental design change that some have struggled to get right. The D6 Documentum Foundation Services is EMC’s second attempt and in general I think they got it right. The challenge though is retooling the rest of the literally hundreds of products to take advantage of it. There are so many legacy product lines just within Documentum to deal with that it strikes me this problem cannot be unique to EMC.

A number of years ago I met with some of the leadership at Documentum and they introduced me to the then new concept of the Java dmcl. It took several years for it to become a reality. It’s important to note that this was not a reimplementation of the content server, just a new client library. DFS alters client access yet again and is a far more fundamental change than just swapping out a library. The change is so dramatic that it suggests starting over for most existing applications if you want to take advantage of it at all. I raised the concern then, how long will it take before java dmcl is as stable as the old client libraries? The same question applies here.  When will my job be safe if I build this app on DFS?

For those responsible for upgrading existing systems, 2.0 architectures present a few key challenges from people building new. The most daunting challenge is how to pay for it. The monumental ROI we claimed years back was exhausted in the first implementation and keeping up with supported versions has little or no return to offset the colossal cost of retooling. Despite the rhetoric, Web 2.0 functionality does not necessarily generate ROI when replacing a process that is already electronic.

More and more I am seeing implementations where it is cheaper and better in the long term to start over than upgrade in place. With that expense in mind, ECM 2.0 often does not bring enough to the table on it’s own to justify the expense of replacement. In fact, you should not force the features on the application unless there is sound business justification to do so. Vendors as well need to settle on where and how they will support 2.0 features – as core components, bolt-ons or separate stacks. Despite EMC’s platform position I do not believe this rationalization is complete. So, regardless of the vendor, with ECM 2.0 it might make more sense to wait for SP2.

Exit mobile version