Alan Pelz-Sharpe always has a way of getting me thinking. His latest post, ECM Coexistence and the Vuvuzela, remarks how vendors and customers alike are looking to integrate new content systems with legacy content systems rather than replacing them. Connectors and API’s and standards are all the rage. What I wonder though is why the change and why now. With apologies to Sir Isaac I propose the principal reason is Newton’s First Law of Content.
Content At Rest
A liberally paraphrased law as it applies here might read “All content will persist in a state of rest (the same system) or of uniform motion (the same process) unless it is compelled to change.” What this tells us is that a content management system remains online forever unless there is some great compelling need to turn them off. For the better part of all of our careers “change” in the IT landscape has been compelled by a few primary drivers.
- The need to serve – The system must adapt to external initiated changes in operational and regulatory requirements.
- The need to scale – The system must be sustainable and expandable without degradation of the ability to meet the original requirements over time even when those functional requirements don’t change.
- The need to save – The cost to maintain a system cannot exceed the margin generated by the business processes the system supports.
In years past the change pressures from these three drivers tended to create scenarios where the case could be made for justifying new systems based on new needs but paid for on ROI from the retirement of the old. This is based on the assumption that the cost of support threatened the balance of the business model.
Despite the fact that so many systems were cost justified on retirements, reality slaps planning in the face and they rarely happen. The cost, complexity and general lack of interest in doing the mundane work of decommissioning combine to move these projects ever lower on the priority list. It requires an organizational will or some other greater force of compliance or cost to make it happen.
Interia at Work
Advances in sustaining technology like virtualization, low-cost outsourcing of support and creative status reporting combine to maintain these multiple systems until the a greater force drives consolidation and decommissioning. Consequently the market has responded by seeking to enable heterogenous access. Aging systems in production longer is a force compelling the market to create more access to multiple systems but it does nothing to move the content from where it was to where it needs to go.
The dynamics of leaving older systems around longer affect many things not the least of which being how project budgeting is managed. Removing retirement as a potential source of ROI to justify the new systems forces the owners to prove the value for the new system on its own functional merits.
This is a good thing because it can lead to a more rational evaluation of the real value. The resulting smaller baseline budgets then force projects out of the capital expense categories into operational expense lines. That then promotes the use of SaaS or other cloud models because of the lower cost of entry.
The Greater Force
As the number of older systems increases a tipping point is inevitably reached and approaches to retirement must evolve. In my day job I recently observed the use of xDB for multi-system application decommissioning. The method extracts the data, normalizes it to XML (which is more “time tolerant”) and then creates dynamic searchable archives that provide access to the data until it finally ages out. Very cool use of the technology.
Also the EMC SourceOne archiving line continues to expand with the addition of file archiving to the SharePoint and Email offerings. Both of these examples are ultimately responses to aging data and applications. As Alan suggests we have to support heterogenous access to multiple products but data and system retirement must be a part of the overall picture.
I would argue that for many large organizations the need for decommissioning capabilities is now large enough to warrant systems,projects,budgets and practices their own. These application archives need ultimately to be independent of the array of systems generating the data they store. As the market moves forward the ROI for retirement shifts from new system creation and is aggregated into broader retirement and archiving initiatives.
Alan is correct in the observation that the myth of the monolithic repository is just that. An unattainable ideal. Reality however is not an ever-expanding collection of different products within the same customer either. At some point there must be consolidation to a manageable set of tools that is combined with a long-term content and application archival strategy to maintain the balance between service,scale and savings.