ECM and the FLATNESSES model

So, a while ago I decided to assess current generation (ECM 1.0) Enteprise Content Management against Dion Hinchcliffes SLATES model of Enterprise 2.0, and Chris McGath of ThoughtFarmer seemed a bit shocked that ended up rating the current state of play quite highly :-)

So Chris decided to assess his product agains Dion's MK II model, with the acronym FLATNESSES; so I have to take up the challenge and look at the additional elements of this new Entperprise 2.0 model and how ECM systems fit in. This is Chris version of Dion's diagram, showing the additional elements which we want to examine:

I have added some comments on some of the elements originally assessed against SLATES too.

• Freeform: Well in the biggest sense of the word, an ECM 'platform' such as EMC Documentum, IBM FileNet or OpenText Livelink can be considered free form, as you can build whatever ECM enabled application you like on top of said platform, with the components and interfaces it supplies. However in the Enterprise 2.0 context 'freeform' appears to speak to Dion's idea of an egalitarian experience for users. In order for ECM 2.0 to be freeform in this context, it needs to support applications like blogs and wiki's where the content of pages is hosted in the ECM repository, manged by its WCM sub-system or provided with the benefits of federated services, such as records management etc. Personally I would also like to interpret 'freeform' as the ability to utilise AJAX, FLASH, Silverlight, PHP or whatever we want to quickly build simple, contextual interfaces that offer true browser based, user centric design.

• Links: dealt with under the SLATES posting, however moving forward the ability to link to any piece of content anywhere in the repository via URL / URI mechanisms needs to be linked to the increaslingly popular use of RDF / RSS technology for linking to other enterprise systems.

• Authorship: dealt with under the SLATES posting, but since then IBM have release their free 'Lotus Symphony' toolset, based on OpenOffice.Org components creating Open Documentum Format (ISO) standard XML files. So why not build in such tools to replace the simple inline text editor offered for quick content creation in many current ECMS' - well because it might damage relationships with MS by reducing licensing of their cash cow ? Well no one said you can't do the above and still provide Office connectors !

• Tagging: dealt with under the SLATES posting, it should be easy to such 'consumer web' oriented aspects such as 'tag clouds' if users find them an easy way to browse the metadata held in your ECMS

• Network Oriented: "Application content must be fully web oriented, addressable and reusable". Well we have already mentioned that many ECMS' have moved away from 'fat client' or even 'thin client' type software integration with the desktop towards web browser based interfaces, giving URL access to content objects. This move should continue, but care and attention needs to be applied to these interfaces, especially as they can be quite slow compared to MS Windows Explorer integrations provided by the old client install technology. Dion's diagram also mentions "small pieces" - I am presuming this means pieces which are easily consumable over the network. Well that depends on your business context and what content your producing, but actually we are looping back to authoring and XML; if you can do a good job of managing small 'chunks' of XML content as a core competence, then it should be easy to push that to a web browser, or transform it into a paginated PDF for third party printing, or to a talking eBook for your visually impaired customers etc.

• Social: so where does this fit into the ECM universe. It might be via links to other systems, for example if you leverage a seperate Directory Service to provide authentication for your ECMS, then it may be a case of surfacing useful information from your directory alongside the content in a particular context. It may be a case of leveraging audit trail info, i.e. these people have read this document, other people who read this document also accessed this document... etc. However social aspects are very definately contextual to every different organisation, so the lowest common denominator might be linking people to content by surfacing metadata already held, and doing it in a way thats useful to your users.

• Emergence: using Prof. J. Goldstein's definition of "the arising of novel and coherent structures, patterns and properties during the process of self-organization in complex systems." we might argue that ECMS technology could support emergence, but that acually ECM is usually (or should be) part of wider Information Management strategies, and as such should include well worked out structures such as metadata schema's, taxonomies, business classification schema's and folder structures. This suggests ECM strategies and emergence are to each other as matter and anti-matter. However it might not be that simple. For well structured information stores, the marriage of ECM and BI to spawn 'Content Intelligence' might find patterns in content creation or use. An interesting topic for further thought.

However might bottom line on this is the same as always, 'ECM' is a strategy not a product or suite of products - its about information management not about shiny new software :-)

Interesting times

So it all started with a quick session of blog reading last night, and a comment by my friend Laurence over on Word of Pie; he quite rightly, and very gently takes me to task for being a tad cynical in a previous posting. However, although I am very interested in most of the phenomena that currently have a 'two point oh tag' added to them, I truly hate the trend / fashion to make everything a ' 2.0 ' - so in many respects Laurence is absolutley right, the definition of ECM 2.0 is whatever we want it to be, and it can be delivered whenever the vendors and more importantly the customers, decide they are ready for it.

But... (and there is always a 'but' somewhere eh ?) its not quite that simple is it ? Nah, it never is......

So, I am attending CMF2007 in Aarhus next week to present a case study on the work we have undertaken so far on redeveloping our Intranet. Our CIO has always said that he see's the Intranet to be completely wound up with ECM, and I agree, its all simply different facets of 'information management', and as I always say in presentations on ECM, collaboration or whatever, its absolutely not about the 'tech' its about policy and procedures and most of all people, i.e having a policy on what works best presented on a static web page, a wiki, or as an MS Office 'document' in the ECMS.

So, notwithstanding the fact that the whole " Web 2.0 / Enterprise 2.0 " bubble could burst at any moment (follow the link to Steve Rubels take on this), lets examine what is out there, that we can add to our ECMS to make it ECM 2.0 - and I mean stuff available now, not in even the near future.

For a start John McKendrick over at FASTforward asks if 'web 2.0' is the business persons revenge for SOA ? Because SOA projects have become so large and complex, is the emergent goodness of Web 2.0 tools the only way business people can actually get their hands on useful collaborative tools ?

Well, to progress, let me give you three links to postings from Dion Hinchcliffe at ZDNet on Enterprise 2.0, in order:

• The State of Entperprise 2.0
  

• SaaS and Office 2.0 evolving towards Enterprise 2.0
  

• Significant workplace inroads for Entperise 2.0
  

The first posting gives a good review of Andrew McAfee's SLATES acronym. Lets see where ECM systems fit in 'SLATES' shall we ?

S = Search; one of the main arguements often given in the business case for ECM is the improvements in findability brought by having your content in a system which can manage metadata and provide metadata and full text indexing search - so 1 for ECM ?

L = Links; the deep linking between information using URI's - well a lot, of not most (or even all) ECMS now use persistent URI / URL to link to all content items - so thats 2 for ECM.

A = Authoring; ensuring all users have access to easy authoring tools - well not necessarily a good score for ECMS' on this one, even with built in MS Office connectors etc, or the ability to write small pieces of content using inline editing tools in the browser, when you add the burden of user created metadata etc then we might be slightly failing on this one - so ECM score now at 2.5 ?

T = Tags; or Metadata in other words. The more users get used to web 2.0 consumer systems, or even tagging their MP3 collections, the easier it is for them to understand the use of metadata in the enterprise. So, ease of use might not be overwhelmingly great, but ECMS have been doing 'tagging' for a long time, so ECM 3.5 for 4 ?

E = Extensions; extend knowledge by mining patterns and user activity. Well in my opinion any good ECMS should let you mine user activity data and report on whats going on in the system, even provide hooks for MI reporting or BI tools. So, 4.5 ?

S = Signals; make info consumption efficient by 'push', RSS feeds etc Well some ECMS are better at this than others, but even if they revert to the 'old fashioned' technology of email for their notifications and alerts, again a standard feature of ECMS' is the ability to subscribe to content and be informed if the content is changed updated etc, so final score could be 5.5 out of 6 ?

So what do you think ? Do current generation ECMS at meet the SLATES test of Web 2.0 goodness, within the context of enterprise information management ?

If you want, go to the second of Dion's postings, and do a your own checklist against the longer acronym of FLATNESSES - there is overlap with SLATES, but where the ECMS probably falls over is on F for 'Free form' and E for 'Emergent' and the S for 'Social' but then that all depends on how strident you are at applying Prof. McAfee's criteria. For example Documentum's eRooms is a collaborative system which part of EMC's ECMS (I could not resist that concatenation of our favourite 4 letters), and you can absolutely build social networking and social interactions in an eRoom, even if its not exactly Facebook.

The final of the 3 posting's in examining the take up of tools in the enterprise addresses the cultural element, and this should not be underestimated. The majority of the core 4,500 staff of my employer would probably be characterised as 'knowledge workers' under most definitions of the term, and yet I.T literacy is (IMHO) quite low. So getting alerts and notifications from the ECMS via email may add even more crud to your inbox, but are your staff ready for a mass migration to RSS feeds and feed readers ?

Back to a tenous link between this and my attendance at CMF2007 to talk about our intranet. We have been examining portal software technology, and have seen's demo's from big players like IBM and BEA. Over at the BEA en.terpri.se blog there is a posting on "why enterprise 2.0 matters, its about how people interact" and to me this brings us back round, full circle. It does not matter if your talking about an enterprise portal, an eRoom for a discrete group, the blogs used as 'news' sites by project teams, a wiki used by developers to document their software, in the end it should all be easy for the average end user to get to grips with, and should facilitate content centric collaboration and improved findability, and that call is taken up by Dan keldsen with this posting, where he asks the question 'who puts the C in ECM'.

So, to complete this long posting, and go round full circle, even if my cynism (or British pessimism) is sometimes misplaced, to truly get all of what is discussed above into a truly 'enteprise' wide content management, probably requires the SOA / standards based approach as espoused by Laurence. But even if the technology was perfect, your people might not be, so stop to consider how you get them to use this stuff ! (see comments on this at the Tower Software blog).