16 February 2011

Back to my roots

Lately I've dived back into document control, content management and intranets and data ware-housing in general, a swarm of concepts and technologies I haven't played with in a while. Taking these breaks from certain areas is a good thing; it allows me to return to it at a later stage and witness how they've changed, often for the better.

But my return to this field hasn't been the happy smooth ride I was hoping for. Put briefly, the technologies and intellectual concepts within seems to me to still, uh, suck. I'm constantly living on the edge between people and policies in various organisations, and there's nothing more frustrating(1) than seeing the divide between good intentions and good actions, ever growing despite that people should know better.

The first thing I notice is that the concept of persistent identification management is no further than when I left it. Documents are still named silly things, and identifiers in systems that tries to deal with it still apply internal identifier schemes, and have no external facility for managing them, either sharing nor changing them. It seems to me that most of the technologies in the DC / CMS / KM space still has no good solutions to perhaps the problem that sit at the core of each of them.

Some user interfaces have improved, but not in the way I was hoping. Sure, things look smoother and prettier, and surely the misnomer "web 2.0" have injected it healing juices into old battered ideas, but they're still bad ideas at the core no matter what. It doesn't matter that the buttons look good or are big and shiny if the functionality behind them are fundamentally flawed. A document without external identity is still an internal abomination, even when the buttons are easy to hit.

Workflows are still sequential stop-points along a list of basic logic that only triggers on what we might think of, never of what should happen. The ontological concepts are often completely missed, if present at all. Roles and permissions are all ACL where, perhaps, human organisations don't function that way. Maybe that ACL model fits your network, but does it fit how we humans deal with information and knowledge? We humans might have permissions and access through it, but they do not cover our roles and positions in compound situations. I would have thought this was done better these days.

Wiki's have come a long way. I'm excited about using Semantic MediaWiki with Semantic Forms and a few other plugins, for doing essential super-rapid application development, but the Wiki user-interface have not improved much. The concept of every page simply being a point to which to hang knowledge is still mostly if not entirely missed, even by Wiki makers. People still think a Wiki page is just a page, when it should be a point of data to hang our other bits on to, including, ahem, identity management.

And don't get me started on the various and numerous CMS out there. There's now simply too many of them, and they all do pretty much the same with minor differences. It's almost a bit embarrassing that the field haven't progressed beyond the basic model they pretty much all offer. Even Drupal which claim to be different still, to me, feels and works, well, to a poor model.

So let's talk about the models. They're still the same. They haven't moved. It's all still based around the concepts of content, structure, users and plugins, plopped into a relational model that have to have a specialized table for any ontologically interesting concept you want to introduce. I don't understand, are most of these systems being developed with bribed from the DBA association or something?

Where are the semantic options? Where's the ontology editors? Where's the conceptual modeling tools for content and knowledge management? Where's the persistent identification schemes implemented so we can actually do, you know, integration of systems? Where's the simple data mergers that allow distributed and / or federated reporting?

None of these things are tricky in and of themselves, not even when merged into larger systems, unless, of course, they're rooted in old and stale ideas. Where's the new ideas? The bright and new ways of solving some of these old problems? What have I missed?