Thanks for commenting and for providing references to your definitions and your services, both of which provide important context for the thoughts in my post. We seem to be on similar wavelengths.

In general, I believe, the effectiveness of external (system) support for any activity depends on the model of the system providing an accurate match with the model of the activity. This is inherently “obvious” and, when described in detail, is based on enabling understanding, change and reuse.

I think that your SRM definition captures many aspects of this area, and I see that you have considerable experience in applying it in your XeeSM system. I wish you well with that system and would like to learn more about it. By the way, I’ve also applied to join your (sub)group on LinkedIn.

The emerging social media/networking services are demonstrating new and more complex models. Despite, or perhaps because of, their volatility, these models will almost certainly provide better matches with real-world models of many kinds of relationships, than have previously been available. This is likely to lead to their longer-term application for more robust and professional purposes, whether or not the current services achieve that transition.

While the detailed operation and management of many areas of required behaviour tend to iterate relatively smoothly towards useful models, the thinking behind my comments on future directions is that the same cannot always be said of the architectural models. Silos of behaviour tend to grow up to serve emerging needs, but then to sink under their own weight for lack of sufficiently firm foundation layers that provide the necessary peer-to-peer protocols. Whether the necessary underpinning can arrive in time to support the current early infrastructures, and the extent to which rebuilding occurs from the bottom up, remains to be seen.

An interesting discussion in interesting times! What do you think?

John

It should be undone.

Thanks for identifying that implementation of relationships. I’d be interested to hear more about how flexible the Freebase construct is.

At present, in most systems, fundamental relationships are available in a very limited number of fixed types, as was the case for variable types in old fashioned programming languages. Fabrications of more sophisticated relationships usually end up with the implementation fragmented among the classes involved in the relationship. Surely we can do better!

It will be interesting to see how comprehensively it is feasible for relationships to represented.

John W Lewis

Very few people have yet understood the trends in relationships which transcend traditional corporate and individual boundaries.

However, in the past, such significant shifts have proved insurmountable by organisations founded on earlier paradigms. In my view: forget IBM, Microsoft, Oracle, etc, and look to Google, LinkedIn, Facebook/FriendFeed, Twitter and other “connection/relationship oriented” outfits … maybe even Nokia, once they get over their shock and unfounded fear of Apple!

Yes knowledge management, in general, and data/knowledge mining/harvesting, in particular, are all available once appropriate relationships are established.

Encouragement, rewards …? Why, isn’t it all good old innovation?!

John W Lewis