« Social Software - a twitter moment | Main | Hugh's hat »


Balaji Sowmyanarayan

Sounds more like segregation of Content and Presentation in html, CSS sense. Well, the segregation freed up websites from the shackles of the geeks. Designers ruled(as in 9Rules) - and suddenly world(wide web) is with rounded corners!

Ok, back to DID, PID. So you model the information sources( patient walking in with a broken arm, or X-ray machine utilization data, or a Knowledge workers expense stream) and then on top of it model their(info source's) inter-connections. This inter-connection modeling is where reuse of social graph manipulation engines from 'Social Software' is now irresistible( call it romantic charm of Social Software).

Now, just like static content is a subset of dynamic-on-the-fly served content( like this blog
for example) you are saying Precise Information will do. All the old habit Dubious Information can be generated out of PI.

As I see it, Precise Information is more liquid. Whereas DI is colored - like drinking water that has passed through a coffee machine's filter. There is no way to get water out of coffee - there is loss of liquidity( as in cannot be converted to another desired from) here. Though there is a value add.

Common thread in your discourse all along is 'Hey technology affords that we can model the world and economic activity around it with PI as starting point. I'm developing tools(thingamy) for making this paradigm shift easy'

My question as before is, can you make a hosted version of thingamy available as is for people at large( not necessarily Enterprise Software audience) to play around? I'm sure the extra overhead and distraction will be worth the trouble. A paradigm shift needs a lot of learning, applying and re-learning. The above request will serve the purpose of learning ground well.



last one first: yep, will do - and it could allow some experiments I have not thought of using the modelling tool...

As to PI vs DI and real world modelling:

If you copy the real world directly you would have very small and finite objects in the model representing the real world objects combined with the relationship between such objects. That would limit the complexity of the model to the complexity of the world something current models do not, they increase the complexity many fold with their event and transaction objects representing one single real world object.

A patient would only be represented with name, DOB and social security number - related to a house with street and zipcode, his abode and a medical condition (allowing for virtual objects), this time a broken arm.
Let then the objects capture what happens to them in a flow.

The combination of singular objects and relationships (plus capture of what happens to them) would be the fact layer of PI.

Then the logic layer:

1. Often it's as simple as using a report template - which usually reflects some logic - one for me, one for you, one for whoever.
2. Or treat the logic - as in analysis or arguments - as a singular and virtual object related to one or more factual objects. This is in reality just a elaborate template but could be (too?) hard to model in a system, but perhaps, perhaps... will give that one a think!

The comments to this entry are closed.

My Photo


  • Phone: +33 6 8887 9944
    Skype: sigurd.rinde
    iChat/AIM: sigrind52

Tweet this

Thingamy sites

  • Main site
  • Concept site

Tittin's blog


Enterprise Irregulars


Twitter Updates

    follow me on Twitter


    • Alltop, all the cool kids (and me)


    Blog powered by Typepad
    Member since 01/2005