2007-07-26 Developers Conference Call

<html><head><title></title></head><body>h1. Date

26 July 2007

In Attendance

  • Ben Wolfe
  • Paul Biondich
  • Burke Mamlin
  • Darius Jazayeri
  • Justin Miranda
  • Vladamir



Burke provided an introduction about logic service. The basic idea is that we are trying to take the atomic data points and try to provide a means to put some business logic in between.

Discussion about logic service, in terms of making sure everyone is on the same page. As a low level service it’s going to be consumed by a lot of different services, and especially with reporting. There is a big need for pulling out higher level concepts. Vladamir has been doing an excellent job of solidifying the ideas and turning it into actual code. Vladamir gave an overview and some discussion with the group on what he has been working on. More discussion.

Justin: What I want for the ODA Driver is for it to have sort of a name space, so that we can allow users to select from one of those name space or token groups and then they get back a limited number of tokens. For instance, the patient demographics would be one, medical observations, lab tests and things like that where you don’t have everything in one big lump. Burke: I think the way we would handle that is by a grouping or to create a tagging mechanism for tokens.

Burke: For the “to do†list for Vladamir and I, we want to make the data service needs to be public. Number two, we need to find a third way to register rules and that is by reference, which will be a string. Number 3 would be to add tags to tokens.
Also, one of things we are going to have to introduce to the logic service is a concept of a, whatever we want to call it, but an index date for all criteria to be applied.

Meeting Adjourned

  1. DataService needs to be public
  2. Add 3rd way to register rules: by reference (String)
  3. Add tags to tokens so people can find tokens by tag(s)
  4. Watch out over-caching (across requests) per Darius' experience
    • (cache rule results maybe...but not data results)
  5. Get unique list of tags
  6. Need to support "index" date so criteria can be applied relative to that date