Decision Incubator

05 Aug 2010


See also: IRC log


Jeff Waters, Don McGarry
Jeff Waters


<jeff_> Scribe: Jeff Waters

<jeff_> ScribeNick:jeff


<jeff_> jeff: Thanks to everyone for joining us for our 10th meeting of the Decisions and Decision-Making Incubator. We are almost half-way through our allotted one-year time frame for our incubator work. This is a good opportunity to take stock and streamline a way-ahead to complete our work. If you don't mind me briefly quoting from our charter ...

<jeff_> jeff: The goal of our incubator is "exploring and determining the requirements, use cases, and a potential standard format for representing our decisions efficiently and effectively in a collaborative networked environment for the purpose of information exchange for situational awareness". So our incubator products are requirements, use cases and potential standard format. And the scope is information exchange, situational awareness, in a networked environ

<jeff_> jeff: "The work performed by this incubator activity is designed to help organizations improve integration of human decisions into computer systems, ..., to improve information sharing, and ultimately, to support better, rapid, and agile decision making."

<jeff_> jeff: The potential standard format should provide concise, generic, structured assessments and decisions...." At the most basic level, the goal is to move decisions beyond just stating the results in phone calls, announcements and emails, to a structured, machine-understandable format with the basic info needed to understand the decisions. But concise and simple (elegant) is good.

<jeff_> jeff: So we have identified initial use cases and defined required components. We've begun modeling. Following eXtreme Design approach, we are looking for patterns that can be reused so that we don't reinvent the wheel. One thing that can help us, is a mapping of those components to implied patterns. I looked into this and it's one of our agenda items. With this mapping, we can perhaps efficiently schedule our remaining design work.

<jeff_> jeff: Also I've been motivated by the Open Linked Data example of simply taking basic info and converting it simply to RDF. In this vein, I have on the agenda a suggestion for a simple xml format for a decision that we can convert in this same manner.

<jeff_> jeff: A third agenda item is how the design and simple conversion format can proceed in parallel and the former upgrade the latter. Also the concept of instrumenting a tool for use with an infrastructure backbone in an interoperability testbed environment will allow us to explore a near-term capability and the applicability of our designs. Documentation is a part of this.

<jeff_> jeff: Finally, there are interesting possibilities for pushing as well as pulling our decisions and other semantic data. In the area of emergency management in particular, we have an opportunity to explore this with one of our partners.

<jeff_> jeff: So all the agenda items come together serve as mechanisms (design, simple format, interop testbed with tool, push and pull) to help us finish our effort. So let's begin with a mapping of components to patterns.

<jeff_> http://www.w3.org/2005/Incubator/decision/wiki/Decision_Mtg_10_Agenda

<jeff_> don: I think that sounds like a good plan

<jeff_> eva: another thing, we are arranging a workshop on ontology patterns at the semantic web conference and Aug 20 is the deadline for submission

<jeff_> eva: it struck me the other day that we could submit on our incubator, etc. and we could discuss this as well

<jeff_> Jeff: sounds good to me

<jeff_> eva: Aug 20 is a short deadline, 5 or 6 pages describing your ideas for ongoing work, not a full research paper, but a position paper

<jeff_> don: sure, when's the conference?

<jeff_> eva: the workshop is on the 8th of November, it's in Shanghai and I'm going and I'm chair of pattern section

<eblomqvi> http://ontologydesignpatterns.org/wiki/WOP:2010

<jeff_> jeff: how do we get started? An outline?

<jeff_> eva: I could post a few ideas

<jeff_> jeff: do you have an example of a position paper?

<jeff_> eva: not right now, but maybe I can find one

<jeff_> eva: what's the idea, what's the goal, etc.

<jeff_> don: we should do email probably to work the paper

<jeff_> don: with stuff that we've been doing Jef, we have some good examples of what's been done so far and what we want to apply the decision-making stuff to

<jeff_> jeff: (summarized status of good email contacts with Jim Hendler)

A Decision Format as a Set of Patterns

<jeff_> I'm pasting in a link to a page that I created and it's linked from the main page as well. This is a page discussing the implied patterns underlying the decision components. This is just my thinking but I'm hoping that we can all edit or adjust this as needed. My goal with this is to have a defined, limited list of the patterns that we need to go find or implement to complete our basic decision format in the form of a set of patterns. So briefly, here th

<jeff_> Here is a link to a mapping of decision components to patterns: http://www.w3.org/2005/Incubator/decision/wiki/Final_Report_Decision_Components_And_Patterns

<jeff_> eva: agree with the idea that we try to focus on the underlying patterns, but it's important to not lose the connection to the use case, cause if too general, we may construct too big or complex a pattern

<jeff_> jeff: yes we just need to keep them in sync

<jeff_> jeff: The Question implies a "Statement with Variable" pattern.

<jeff_> The Options suggests "Filter" and "Aggregation" patterns.

<jeff_> jeff: The Metrics imply a "Normalizer", "Weighting" and "Aggregation" pattern.

<jeff_> jeff: The Assessment is an "Ordering" pattern and the Choice is a "Selection" pattern.

<jeff_> jeff: Please consider and let's edit, if this makes sense, agree on a set of patterns and perhaps divvy up the design modeling of these.

<jeff_> eva: sounds good

A Simple XML format for Conversion as Open Linked Data

<jeff_> Motivated by the Open Linked Data sets and in an effort to reach out to those still learning or not quite seeing eye-to-eye with a purse rdf format, we can consider a simple XML format, similar to the types of formats produced by governement organizations for their open data. Then we can work the conversion, initially, in the same simple way that RPI is doing for the open linked data work. (Nothing too fancy.) This approach is also consistent with GRDDL.

A Way Ahead via Interoperability Testbed and Open Linked Data Example

<jeff_> A way ahead via Interoperability Testbed and Open Linked Data. So having the design work in parallel with the simple reachout, I'm interested in exploring instrumenting a tool to read/write this format and explore in an interoperability testbed. See for example icbrne.org. The emergency data is a good opportunity for this.

<jeff_> don: I'm going to suggest on the status of my server today, I've got plenty of horsepower on this thing

<jeff_> don: a good goal is whatever we are going to do have it run over http and https

<jeff_> don: asking for more ports

<jeff_> jeff: any server interaction on your part eva?

<jeff_> eva: I don't have any details, but maybe we can hook on stuff later

Implications for Routing

<jeff_> Routing is a push approach to the pull approach of what we normally do with SPARQL. As we develop our decisions, there is the opportunity to push them as needed. But the modularity and flexibility of the triple format, and the implied decomposition of stovepiped XML schema formats, allows for some interesting opportunities to push more granular information. We might have an opportunity to explore some of this with our parther Solace Systems.

<jeff_> jeff: I will find some information and send it to you, eva

<jeff_> jeff: So I believe there are interesting opportunities for us as we move ahead.

<jeff_> jeff: So I have action items to do a title and outline for our proposed position paper for semantic technology conference and also to send Eva some information on routing with policies

<jeff_> jeff: Eva will send what she can about the position paper example, if she finds one

Summary of Action Items

[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.135 (CVS log)
$Date: 2010/08/05 15:05:33 $

Scribe.perl diagnostic output

[Delete this section before finalizing the minutes.]
This is scribe.perl Revision: 1.135  of Date: 2009/03/02 03:52:20  
Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: RRSAgent_Text_Format (score 1.00)

Found Scribe: Jeff Waters
Found ScribeNick: jeff
WARNING: No scribe lines found matching ScribeNick pattern: <jeff> ...

WARNING: 0 scribe lines found (out of 75 total lines.)
Are you sure you specified a correct ScribeNick?

WARNING: No "Present: ... " found!
Possibly Present: ScribeNick dmcgarry don eblomqvi eva inserted jeff jeff_
You can indicate people for the Present list like this:
        <dbooth> Present: dbooth jonathan mary
        <dbooth> Present+ amy

Agenda: http://www.w3.org/2005/Incubator/decision/wiki/Decision_Mtg_10_Agenda
Got date from IRC log name: 05 Aug 2010
Guessing minutes URL: http://www.w3.org/2010/08/05-decision-xg-minutes.html
People with action items: 

[End of scribe.perl diagnostic output]