Chandler Virtuality Discussion

Mimi presented the research behind Chandler's virtuality vision and the design choices that were made along the way.

Attendees

Mitch, Katie, Mimi, Lisa, Sheila

Goals

• Put a stake in the ground for Chandler's virtuality
• Demonstrate how Chandler's virtuality will:
  • Feel familiar to users
  • Meet their organizational needs
  • Scale to deal with a lot of data
  • Make room for Chandler as an extensible platform
• We will accomplish this by:
  • Presenting research and user-based studies
  • Explaining the conceptual model behind the design
  • Demonstrating how the conceptual model is realized in the UI
  • Comparing and contrasting our design with alternatives

Notes/Comments

• Mimi was asked if she considered no classification system and doing everything with tags?
  • Yes she had but feels tags are a flavor of facetted classification.

• There was group agreement this is the first step towards an external presentation.
• 1st section (intro) -> would be comprehensible to an audience of interaction designers.
• 2nd section -> language might not be ready for a conventional audience yet
• Overall there were a number of well put insights that we haven't seen articulated anywhere.
• Needs some editing for a bigger audience.
• It's understood that this presentation will evolve into a paper and a presentation for a wider audience would be more concise.

• Comments on KINDS:
  • subkinds - we should allow people to add the kinds of things in grey (not the high level kinds but the examples underneath).
  • These sort of things exist in the same kind of way as the kinds themselves.
  • Might want to consider a hybrid approach around the grey (some added freely but others that are standard)

• Comments on COLLECTIONS:
  • none

• Comments on ATTRIBUTES:
• Are we going to talk about how this facetted hierarchy will be applied to the Chandler UI specifically? (NO - not in this presentation)

• Mitch's feedback
  • Mitch believes it's the right design strategy to tease out distinctions between kinds and attributes wrt the broad class of problem that we are trying to solve here.
  • Pretty strong agreement with the fundamental distinctions that Mimi is making.
  • For something that deserves to be a kind - it has to be fundamental, logical as a function of the domain (PIM) ie: email is fundamental.
  • Similarly wrt collections - why should something be a collection? Focus not universal - personally material - relates specifically to your life.
  • Collections have different philosophical grounding than a kind which has a universality.
  • Tree of facets makes sense and very promising.
  • Development strategy - need a way to experiment with designs that realize the view of this world without pining success on it.
  • The attributes slide needs discussion but out of scope for this discussion. We will make a note and go back to it.
  • What mimi is showing is that - a way to present data that is outside the scope of
  • Interested in exploring the concept of a new view for multiple facets - scrolling windows of items (all the kinds views in one window). Not now but maybe in the future.
  • The Action Mode are always likely to be single (schedule dinner party via email)
  • Dashboard is not really a "non facetted view"
  • We need a clear expression/terminology for describing the dashboard in 0.7
  • If I am looking at an arbitrary view - some kind/some collection
  • Can we envision supporting sections of some kind for any arbitrary attribute - some column or other value -> yes we are but we are starting with triage status first and sort.
  • Most people can handle is probably 3 dimensions, good that people who don't need spheres, don't have to do this.
  • Design is revolution in info management - big and it's good.

• Lisa feedback
  • Should revisit term "stable platform" -> stable structure or frame of reference
  • DDC analysis should be a standalone blog entry or a wiki page write up.
  • Put together a list of required reading before presentation (when we present this to broader group)
  • Add a diagram for the hierarchy of facets

Next Steps and Thing to Think About

• Revisit 0.6 virtuality with Mitch to address a few minor questions (hopefully with app and sidebar complete).
• Characterizing where we are now in relation to our overall virtuality and work on roadmap
  • Virtuality is partially implemented in 0.6
  • We have a provisional design for aspects of the virtuality that we haven't implemented yet. We have a set of design requirements of where we want to get to.
  • Work on virtuality implementation roadmap - In 0.7, the work in the virtuality will be the implementation of the following things, what will not be done.
  • How to identify what meaningful progress is on this.
• We could think about a really conventional imap mode - no cross kind - very conventional
  • Is there a safe path to let people use their email in the new world.