Lias Architecture Meeting 2005-05-05

• Let's not get too bogged down in how all the searches are going to go, just yet. Even just rudimentary searching of notebooks is leaps and bounds beyond the current state of things.
• Harvey's much more comfortable with LaTeX instead of WYSIWYG. I'm not sure how wide-spread this is... or how it will really work with pulling stuff in from a variety of sources.
• Phases:
  • Formulation
    • Finding references
    • Keeping a bibliography of the things you browsed
    • Keeping annotations of the things you browsed
  • Proposing
    • Preparing a proposal
  • Experimenting
    • Finding algorithms
    • Writing algorithms
    • Running algorithms
  • Reporting
• What if the notebooks were supported in multiple formats vs. something like Scientific Workplace?
• How much structure do we need in a notebook? The notebook is a smart container. One of the things it contains is scientist's journal entries... those could be in LaTeX or Word or PDF or whatever you wanted.
• Argument for a very web-ish interface so you can use it from anywhere.
• Tangent: Notebook in a subversion or CVS tree? Put some thought into scoping of comments. Easy export for when someone wants to package up their data and go. It could be easy to merge together things.
  • need to keep track of meta-data for items
  • need to keep track of environment needed for algorithms
• WebDAV is a good start
• Notebook can watch a Folder... or you can something drag and drop...

Questions for next week:

• Notebook design questions
• How does the notebook relate to Kepler?

-- PatrickStein - 05 May 2005