Personal tools
You are here: Home Meeting Notes Able Meeting 2003-10-03
Document Actions

Able Meeting 2003-10-03

by Shang-Wen Cheng last modified 2004-01-20 04:00 PM

ABLE Meeting: Fri, October 3, 2003 @ 1330-1500

Styles

  1. Style composition--calculus (''mix-ins'') of arch styles ("horizontal" relationship)
    1. To compose, or mix-in styles of independent domains into a particular system.
      1. E.g., styles for performance, reliability, security, that are independent from Pipes-and-Filters, Client-and-Server, etc.
      2. Draw from formal world: Larch traits, initial vs final algebras
    2. (Jonathan) Style semantic conflict (David: exists in the types as lack of instance that satisfies the styles)
    3. Different ways of composition (would need to have hooks/parameters for composition)
      1. S2 is a refinement of S1
      2. S1 union S2
      3. S2 mixed in to S1 (i.e., no new component types introduced, just addition of properties)
  2. Tooling:
    1. Extended notion of style in AcmeStudio
      1. Style-dependent analysis
      2. UI elements
      3. Operations
  3. Architectural refinement--need better support ("vertical" consistency)
    1. So far, punted, via "representation"
    2. We need to be able to maintain consistency and analysis capability from one refinement level to the next, and be able to move from level to level
    3. The need was observed in at least two examples:
      1. Ford project: the need for abstract and concrete components, where choice of component occurs when concrete instantiation occurs
      2. NASA MDS project: things you need in the more concrete level that you don't want to have to worry about in the more abstract level
    4. (JungSoo) Is this vertical consistency only in terms of structure? (David: essentially yes.) At what point do we introduce computational information? (next item.)
    5. Link w/ ArchJava (consistency between arch and code)
  4. Computational, behavioral aspects
    1. Link with behavioral formalism not made first-class
      1. E.g., Wright, FSP
    2. Low-hanging fruit:
      1. Kramer, Magee interested in integrating LTSA
      2. Xinhua's work---takes FSP-prop'd Acme and pull out FSP to analyze, but no hook to LTSA
  5. Architectural patterns/idioms/templates
    1. George's work
    2. Being able to represent these templates, plus the tool support
    3. Language is the universe of discourse, and the ''patterns'' are reusable chunks of the discourse
    4. Pattern composition, might be similar to issues of style composition
    5. Reuse library that one could create and use, e.g., a Coleman Filter for Pipes-and-Filter style
  6. Dynamism
    1. Architecture is not static, and we might have to think of representation of dynamism from scratch---language
    2. Operators
      1. Per style
      2. Common across styles
  7. First-class connector areas
    1. Connector operators
    2. Operations on connectors (Bridget)
« January 2009 »
Su Mo Tu We Th Fr Sa
1 2 3
4 5 6 7 8 9 10
11 12 13 14 15 16 17
18 19 20 21 22 23 24
25 26 27 28 29 30 31
 

Powered by Plone, the Open Source Content Management System

This site conforms to the following standards: