[R6RS] my notes on today's conference call (19 April 2006)

[William D Clinger]

Conference call April 19 2006 12:45pm
All present by about 12:45pm:
Kent, Anton, Matthew, Mike, Will


0. finalize agenda (1 minute)

1. action items from 4/11/2006 (5 minutes)
   - publish new Unicode SRFI (Matthew)
        Marc Feeley doesn't agree with all of Matthew's revisions
        action: withdraw Unicode SRFI
        Matthew will post message as individual
   - publish new Arithmetic SRFI (Will)
        drafted and checked into Subversion but not published
        problems identified: shift operations, fx+ etc
        will be revised again today
   - update syntax-case SRFI, with reference implementation (Kent)
        Kent will post note on opacity issue
   - propose r6rs eval (Mike)
        done
   - refine core/language split proposal (Anton)
        done, but refining continues
   - carried over:
     - look over proposed condition hierarchy and details; comment
       via mailing list (All)
        Will will stimulate discussion by posting annoying email
     - propose library modifications to allow subsetting, supersetting,
       defining new libraries, and to allow for lexical syntax extensions
       or future changes (Anton, Matthew, Kent)
        Anton will send email

voting on issues, most fairly minor
    couple of votes outstanding
    mv binding vote, tied 50/50 (Will hasn't voted)
    eof-object
    Kent wants to conduct votes in weekly conference call
        two weeks from now (yes/no/abstain, or maybe forced yes/no)

records
    no major issues raised in SRFI discussion
    Marc Feeley urged us to consider using his keyword SRFI
        in the record SRFI
    that would commit us to putting keywords into R6RS
    but SRFI discussion of keywords does not yet show consensus
        even on lexical syntax

2. enumerations (15 minutes)
   - discuss Mike's proposal
        Will will recast Mike's proposal using symbols
        others will propose interesting examples where
            enumeration issues interact with efficiency
            or quality of error checking

3. r6rs eval (15 minutes)
   - discuss Mike's proposal
        ditches null-environment
        adds eval-library
        scheme-report-environment might be in the r5rs library
        restrictions proposed for eval:
            must raise specified exception if first arg is definition
            cannot assign to variables imported from its second arg
        library namespace (which we have to face anyway)
        how to restrict libraries available through library-environment?

4. adjourned about 1:44pm

Will