IETF 65 Calsify WG

1
IETF 65 Calsify WG

• March 21, 2006
• Dallas, TX

2
Agenda

• Agenda bashing (Chairs, 5 mins)
• Discuss 'simplification' vs 'clarification' as
  the way forward for this WG. (Daboo, 15 mins)
• 2445bis issues discussion (Desruisseaux, 15 mins)
• 2446bis issues discussion (Daboo, 10 mins)
• 2447bis status - Ready for WG last call?
  (Melnikov, 5 mins)
• Report from CalConnect on recurrences (Daboo, 5
  mins)
• Report from CalConnect on last interop event
  (McCullough, 5 mins)

3
Calsify

• Simplification vs fixing?
• Proposal simply fix the things that dont work
  now rather than removing/splitting of features.
• Worry about draft status after we do that.

4
RFC 2445 Issues Discussion

• Bernard Desruisseaux

5
RFC2445 Issues (1)

• Need to clarify that the only character sets
  allowed are UTF-8 and US-ASCII.
• The NON-US-ASCII production rule should make
  reference to the UTF-8 production
  rules NON-US-ASCII UTF8-2 / UTF8-3 / UTF8-4
  UTF8-x are defined in RFC 3629
• Need to clarify how to handle the method
  parameter for sequence of iCalendar objects with
  different METHOD values.
• Components should allow iana-prop in addition
  to x-prop.

6
RFC2445 Issues (2)

• The semantic of PERCENT-COMPLETE is only defined
  when used in an iTIP message with METHODREPLY.
• Could allow PERCENT-COMPLETE to be used as a
  parameter to the ATTENDEE property.
• Not clear that iCalendar should specify when the
  value of SEQUENCE should be incremented.
• Should only be specified in iTIP.

7
RFC2445 Issues (3)

• The DELEGATED-FROM, DELEGATED-TO properties and
  the SENT-BY parameters require mailto URI values.
• Any kind of URI should be allowed.
• Some properties are listed as OPTIONAL in the
  ABNF of some components but as REQUIRED in those
  components in their own Conformance section,
  e.g., DTSTAMP, DTSTART, UID.
• ABNF needs to be fixed.

8
RFC2445 Issues (4)

• Not possible to specify that a VALARM has been
  processed, nor which instance of a repeating
  VALARM has been processed.
• Updating/deleting VALARM is the only
  interoperable way right now.
• The ACTIONPROCEDURE of VALARM doesnt
  interoperate well and raises security issues.
• Its use should be deprecated.

9
RFC2445 Issues (5)

• Needs to clarify how to specify the name of an
  inline attachment.
• The FMTTYPE parameter could follow the same
  syntax as the Content-Type MIME header field and
  make use of the name parameter, e.g.,
• FMTTYPEapplication/mswordnamereport.doc
• Not possible to specify a DQUOTE (") in a
  parameter value, e.g.,
• ATTENDEECN"RL "Bob" Morgan"
  mailtorlmorgan_at_washington.edu
• ATTACHFMTTYPE"text/plainname"foo.txt"
  charset"ISO-8859-1"wuajsfkfasdfkjfl

10
RFC2445 Issues (6)

• Need to clarify if VFREEBUSY components can be
  used to block off time in a calendar.
• Need to clarify if DTEND is inclusive or
  exclusive when both the DTSTART and DTEND are
  DATE values.
• Is a DATE DTEND allowed with a DATE-TIME DTSTART
  in a VEVENT? If so, what does it mean?

11
RFC 2446 Issues Discussion

• Cyrus Daboo

12
RFC2446 Issues (1)

• Change tables to list the exceptions to 2445
  component/property rules only, or make current
  set of tables complete wrt items in 2445.
• Allow REFRESH of a published event?
• Concept of forwarding needs clarification.

13
RFC2446 Issues (2)

• SEQUENCE changes?
• Yes for any change to time/duration of event.
• What about addition/removal of attendees?
• What about change to location?
• Addition/removal of attachments?
• Should we create a table or properties and
  indicate when changes to each MUST/SHOULD/MAY
  result in a SEQUENCE change?

14
RFC2446 Issues (3)

• Should we use examples as a way of describing the
  right way to do scheduling?

15
RFC 2447 Status

• Alexey Melnikov

16
RFC 2447 Issues/To-do

• Should infinite multipart/mixed nesting be
  allowed?
• Need feedback to know if this is an interop
  issue.
• Add examples of 8bit and quoted-printable.
• Address IANA considerations section.

17
Report from CalConnecton recurrences

• Cyrus Daboo

18
Calconnect Recurrence Recommendations Document

• http//www.calconnect.org/publications/icalendarre
  currenceproblemsandrecommendationsv1.0.pdf
• Describes problems related to handling
  recurrences in iCalendar (mostly iTIP) and
  recommendations on how to deal with that.

19
Summary of Results

• Are multiple RRULEs and EXRULEs really useful,
  could we do without them?
• Are EXRULEs really useful, could we do without
  them?
• Removal of THISANDPRIOR, since THISANDPRIOR
  always refers to a finite number of occurrences
  it could be done with exceptions.

20
Report from CalConnecton last interop event

• Jeff McCullough

21
Report from CalConnecton last interop event

• Basic iCalendar Interop improving
• Scheduling still has problems
• CalDAV implementations improving