[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Merged Draft containing feedback from all of you.



> As requested by the working group at the BOF held earlier this year, 
> Darren Shakib, Derik Stenerson and I have attempted to merge Frank 
> Dawson's draft and our previous draft into a single document (see URL 
> below).  There are still a couple of issues that we have been unable to 
> resolve that have not gotten much coverage on the list; I've included 
> them at the end.  The draft has also been updated to reflect the 
> group's discussion over the last few weeks.  
> 
http://www.imc.org/ietf-calendar/draft-calsch-sch-00.txt

> Also, here is the URL to the MIME-DIR draft since this draft refers to 
> it:
> http://ds.internic.net/internet-drafts/draft-ietf-asid-mime-direct-02.tx
> t
> 
> There are many topics in the draft where multiple solutions were 
> discussed on the list.  Where this is the case, Derik has tried to 
> represent the best concensus of the group.  This was merely done for 
> the sake of completeness so we'd have a draft to work from, not to say 
> "this absolutely the way it must be".  If there are errors (items that 
> do not properly represent the opinion of the group), then the draft 
> will be iterated until it accurately reflects the opinion of the group.
> 
> One new addition to the draft is an overview which discusses the 
> timezone issues for recurring meetings; this will hopefully allow folks 
> that are new to the recurring meeting issues to come up to speed 
> quickly on the complexities/issues.  Several examples have been added 
> to the draft, which should make it easier to discuss the issues.
> 
> To help keep the meeting running smoothly, we've drawn up a list of 
> open issues that might be of use on the agenda:
> 
> 
> -------------------------------------------------
> 
> TIMEZONE AND DAYLIGHT SAVINGS ISSUES FOR RECURRING MEETINGS
> - How important is it to allow a recurring meeting that lasts 2 years? 
> 5 years? 10 years?  should we bound this part of the problem?
> - What TZ should be used when scheduling a meeting.  creator location?  
> event location?
> - How should a change in TZ rules be handled?  sender re-issues meeting 
> request?
> - Is a server which is a world-wide authority on DSTs changes feasible?
> 
> Also, there are multiple solutions to the problem.  I propose that each 
> unique working solution should get a time slot for discussion.  Here is 
> a very brief description of the solution (which I think has some  
> limitations) that is in the draft:
> - DST shifs are explicitly specified in the meeting request by the 
> creator of the meeting and are relative to the timezone of the location 
> of the creator of the meeting.
> 
> 
> GENERAL RECURRANCE PATTERN ISSUES
> - How important are non-Gregorian calendars?
> 
> 
> DRAFT CONSOLIDATION ISSUES
> - begin/end  vs.  MIME
> - recurrance pattern format
> - recurrance pattern capabilities
> - property names
> - normalization of properties
> 
> 
> Thanks,
> Alec.
> 
> 
> ----------
> From: 	owner-ietf-calendar@imc.org[SMTP:owner-ietf-calendar@imc.org] on 
> behalf of Alec Dun
> Sent: 	Tuesday, November 5, 1996 1:59 AM
> To: 	ietf-calendar@imc.org
> Cc: 	fdawson@earthlink.net
> Subject: 	Merging the drafts.
> 
> Folks,
> Frank and I met and we tried to work out the differences in the drafts. 
>  At this point, I would say we're back to square one on merging our 
> drafts.  The right thing to do at this point is push the issues to the 
> group so we can resolve them.
> The most important first milestone for the working group is to be able 
> to exchange meeting requests and responses in e-mail, so I've focused 
> on that in my discussion of the issues below.
> First, I've included some examples of what meeting requests and 
> responses look like in both the Microsoft and Versit drafts.  It helps 
> to see examples when going thru the issues.  Below each set I have a 
> list of issues that need to be resolved.
> At the end of this message, I have included a list of generic issues 
> that are not discussed in the examples.
> Lastly, I am working with Tim Howes in the ASID working group to merge 
> the MIME-PROP draft into the MIME-DIR draft.  This merge is going well. 
>  At this point, both the vCalendar and PROP-CAL drafts are built on top 
> of the MIME-DIR draft.
> Hopefully we can get lots of feedback on these issues.
> Thanks,
> Alec.
> ************************************************************************
> ********************
> MEETING REQUEST
> ---------------Microsoft Draft.------------------
> Mime-Version: 1.0
> Date: Thu, 4 Nov 1996 01:30:00 -0400
> To: ietf-calendar@imc.org
> From: anik@ontime.com (Anik Ganguly)
> Subject: Meeting during Dec. IETF meeting
> Content-Type: text/calendar; charset="us-ascii"; profile="appointment"
> Profile: appointment
> Variant: OneTime
> Description:First IETF-Calendaring and Scheduling Working Group Meeting
> Summary:IETF Calendaring Working Group Meeting
> Location: San Jose, CA - Fairmont Hotel
> Start: 10 Dec 96 21:00:00 Z
> End: 10 Dec 96 22:00:00 Z
> Confidentiality:Public
> Category:Meeting
> ID:<foo@bogus.com>
> AppointmentVersion:4 Nov 96 14:00:00 Z
> --------------- Versit Draft.---------------
> Mime-Version: 1.0
> Date: Thu, 4 Nov 1996 01:30:00 -0400
> To: ietf-calendar@imc.org
> From: anik@ontime.com (Anik Ganguly)
> Subject: Meeting during Dec. IETF meeting
> Content-Type: text/calendar; profile="event/request"
> BEGIN:VCALENDAR
> DAYLIGHT:TRUE;-06:00;19960407T025959;19961027T010000;EST;EDT
> PRODID:-//RDU Software//NONSGML HandCal//EN
> TZ:-05:00
> VERSION:1.0
> BEGIN:VEVENT
> ATTENDEE;EXPECT=REQUEST:ietf-calendar@imc.org
> DESCRIPTION:First IETF-Calendaring and Scheduling Working Group
>    Meeting
> CATEGORIES:MEETING
> CLASS:PUBLIC
> DCREATED:19961022T083000
> SUMMARY:IETF Calendaring Working Group Meeting
> DTSTART:19961210T210000Z
> DTEND:19961210T220000Z
> LOCATION:San Jose, CA - Fairmont Hotel
 UID:guid-1.host1.com
> END:VEVENT
> END:VCALENDAR
> --------------- Issues ---------------
> 1.  Grouping/nesting the properties into VCALENDAR and VEVENT parts is 
> unnecessary.  We can reduce complexity and not detract from the meaning 
> by simply omitting all the BEGIN/END blocks completely from the above 
> example.
> 2.  MIME-DIR and begin/end.  The begin/end blocks are non in the spirit 
> of the MIME-DIR draft.  The correct way to delimit content in a message 
> is to use MIME.  By inventing a new content delimiting scheme, folks 
> are forced to write additional parsers for no reason.  Instead they 
> could just use their existing mime parsing code-base.
> 3.  Date/time formats.  If we use the existing RFC 822 date format, 
> then developers only need one date/time parsing routine.  If we use 
> something else, they have to have two date/time parsing routines.  The 
> ISO 8601 draft isn't any better, just different.  Both allow UTC times, 
> so there is no ambiguity either way.
> 4.  Attendee.  We should not need to repeat the recipient list in the 
> message; that's what the 822 headers are for.
> 5.  Attendee: parameters before the ":".  This does not follow the 
> spirit of the MIME-DIR draft.  The parameters before the ":" should be 
> used for attributes which give some information on how to parse the 
> content after the ":".  Some examples are charset and encoding.  In 
> examples like this, this should be broken into two properties, or the 
> data should be placed behind the ":":
> ATTENDEE: ietf-calendar@imc.org
> REQUESTTYPE: REQUEST
> or
> ATTENDEE: ietf-calendar@imc.org, REQUEST
> ************************************************************************
> ***********************
> RECURRING MEETING REQUEST (3rd Wednesday of every month at 10-11am)
> ---------------Microsoft Draft.---------------
> Mime-Version: 1.0
> Date: Thu, 4 Nov 1996 01:30:00 -0400
> To: ietf-calendar@imc.org
> From: anik@ontime.com (Anik Ganguly)
> Subject: Meeting during Dec. IETF meeting
> Content-Type: text/calendar; charset="us-ascii"; profile="appointment"
> Profile: appointment
> Variant: Recurring
> Description: Monthly staff meeting
> Summary: Monthly staff meeting
> Location: Conference room 12
> DaysOfMonth: 15, 16, 17, 18, 19, 20, 21
> DaysOfWeek: Wed
> MonthInterval: 1
> InstanceStartTime: 10:00
> InstanceEndTime: 11:00
> StartDate: 6 Nov 96
> ReferenceTimeZone: EST
> PatternDescription: "3rd Wednesday of every month at 10-11am"
> Confidentiality:Public
> Category:Meeting
> ID:<foo@bogus.com>
> AppointmentVersion:4 Nov 96 14:00:00 Z
> ---------------Versit Draft.---------------
> Mime-Version: 1.0
> Date: Thu, 4 Nov 1996 01:30:00 -0400
> To: ietf-calendar@imc.org
> From: anik@ontime.com (Anik Ganguly)
> Subject: Meeting during Dec. IETF meeting
> Content-Type: text/calendar; Content-Profile:event/request
> BEGIN:VCALENDAR
> DAYLIGHT:TRUE;-06:00;19960407T025959;19961027T010000;EST;EDT
> PRODID:-//RDU Software//NONSGML HandCal//EN
> TZ:-05:00
> VERSION:1.0
> BEGIN:VEVENT
> ATTENDEE;EXPECT=REQUEST:ietf-calendar@imc.org
> DESCRIPTION:First IETF-Calendaring and Scheduling Working Group
>    Meeting
> CATEGORIES:MEETING
> CLASS:PUBLIC
> DCREATED:19961022T083000
> SUMMARY:IETF Calendaring Working Group Meeting
> DTSTART:19961106T140000Z
> DTEND: 19961106T150000Z
> RRULE: MP1 3+ WE #0
> LOCATION:San Jose, CA - Fairmont Hotel
> UID:guid-1.host1.com
> END:VEVENT
> END:VCALENDAR
> --------------- Issues ---------------
> 1.  Daylight property.  Daylight savings time is not the same date 
> every year.  The property described as is just doesn't work.  
> Furthermore this is really only important for a recurring meeting.  
> Also, the TZ does not work alone since there could be various daylight 
> savings schemes in a particular time-zone.  We should just stick with 
> the three letter time-zones.  There are problems in that some 
> time-zones have duplicate names, but we should work to fix that.  One 
> nice thing about time-zones is that they encapsulate all the rules for 
> the specific region.
> 2.  Recurring format readability.  The recurring format "MP1 3+ WE #0" 
> is not human readable.  I've been told (ok, beaten up) that readability 
> is important when constructing and receiving mime messages.  Nobody is 
> going to figure out when this meeting is without a vCalendar spec.
> 3.  Recurring format problems.  There is no way to specify certain 
> dates in the vCalendar recurrance format, for example election day, 
> which is the first tuesday after a Monday in November every four years.
> 4.  Recurring format localization problems.  There are other calendar 
> systems used by other countries, for example Hijri and Lunar calendar 
> systems.  Many cultures want to be able to schedule recurring meetings 
> based on these other systems.  By breaking the recurrance pattern into 
> several properties, it is simple to add simply add properties to 
> support these other patterns, for example:
> DaysOfLunarMonth: 15
> LunarMonthInterval: 1
> ************************************************************************
> ********************************
> APPOINTMENT RESPONSE
> ---------------Microsoft Draft.---------------
> Mime-Version: 1.0
> Date: Thu, 4 Nov 1996 01:30:00 -0400
> To: anik@ontime.com 
> From: alecdu@microsoft.com
> Subject: RE: Meeting during Dec. IETF meeting
> Content-Type: text/calendar; charset="us-ascii"; 
> profile="appointmentresponse"
> Profile: appointmentresponse
> Variant: OneTime
> Response: Accept
> AppointmentID: <foo@bogus.com>
> AppointmentVersion:4 Nov 96 14:00:00 Z
> Start: 10 Dec 96 21:00:00 Z
> End: 10 Dec 96 22:00:00 Z
> Summary:IETF Calendaring Working Group Meeting
> ---------------Versit Draft.---------------
> Mime-Version: 1.0
> Date: Thu, 4 Nov 1996 01:30:00 -0400
> To: anik@ontime.com 
> From: alecdu@microsoft.com
> Subject: Meeting during Dec. IETF meeting
> Content-Type: text/calendar; profile="event/request"
> BEGIN:VCALENDAR
 DAYLIGHT:TRUE;-06:00;19960407T025959;19961027T010000;EST;EDT
> PRODID:-//RDU Software//NONSGML HandCal//EN
> TZ:-05:00
> VERSION:1.0
> BEGIN:VEVENT
> ATTENDEE;ROLE=ATTENDEE;STATUS=CONFIRMED: alecdu@microsoft.com
> DESCRIPTION:First IETF-Calendaring and Scheduling Working Group
>    Meeting
> CATEGORIES:MEETING
> CLASS:PUBLIC
> DCREATED:19961022T083000
> SUMMARY:IETF Calendaring Working Group Meeting
> DTSTART:19961210T210000Z
> DTEND:19961210T220000Z
> LOCATION:San Jose, CA - Fairmont Hotel
> UID:guid-1.host1.com
> END:VEVENT
> END:VCALENDAR
> --------------- Issues ---------------
> ***  I couldn't really find any examples or mention of what a meeting 
> response looks like in the vCalendar spec, so I just took a wild guess 
> and I'm probably wrong.  Perhaps there is no spec for this in 
> vCalendar.  Regardless, the response needs to be a little more compact 
> than I've made it out to be, and maybe it is.
> In any event, being able to respond to a meeting request is important 
> functionality as a deliverable for the working-group, and the final 
> merged draft needs to clearly describe how to respond to meeting 
> requests to say if you are attending a meeting or not.
> ************************************************************************
> ***************************
> OTHER ISSUES
> 1.  In the final draft, there need to be good clear examples of what 
> meeting requests and responses should look like.
> 2.  Attachments.  Attachments should just be another body part in the 
> message, or attach an HTML body part which contains meeting information 
> and one or more relative/absolute URL.
> 3.  Alarms.  Alarms can be most simply expressed and easily read as a 
> set of properties for which parse code exists rather than a complex 
> string that you have to write new code to parse.  It would be as below. 
>  In a recurring meeting request, I would omit the Alarm Date.  For 
> example, an audio alarm would look like this:
> AAlarmDate: 10 Dec 96
> AAlarmTime: 20:45:00 Z
> AAlarmAudio: http://www.ibm.com/taps.wav
> AAlarmSnoozeTime: 15
> AAlarmSnoozeRepeat: 4
> 4.  String-list delimiters.  Whichever string-list delimiter we choose, 
> it should be consistent so we don't end up writing multiple sets of 
> parse code.  semicolon seems like a good choice, especially since it 
> allows us to share parsing code with RFC 822 header parsing code.
> 5.  Exceptions to recurring patterns.  We should simply list the dates 
> of exceptions.  If we need a recurrance pattern for exceptions for 
> recurrance patterns, then that means we've just done a crappy job of 
> defining the recurrance patterns.
> 
> 
> 
> 
>