[uf-discuss] Re: To-do items?

[Brian Suda]

2008/2/24, Toby A Inkster <mail at tobyinkster.co.uk>:
>  But that is just an example of the guesswork and interpretation needed to
>  implement the hCalendar spec. The spec is incredibly vague and even notes
>  its own incompleteness on some topics.

--- Lets not through the baby out with the Bath water. The hCalendar
spec is modeled off of the iCalendar spec and rather than spend years
matching every last useless item, it is better to get things out there
and iterate on them. If there is a big enough push and need for the
other 4-5 items to be mapped from iCalendar to hCalendar, then we can
examine those as they appear.

The wiki could use some clean-up and better wording about exactly what
1:1 means. There is an ICAL-BASIC which is in progress which cuts out
most of the un-used and un-supported "features" at one point hCalendar
was to be modeled off of that, but ICAL-BASIC is not finished.

There has been work on a Task Microformats which is also modeled after the VTODO
http://microformats.org/wiki/task-formats

The best course of action is to take examples from the wild and use
VTODO as a boiler plate, cut out the cruft, then it can be folded back
in. No need to drag a working format down to get edge-cases worked it
from the start.

> It contains some blatant
>  contradictions with the iCalendar spec which it normatively references --
>  such as stating that dtstart and summary are required children of a
>  vcalendar, when according to the source specification they are required
>  children of a *vevent* and are not allowed directly inside a vcalendar.

--- can you be more specific, i can't find the reference "dtstart and
summary are required children of a vcalendar". Some of these quirks
are also issues with consuming applications such as Outlook requiring
DTSTAMP which is actually optional.

-brian

-- 
brian suda
http://suda.co.uk