[Telepathy] Comments on telepathy-spec-contact-info branch
Xavier Claessens
xclaesse at gmail.com
Tue Apr 10 09:30:32 PDT 2007
The API seems good to me and easy to use for a client point of view. For
the rest I agree we should be near of VCard fields, but the most
important is to have the same set of information than GNOME about-me
(evolution-data-server), but that's only the opinion of a GNOME hacker,
maybe KDE has something else.
Xav.
Le mardi 10 avril 2007 à 13:44 +0100, Simon McVittie a écrit :
> Comments welcomed from client authors (Zdra?) and anyone with
> involvement in Evolution/KAddresbook/etc., since they'll be the ones
> using this...
>
> (See <http://projects.collabora.co.uk/~monkey/telepathy-spec-contact-info/>)
>
> + <tp:copyright> Copyright (C) 2005, 2006 Collabora Limited </tp:copyright>
> + <tp:copyright> Copyright (C) 2005, 2006 Nokia Corporation </tp:copyright>
> + <tp:copyright> Copyright (C) 2006 INdT </tp:copyright>
>
> I don't think these are correct.
>
> + <tp:docstring>
> + An array of structs containing:
> + <ul>
> + <li>a string information field</li>
>
> "A string field name"?
>
> + <method name="GetContactInfo">
> + <arg direction="in" name="contact" type="u">
> + <tp:docstring>
> + An integer handle of the contact to request his informations for
>
> "An integer handle of the contact whose information is to be requested"?
>
> + <p>An interface for requesting information for contacts on a given connection and publishing
> + your own information.</p>
> +
> + <p>The following types and field names should be used where appropriate: </p>
>
> Where did you get these particular fields from? It seems sensible
> to get them from some sort of reference rather than inventing Yet
> Another set of conventions. I'd go for (a simplified form of) vCard,
> probably; RFC2426. hCard <http://microformats.org/wiki/hcard> might also
> be worth a look.
>
> As a general comment: we're not supporting multiples here. Multiple
> names would obviously be silly, but multiple mobile phone numbers are
> entirely reasonable to have. Perhaps we should throw some arrays in
> somewhere...
>
> + <dl>
> + <dt>s:full-name</dt><dd>Full name</dd>
>
> Same as vCard's FN, for what it's worth
>
> + <dt>s:family-name</dt><dd>Family name</dd>
> + <dt>s:first-name</dt><dd>First name</dd>
>
> First/last and given/family are distinct; in principle, a particular protocol
> might have both, one or neither. vCard uses family and given names; I think if
> we just pick one or the other, we should follow its lead.
>
> (Consider Chinese and Japanese names where the family name is written
> first, to see the distinction.)
>
> vCard also has additional (middle) names, honorific prefixes and
> honorific suffixes. I think we can probably skip those...
>
> + <dt>s:url</dt><dd>URL of a personal webpage</dd>
>
> URL in vCard
>
> + <dt>s:birthday</dt><dd>Date of birthday. Format used is YYYY-MM-DD</dd>
>
> vCard uses BDAY, but I think expanding the name is reasonable.
>
> + <dt>s:email-home</dt><dd>Personal email</dd>
>
> vCard equivalent is EMAIL; vCard makes no distinction between personal
> and work e-mail, but I think this is a useful distinction to make.
>
> + <dt>s:tel-home</dt><dd>Home phone</dd>
>
> vCard equivalent is TEL;TYPE=home,voice.
>
> + <dt>s:fax-home</dt><dd>Home fax</dd>
>
> vCard equivalent is TEL;TYPE=home,fax.
>
> + <dt>s:street-home</dt><dd>Home address: street</dd>
> + <dt>s:locality-home</dt><dd>Home address: locality</dd>
> + <dt>s:region-home</dt><dd>Home address: region</dd>
> + <dt>s:postal-home</dt><dd>Home address: postal code</dd>
> + <dt>s:country-home</dt><dd>Home address: country</dd>
>
> vCard equivalent is ADR;TYPE=home, which is divided into post office
> box, extended address (e.g. flat number), street address, locality, region,
> postal code and country. Perhaps we should allow the smaller
> subdivisions too, just to be consistent.
>
> + <dt>s:email-work</dt><dd>Business email</dd>
> + <dt>s:tel-work</dt><dd>Business phone</dd>
> + <dt>s:fax-work</dt><dd>Businnes fax</dd>
>
> EMAIL, TEL;TYPE=work,voice and TEL;TYPE=work,fax.
>
> + <dt>s:company-work</dt><dd>Company</dd>
> + <dt>s:department-work</dt><dd>Department</dd>
>
> -work is redundant, and "company" is misleading e.g. if someone works
> for a charity or university. vCard has ORG (expanding that to
> organization would be reasonable).
>
> vCard's ORG is split into "the organization name, followed by one or
> more levels of organizational unit names". I think we can simplify that
> to org-name and org-department (or something) with minimal loss of
> information.
>
> + <dt>s:title-work</dt><dd>Position</dd>
>
> vCard TITLE. -work is a bit redundant, but does help to indicate what
> this field means.
>
> <dt>s:title</dt><dd>Job title or functional position</dd> mirrors the
> language in the vCard RFC. The example the RFC gives is "Director,
> Research and Development".
>
> + <dt>s:role-work</dt><dd>Role</dd>
>
> vCard ROLE, "to specify information concerning the role, occupation or
> business category of the object the vCard represents". The example the
> RFC gives is "Programmer".
>
> + <dt>s:street-work</dt><dd>Work address: street</dd>
> + <dt>s:locality-work</dt><dd>Work address: locality</dd>
> + <dt>s:region-work</dt><dd>Work address: region</dd>
> + <dt>s:postal-work</dt><dd>Work address: postal code</dd>
> + <dt>s:country-work</dt><dd>Work address: country</dd>
>
> vCard ADR;TYPE=work. See home for comments.
>
> + <dt>s:tel-mobile</dt><dd>Mobile phone</dd>
>
> vCard TEL;TYPE=cell. Perhaps tel-cell to be consistent with vCard?
> tel-mobile is a more obvious phrasing, though.
>
> + <dt>s:description</dt><dd>Description</dd>
> _______________________________________________
> Telepathy mailing list
> Telepathy at lists.freedesktop.org
> http://lists.freedesktop.org/mailman/listinfo/telepathy
More information about the Telepathy
mailing list