Re: [VCARDDAV] wg concensus to publish? NO

[Rohit Khare <[e-mail]>]

Hi, I’m Rohit Khare, and I apologize for being so late to this list.  
Perhaps like other folks, I was aware that there was some work going  
on under the brand name of “vCard”; and checked in on the drafts and  
this listserv from time to time; but didn’t realize the community was  
at WG Last Call until now.

I am concerned that vCardDAV is reusing the brand name and credibility  
of a very widely used non-IETF ‘standard’ to advance several dramatic  
changes, without taking full advantage of the unique opportunity to  
solve some extremely important adjacent problems at the same time.

Almost every successful data storage format has been human-readable  
and incrementally editable (see also ‘ASN.1’). Merely using XML  
doesn’t make that so, especially when there are opportunities to learn  
from formats fielded for exporting 100M+ contacts that could be  
considered and rejected for better reasons than “Standards are  
good” (w.r.t. sex=1 vs. gender ‘male’ — we’re dealing with social data  
rather than medical data, and there are far, far more culturally- 
defined gradations of gender than sex alone).

Similarly, DEATH. I stand ready to be corrected, but I don’t know of a  
significant fielded system for “contacts” or “social graphs” that has  
user experience with such fields. Names are defined by their uses, so  
we ought to be able to identify more utility than merely demonstrating  
round-trip interop to accept an expansion of the core vocabulary.

Finally, there’s the role of standards as a social forum to solve  
industry-wide problems. There are a lot of “investors” in the existing  
vCard, but I haven’t been convinced by my (admittedly cursory!)  
readings of the list and docs that organizations controlling large  
amounts of contact data have put these new ideas through their paces  
by exporting (much less, importing) data in vCard4. In this, I appeal  
to the circular fact that this work is trying to brand itself as a  
vCard successor, so it bears the additional burden of soliciting  
support from existing vCard users before it can assume that mantle.

Of course, there’s no end to demands to “coordinate” with other groups  
and standards, especially for something as foundational as contact  
information. I realize that volunteers who have poured their time and  
effort into this process deserve better than an endless stream of  
objections or complaints.

I am speaking for myself, in the IETF tradition of individual  
participation. Yes, I’ve worked with Tantek on microformats.org ; I  
currently work with Joseph Smarr; and I recently worked for the  
division of BT that Kevin Marks is at. Heck, DAV itself came out of  
our Web research group at UCI (as did REST).

In each of these cases, though, my role was based on more than a  
passing familiarity with standards, markup, Web architecture, and  
“social” computing.  It has been a few years since I’ve authored an  
IETF standard, but I don’t believe the norms have changed enough to  
believe that ‘voting’ is any way to come to ‘rough consensus and  
running code.’

I’m under no illusion that toting up YES and NO replies to the list is  
definitive, or even deserving of equal weight. I also have a lot of  
sympathy for the years of work that have gone into these docs and the  
frustration that might result from a late-comer voicing ‘principled’  
concerns.

However, and with sincere apologies to folks I haven’t even met f2f  
yet (including Cyrus, Marc, and the rest of the WG) I have to state  
that NO, I’m not comfortable with a Last Call for these two documents  
at this time.

Respectfully,
Rohit Khare
[e-mail]
Hacker, KNX.to (former)
Product Manager, Google (current)