W3C home > Mailing lists > Public > public-device-apis@w3.org > June 2010

Re: [contacts] Removal of serviceId from API

From: Rich Tibbett <rich.tibbett@gmail.com>
Date: Wed, 30 Jun 2010 09:10:05 +0100
To: "Dominique Hazael-Massieux" <dom@w3.org>, "Suresh Chitturi" <schitturi@rim.com>
Cc: public-device-apis@w3.org
Message-ID: <op.ve3sy3i241t9di@rich-mac.local>
On Thu, 24 Jun 2010 23:22:20 +0100, Suresh Chitturi <schitturi@rim.com>  
wrote:

> I was thinking we could still support the notion of unified address  
> book, but use the serviceId attribute as defined now to distinguish the  
> contact entries.

The serviceId as specified now is incapable of defining a contact object  
containing information from multiple sources.

point 1: serviceId as it is currently included in the spec is not well  
specified or fit for purpose.

> I think it would still work unless you are missing something. But we  
> would likely need to a getter for available service from the  
> device/system, e.g. DOMString getContactServices();

I'd only share the names of my contact databases on a need to know basis  
and I'd suggest that webapps don't need to know where I store my  
information, just that I have information to share - hence the unified  
model.

point 2: serviceId is entirely unnecessary and superfluous information to  
provide to web applications at the Web API level.

> so the user can set the serviceId accordingly.

Let me say that the 'user' might not have any say in such a decision and  
suddenly find that a website (of all things!) has pushed their contact  
records from one database to another. A horrible horrible thought I hope  
you agree.

point 3: serviceId introduces 'who owns the data' wars. A conforming  
implementation may be capable of maintaining sync to a plurality of  
sources. This doesn't need to be exposed to the API and the web.

>
> On Thu, 24 Jun 2010 at 2:33 AM Dominique Hazael-Massieux wrote:
> >> I don't think what we have in the spec can be tightened into something
> > workable for a source-aware contacts API, but if you can provide a
> > proposed addition to the spec that you think would fit that  
> requirement,
> > it would be help figuring it out.
> >
> >
> > Dom
>

+1. Dom also makes some good points in previous emails.

I urge you to re-consider your position on the need for such an attribute  
and the need to distinguish between address books from a web developer's  
standpoint.

In the meantime, this attribute will be removed from the Contacts API.

- Richard
Received on Wednesday, 30 June 2010 09:05:12 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 9 May 2012 00:14:10 GMT