[uf-discuss] multiple tel and adr types

[Tantek Ç elik]

On 4/7/06 7:24 PM, "Tantek Çelik" <tantek at cs.stanford.edu> wrote:

> On 4/7/06 3:11 PM, "Ryan King" <ryan at technorati.com> wrote:
> 
>> While writing test cases for the TEL property of hCard, I realized
>> that we could have two different constructions of the type sub-
>> property. Examples:
>> 
>> <div class="tel">
>>  <ul>
>>    <li class="type">home</li>
>>    <li class="type">msg</li>
>>  </ul>
>>  <span class="value">+14155551234</span>
>> </div>
> 
>> <div class="tel">
>>  <ul class="type">
>>    <li>home</li>
>>    <li>msg</li>
>>  </ul>
>>  <span class="value">+14155551234</span>
>> </div>
>> 
>> I think both of these should be allowed (and not just on <ul>'s).
> 
> Why?
> 
> 
>> Brian seems to think it would be plausible to implement in X2V. Any
>> objections?
> 
> Though it is implementable, it appears to introduce an unnecessary
> additional chunk of code in all implementations, and thus IMHO is
> objectionable.
> 
> Let's keep it as simple as possible.

To be clear, let's stick with how to do multiple tel and adr types per the
hcard-examples, which basically means the first example, with multiple
elements with class name of "type".

http://microformats.org/wiki/hcard-examples#3.3.1_TEL_Type_Definition

http://microformats.org/wiki/hcard-examples#3.2.1_ADR_Type_Definition


> I am opposed to adding more flexibility without authoring benefit.
> 
> As far as I know, there is no authoring benefit to doing the second over the
> first.
> 
> If someone has found (or authored) an actual real world example in the wild
> that demonstrates as such, let's start with that.
> 
> We need to be very diligent about adding flexibility to the format *only*
> when it provides a concrete benefit to the author, driven by actual use
> cases.

Thanks,

Tantek