[uf-discuss] URI schemes vs. visible data (was Re: communications log, "tel" microformat?)

brian suda brian.suda at gmail.com
Sat Dec 3 12:25:49 PST 2005


Benjamin Carlyle wrote:

>The thing that really triggered me to post was reading the geo
>specification on the microformats wiki:
><div class="geo">
> <abbr class="latitude" title="37.408183">N 37° 24.491</abbr> 
> <abbr class="longitude" title="-122.13855">W 122° 08.313</abbr>
></div>
>
>It seemed to me that this would be simpler described as something like:
><a href="geo:37.408183,-122.13855">N 37° 24.491 W 122° 08.313</a>
>  
>
If you look through the archives you will find a post about the problems
with strange protocols, like geo: and tel: and isbn:, etc. While they
may be correct, no browser supports them, and your visitors will be
clicking on links that error. Not to mention then any transforming
applications would have to be aware of every type of protocol possible
to point to a lat/lon pair - not even browsers do that.

>What struck me really was that the visible data is ignored in this
>specification, so it seemed that URIs might be a way forward. 
>  
>
I'm not sure what you mean by "visible data is ignored"? Just like the
date-time issues, you can put the machine data in the TITLE attribute of
an ABBR element and the human readable data as the node value, or put
the machine data AS the node value of any element.

<abbr class="latitude" title="37.408183">N 37° 24.491</abbr>
is the same as
<span class="latitude">37.408183</span>

><abbr class="geo" title="37.408183,-122.13855">N 37° 24.491 W 122° 08.313</a>
>or <span class="geo">37.408183,-122.13855</span>
>  
>
The advantage to explictly wrapping each value in its own container
would avoid ambiguity to what the delimiter is. The ICBM protocol (i
think) uses ';' semicolon as a seperator, not a comma. Others may use
different delimiters. So
<span class="geo"><span class="longitude">-122.13855</span>,<span
class="latitude">37.408183</span></span>
could even be put out of order and still correctly transformed by
applications.



More information about the microformats-discuss mailing list