Re: [Geopriv] Draft Updates "GEO TAGS for HTML"

From: Henning Schulzrinne ^lt;hgs@cs.columbia.edu>
Date: Mon Aug 06 2007 - 13:30:30 EDT

On Aug 6, 2007, at 1:26 PM, Andrew Daviel wrote:

>
> On Mon, 6 Aug 2007, Henning Schulzrinne wrote:
>
>> For the HTML tags: Why not adopt the names and definitions of the
>> XML civic tags, rather than creating yet another labeling
>> hierarchy? That way, translating, say, a HELD or DHCP result into
>> the HTML tags becomes much easier. Example:
>>
>> civic.hno = 17
>
> Basically, because my drafts predate HELD by some considerable
> time, and I have always been looking for simplicity, such as the
> TLD country names that everyone knows. But if I understand
> correctly, civic.a1 maps to the LSB of geo.region. Splitting
> geo.region into geo.country and geo.xx (as in geo.a1, geo.hno)
> seems not unreasonable. It was really only intended as a check
> against human error in geo.position, anyway (getting the sign wrong
> in longitude, mostly - a common mistake)

This has nothing to do with HELD. The labels are defined in RFC 4119
(and the current pdif-lo draft), which predate your draft. You may
want to take a look at it.

>
>
>>
>> For geo, using the space-separated version of http://georss.org
>> seems cleaner as it avoids the various separator-white space
>> permutations and can easily be parsed with sscanf, which your
>> scheme can't.
>>
>> As in
>>
>> geo.position="45.256 -71.92"
>
> I see no particular problem parsing commas or semicolons; I have
> some code
> sscanf(position, "%lf ; %lf", &DocY, &DocX)
> "White space .. in the format string match any amount of white
> space,
> including none, in the input"
> while semicolon is used e.g. in existing "Content-Type: media;charset"
> so I see no advantage in changing to a space.

The space-only version simply conforms to the XML version. The
semicolon adds no particular value here.

>
>
> - Martin Kofahl's proposed update of the HTTP spec is available at
> http://geotags.com/geo/draft-daviel-kaegi-http-geo-header-06.txt
> I have not finished reviewing it.
>
> (I'm off sailing, so further replies in the next 2 weeks may depend
> on fair weather and cellphone coverage)
>
> regards
>
> Andrew Daviel
> TRIUMF/Vancouver Webpages

_______________________________________________
Geopriv mailing list
Geopriv@ietf.org
https://www1.ietf.org/mailman/listinfo/geopriv
Received on Mon, 6 Aug 2007 13:30:30 -0400

This archive was generated by hypermail 2.1.8 : Mon Aug 06 2007 - 13:30:38 EDT