Re: [Geojson] Ben Campbell's No Objection on draft-ietf-geojson-03: (with COMMENT)

["Ben Campbell" <[e-mail]>]

Thanks for the response. Comments inline:

On 2 Jun 2016, at 0:49, Tim Schaub wrote:

> Thanks for the comments Ben.  My thoughts on one of them below ...
>
> On Wed, Jun 1, 2016 at 7:06 PM, Ben Campbell <[e-mail]> wrote:
>>
>> ----------------------------------------------------------------------
>> COMMENT:
>> ----------------------------------------------------------------------
>>
>
> ...
>
>> - 3.1.6, 4th bullet: Why SHOULD? Can you imagine situations where it
>> would be reasonable to not follow the right-hand rule?
>
> There has been an effort throughout the IETF process to reduce the
> likelihood that GeoJSON in use today (pre-IETF GeoJSON) will not be
> rejected as invalid by strict-parsers developed in the future.  The
> GJ2008 document didn't discuss winding order of rings.  So while it
> would be convenient for consumers if they didn't have to check the
> winding order (i.e. if the spec said MUST follow the right-hand rule),
> I think this is a recommendation only in hopes of improving
> interoperability while retaining some backwards compatibility.

That's a reasonable motivation for a SHOULD. I'd be fine if you left it 
htat way. A few words of explanation to that effect might be useful.

>
> I wouldn't be opposed to making this MUST.  People will write and use
> lenient parsers until old (left-hand) GeoJSON falls out of use.  With
> SHOULD, we'll have mixed-handedness forever.

Another possible approach would be to say that implementations of this 
spec MUST use right-hand windings, but that parsers SHOULD not reject 
left-handed windings for reasons of backwards compatibility.

Thanks!

Ben.