[Tiling] Fwd: [OSGeo-Standards] Irregular tile coverage in TMS?

[Rushforth, Peter]

Hi,

Please consider using AtomPub as the base protocol for tiled services.  

As pointed out here: 

http://dret.net/netdret/docs/icwe2010/

it might be useful for vector data (KML) too.  

The markup echos what
Allan Doyle was calling for on the Geoweb-REST list, ie @rel="east" etc,
to guide clients in selecting the geographic 'next' page of entries,
which is nicely compatible with REST IMHO.

AtomPub is a strong contender for a standard RESTful interaction
model, widely used.

Of course, as Paul points out, implementation does trump theory.

Cheers,
Peter


> -----Original Message-----
> From: tiling-bounces at lists.osgeo.org 
> [mailto:tiling-bounces at lists.osgeo.org] On Behalf Of 
> christopher.schmidt at nokia.com
> Sent: September 13, 2010 13:26
> To: pramsey at cleverelephant.ca
> Cc: tiling at lists.osgeo.org
> Subject: Re: [Tiling] Fwd: [OSGeo-Standards] Irregular tile 
> coverage in TMS?
> 
> 
> On Sep 13, 2010, at 1:06 PM, ext Paul Ramsey wrote:
> 
> > It looks like these discussions might add some notes and 
> alterations 
> > to the existing document, so I've set up a page to track them and 
> > their status. Eventually this could result in a TMS 1.1 page.
> > 
> > 
> http://wiki.osgeo.org/wiki/Tile_Map_Service_Specification_Change_Reque
> > sts
> 
> It might make more sense to go to a '2.0' and align TMS with 
> the 0,0 top left that everyone else has gone with. (WMTS, 
> Commercial Providers, OSM, etc.)
> 
> -- Chris
> 
> > P.
> > 
> > On Mon, Sep 13, 2010 at 12:21 AM, Oliver Tonnhofer 
> <olt at bogosoft.com> wrote:
> >> Hi,
> >> 
> >> I think we need to differentiate between active and 
> passive TMS implementations. When a tile is missing for an 
> active implementation, it often means a cache-miss (unless 
> the server knows more about the coverage) and the tile needs 
> to be created. For passive implementations, i.e. bunch of 
> static files, there is no way to create that tile.
> >> 
> >> On 13.09.2010, at 04:12, Matt Priour wrote:
> >>> I think requiring servers to actually cache individual 
> empty images for each empty tile is not useful.
> >> 
> >> Well, that's an implementation detail. MapProxy can store 
> symbolic links for all tiles that only contain one color 
> (empty, water, forest, etc.). So you don't waste space, only inodes.
> >> 
> >>> The server response, whether it is an no data image for 
> each request, a redirect to a common no data image, or a 
> simple 404; is not really all that important either. Clients 
> should be able to handle any of those type responses and 
> behave appropriately.
> >> 
> >> 
> >> Yeah, I agree with you, clients should be prepared to get 
> different replies. But I would like to suggest that returning 
> empty images should be 'best practice'.
> >> 
> >> Regards,
> >> Oliver
> >> 
> >> 
> >> 
> >> _______________________________________________
> >> Tiling mailing list
> >> Tiling at lists.osgeo.org
> >> http://lists.osgeo.org/mailman/listinfo/tiling
> >> 
> > _______________________________________________
> > Tiling mailing list
> > Tiling at lists.osgeo.org
> > http://lists.osgeo.org/mailman/listinfo/tiling
> 
> _______________________________________________
> Tiling mailing list
> Tiling at lists.osgeo.org
> http://lists.osgeo.org/mailman/listinfo/tiling
>