MapML was: Cargo Or Cult

Hi Alistair,

Very interesting work!  And thanks for sharing it with us.  Were you at the workshop in London?  I don't recall meeting you; I apologize if you were there.  I was a bit excited.  It would be great to capture what your geo-map design is on the wiki.  Although we mentioned Web Components as a candidate technique for a geo-map proof-of-concept, I don't recall anyone at the workshop saying they had polymer or other experience.  So you are showing us the way!

I really like the simplicity of the geo-map tag that you present i.e. a simple zoom x y attribute group.  Can you tell us more about your thinking there?  It seems like kismet :-).

Have you thought about encapsulating map semantics in a media type before?  My thinking is that in order to be generally useful, servers running services like your wms and open street map tiles would have to forego dictating to the world what their URI structure should contain, and instead 'map' their service / interaction details to a file format that could be presented in the browser DOM.

It might look something like this:

<map>
  <head>
    <meta http-equiv="Content-Type" content="text/map;projection=epsg:nnnn;scale=nnn.nnn"/>
  </head>
  <link rel="east" href="link to a map document to the east"/>
  <link rel="west" href="link to a map document to the west"/>
  etc
  <feature>
     <!-- either an inline feature or a link to a tile/map image with embedded/enveloping georeferencing -->
     <!-- requests of course could be URIs to WMS-s, WFS-s, openstreetmap / wmts / geoservices tiles  etc -->
  </feature>
  <feature>...</feature> <!-- as many as necessary -->
 <link rel="next" href="link to next set of tiles or features to fulfill the request in the view of the server"/>
</map>

In the case of link@href to documents of the same media type as this one, @type could be omitted.  Not so for other links, where @type to image/png etc would be necessary to convey to the browser if it should dref something it understands or not.

I could try to sketch this out on the wiki in your stated timeframe too.

What do you / anybody in the group think?

Cheers,
Peter

________________________________
From: Al Phillips [alphillips101@gmail.com]
Sent: April 5, 2014 10:01 PM
To: Rushforth, Peter
Cc: public-maps4html@w3.org
Subject: Re: Cargo Or Cult

Hi Everyone,

Thanks for starting this Peter.
I have also been thinking about this for some time.

Last year I experimented with the idea using Web Components (Polymer) which I called funnily-enough <geo-map>. You can see it here https://github.com/alphillips/geo-map
Looks like there are similarities with some of the ideas at the Bar Camp workshop.

I will try to put some of my ideas together on the wiki over the next week or so.

cheers
Alistair Phillips


On 5 April 2014 00:45, Rushforth, Peter <Peter.Rushforth@nrcan-rncan.gc.ca<mailto:Peter.Rushforth@nrcan-rncan.gc.ca>> wrote:
Hi folks,

Thanks for participating in the Maps 4 HTML CG.

I have been thinking about this for quite a while now, and some ideas have come and gone for me.  The way I would like to proceed is a little bit casual to begin with.  It is a small group so far, which is great.  I think that we need to have some casual discussions to start with.  As a way of getting things going, I've posted the notes I made about the Bar Camp proceedings from the Linking Geospatial Data workshop of March 2014 on our wiki.    I am open to doing Google hangouts or other means of real-time communication.

Also, I have created a pair of pages which I have called "Cargo" and "Cult".  Although slightly whimsical in name, the ideas listed thereon may also be controversial to some, less so to others.  Bring the discussion to this list, at least for the time being.  Note that anyone can post and read the archives, member or not.  Only members can edit the wiki.  If you do so, please sign your edits with ~~~.

Also, anything else you'd like to discuss related to the practicalities or process of bringing mapping to the mainstream Web, please discuss here.

Regards,
Peter Rushforth

Received on Sunday, 6 April 2014 16:52:27 UTC