[Mass-Market-GEO] Fwd: Re: [Tc] OWS Context Documents

Rushforth, Peter Peter.Rushforth at NRCan-RNCan.gc.ca
Fri Jan 14 10:47:52 EST 2011


Simon,

This late! reply is more of a question: are the MIME types generated in this 
manner going to be registered with IANA, and if not will they be listed
in the OGC naming authority web pages or somewhere central?

I see that GML has a mime type, but it is not currently in the IANA registry.

http://www.iana.org/assignments/media-types/application/

Thanks
Peter

> -----Original Message-----
> From: Simon Cox [mailto:simon.cox at jrc.ec.europa.eu] 
> Sent: July 8, 2010 09:27
> To: Rushforth, Peter; 'Pat Cappelaere'
> Cc: mass-market-geo at lists.opengeospatial.org
> Subject: RE: [Mass-Market-GEO] Fwd: Re: [Tc] OWS Context Documents
> 
> Note that the TC recently approved a policy change to require 
> editors to specify a MIME-type for OGC content. 
> 
> --------------------------------------------------------
> Simon Cox
> 
> European Commission, Joint Research Centre Institute for 
> Environment and Sustainability Spatial Data Infrastructures 
> Unit, TP 262 Via E. Fermi, 2749, I-21027 Ispra (VA), Italy
> Tel: +39 0332 78 3652
> Fax: +39 0332 78 6325
> mailto:simon.cox at jrc.ec.europa.eu
> http://ies.jrc.ec.europa.eu/simon-cox 
> 
> SDI Unit: http://sdi.jrc.ec.europa.eu/
> IES Institute: http://ies.jrc.ec.europa.eu/
> JRC: http://www.jrc.ec.europa.eu/
> --------------------------------------------------------
>  
> Any opinions expressed are personal unless otherwise indicated. 
> 
> -----Original Message-----
> From:
> mass-market-geo-bounces+simon.cox=jrc.ec.europa.eu at lists.openg
> eospatial.
> mass-market-geo-bounces+org
> [mailto:mass-market-geo-bounces+simon.cox=jrc.ec.europa.eu at lis
> ts.opengeospat
> ial.org] On Behalf Of Rushforth, Peter
> Sent: Thursday, 8 July 2010 15:17
> To: Pat Cappelaere
> Cc: mass-market-geo at lists.opengeospatial.org
> Subject: Re: [Mass-Market-GEO] Fwd: Re: [Tc] OWS Context Documents
> 
> Pat,
> 
> I'm saying that OWS context documents, if they had their own 
> media type, would be ideal candidates for media link entries 
> in AtomPub feeds.  
> 
> The linking of documents (hypertext) is fundamental to REST.  
> This is the case whenever your browser loads a web page with 
> an img element in it, a "round trip" to the server, as you 
> call it, is performed by the browser.
> 
> If a hypothetical "map browser" loaded an atom feed with 
> links to context documents, it could process those documents 
> appropriately to the situation / user settings etc.  If those 
> documents were large, it might be less efficient (than even 
> the round trip to the server!) to force that client to parse 
> all those documents even when only a list of summaries was desired.
> 
> Cheers,
> Peter
> 
>  
> 
> > -----Original Message-----
> > From: Pat Cappelaere [mailto:pat at cappelaere.com]
> > Sent: July 8, 2010 07:33
> > To: Rushforth, Peter
> > Cc: Panagiotis (Peter) A. Vretanos; Larry Bouzane; 
> > mass-market-geo at lists.opengeospatial.org
> > Subject: Re: [Mass-Market-GEO] Fwd: Re: [Tc] OWS Context Documents
> > 
> > Peter,
> > 
> > So what you are recommending is to have an ATOM feed of 
> "OWS Context" 
> > resource entries that will have referencing links to an alternate 
> > representation XML document (and not binary) and forcing a 
> round trip 
> > to the server every time (rather than putting the semantic 
> > representation inline).  Is that correct?
> > 
> > Pat.
> > 
> > On Jul 6, 2010, at 2:56 PM, Rushforth, Peter wrote:
> > 
> > > Peter, Pat and Larry,
> > > 
> > > My comments are inline. 
> > > 
> > >>> 
> > >>> With all the activity in OGC around ATOM lately, I am
> > working if we
> > >>> are using it as intended and I think this might be an
> > >> example of how not to use it.
> > >>> 
> > >>> My understand of ATOM is that it is a standardized
> > >> syndication format
> > >>> which means that it is an XML encoding of a summary
> > >> description of a
> > >>> resource and not "the" resource itself.
> > >> 
> > >> An atom entry is the resource itself (unless you make an 
> effort to 
> > >> not disclose it).  So you are safe.
> > >> An ATOM feed contains many entries.  It is up to you to
> > decide what
> > >> that is (and you are free to extend the entries to 
> include any XML 
> > >> you want including the OWS context document itself)
> > >> 
> > >>> A big benefits of ATOM, of course, is that it can be used
> > >> to uniformly
> > >>> describe all kinds of content ... GML features, OWS context
> > >> documents,
> > >>> etc ...
> > >> 
> > >> Beyond ATOM, there is ATOMPUB too!  to publicize the resource 
> > >> collections you may want to publish.
> > > 
> > > AtomPub is a great place to start when thinking about
> > publishing media
> > > types, like OWS Context.  Document type semantics are 
> specified by 
> > > MIME types http://www.iana.org/assignments/media-types/.
> > OWS Context
> > > documents and other OGC types could have their own MIME types as 
> > > appropriate.
> > > 
> > >> 
> > >> 
> > >>> In this case you seem to be saying that ATOM is becoming "the" 
> > >>> resource which might not be the best practice.
> > > 
> > > The atom entry is a *representation* of the resource, which
> > contains
> > > some standard metadata, plus either the content or a link to the 
> > > content.  There might be other possible representations of
> > the same resource.
> > > 
> > >> 
> > >>> For example, someone querying a catalogue for context
> > >> documents could
> > >>> get the response expressed in ATOM.  Each entry in the
> > >> resulting feed
> > >>> would describe an OWS context document and also contain a
> > >> link so that the "actual" resource (i.e.
> > >>> an OWS Context Document) can be retrieved.
> > > 
> > > If the Context document has it's own media type, then 
> AtomPub says 
> > > that the atom entry is a 'media link entry' which links to
> > the content.
> > > 
> http://bitworking.org/projects/atom/rfc5023.html#media-link-entries
> > > 
> > > The atom:content at type attribute could specifiy the media
> > type expected
> > > in response to de-referencing the link contained in
> > atom:content at src.  
> > > In this manner, a client can choose to retrieve only the
> > media types
> > > it understands how to process.  Of course,  one could
> > always use text/xml for a Context document, but that doesn't imply 
> > context semantics.
> > > 
> > >>> 
> > >> 
> > >> You certainly could have a link to an OWS document for 
> each entry. 
> > >> But what would this buy you?
> > >> Might as well have it inline.
> > > 
> > > Sometimes the size of several documents in a feed makes the
> > transmission bulky.
> > > Also, if the media type is such that it is best represented
> > as binary,
> > > a media link entry annotated with the mime type works well.
> > > 
> > > Peter Rushforth
> > > Technology Advisor / Conseiller technique GeoConnections / 
> > > GéoConnexions
> > > 650-615 Booth St. / rue Booth
> > > Ottawa ON K1A 0E9
> > > E-mail / Courriel: Peter.Rushforth at NRCan-RNCan.gc.ca Phone /
> > > Télephone: (613) 943-0784 Fax / telecopier:  (613) 947-2410
> > 
> > 
> 
> _______________________________________________
> Mass-Market-GEO mailing list
> Mass-Market-GEO at lists.opengeospatial.org
> https://lists.opengeospatial.org/mailman/listinfo/mass-market-geo
> 
> 


More information about the Mass-Market-GEO mailing list