[georss] Intended lat/long order

[Marten Hogeweg]

If only earth was flat again... 

-----Original Message-----
From: Ron Lake [mailto:rlake at galdosinc.com] 
Sent: Thursday, November 08, 2007 10:06 PM
To: Marten Hogeweg; Christopher Schmidt
Cc: georss at lists.eogeo.org
Subject: RE: [georss] Intended lat/long order

Yes - indeed - doing 3 things at once is not always wise.

Of course the set of questions in general may include:

- what is the coordinate order
- what is the datum
- what is the prime meridian 

So a definition is needed.

Any excessive back and forth was my fault by typing the wrong code and
confusing poor Chris.

R

-----Original Message-----
From: Marten Hogeweg [mailto:mhogeweg at esri.com]
Sent: November 8, 2007 9:57 PM
To: Ron Lake; Christopher Schmidt
Cc: georss at lists.eogeo.org
Subject: RE: [georss] Intended lat/long order

Actually, starting from CRS EPSG=4326
http://www.epsg-registry.org/export.htm?gml=urn:x-ogc:def:crs:EPSG:4326

It seems that this uses ellipsoidalCS EPSG=6422
http://www.epsg-registry.org/export.htm?gml=urn:x-ogc:def:cs:EPSG:6422

The difference being that EPSG 6422 is 2-dimensional, while EPSG 6423 is
3-dimensional.

All this back and forth may have been avoided if the GeoRSS site simply
stated the order of the coordinates in addition to providing samples.

Marten 

-----Original Message-----
From: georss-bounces at lists.eogeo.org
[mailto:georss-bounces at lists.eogeo.org] On Behalf Of Ron Lake
Sent: Thursday, November 08, 2007 9:42 PM
To: Christopher Schmidt
Cc: georss at lists.eogeo.org
Subject: Re: [georss] Intended lat/long order

Chris:

The axis order is determined always by the CS (in this case is
EPSG:6423).  The CS is just a part of the CRS.  The definitions are
normalized so that a CS (like 6423) can be used in MANY CRS.  Now geoRSS
uses CRS 4326 which uses CS 6423 - but the CRS also determines many
other things like the datum, prime meridian etc.  Axes are just part of
the definition and because things are normalized it is buried a bit
deeper. You can of course use XSLT script to assemble the pieces in
querying the registry.

The UI makes this clearer as you just expand the CRS by clicking the
plus-minus buttons.  I sent screen shots in the previous e-mails.

Does that make more sense?

You are write the URN's are NOT links - they are URN's. You can refer
directly (a URL) to the CS definition as in
http://www.epsg-registry.org/export.htm?gml=urn:x-ogc:def:cs:EPSG:6423

Cheers

Ron

-----Original Message-----
From: Christopher Schmidt [mailto:crschmidt at crschmidt.net]
Sent: November 8, 2007 9:30 PM
To: Ron Lake
Cc: Christopher Schmidt; georss at lists.eogeo.org
Subject: Re: [georss] Intended lat/long order

On Thu, Nov 08, 2007 at 09:16:05PM -0800, Ron Lake wrote:
> Hi,
> 
>  
> 
> The order is specified in the GML encoding for the CS (Note that the
CRS
> is made of many pieces of which the CS is one).  If you expand the CRS

> in the UI you will see:

Okay. I'm going to start here:

http://www.epsg-registry.org/export.htm?gml=urn:x-ogc:def:crs:EPSG:4326

This is a CRS. I think.

>From there, I can get to
http://www.epsg-registry.org/export.htm?gml=urn:x-ogc:def:cs:EPSG:6422

but that's 6422, not 6433. 6422 seems to have the information I want,
but you said that I should be looking at 6423: How should I have gotten
there from
http://www.epsg-registry.org/export.htm?gml=urn:x-ogc:def:crs:EPSG:4326
?

Oh, I think maybe I see: I should have followed
<epsg:sourceGeographicCRS xlink:href="urn:x-ogc:def:crs:EPSG:4979"/> to
http://www.epsg-registry.org/export.htm?gml=urn:x-ogc:def:crs:EPSG:4979
from which I could get to
http://www.epsg-registry.org/export.htm?gml=urn:x-ogc:def:cs:EPSG:6423.

For the record, I have no clue how I'd have found that without knowing
it was there and trying every link. (And of course, they're not actually
links, so I can't actually dereference them in a machine-friendly way.)

So which does GeoRSS use? EPSG:4326, as I've always thought? Or
EPSG:4979, if you think that we're talking about EPSG:6423 for a CS? 

> This is the CS definition that is reflected in the UI.  You will see
Lat
> precedes Lon.  Note that many other CRS could use the same CS.
> 
> Sorry if I was not clear the first time.

Well, from what I'm told, there was a screenshot or something in your
post, which would change things for me. I wasn't looking at the UI, I
was looking at the machine-readable metadata. 

Regards,
--
Christopher Schmidt
Web Developer
_______________________________________________
georss mailing list
georss at lists.eogeo.org
http://lists.eogeo.org/mailman/listinfo/georss