[MetaCRS] Using CS-Map to retrieve WKT for EPSG CRS codes: Help?

[Christopher Schmidt]

On Wed, Dec 17, 2008 at 03:21:42PM -0800, Norm Olsen wrote:
> Hello Christopher . . .
> 
> Two points, quickly, as I need to earn my paycheck:

What, we don't get paid to sit around and talk shit about WKT ... or
SRSes... all day? :)  (How can we get that job?)

> 1> the newer Oracle WKT has references to OPERATION CODE, not OPERATION METHOD CODE.  Thus, that is rather painful, especially if you are not keen on becoming reliant on have a copy of the EPSG database available (not to mention that EPSG is in ACESS 2000 format).

Well, certainly this kind of thing is hard :) I'm hopeful that what can
be done is something like, create a list of (Flavor,PROJECTION String),
and map that to some generally agreed upon projection string... but no
need to look at that more until I get deeper into it. For now, I'll
accept that all this stuff is a lot harder than I ever dreamed it was
when I hacked together sr.org in a couple hours a year ago :)

> 2> The Bursa/Wolf is an approximation of the Seven Parameter
> transformation.  As the rotation angles get beyond one or two seconds
> and/or the position being converted moves away from the origin point,
> the differences can become very noticeable, i.e. in the one or more
> meter range.  In the oil business (i.e. the P in EPSG is for
> petroleum) this may not be a big deal, but for many others it is a big
> deal.  The TOWGS84 feature of WKT just doesn't cut it for many users.

Gotcha.

So, I guess the question I have is: If WKT doesn't cut it as is, are
there *already* alternatives? Is there a better serialization format for
SRSes?

I *think* the answer is "no"; at least, I'm not aware of one that I
think of as generally better. There are clearly other alternatives --
the GML CRS description language comes to mind -- but I'm not convinced
these are actually anything more than a more verbose translation of the
WKT.

If there isn't a solution to programatically describing projections that
is sufficient to accurately describe the projections in a way that can
be interchanged between applications, it seems to me that is something
that is important to improve upon in the long run. Unless we just want
everyone to switch to Web Mercator, or EPSG:4326 :)

Regards,
-- 
Christopher Schmidt
MetaCarta