[Proj] Old problems

Gerald I. Evenden gerald.evenden at verizon.net
Sat Dec 31 19:55:04 EST 2005

On Saturday 31 December 2005 05:29 pm, Paul Kelly wrote:
> Hello Gerald
> There is some documentation on the discrepancies between PROJ.4 and
> libproj4 and the versions used in GRASS on Thierry Laronde's KerGIS site
> at: http://www.kergis.org/doc/proj.html

Indeed, you bring up the "issues" debated to some length many months ago.

My position, as then, is that the operation of projections is separate from 
the activity of datum changes.  For those whose data is in their desired 
datum (if they even know about it) and those working with small scale mapping 
are not concerned with datums.  In addition, the concepts and math 
underlaying the two issues do not intersect.

I think everyone came same conclusion but some effort had gone into the 
merging of the two processes in the remotesensing version and nobody had the 
time nor inclination to undo the merging.  Alas, we are still in the same 
situation today.

BTW: yes, "small is beautiful."  I spent several years working on a system 
where there were extra charges if your program exceeded 100k. (Seems 
unbelievable)  It was considered cheaper if I spent extra time becoming a 
whiz with JCL, tight FORTRAN coding and hours dreaming up ways to shoehorn a 
big computational effort into 100k.

> I provided a lot of the information and assumptions contained in the
> description there so I'd be interested to hear if you agree or disagree
> with it.

I think we are on the same page.

> KerGIS is kind of a spin-off GIS from GRASS, based on the last public
> domain version of the source code before it was licenced under the GNU GPL.
> Best regards,
> Paul

What I would really like to see is PROJ.4 and libproj4 merged into one 
distribution.  I would also like to see a libdatum.
Jerry and the low riders:Daisy May and Joshua
"The being cannot be termed rational or virtuous,
who obeys any authority, but that of reason."
---Mary Wollstonecraft 1792

More information about the Proj mailing list