On Feb 9, 2010, at 10:52 PM, percy wrote:
Interesting discussion...
I was asked to author a couple of chapters on Python for GIS for a
"GIS
Programming: concepts and applications" book a while back, after the
original co-author stepped down.
I agreed to do it on the condition that I could at least include side
boxes or extra material on how to do things with open source tools,
in
addition to the ESRI arcgisscripting library. Happily, the publisher
thought that was a great idea.
I have been meaning to ask Sean and Howard to review my chapters when
they are ready, and in light of the current discussion it will be
even
more important. I want to make sure the audience gets the best set of
tools possible!
It will be a few more months before my chapters are ready for review,
but I look forward to your suggestions for improvement.
Cheers,
Percy
Percy,
Good luck on the book. I'd be willing to review, but since I don't
have ArcGIS installed I may not be of much use.
The ArcGIS scripting API simply sucks. A good book could go beyond
documenting the suckiness and show users how to make it suck less.
Maybe even give them open source code to make it suck less. For
example: a package of adapters like the one I wrote about at:
http://sgillies.net/blog/953/iterators-again/
The GeoEco project looks like it's full of ways to make
arcgisscripting suck less:
http://code.env.duke.edu/projects/mget
although I can't recommend its module and method naming practices. PEP
8, people.
Showing people how to distribute and install Python tool box
components using distutils/distribute and pip (assuming its possible)
could be neat too.
Cheers,
--
Sean
