[postgis-devel] PostGIS Development Roadmap

Mark Cave-Ayland mark.cave-ayland at siriusit.co.uk
Tue Apr 22 04:39:20 PDT 2008


Kevin Neufeld wrote:
> These all sound good to me.  Can we add to this list a documentation 
> overhaul?
> Personally, I would love to see the documentation become much more 
> exhaustive in it's method descriptions (a more professional look 
> wouldn't hurt either).  I'm envisioning something where we have a good 
> description and a comprehensive example (complete with pictures) for 
> *every* method.  When I say "good description", I mean something that 
> includes all the nuances and subtleties associated with a particular 
> method.  IE. ST_BuildArea(Geometry) should at least mention that it 
> quietly discards all unused linework (dangles and cutlines) that aren't 
> used in the resultset :)  As for a professional look ... honestly, 
> DocBook doesn't do it for me.  I'm not sure what to suggest here, but 
> maybe in the interim we could use something like MediaWiki to gather our 
> thoughts together (aka PostGISpedia).  I realize we already have a wiki, 
> but I personally think it has a clumsy and awkward feel to it.  It's 
> also next to impossible to quickly attach pictures to enhance method 
> descriptions.

For me, docbook works well for the reasons that the XSLT transformation 
can be customised fairly well, plus the documentation builds under both 
Unix types and Win32 without too much fuss. So if we were to find a 
replacement, it would need to work on all platforms in this manner.

> Additionally, to expand the Regression Tests bullet item, can we add 
> Unit Tests for every method?  This would save us the embarrassment of 
> breaking something simple we really should have caught in the first 
> place (i.e. reference ST_EndPoint() that broke in 1.3.0 or 1.3.1).

Well, sure ;)

The development roadmap was my personal list of things we really need to 
get done, so if you'd like to volunteer to redesign the regression 
tests/rewrite documentation then I have no isue with that.

With regards to the particular ST_EndPoint() bug, this was unusual in 
that the bug had been there a long time - it was only due to some change 
of memory alignment within the 1.3 series that we started to notice it 
in the first place :(

Also, if everyone is in agreement with the development roadmap, it's 
probably worth a brief email on -users explaining how we plan to move 
forward.


ATB,

Mark.

-- 
Mark Cave-Ayland
Sirius Corporation - The Open Source Experts
http://www.siriusit.co.uk
T: +44 870 608 0063



More information about the postgis-devel mailing list