[mapserver-dev] Tile Access API

[Stephen Woodbridge]

Christopher Schmidt wrote:
> On Thu, Apr 17, 2008 at 02:30:44PM -0500, Stephen Woodbridge wrote:
>> Christopher Schmidt wrote:
>>> On Thu, Apr 17, 2008 at 02:02:12PM -0500, Stephen Woodbridge wrote:
>> No one said IT was smart, look at the blind trust of .NET code. ;)
>>
>> But seriously, having a small purposed C application is easier to vet, 
>> if only emotionally, than loading 20+ additional packages and worrying 
>> about how each of them is configured from a security point of view.
> 
> How does TileCache result in "20+ additional packages?" I think you're
> jumping into the GDAL topic, where you've already been told that this is
> a packaging problem, not a GDAL problem: TileCache is the same,  
> 
> Depends: ${python:Depends}
> Recommends: python-imaging
> Suggests: python-mapscript (>= 4.10), python-mapnik, memcached,
>           python-memcache, python-boto, python-flup, python-paste, 
>           python-wsgiref
>   -- http://svn.tilecache.org/trunk/debian/control
> 
> You can install any of those, and you get support for various parts of
> the code, but there is no need to install them as dependancies for the
> code.
> 
> Anyway, we're arguing in circles. I understand the problem, I just don't
> think that the described solution (mapcache.cgi) is really the most
> sane solution to the problem. 
> 
> Regards,

OK, point taken and we should move on.
BTW, I'm not against this RFC per say, just think that it has limited 
value but would not block it. Clearly there are people I respect that 
feel there is value in moving this RFC forward.

Thanks,
   -Steve