dhis2-devs team mailing list archive

[Jason Pickering <jason.p.pickering@xxxxxxxxx>]

I think the biggest problem is that the underlying shape files
themselves have topological issues. This is a big problem specifically
with the WHO dataset, which has been composed of individual
submissions of country level data, and then sort of seamed together.
The GAUL dataset from FAO is supposedly topologically integral, and
should not suffer from spilters/gaps etc when simplified, IF, the
topology is preserved. This assumes that the original shape file is
topologically correct from the beginning, which is often not the case
with a lot of the data we deal with. As noted in the thread you sent
the DP algorithm does not maintain toplogy (assuming that exists from
the beginning).

You probably need to fix the topology on the underlying data, a
daunting task that is still on my old WHO to-do list from 2006. Good
luck.

Regards,
JPP


On Fri, May 21, 2010 at 3:40 PM, Knut Staring <knutst@xxxxxxxxx> wrote:
> Or rather, there is an open soruce python implementation that will
> work inside PostGIS (the first link below).
>
> We could use this here instead of FME since we have all our data in
> PostGIS, and as I said in an email last night, I actually think it is
> a good idea in the longer term to provide a version of DHIS 2 which
> runs on top of PostGIS (with full precision data) and also has
> Geoserver somewhat integrated. So people who have this version of
> DHIS2 could have a third option in addition to the two I outline
> below.
>
> On Fri, May 21, 2010 at 3:30 PM, Knut Staring <knutst@xxxxxxxxx> wrote:
>> Unfortunately, the links below and my earlier research indicates that
>> the "preserve shared boundaries" feature may not be available outside
>> of proprietary tools. Still, for many purposes, it will be fine to not
>> have this, especially at low levels of simplification (resulting in
>> larger GeoJSON files, which will of course be heavier to get over the
>> internet and for the browser to handle). It is largely an empirical
>> question - I noticed the cracks when trying to create GeoJSON that was
>> small enough for world maps, which was a true struggle. One option
>> would be to provide users with a choice of whether the shapefile they
>> upload should be simplified or not, and hardcode a reasonable
>> simplification. Their other option would be to use Mapshaper (or ask
>> someone with a proprietary tool) to simplify before uploading.
>>
>> Knut
>> ---------- Forwarded message ----------
>> From: Johan Lemarchand <johan.lemarchand@xxxxxxxxx>
>> Date: Fri, May 21, 2010 at 3:15 PM
>> Subject: Douglas-Peuker and Simplification Program
>> To: Knut Staring <knutst@xxxxxxxxx>
>>
>>
>> http://www.cartoweb.org/downloads/vertexsimplification/documentation.html
>>
>> http://groups.google.com/group/Google-Maps-API/browse_thread/thread/e5e865ddeeaa81ea?pli=1
>>
>>
>>
>> --
>> Cheers,
>> Knut Staring
>>
>
>
>
> --
> Cheers,
> Knut Staring
>
> _______________________________________________
> Mailing list: https://launchpad.net/~dhis2-devs
> Post to     : dhis2-devs@xxxxxxxxxxxxxxxxxxx
> Unsubscribe : https://launchpad.net/~dhis2-devs
> More help   : https://help.launchpad.net/ListHelp
>



-- 
--
Jason P. Pickering
email: jason.p.pickering@xxxxxxxxx
tel:+260968395190