[Talk-us] NHD: python or java?

[James Umbanhowar]

On Friday 07 August 2009 04:20:11 Sam Vekemans wrote:
> On Fri, Aug 7, 2009 at 1:05 AM, Dale Puch <dale.puch at gmail.com> wrote:
> > On Fri, Aug 7, 2009 at 1:29 AM, David Carmean<dlc at halibut.com> wrote:
> > > Which tool is most used, at the moment, for importing NHD data? 
> > > Chris's
> >
> > python stuff
> >
> > > or Ian's java stuff? :)
> > >
> > > For those using Ian's stuff, who's developed the most complete rules
> >
> > file?
> >
> > > _______________________________________________
> > > Talk-us mailing list
> > > Talk-us at openstreetmap.org
> > > http://lists.openstreetmap.org/listinfo/talk-us
> >
> > I got the impression that Chris's python is tailored made for the NHD
> > where Ian's java is geared towards making generic shape imports a lot
> > easier for the novice or non programmer.
>
> Exactly, the java version, is really best for manual importing of the data,
> and the python is for when your using bulk_upload.py script.  For when you
> dont need to mess with what is already existing. ie. in many areas, were
> there are no rivers / waterbodies to conflict with, a full upload is fine.
> But for this canvec stuff i'm working with (importing rail-lines,
> powerlines, sport_park_fields etc, i need to watch for what other have
> done).  So bulk_uploading the data wont help. (ie. importing rivers in the
> middle of a city, where other mappers might have done just parts of the
> river. (perhaps in more detail than the import data available)
>

I've been using the python scripts and uploading with the balrog scripts 
(http://www.openstreetmap.pl/balrog/bulkupload/) , they seem to work more 
reliably for me.

To avoid collison with others' data, I have noted the presence of previously 
uploaded stream and waterbody data and gone back after import and deleting the 
usually less correct NHD data.  This is actually not that big a deal as other 
users usually focus primarily on major rivers and large bodies of water.

> I recommend maintaining just 1 version of the data for each script, with a
> comparision chart (like i have with the Google Docs chart), as they both
> need to have all the same tag-matches, and just keep track of the script
> version.
>
> How are you keeping track of the National Dataset?
> For Canada, Im thinking at using a GoogleDocs spreadsheet might be easier,
> since there are so many different tiles (canada is made up of an NTS Grid
> of only 4,984 unique tiles) where some will be uploaded at different levels
> of completness... and different levels of data available.  So i have a
> chart with all the types of data..  then what i can do is give access to it
> for whoever wants it.
> http://spreadsheets.google.com/pub?key=tmY7V02fBT8C8vimCf8ioXg&output=html
>

The completed subbasins are being tracked on the NHD Wiki page 
(http://wiki.openstreetmap.org/wiki/NHD#Completed_Basins).  In the case of 
NHD, I think most (all) are uploading the high resolution data, though these 
seem to not be available for all subbasins.

A spreadsheet might be a better solution...

>
> Hope that helps,
> Sam
>
> > --
> > Dale Puch
> >
> > _______________________________________________
> > Talk-us mailing list
> > Talk-us at openstreetmap.org
> > http://lists.openstreetmap.org/listinfo/talk-us

James