[talk-au] Proposed import of SA waterbodies data

Wed Jan 21 04:14:01 UTC 2015

Daniel,

The detail is very good. Here is a screenshot from JOSM:

https://drive.google.com/open?id=0B1JwNHL1bER0UzlIcFJzTmNFdms&authuser=0

The cyan lines are the datasa waterbodies, overlaid on an OSM map of Mount Barker. (By the way, this is showing part of the complete datasa waterbodies data, not the subset I’m proposing to upload.)

Importing the different types of features separately would be more technically difficult, because the different types of features share nodes where they touch each other. (It would be hard not to duplicate the nodes over the multiple uploads). But doing the largest objects first is definitely just as easy as what I’m proposing, so long as I include any features connected to those large objects at the same time.

I could include all the small unnamed dams. Perhaps they could be part of future stages of the import. My idea was to keep the first import relatively small, by including only the “important” features. I agree that my definition of important is fairly arbitrary, and I would be willing to change it. I’m a little bit scared by the fact that there are 90,000 dams in the dataset. I don’t want to be accused of spamming OSM!

Henry

From: Daniel O'Connor [mailto:daniel.oconnor at gmail.com] 
Sent: Tuesday, 20 January 2015 10:33 PM
To: Henry Haselgrove
Cc: talk-au at openstreetmap.org
Subject: Re: [talk-au] Proposed import of SA waterbodies data

Secondly, is it better to do 4x imports for each feature set; starting at the most high level feature layer?

I mean there are commonly things you just don't expect, so a trial run with large obvious features is probably best; pushing further into detail.

Third one; why just features with names?

Part of the motivation for mapping smaller dams has been for things like XPlane (realistic terrain); and excluding unnamed features would potentially make the data set less useful than it could be

On Wed, Jan 21, 2015 at 1:58 PM, Daniel O'Connor <daniel.oconnor at gmail.com <mailto:daniel.oconnor at gmail.com> > wrote:

What's the level of detail like for small bodies of water?

For example, I've put in some manual effort around small dams in places like http://www.openstreetmap.org/#map=14/-35.0894/138.8376 - i'd be curious to see a preview of this sort of area with the merged data.

On Wed, Jan 21, 2015 at 1:30 PM, Henry Haselgrove <haselgrove at gmail.com <mailto:haselgrove at gmail.com> > wrote:

Hi All,

The South Australian government provides a dataset called “Waterbodies” on the website data.sa.gov.au <http://data.sa.gov.au> . 

It contains information about approximately 150,000 lakes, reservoirs, wetlands, and dams throughout the state (and in some margin around the state). The SA government gave explicit permission for data from data.sa.gov.au <http://data.sa.gov.au>  to be used in OSM.

I  have created the following wiki page to describe a plan I have to import some of the waterbodies data:

https://wiki.openstreetmap.org/wiki/Import/South_Australian_Waterbodies

Any feedback you have about the plan would be greatly appreciated!

Thanks,

Henry Haselgrove 

_______________________________________________
Talk-au mailing list
Talk-au at openstreetmap.org <mailto:Talk-au at openstreetmap.org> 
https://lists.openstreetmap.org/listinfo/talk-au

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/talk-au/attachments/20150120/9290aff1/attachment-0001.html>