QLD DNRMMRRD CC BY Waiver

We have just received the CC BY waiver from Department of Natural Resources and Mines, Manufacturing and Regional and Rural Development (DNRMMRRD) which covers all of their CC BY 4.0 licensed open data published at https://www.data.qld.gov.au/dataset/?organization=resources.

Satellite and Aerial Imagery published by the department is not covered.

This is something the community has been hoping to obtain for a while, I sent a formal request for the waiver in May 2017 (8 years and 6 months ago) and at the time the department (DNRM) stated that they believed that a CC BY license is sufficient for use of their data in OpenStreetMap and they did not accept that OpenStreetMap could not use their data under CC BY, which is fine that’s their interpretation of the license but the issue was that the OSMF LWG took a different interpretation that we cannot use CC BY data directly in OSM, and this issue became a sticking point that we could not get past after considerable efforts, including a follow up request in 2018.

Following this, Nemanja Bracko continued to work on obtaining the waiver and eventually in 2020 they received a letter Naming roads/streets in Queensland, Australia · Issue #49 · microsoft/Open-Maps · GitHub however subsequent review by the LWG determined that this was still not sufficient, leading to the matter once again being stalled.

Following this, Graeme Fitzpatrick continued working on the issue around 2023, instead trying to work through the Premier and continuing into 2024 through Mark Wheeley at the Department of Communities, Housing and Digital Economy who manage the Queensland Government Open Data Portal.

It was through this approach and the ongoing work behind the scenes last year and throughout this year that we have now received the completed CC BY waiver.

Thank you to everyone involved in making this happen.

I’ll be working through what layers we can get into the editors as background layers and overlays, at the same time it opens up using these datasets in import proposals.

4 Likes

Great effort folks - well done to all involved.

1 Like

I’ve added overlays for Addresses, Land Parcels and Roads and Tracks and base maps for Lite and Topo basemaps into Editor Layer Index (iD) and into JOSM’s index.

[edit] the iD preview site https://ideditor.netlify.app/ is showing the updated layers

1 Like

Further correspondence with LWG and NRMMRRD (https://wiki.openstreetmap.org/wiki/Australian_Data_Sources/Queensland/NRMMRRD) has given the OK to use their CC BY 4.0 licensed imagery to trace from, so I’ve now added that to JOSM and iD, though not yet as “Best” / the default, any thoughts on that?

Andrew, just checked & “Qld Imagery” is showing in JOSM, but not iD?

Unfortunately, it’s also rather old which surprises me a bit - in my area, definitely prior to April 2023!

It’s only on the iD preview site https://ideditor.netlify.app/ until iD do a new release to openstreetmap.org.

Yes that’s expected. They collect imagery through the Spatial Imagery Services Program and retain that under commercial terms - though they also publish this to Queensland Globe so the public can view it. The imagery we can use is the SISP imagery which is greater than 3 years old that they publish as CC BY. It’s this older CC BY imagery which is in the editors, not the newer imagery you see on Queensland Globe.

The newest imagery we can use is from 2022, but it ranges from 2013 to 2022, in rural areas it’s mostly 2017-2020, urban areas are mostly 2021 (I documented where I sourced this vintage information from on the wiki).

You’ll still want to review this against Bing and Esri to decide what’s going to be best to traces from on a case by case basis.

Hard to say. I can think of at least three criteria you might consider; accuracy, clarity, and recency.

Having looked around QLD it seems that in metro area Bing and SISP are much the same in terms of accuracy and clarity, but the SISP images are going to be at least three years old. In regional areas the SISP images are clearer, but if you look at very remote areas SISP might be satellite imagery and weirdly Bing has quiet clear imagery (clearer than the stuff they have for some QLD towns).

Given that most editing is going to be in metro areas, maybe just leaving Bing as best would be suitable.

EDIT: I just used your script to download the metadata. It confirms that the “2017” area is 2.4m resolution satellite. It does look like you could extract boundaries for the various outback towns that have 10cm resolution to make a “best” layer, but that would require two entries in the ELI.

1 Like

I reckon it’s not worth the effort… and based on your comment best to just stick with Bing as the default.

1 Like