[OSGeo-Edu] Superficial review of copyright issues related to collection and publication of education material on OSGeo Website

Simon Cropper scropper at botanicusaustralia.com.au
Mon Aug 2 22:35:57 EDT 2010


Hi Everyone,

          (C) Simon Cropper 2010. 
          CC-BY-NC-ND (http://creativecommons.org/licenses/by-nc-nd/3.0/)

This mega-email is aimed at encapsulating a range of issues related to 
copyright of educational material collected and/or published by the 
educational group of OSGeo.

These views, ideas and comments come out of a week or so of asking questions 
and attending the recent IRC Chat on Friday. Further down I summarise the few 
responses to my question about whether people have refrained from publishing 
material due to licensing issues. Following this I make an enormously 
presumptuous attempt at outlining what I think needs to be done.

Please don't flame me. I am including this feedback in good faith and in an 
attempt to value add to a topic where I think their is a lot of ambiguity and 
possibly a degree of naivety by prospective or existing authors/contributors. 

***************************************************************
PREAMBLE
***************************************************************

First I would like to review a number of basic concepts so that there is no 
ambiguity about what I am saying.
* published documents consist of two primary components: work prepared by the 
author and work that is not.
* it is expected that any published document that contains material not 
prepared by the author has permission to do so -- that is, it is not 
infringing someone's copyright.
* when a document is published the author indicates that it is released under 
a certain licence, and it is implied that all material contained within that 
document are also covered by the same licence or if not that some sort of 
notice would appear within the text.
* it would also be expected that tutorials that used data and encouraged the 
reader to reproduce the demonstrated tasks on the same dataset, would ensure 
that the data is covered by the same copyright or at least could be used 
without infringing copyright.

So far, I am sure, all of you are nodding your heads.

My personal objective at the moment is to prepare a series of tutorials on the 
use of gvSIG to complete set workflow processes I typically do on a day to day 
basis. My tutorials consist of text and images that illustrate a certain task. 
Three sets of copyright, or at least potential copyright, issues relate to 
these tutorials -- (1) my copyright related to the text, (2) images of the GUI 
interface of gvSIG, and (3) images of data before and after a certain task is 
performed.

Obviously a decision needs to be made by me about how I want to release my 
text. Typically at the moment I publish under a CC-BY-ND-NC licence. 
Essentially I release work for use by others but without the expectation it 
would be modified and updated in anyway.

Reproduction of the images of the gvSIG GUI is covered under the fair use 
provisions of the Australian Copyright Act 1968 (i.e. use for research or 
critical review). Regardless, I breached the issue with people in the project 
prior to starting my tutorials.

Data is another matter though. The copyright of all the data in Victoria, or 
at least the datasets I commonly use, is vested in the State of Victoria. So I 
need to get permission to reproduce these images. Usually permission is given 
but requires individual documents to be reviewed and vetted. That is, 
permission is 1:1. Each and every use is individually licensed.

At this stage, publication of a document would require that all these issues 
are addressed and resolved. The actual licence that the document is released 
under is usually dictated by the most restrictive copyright. For example, in 
my case release into the Public Domain is impossible because the State of 
Victoria has a strict 1:1 policy as described above.

OK, lets introduce another issue...

Tutorials are best accompanied by data that the reader can use to reproduce 
the techniques described in a tutorial.

It is implied in most situations that this data is also covered by the same 
licence as the accompanying tutorial and that the reader can download and use 
the information without infringing someone's copyright. In fact, this should 
not be assumed by the reader.

Copyright related to the dataset accompanying a tutorial should also be
explicitly documented, and allow for the uses proposed in the tutorial. 
Otherwise the author of the tutorial is exposing the reader and themselves 
litigation for infringement of copyright, if the copyright owner becomes aware 
of the tutorial and takes umbrage at the use or misuse of the data.

Bringing it back to me (no I am not narcissistic, it is just easier to 
illustrate an issue using personal experiences)...

First let me expand on my personal objective -- I want to illustrate the tasks 
I typically conduct on spatial data at work. This involves the collation of
disparate datasets, standardisation of these datasets (importing tables as 
event layers, transforming/reprojecting/warping shapefiles) and preparation of 
cartographic material for inclusion in my reports.

Although the type of data is not unique (e.g. ECW images, shapefiles, DWG files) 
the area they represent, the Spatial Reference System they are provided in, 
inherent errors in the data, as well as the features they represent are 
unique.

To simply illustrate a point. Why would I demonstrate to an Australian how to 
transform an American datum. What they need to know is how to reproject 
between AGD66 and GDA94 -- the theory, the techniques and the nuances.

My desire is to provide tutorials that demonstrate typical tasks conducted in 
the Natural Resource Management Industry in Australia on Australian data for 
Australians. If others can benefit from the tutorials, all the better, but this 
is not the primary objective.

Where my personal journey has come unstuck is that I am having real problems 
in getting even a small snippet of data* released by the State of Victoria. At 
a person-to-person level, all I get is encouragement. At a senior government 
level, there is an desire to move towards release of information to the public 
but when it comes to the crunch and data is provided it comes with numerous 
constraints preventing its release to the public.

So for me, the inability to secure suitable datasets, has resulted in my 
inability to create and publish workflow specific tutorials for my industry.

      * data in this context is base aerial photos, contours, cadastral data,
         vegetation maps, geology maps, waterways, flora/fauna point data 
         for a single area. 

***************************************************************
RESPONSE BY MAIL LISTS
***************************************************************

Last week I canvassed various forums, asking (1) under what licence people 
publish their data, and (2) whether anyone had not published something due to 
licensing issues, and if so why.

I had 5 responses plus my own input. Not a very big sample but enough to 
identify some issues.

Licences used for documents published
- Creative Commons Attribution - Non Commercial 2.5 Italy
- Copy Straight 
- Creative Commons Attribution - No Derivatives
- Creative Commons Attribution - Non Commercial - Share Alike
- Creative Commons Attribution 
- Creative Commons Attribution - No Derivatives - Non Commercial

Licenses used for miscellaneous posts, thread or blogs (like this one)
- not specified
- accept terms of service provider
- not specified
- not specified
- GNU Free Document License
- accept terms of service provider

Work not published due to licensing issues
- yes; licence of dataset prevented distribution
- no
- yes; no time to resolve copyright issues associated with data used
- no
- yes; licence of dataset prevented distribution
- yes, licence of dataset prevented distribution

What I glean from this sample is...
1. The Creative Commons licences are commonly used by authors. The bulk of the 
authors in this sample like to have works attributed to them and most don't 
want others to commercially exploit their works.
2. The bulk of the people sampled don't give much thought on what licence 
posts, threads, blogs, etc are published under. Respondents either said they 
did not know or just accepted the terms of service provider; or didn't answer 
the question. Personally, when I checked the places where I contribute, 
the licence was not specified. I presume then that the published work then 
falls back to the relevant copyright legislation in the country where the 
server that contains the data is situated. Messy, very messy.
3. It would appear that a lot of good work is not being made available because 
copyright issues relevant to the underlying datasets used in preparation of 
the works have not been resolved.

***************************************************************
OSGEO EDUCATION REPOSITORY
***************************************************************

A superficial review of the OSGeo Education Resource repository shows...

(a) that the CC-BY-SA is quite a common licence. In my mind it reaffirms my 
belief that one of the main means of gratification for contributors of the open 
source community is recognition of their input. That why it puzzles me why 
people are happy with allowing their works to be modified (that is, derivatives 
created) without ensuring the quality of the derived work is maintained.

(c) text based documents rarely, if at all, provided clear evidence that 
copyright permission has been obtained for the images included in the 
documents.

(d) where authors pointed to datasets, and in some cases provided datasets, 
the license under which this data was provided was at best ambiguous. Careful 
tracing of the data sources for 3-4 tutorials found the datasets were covered 
by copyright and could not be used without explicit permission by the owner. 
None of the associated documents had statements clarifying copyright ownership 
or that permission had either been sought or given (e.g. Reproduced with 
permission from the Commonwealth of Australia 2010). None of the data have 
associated copyright details in the metadata or even distributed with the 
files. So even, if the author had clarified the copyright issues they have not 
provided enough detail to satisfy me that by downloading the data used in 
their tutorial I would not be infringing copyright. In one instance the author 
pointed me to a web page, which explicitly excluded the use of the material to 
be downloaded for the use the author wanted me to use it.

***************************************************************
COMMENTS
***************************************************************

OSGeo, as the central industry body for a group that's existence specifically 
relates to freedom of use (FOSS4G), should take a  lead role in developing 
best industry practice when it comes to the use of copyrightable material.

Some of the ways OSGeo could address these issues, is by...

(a) updating there education repository to ensure all existing publications 
and the underlying datasets used are appropriately licensed (i.e. conduct an 
audit).

(b) changing the educational resources data entry form to explicitly ask 
authors submitting works that they have addressed copyright for all work used 
in the tutorials and data distributed with tutorials (or hyperlinked to).

(c) changing details on the mail servers to explicitly specify the license 
under which posts are provided or have the user choose a license when 
submitting a post or registering as a subscriber to a list. 

(d) creating a repository of data suitable for use with educational 
material for which clear licence agreements have been negotiated/acquired.  
Datasets need to vary in type (polygon, line, point), format (images, 
shapefiles, CAD files, etc), jurisdiction (USA, UK, Australia, etc) and 
resolution (large and small scale). An active effort should go into collating 
existing material then actively chasing missing datasets.

(e) developing resources for authors to help them acquire permission from 
clients, employers and other stakeholders to use copyright material in the 
preparation of educational works. This could be in a form of a brochure by 
OSGeo, draft licence agreement clauses, development of a logo/badge that could 
be displayed by the copyright owner on their website stating how they have 
contributed data to the public domain (for the betterment of mankind, yada 
yada yada). For those commercially oriented data suppliers offer a hyperlink on 
the repository to their main page as 'payment' for the release of a small 
dataset into the public domain (increased exposure, form of marketing).

***************************************************************
FINAL STATEMENT
***************************************************************

Thanks for reading this post. 

I know it was long; but I had a lot to say.

-- 
Cheers Simon

	Simon Cropper
	Botanicus Australia Pty Ltd
	PO Box 160 Sunshine 3020
	P: 03 9311 5822. M: 041 830 3437
	W: http://www.botanicusaustralia.com.au


More information about the Edu_discuss mailing list