[OSM-legal-talk] legal-talk Digest, Vol 19, Issue 1

[John Wilbanks]

First to the stallman quote - I'm not surprised at all that RMS likes 
copyleft on maps. FWIW, I do too. When you frame the question as "should 
<copyrighted thing X> be licensed under copyleft" I would tend to answer 
"yes".

I was instead encouraging you to engage in a conversation with him about 
the freedom and morality of database and data licensing, about the 
potential implications both for good and for bad of the various choices 
at the table here. I haven't done so on this topic with him, so I've not 
salted this earth to ambush you.

on to 80n's question:

 > I appreciate the effort you have gone to.  It would be helpful if you 
could
 > elaborate on the reasons why each of these lawyers came to the 
conclusions
 > they did.

Thanks for the kind comment - it's appreciated.

I have pretty much laid out the answers as they came to me  in my 
previous emails, in most cases directly quoting the answers. This wasn't 
treated as a complex issue to me - "Feist solved that" was the common 
theme followed closely by "in databases, the only things that matter are 
arrangement and sweat of the brow" and thus this type of thing was 
irrelevant. There was a general sense that this was a sidebar matter, 
not a central one - answers were in short staccato sentences, not 
paragraphs.

> My suspicion is that you asked about GPS traces, whereas we are concerned
> with geo-data that is derived, manually, from an aggregate of GPS traces,
> aerial photography, human observation and a tad of cartography.  Theses are
> two very different animals.  Can you confirm which of these your answer
> pertains to?

I asked something in between. I was assuming GPS traces, human 
observation, and aerial, because that's what I found when I dug around 
on geospatial data generally. But I was not talking about the nodes and 
ways and such in OSM, simply the actual factual data. And the answer was 
that it doesn't matter how data gets collected, as long as it's data.

A poster noted earlier that the nodes and ways are pretty likely to be 
found under copyright and the maps are definitely copyrighted, and 
that's what the research turns up also.

Cartography might change things. I can't speak to that, because I didn't 
ask it (the same risk of copyrights propagating from maps to data there 
though, I would assume).

> The fake entries in Rural's directory were random phone numbers.  There was
> no claim that they were creative works.  If there had been then it would
> have set a precedent, but the ruling rested solely on a consideration of
> copyright in the directory as a collection of facts.
> 
> Indeed, if you read the case summary then you'll see that it does
> acknowledge that a directory containing fact *can* be copyrighted if the
> selection and arrangement is novel or creative (although the facts
> themselves cannot).  But in this case, a listing of telephone numbers, it
> failed on both these counts/ there was no selection, one of the conditions
> of Rural's monopoly was that they were required to include *every* phone
> number/ and the arrangment was not novel either - alphabetical order was
> considered to be obvious and unoriginal.

I've read the case summary, it's stimulating stuff :-)

No one made the case that these were creative works, thus they were not 
considered as such. I'm not arguing that my research turned that up. 
What I heard from the lawyers, again, was that the fact that these were 
part of the case, but not even deemed worthy of arguing up or down, 
creates a precedent mitigating against consideration of easter eggs as 
relevant to the copyrightability of the knowledge compilation.

Another direct quote is that "there is no 'gotcha' that can clearly 
transform the copyrightability of a database - feist provides a test 
based on arrangement and originality, and that's what i would argue in 
court, no matter how catchy the easter eggs were."

jtw