[OSM-legal-talk] Hello... and my case for PD

Frederik Ramm frederik at remote.org
Mon Feb 26 23:13:16 GMT 2007


Hi,

    I've finally decided to subscribe to legal-talk as well, if only to 
stop annoying "talk" subscribers with advocacy of any kind.

I have read February's posts on this list, and of course the discussions 
on talk. I am a software engineer, not a lawyer. So that's my 
background. I make a living doing freelance work, but wouldn't hesitate 
to set up a small company that creates and sells some kind of product if 
I had the right idea at the right time. So while I am not a commercial 
user of OSM stuff, I might one day become one - selling, for example and 
completely hypothetical, a car navigation system built on OSM data, with 
paid staff improving data where it is insufficient for routing.

My aim is to make OSM data free (enough) to allow anyone to use it, even 
commercially. This of course implies that I also want to keep it free, 
and protect it from becoming gobbled up by someone.

I am not necessarily making a case for PD but from what I currently see, 
PD is the only honest option that would fulfill these goals.

There are two arguments that are regularly put forward by people 
involved in OSM which I strongly disagree with.

The first goes like this: "Making OSM data public domain will cause 
(profit-minded) people to make small improvements, then distribute that 
on a restricted license, people would flock for that, OSM would not be 
able to keep up and have to close shop."

This is not entirely wrong; the risk exists. If someone took all of OSM 
today, put it on a strong server architecture with some cool software 
thrown in, encouraging people to upload their data on his platform 
rather than ours - many people would probably do it. But then again,
our data is not a key ingredient of this. Someone like Google could
simply strike a deal with TeleAtlas etc. and set up a Wiki-like map 
editing suite pronto. And they - or someone else - will do that some 
time, and the risk to OSM will be exactly the same: People who don't
care about the license flocking to the shinier solution.

In fact, if Google etc. wanted to do something like that right now, and 
OSM were public domain, they would probably still go for TeleAtlas etc. 
because they have better coverage.

The race is on, whether we are PD or anything else. We have to keep up 
with what the commercial competitors are offering. If we fall (too far) 
behind, then we'll lose contributors.

The second thing I want to talk about is: "Copyleft licenses rarely 
prevent people from doing anything, they just prevent people from 
preventing other people from doing anything...". This is somewhat of an 
ethical issue, and who may lay claim to the word "free". In my eyes,
"copyleft" license do not deserve the attribute "free", because they
*do* prevent certain uses of the data.

For example, my hypothetical scenario above. If I were to produce a
navigation system, I would have to improve the data, and I would have
to pay for that (by employing people). I would try to get this 
investment back by selling the software or device, plus a small profit, 
hopefully. If all works well, then it is a nice plan - I get to be a 
company boss, lots of people get employed by me, lots of users can buy a 
navigation system that's better or cheaper or cooler than other 
offerings. Everybody's happy. No damage is done. But this whole plan, 
while viable in a PD context, is prevented if OSM data is under a 
copyleft license because I would have to make my improved data available 
to my competitors, who would then build and sell the same device but 
without paying people to improve the data, and thus being able to make a 
cheaper product.

Now you can argue wheter you want to allow such projects or not, but 
please don't tell me that a copyleft license does not prevent anything, 
that's just splitting hairs.

Of course, and this has been pointed numerous times on this list, the 
whole discussion may be moot depending on what a derived work is and how 
this relates to technical implementations. In my hypothetical scenario, 
I might for example put my improvements in an extra database which is 
queried by the navigation system in addition to the OSM dataset, and 
then arguing that my data is proprietary and not affected by the 
share-alike aspect of the OSM license. This would of course make my 
product slightly more expensive because I would have to have expensive 
consultations with lawyers and an insurance that helps me out if one of 
the copyleft activists starts filing lawsuits ;-)

If the CC-BY-SA license we currently use can be interpreted (or 
clarified) to actually allow something like that, then I can probably 
live with it... but then why again are we not PD in the first place?

It seems to me that PD is a very honest option, with no room for legal
ambiguities. Everybody knows what it means. All cards are on the table. 
No money spent on lawyers. No multi-page unreadable documents. 
Everything nice and simple.

A final note on the general mode of legal discussion in this project. We 
tend to throw terms around ("PD", "CC-BY-SA" etc.), and at the same time 
most of us refuse to give an exact definition. The other day, someone on 
talk shouted "this project is CC-BY-SA, whatever that means, take it or 
leave it and don't come to us asking for legal clarification". This is 
ludicrous, and true at the same time. There are probably a handful of 
people among us who have had a say when the license was selected; 
everyone else just came aboard - none of us was asked whether we wanted 
to join the commercial, the CC-BY-SA, or the PD variant of OSM. For most 
of us, it was buying the pig in a poke - not even the creators of the 
project are able to specify exactly what CC-BY-SA means for us.

I would hope that, in order to have a meaningful debate on the subject,
we could concentrate on what kinds of uses of OSM (and OSM data) we want 
to allow and what kinds of uses we want to prohibit, and THEN find a 
license and matching interpretation for it. Currently, it is too often 
the other way round: "I am for CC-BY-SA because it allows/disallows/ 
encourages this and that", with the notion that CC-BY-SA really 
allows/disallows/encourages that use often open to discussion. I would 
much rather have people honestly say: "I want to make sure that X is 
allowed, and I want to disallow Y".

I tried to make a start with my example of the regular OSM guy starting
a company with a car navigation system - this is one use that I would 
like to allow.

Bye for now
Frederik

-- 
Frederik Ramm  ##  eMail frederik at remote.org  ##  N49°00.09' E008°23.33'




More information about the legal-talk mailing list