[Osmf-talk] New license change proposal status
frederik at remote.org
Wed Dec 2 03:18:54 UTC 2009
Mike Collinson wrote:
> In addition to an update I've just sent to this list, I also want to
> respond to your issue of the vote wording so that the issue is
> adequately discussed.
What percentage of approval among OSMF members do you require for
continuing with the process? (Personally, in such an important matter, I
would not want OSMF to proceed if 49.9% of members were against?)
I am unhappy with several aspects of the process. I don't have the time
to write a concise run-down of everything that's in my head just now
(hopefully at the weekend), but I'll give you an outline of what I think.
I have a feeling that you are trying to steamroll people into
submission, and I find that inadequate giving the diligence that the LWG
has shown over the last months.
You are asking us, the members, to say yes to your proposal and after
that there's no way back. We, the members, take ultimate responsibility.
At the same time, you present a number of documents that are written to
convey the message: "Don't worry, we know what we're doing, simply sign
here and all is going to be well."
There are a number of people on the License Working Group whom I have
considerable respect for and who are, I believe, bright thinkers whose
judgment I would tend to trust. And these people now, as it is written,
"recommend" the ODbL plus surrounding paraphernalia.
However: Even the brightest minds often work within an envelope. The
envelope that you have been working inside, the question that was posed
to you, is not: "Which license would be best for OpenStreetMap", but:
"Which viral/share-alike license would be best for OpenStreetMap".
I am pretty sure that some of you would have reached an altogether
different conclusion had you been asked to work in the wider envelope.
I am prepared to accept your result that ODbL is the best available
share-alike license for Geodata. This does not, however, mean that it is
the best license for OpenStreetMap. This question has been completely
outside your mandate.
In your proposal you write, again, that Creative Commons do not
recommend CC-BY-SA for data, which is true. You expend 8 lines of text
to describe that. Then you use one line of text to say that Creative
Commons now recommend "effectively" to use CC0 for data, only to add: "A
substantial portion of the community however feel that the project
should remain under a reciprocal license in the same spirit as our
current license." - I think that this paragraph is a rhetorical
masterpiece, first using Creative Commons' credibility to make your case
and then dumping them to make a 180 degree turn. Rhetoric, however,
should not determine the future OSM license.
Have you made any effort to find out how "substantial" this portion of
the community really is? No. You are prepared to go ahead with the new
license and steamroll the community to accept it, without first finding
out whether there is "substantial" resistance to ODbL. But at the same
time you ignore the Creative Commons recommendation citing "substantial"
resistance to CC0.
You correctly state in one place:
"This will be read by thousands of contributors, many of whom may not be
overly concerned about detail."
I'm saying that if you would spend as much effort into pushing through
CC0 as you are planning to put into ODbL, the results would be similar.
What we're seeing here is a huge marketing campaign for ODbL. You are
*selling* this license. The License Proposal is a sales brochure, and
not suitable as a basis for members to make an informed decision.
You will presumably be e-mailing this document to all members from an
official foundation E-Mail address, whereas those who do not agree with
your recommendation may voice their opinion on e-mail lists. Nowhere in
your document does it say that not everyone is of the opinion that this
is the best way forward. Opposition doesn't exist. The LWG speaks with
one voice, nobody disagrees, everyone has a happy face.
Once the members agree to your proposal, and I am under no illusion that
you will require more than 50.1% as a suitable majority, you will tell
everyone: "OpenStreetMap is moving to a new license". Again, most of
those receiving that communication will not be informed about the fact
that there might be quite a lot of people who are *not* planning to move
to a new license, or not to this license, or who reluctantly go along
but would have wished for something else.
All the pain I expressed above could have been reduced a bit if you
could have been bothered to put the three-part question to the community
like I suggested:
>> ( ) I release my data under ODbL
>> ( ) I do not release my data under ODbL
>> ( ) I consider all my data PD anyway and don't claim database
>> protection so do whatever you want
However, you believe that even this would overtax the average community
> We finally concluded that unless the basic OSMF member vote
> itself was a simple yes/no, there are simply too many shades of
> opinion to get a question/answer wording that is not loaded (i.e.
> potentially biased) and that everyone could be happy with.
I don't think that what you write here is quite honest. I believe that
what you really wanted to say is this: "We do not want to give people
the impression that they have a choice, because that would make them
think, and perhaps in the end disagree completely".
But this precisely is steamrolling - trying to keep people from
thinking. Making it as easy as possible for people to buy whatever you
want to sell.
I have a bad feeling about this; I think the process is unbalanced,
biased, and worst of all, I think that some of the very people
"recommending" this license actually do have reservations but do not
voice them because they fear that everything else is even worse than the
ODbL. (Which may be true but then it deserves to be said.)
I share your view that CC-BY-SA is unsuitable (but not everyone in the
project does). I also know that any license change is going to be
painful. I have many reservations about the ODbL which are not due to
shortcomings in the license that could be fixed, but which are simply a
general consequence of share-alike licensing. One of them, but by far
not the only one, is that any such type of license requires policing,
and that consumes a lot of work and causes bad blood as we're currently
discovering with all the accusations thrown around on the "lacking
proper attribution" wiki page.
I will try to put together a document with all these reservations ASAP
but unfortunately my day has only so many hours. I have wanted to create
some "straw poll" facility to give OSM community members a way to
express their opinion BEFORE you do anything but I haven't got to it. It
is commendable that all of you have spent so much time on refining ODbL
and defining processes and so on, but I would have hoped that at least
some work would have gone into finding out about these problems.
Unfortunately that was not part of your mandate, and so we now have a
shiny sales brochure telling us about all the good things that come with
ODbL, but not the dark side of it. (The dark side is two-fold - some
dark things come from a license change per se, no matter what the new
license is, and some come from share alike licenses in general. There
are also a few remaining issues with ODbL but I wouldn't brood on them
too much since they could still be ironed out later.)
I really believe that you have done a good job until now, but much as
everyone would like to have the issue resolved, I don't think the job is
quite done already, and further steps must be made carefully.
If we find ourselves as an ODbL licensed project one year from now
(which is *possible* but far from sure; and I would in all likelihood
still be here even if we adopt ODbL) then I want the switch to ODbL to
stand on a rock solid footing and not on some "50.1% of members said yes
and then we pulled a ton of strings to make people look the other way
and now we're all glad it is over" basis.
Frederik Ramm ## eMail frederik at remote.org ## N49°00'09" E008°23'33"
More information about the osmf-talk