[OSM-legal-talk] Is the license change easily reversible?

[Rob Myers]

On 19/02/12 11:17, Frederik Ramm wrote:
> 
> But, even after the switch to ODbL, OSMF could go back to CC-BY-SA 2.0
> at any time - and would, as far as I can see, only need a simple
> majority board decision for that.

Yep. And 2.0 could then be upgraded to a higher version with the next
planet dump. This might be desirable if, for example, 4.0 has
irresistibly wonderful database right handling.

Everyone is following the CC 4.0 drafting process and providing input,
right? :-)

> This puts OSMF in a position of quite some power.

[inserts quote about power and responsibility.]

> Could we - could OSMF - in such a situation simply say: "Know what, Mr
> big guy? Either you play nice and release that data, or we'll simply go
> back to CC-BY-SA 2.0 next month."

Yep. Although they could continue to use the existing data. Which might
make delicensing feel less of a(n immediate) threat.

> I don't assume that we would really *want* to go back but it wouldn't
> exactly kill us, and depending on what is at stake (I assume it could
> easily be a multi million dollar thing) we (the project) would lose much
> less than those we'd be up against. We wouldn't really want to but we
> *could*, and the fact that the big guy would only have to piss off the
> wrong four people at OSMF to ruin his product could balance one thing or
> the other in our favour.

Protecting the freedom of individuals to use the data that OSM gathers
and distributes isn't about pissing people off, etc., but yes there is
apparently a nuclear option there.

The collateral damage would be eye-watering though, in terms of burnt
karma, lost trust, and punishment of innocent actors.

> 1. Is my reasoning correct?

I believe so.

> 2. Are we happy with OSMF board wielding this power - should we (the
> OSMF membership) perhaps curtail OSMF board's powers by creating a rule
> that says that any decision regarding the license under which the data
> is published must be taken by the whole membership and not just the board?

Do you mean the foundation membership or active contributors to OSM?

> 3. If the CTs were changed post-license-change to omit CC-BY-SA 2.0 from
> the list of available licenses, then the above scenario would become
> impractical - we could then not simply go back to CC-BY-SA 2.0 without
> losing data from new contributors (unless going through the 2/3 rule).

The CTs could then be changed back, and data contributed prior to the
initial change could be licenced back down.

> Such a change in the CTs would create more security for anyone investing
> in a product based on our data by taking away the bargaining chip I have
> written about. Does the power to change the CTs currently sit with the
> board alone, and are we happy with that?

 Pass.

> The power to modify the CTs carries with it the power to entrench the
> current license practically forever; someone with liberty to change the
> CT as they see fit could, for example, simply strike out the "future
> license change possible with 2/3 of active contributors" clause and
> therefore create a situation in which no future OSMF can change the
> license without going through what we go through now. Of course the CTs
> cannot be changed retroactively but doing so for new signups is
> effective enough.

And this is part of the problem with listing specific licences. The CTs
should explain the idea, not fossilize the expression. Yes, this will
impose large social and time costs on future decisions by requiring
interminable debate about whether any change fits the spirit and the
letter of what is intended. But that is better than not being able to do
so, or having to change the CTs in order to allow it by fiat. Changing
the CTs doesn't exactly seem to make people feel loved.

- Rob.