Are implicit dual-licensing agreements inherently anti-open?

[Alex Bligh]

Andrew,


>> Technically speaking that's true with any license that has a
>> differentiated ID and contributor grant.
>>
>> In practice, that's far EASIER with the OVPL - if the ID's agree, there
>> is none of the normal ambiguity.
>
> My reading of OVPL is that for any piece of code, there can be one, and
> only one,
> ID.

Yes, just like MPL, CDDL etc..

  In the case of merging two modules of covered code, the IDs, as you
> say,
> need to agree which one of them will be the ID of the merged module.

Yes, just like MPL, CDDL etc..

> Absent
> such agreement, the merged module presumably does not have a valid
> license.
> To my way of thinking, if a negotiation is required to create a valid
> derivative
> of two works which are covered by the same license, then said
> license is not really open source.

Well that applies to MPL, CDDL etc. too. But as far as I know, mergeability
of different project with different initial developers is not an OSI
criterion.

All it is is (yet another) case of projects with different licenses (because
of different ID's) being unable to merge. That already applies with (say)
GPL and $anotherlicense code.

I would suggest you have the above a bit backwards - normally the criteria
people look for is FORKABILITY not mergeability.

Alex