Jan-Peter Homann wrote:
Is it from your point of view allowed, that the manufacturer of the proofing software delivers two different installers (proofing software and argyll) and publishing the UI for Argyll under GPL license ?
If the two programs (or sets of programs) are installed (even separately installed) in a way that is delivering a collective work from the manufacturer to the customer, then I think that this falls under the "work based on the Program" provision of the GNU licence (since a collection is a work, and the work contains the program), and permission to copy any GNU component is dependent on the whole work being made available under GNU compatible licence.
The question comes down to whether it is reasonable to regard the two programs as separate and independent and that they are just being run "side by side", or whether together they constitute a collective work. That's why I think that if (for instance) one program used another specific program to perform some function, even if it communicated via mechanisms like pipes etc. rather than being linked into the same binary, that it's clear that they are intended to work together, and have been supplied to be together, and form a collective work.
It's not clear to me what you mean by "publishing the UI for Argyll under GPL license", since Argyll does not currently have a UI.
If someone wanted to write a UI for Argyll, then they are of course permitted to do so, as long as they comply with the offered GNU licence. If such a UI was implemented in a way that directly links with the Argyll code, then the resulting binaries are a derivative work, and the UI needs to be released under the GNU licence if it is to be distributed. If a UI used other techniques (ie., such as invoking the Argyll command line utilities), then I don't think that this would make the UI a derivative work, but I think that any attempt to distribute the UI with Argyll would form a collective work, and the work would have to comply with the GNU licence.
Both the proofing software and Argyll with UI would be running independently.
Loading of the measureremnt-data into Argyll would be manual process.
The installation of the profile generated by Argyll inside the proofing software would also be a manual process.
If they are running side by side as independent applications, and operating at arms length, and are not supplied as purporting to be part of a single product or solution, then I think this may well be sufficient to avoid being regarded as collective work, and hence to be compliant with the conditions of the GNU licence with regard to the distribution of the GNU program.
These are my current thoughts on this particular subject, and are what would guide me in choosing how I exercise my right to control my copyrighted work.