[KDE/Mac] towards further integration on OS X; accepting arguments through LaunchServices
kde-ml at unormal.org
Thu Sep 4 14:03:30 UTC 2014
Am Donnerstag, 04. September 2014, 12.25:02 schrieb René J.V. Bertin:
> On Thursday September 04 2014 11:44:41 Mario Fux wrote:
> >Sounds really great. Thx! Is this based on Kate based on KF5? I think this
> >is a bigger change (a very positive one) and should be used on the KF5
> >based Kate as it's just cumbersome to first integrate it in the old Kate
> >(based on kdelibs4) and in some months to port it over.
> No, the testing done and reported here is has nothing to do with KDE per
> se. It's something that works with any executable (binary or script) that
> expects arguments in the standard Unix/Posix way.
Ah, then sorry for my misunderstanding.
> It could be integrated
> in the KDE buildsystem, and at a lower level it could even be integrated
> in the startup procedure (presuming KDE has some function you call at the
> start of the main function - haven't even looked at that yet).
I think I meant the integration in the KDE buildsystem which changed quite a
bit recently because of KF5.
> But for
> that approach I'd make sure to check first whether Qt doesn't already
> provide something like that. I understand your concern of starting with
> KF5 first ... but that means the result won't get tested properly for
> quite a while, until KF5 is production/end-user ready (and AFAIC that
> means available through MacPorts).
> >based KDE applications but never forget to port them _always_ to the new
> >KF5 and KF5 based apps otherwise they will be forgotten and great work
> >and effort is lost.
Sorry to state the obvious then ;-).
> >IANAL but if the code doesn't have a license yet then it's proprietary
> >(that's a big weakness of github => not to enforce a license)
> KDE may consider things that way, but I don't agree with that.
First: I'm still not a lawyer.
Second: I spend a bit of time at FSFE booth and reading about licenses and Co
and this is not a KDE consideration it's a general thing.
> The fact
> code is made available in a public repository means it's available for use
> by anyone.
No, it's not. If there is no statement about the license it just means that
you need to ask the authors under which conditions or licenses you can use the
> Adding a license to a (C) John Doe doesn't make code any less
> proprietary IMHO, John Doe still retains the copyright. It just stipulates
> (= limits) how it can be used.
I didn't look at the code. It's just that I understood from you that their is
not license text or copyright statements at all. And I know that I read
several times that it's a problem not to state any license or copyright state
in your (plural) code and thus the code becomes "proprietary" and for sure not
free software or public domain (which is not possible in Europe anyway).
> >other free software projects). But you can "relicense" it and ask all the
> >involved people for agreement. BSD would work for KDE IIRC but as soon as
> As I said, that'd probably be problematic... Supposing I can track down all
> addresses one could take silence for acceptance, but I'm not sure that'd
> stand should there ever be legal action.
No, sorry, silence is legally disagreement.
> Fortunately there's only so many
> ways one can write interface code like this, so I guess one can always
> rewrite it (change variable/function names, move some code around and you
> get to do the same thing "in your own words" O:-) ).
Yes ;-). But be careful.
More information about the kde-mac