[Interest] Contributor agreement rundown

Till Oliver Knoll till.oliver.knoll at gmail.com
Thu Apr 19 07:57:49 CEST 2012



Am 18.04.2012 um 20:01 schrieb Nikos Chantziaras <realnc at gmail.com>:

> On 18/04/12 20:19, Till Oliver Knoll wrote:
>> 
>> ... it is impractical to develop bug fixes (features maybe) in the commercial version [...] and /not/ merging them back into the "main stream" (open source).
>> 
>> ... and yes, with all the contractual fine print and the risk to transmogrify these changes into a locked box). 
> 
> You are not required to merge them back.  ...
> 
> So that argument is moot.

Well, if you read again what I wrote: my argument was NOT that you are required to merge. In fact I actually wrote that you (as an open source contributor) have to live with that risk to "loose" your contribution.

My argument was that it was IMPRACTICAL not to merge them back. Otherwise you would have to rebase your own source tree each time, decide what you /can/ merge back, should you want to merge a particular change back but which might already (partly) depend on your other "closed development".

Believe me: that's a mess. And yes, I've been through that in Qt 3 times, when we extended the QPainter etc. API to support floating point precision for text rendering. Each .1 release required us to merge back our changes (okay, at that time there existed no public Qt source system such as Git...).


So if you want to argue against my argument it would be along the line that it is not that impractical or whatever. But repeating what I already openly admited (that there is no obligation to merge back) is /not/ a valid contra-argument ;)

Cheers, Oliver


More information about the Interest mailing list