I have been doing internal Wireshark releases for years wherever I've been working (as far as I know, they have never been sent outside of the company). I have *never* used the plugin mechanism. I package up the entire program, even if only one file has been changed. My current company has acquired and merged with several other companies and development groups - as far as I can tell, they all have a local Wireshark person who does the same. If people are working closely with me, we sometimes even just keep dissectors as part of the test code for the project that uses them, and team members build it themselves.
Am I allowed to do this? Martin On Mon, Dec 4, 2023 at 2:54 PM João Valverde <j...@v6e.pt> wrote: > > > On 04/12/23 14:52, João Valverde wrote: > > > > > > On 04/12/23 14:32, Anders Broman wrote: > >> Hi, > >> Company plug-ins may have restrictive license as the purpose is to > >> only use them internally no public usage "secret" code for > >> proprietary protocols under patents or IPL. Do we really want to > >> forbid that? In that case why should companies provide code to > >> Wireshark rather than just fork and build internally. > > > > I understand the argument and why that is a point of contention, but > > that does not change the terms of the GPL which must be abided by even > > if this commit was never merged in the first place. > > > > I don't think it is a question of whether we want to forbid it, it is > > whether we can allow it. I believe the answer to that is a clear no if > > we want to respect the terms of the GPLv2 (and I'm fine with that). I > > am not a license lawyer so this is just my understanding of the > > legalities involved. > > > ...nor any other kind of lawyer. :-) > > > There are many reasons why companies may choose to contribute or not. > > Other companies may choose not to contribute to projects not using the > > GPL. And individual developers may or may not want to contribute. Etc. > > We can also debate that but it might veer off-topic. > > > >> Best regards > > > >> A ders > >> > >> Den mån 4 dec. 2023 15:22João Valverde <j...@v6e.pt> skrev: > >> > >> > >> > >> On 04/12/23 13:42, Anders Broman wrote: > >> > Hi, > >> > Maybe you are missing the point that someone may wish to develop > >> an in > >> > house plug-in not meant for distribution which in my > >> understanding is > >> > permissible under GPL. > >> > >> My understanding is that this is permitted under the GPL if using a > >> GPL-compatible license for your software. It's the main difference > >> between the GPL and so-called "permissive" licenses. > >> > >> > > >> > As I understand it that is no longer possible? To me that's an > >> > unnecessary restriction which we do not need to put on our users > >> and I > >> > see no point/gain in doing so. > >> > >> If you don't want to use the GPL you can choose a GPL-compatible > >> license > >> (BSD for example, there are many) and register your plugin with > >> WS_PLUGIN_IS_GPLv2_COMPATIBLE. The SPDX ID is optional but helpful. > >> > >> You may also use GPLv2 and just not distribute your binary (in the > >> case > >> of businesses and corporations outside of the collective entity that > >> legally comprises it). > >> > >> So it's not really restricting your freedom to use Wireshark, it's > >> just > >> respecting the terms of the GPL under which developers contribute > >> to the > >> project. > >> > >> This is my understanding of the terms under which I choose to > >> contribute > >> to Wireshark. If anyone has a better understanding or reason why > >> this > >> interpretation of the GPL, that matches the FSF FAQ, is wrong, > >> please do > >> share. I'm very open to a good-faith discussion. > >> > >> > >> > Best regards > >> > Anders > >> > > >> > Den mån 4 dec. 2023 14:05João Valverde <j...@v6e.pt> skrev: > >> > > >> > Confused was not an offense, "GPL license" is patently not the > >> > same as > >> > "GPL-compatible license" so it is a legitimate reason to be > >> confused. > >> > Please avoid unnecessary and unfair characterizations of my > >> words. > >> > > >> > And I will not revert it on that basis. I will revert it if my > >> > understanding of the our license requirements is wrong or > >> flawed. > >> > It is > >> > not OK for you to exempt some use-case from the license > >> terms under > >> > which every developer contributes to this project. > >> > > >> > Gerald can revert it if he wishes and I will respect it. As > >> > project lead > >> > he can make that call. > >> > > >> > On 04/12/23 12:35, Roland Knall wrote: > >> > > I do not think there is a need for calling someone confused. > >> > > > >> > > The whole discussion is not in any way useful for our users. > >> > There is > >> > > the explicit corporate usecase, where in-house versions do > >> exist > >> > with > >> > > their own protocols and plugins. Often times those > >> versions do not > >> > > even deal with licenses for those modifications at all, and > >> > going from > >> > > the point that they change the CMakeListsCustom.txt > >> files, one > >> > could > >> > > argue, that this is not a source code modification in the > >> sense > >> > meant > >> > > by the gpl license. > >> > > > >> > > Joao, I agree with having a clear path for license > >> application, > >> > and I > >> > > also agree that we should be prudent on what parts a user > >> can > >> > use and > >> > > which he can't. I would even be ok if we have a warning > >> in the > >> > > build-process, explicitly stating that the code being > >> linked is not > >> > > fully compliant and therefore not allowed to be > >> distributed. But I > >> > > strongly disagree cutting off the leg we are standing on > >> just on > >> > pure > >> > > principle. The corporate users are a HUGE part of our > >> userbase. > >> > And if > >> > > we go down this route, we need to have a proper discussion > >> about > >> > this. > >> > > Just adding license enforcement without having the > >> discussion is > >> > NOT > >> > > the way to move forward here. > >> > > > >> > > Please add another patch, which keeps the ABI versioning in > >> > (which I > >> > > really appreciate and think is a good thing to do), but > >> reverts the > >> > > enforcement of the licenses. Then we can start to properly > >> > discuss how > >> > > to move forward with this topic. It will - most likely - > >> require a > >> > > vote by the technical steering comittee. > >> > > > >> > > kind regards > >> > > Roland > >> > > > >> > > Am Mo., 4. Dez. 2023 um 13:23 Uhr schrieb João Valverde > >> <j...@v6e.pt>: > >> > > > >> > > > >> > > > >> > > On 04/12/23 12:19, João Valverde wrote: > >> > > > > >> > > > > >> > > > On 04/12/23 12:12, Bálint Réczey wrote: > >> > > >> João Valverde <j...@v6e.pt> ezt írta (időpont: 2023. > >> dec. 4., H, > >> > > 12:59): > >> > > >>> > >> > > >>> > >> > > >>> On 03/12/23 23:25, João Valverde wrote: > >> > > >>>> Hi, > >> > > >>>> > >> > > >>>> There are some changes in progress to the plugin > >> > registration > >> > > API that > >> > > >>>> break compatibility and require manual intervention > >> > from plugin > >> > > >>>> authors maintaining plugins out-of-tree. These > >> changes > >> > are rather > >> > > >>>> minor and concern only plugin registration, not > >> other APIs > >> > > accessible > >> > > >>>> to plugins. > >> > > >>>> > >> > > >>>> See MR 13524: > >> > > >>>> > >> > https://gitlab.com/wireshark/wireshark/-/merge_requests/13524 > >> > > >>>> > >> > > >>>> Changes required are rewriting the registration > >> code (very > >> > > easy to do > >> > > >>>> [1]) and declare (using a C enum) that the > >> plugin is > >> > released > >> > > either > >> > > >>>> under GPLv2 or later, or a GPLv2 compatible > >> license. The > >> > > other changes > >> > > >>>> to the ABI version number are > >> > > >>> The choice of the word "released" here was > >> unfortunate, > >> > > because it may > >> > > >>> imply distribution. Please consider "licensed" > >> instead. > >> > > >>> > >> > > >>> The license declaration field just affirms what > >> was already > >> > > implicit: > >> > > >>> Wireshark plugins must use licensing terms > >> compatible > >> > with the GPL > >> > > >>> version 2, so there is no policy change there. > >> > > >> GPL allows linking and using GPL-licensed software > >> with > >> > > >> non-GPL-licensed software locally. This is an > >> important > >> > use case of > >> > > >> many Wireshark users who do not wish releasing > >> their plugins > >> > > and your > >> > > >> change broke that. Please revert it. > >> > > >> > >> > > > > >> > > > > >> https://www.gnu.org/licenses/gpl-faq.html#IfLibraryIsGPL > >> > > > >> > > Also it does not require a GPL license, it requires a > >> > GPL-compatible > >> > > license, so you may just be confused. > >> > > > >> > > > > >> > > >>>> currently not relevant to plugin authors (no policy > >> > change is > >> > > >>>> implied), it just uses less boilerplate with > >> macros. > >> > > >>>> > >> > > >>>> This should improve the plug-in experience for both > >> > > developers and > >> > > >>>> users and may improve compatibility in the future. > >> > > >> > >> > > >>>> Comments welcome. > >> > > >>>> > >> > > >>>> Regards, > >> > > >>>> > >> > > >>>> João > >> > > >>>> > >> > > >>>> > >> > > > >> > > >> [1] > https://gitlab.com/wireshark/wireshark/-/commit/90b16b40921b737aadf9186685d866fd80e37ee6#4a1fe9011e8240918e5fc6230c0bcd2e4d3b9c34 > >> > > > >> > > >>>> > >> > > >>>> > >> > > >>>> > >> > > > >> > > >> > ___________________________________________________________________________ > >> > > > >> > > >>>> > >> > > >>>> > >> > > >>>> Sent via: Wireshark-dev mailing list > >> > > <wireshark-dev@wireshark.org> > >> > > >>>> Archives: > >> https://www.wireshark.org/lists/wireshark-dev > >> > > >>>> Unsubscribe: > >> > > https://www.wireshark.org/mailman/options/wireshark-dev > >> > > >>>> > >> > mailto:wireshark-dev-requ...@wireshark.org?subject=unsubscribe > >> > > >>> > >> > > > >> > > >> > ___________________________________________________________________________ > >> > > > >> > > >>> > >> > > >>> Sent via: Wireshark-dev mailing list > >> > > <wireshark-dev@wireshark.org> > >> > > >>> Archives: > >> https://www.wireshark.org/lists/wireshark-dev > >> > > >>> Unsubscribe: > >> > > https://www.wireshark.org/mailman/options/wireshark-dev > >> > > >>> > >> > mailto:wireshark-dev-requ...@wireshark.org?subject=unsubscribe > >> > > >> > >> > > > >> > > >> > ___________________________________________________________________________ > >> > > > >> > > >> > >> > > >> Sent via: Wireshark-dev mailing list > >> > > <wireshark-dev@wireshark.org> > >> > > >> Archives: > >> https://www.wireshark.org/lists/wireshark-dev > >> > > >> Unsubscribe: > >> > > https://www.wireshark.org/mailman/options/wireshark-dev > >> > > >> > >> > mailto:wireshark-dev-requ...@wireshark.org?subject=unsubscribe > >> > > > > >> > > > > >> > > > >> > > >> > ___________________________________________________________________________ > >> > > > >> > > > > >> > > > Sent via: Wireshark-dev mailing list > >> > > <wireshark-dev@wireshark.org> > >> > > > Archives: > >> https://www.wireshark.org/lists/wireshark-dev > >> > > > Unsubscribe: > >> > https://www.wireshark.org/mailman/options/wireshark-dev > >> > > > > >> mailto:wireshark-dev-requ...@wireshark.org?subject=unsubscribe > >> > > > >> > > > >> > > >> > ___________________________________________________________________________ > >> > > Sent via: Wireshark-dev mailing list > >> > <wireshark-dev@wireshark.org> > >> > > Archives: https://www.wireshark.org/lists/wireshark-dev > >> > > Unsubscribe: > >> > https://www.wireshark.org/mailman/options/wireshark-dev > >> > > > >> > > > >> mailto:wireshark-dev-requ...@wireshark.org?subject=unsubscribe > >> > > > >> > > > >> > > > >> > > >> > ___________________________________________________________________________ > >> > > Sent via: Wireshark-dev mailing list > >> > <wireshark-dev@wireshark.org> > >> > > Archives: https://www.wireshark.org/lists/wireshark-dev > >> > > Unsubscribe: > >> https://www.wireshark.org/mailman/options/wireshark-dev > >> > > > >> > mailto:wireshark-dev-requ...@wireshark.org?subject=unsubscribe > >> > > >> > > >> > ___________________________________________________________________________ > >> > Sent via: Wireshark-dev mailing list > >> <wireshark-dev@wireshark.org> > >> > Archives: https://www.wireshark.org/lists/wireshark-dev > >> > Unsubscribe: > >> https://www.wireshark.org/mailman/options/wireshark-dev > >> > > >> > mailto:wireshark-dev-requ...@wireshark.org?subject=unsubscribe > >> > > >> > > >> > > >> > ___________________________________________________________________________ > >> > Sent via: Wireshark-dev mailing list > >> <wireshark-dev@wireshark.org> > >> > Archives: https://www.wireshark.org/lists/wireshark-dev > >> > Unsubscribe: > >> https://www.wireshark.org/mailman/options/wireshark-dev > >> > > >> mailto:wireshark-dev-requ...@wireshark.org?subject=unsubscribe > >> > >> > ___________________________________________________________________________ > >> Sent via: Wireshark-dev mailing list > >> <wireshark-dev@wireshark.org> > >> Archives: https://www.wireshark.org/lists/wireshark-dev > >> Unsubscribe: > https://www.wireshark.org/mailman/options/wireshark-dev > >> > >> mailto:wireshark-dev-requ...@wireshark.org?subject=unsubscribe > >> > >> > >> > ___________________________________________________________________________ > >> > >> Sent via: Wireshark-dev mailing list <wireshark-dev@wireshark.org> > >> Archives: https://www.wireshark.org/lists/wireshark-dev > >> Unsubscribe: https://www.wireshark.org/mailman/options/wireshark-dev > >> mailto:wireshark-dev-requ...@wireshark.org?subject=unsubscribe > > > > ___________________________________________________________________________ > Sent via: Wireshark-dev mailing list <wireshark-dev@wireshark.org> > Archives: https://www.wireshark.org/lists/wireshark-dev > Unsubscribe: https://www.wireshark.org/mailman/options/wireshark-dev > mailto:wireshark-dev-requ...@wireshark.org > ?subject=unsubscribe >
___________________________________________________________________________ Sent via: Wireshark-dev mailing list <wireshark-dev@wireshark.org> Archives: https://www.wireshark.org/lists/wireshark-dev Unsubscribe: https://www.wireshark.org/mailman/options/wireshark-dev mailto:wireshark-dev-requ...@wireshark.org?subject=unsubscribe
