Re: [ESPResSo-devel] New GitHub Terms of Service

[Kai Szuttor]

Since that jobs are running directly on the webserver we don't want to
use that right away. Having to build all PRs is quite a large load.
On Thu, Mar 09, 2017 at 10:17:25AM +0100, Henri Menke wrote:
> One can use GitLab CI on our GitLab.  Some projects already do, e.g. this one
> https://gitlab.icp.uni-stuttgart.de/paper/paper01
> 
> (Sorry, link is only ICP internal)
> 
> On 03/09/2017 10:13 AM, Florian Weik wrote:
> > Hi all,
> > Welcome to the cloud. Just to ad my 2 cents: I think at least technically 
> > Ulf is correct. This makes it impossible to put code that you don't have 
> > the copyright for (as opposed to have been granted a licences to use it) to 
> > github. This is especially ill-fitting for jurisdictions like Germany where 
> > copyright can not be transferred. This is not so much a problem for ongoing 
> > development on github (the users have to agree to githubs ToS to get any 
> > code in, so that they automatically grant github the required licence) but 
> > impossible to fulfill for legacy code. We would have to contact all authors 
> > and ask for permission to use their code on github, which we obviously will 
> > not do. The analysis in the SO thread that Kai posted seems to come to a 
> > similar conclusion. One can only speculate what their rational behind this 
> > is (I think use in search and so on would have been fair use...), but I 
> > don't think we should go with this. Since we are running a gitlab instance 
> > in Stuttgart anyway,
> > one way would be to continue development there, which would avoid such 
> > problems also in the future. A downside might be that tool integration is 
> > not as good, e.g. for travis, but I'm not sure about that. Alternatively we 
> > could just wait, assuming that this particular problem was overlooked and 
> > github will change its ToS. After all this affects all open source projects 
> > that do not have a contributors agreement that transfers all rights, 
> > including e.g. the Kernel.
> > 
> > Cheers,
> > Florian 
> > 
> > 
> > 
> > On Mon, Mar 6, 2017 at 9:47 AM Kai Szuttor <address@hidden 
> > <mailto:address@hidden>> wrote:
> > 
> >     Maybe we should watch this SO discussion:
> > 
> >     
> > http://opensource.stackexchange.com/questions/5193/are-the-new-github-terms-of-service-a-kiss-of-death-for-open-source-projects
> > 
> >     The current opinion seems to be that the new ToS grants gihub that you
> >     have the rights for the content to give github the rights to use your
> >     content without displaying your license.
> > 
> >     Cheers,
> > 
> >     Kai
> >     On Fri, Mar 03, 2017 at 06:09:18PM -0500, Ulf Schiller wrote:
> >     > I had a chance to discuss with our IP officer, and here's my current
> >     > understanding.
> >     >
> >     > Section D.4 of the GitHub terms state: "If you set your pages and
> >     > repositories to be viewed publicly, you grant each User of GitHub a
> >     > nonexclusive, worldwide license to access your Content through the 
> > GitHub
> >     > Service, and to use, display and perform your Content, and to 
> > reproduce your
> >     > Content solely on GitHub as permitted through GitHub's functionality. 
> > You
> >     > may grant further rights if you adopt a license."
> >     >
> >     > It is important to note that, e.g., the GPL does not simply grant 
> > rights, it
> >     > grants rights under certain provisions. The above phrasing does not
> >     > explicitly state that the rights granted each User of GitHub are 
> > subject to
> >     > the terms of the license adopted, if so chosen, thus potentially 
> > creating a
> >     > loophole for other users to strip off the adopted license. This is 
> > exactly
> >     > what the Copyleft of, e.g., the GPL seeks to prevent (cf. Section 4 
> > of the
> >     > GPL). The problem here is that one effectively grants two licenses 
> > which may
> >     > have incompatible provisions. Given that many people have contributed 
> > to
> >     > ESPResSo under the terms of the GPL and may thus still own copyright 
> > for
> >     > their contributions, it may be legally problematic to convey the 
> > whole or
> >     > parts of ESPResSo under any other license (again, this is the intent 
> > of the
> >     > GPL).
> >     >
> >     > I am not saying that there are any sinister intentions on the part of 
> > GitHub
> >     > nor that any rashly action is necessary (as neither did the article I
> >     > linked, in fact). I am just pointing out that there is a loophole in 
> > the
> >     > GitHub Terms of Use that one should be aware of. If my understanding 
> > is
> >     > wrong, I'll be glad to be corrected.
> >     >
> >     > Thank you,
> >     > Ulf
> >     >
> >     > On 03/03/2017 02:47 PM, Ulf Schiller wrote:
> >     > >First of all, apologies for the "tracking cancer". I find it equally
> >     > >annoying but it is unfortunately beyond my control.
> >     > >
> >     > >Second, I have shared the link to create awareness of the potential
> >     > >issue and initiate a discussion.
> >     > >
> >     > >Third, I think most subscribers of this list will appreciate
> >     > >substantiating evidence for the claims that the "article is 
> > completely
> >     > >exaggerated" and that this case "is completely unrealistic". It may 
> > be
> >     > >the case, I simply don't know. Unsubstantiated claims, however, are 
> > by a
> >     > >vast majority of the scientific community considered bad practice.
> >     > >
> >     > >Thank you,
> >     > >Ulf
> >     > >
> >     > >On 03/03/2017 11:59 AM, Henri Menke wrote:
> >     > >>First of all the link without the tracking cancer:
> >     > 
> > >>https://urldefense.proofpoint.com/v2/url?u=https-3A__www.mirbsd.org_permalinks_wlog-2D10-5Fe20170301-2Dtg.htm&d=DwIDaQ&c=Ngd-ta5yRYsqeUsEDgxhcqsYYY1Xs5ogLxWPA_2Wlc4&r=9HKBLELzR4cr8VDrFLGdom3-YW9ayOJQwH65Vt5eFQc&m=-zZTLcpCDdmUGfEFJcPUujvKxlXml-od5pY77JpNPsU&s=EebocPPCx5iyFtrgZ9Nht_NMVEvO34OfFIQbcImKyZo&e=
> >     > >>
> >     > >>Second, we don’t have a problem and this article is completely
> >     > >>exaggerated, especially because this would mean that approximately 
> > 90%
> >     > >>of all projects on GitHub would be taken down.
> >     > >>
> >     > >>Third, if this is actually the case (which is completely 
> > unrealistic)
> >     > >>we just move to 
> > https://urldefense.proofpoint.com/v2/url?u=https-3A__gitlab.com_&d=DwIDaQ&c=Ngd-ta5yRYsqeUsEDgxhcqsYYY1Xs5ogLxWPA_2Wlc4&r=9HKBLELzR4cr8VDrFLGdom3-YW9ayOJQwH65Vt5eFQc&m=-zZTLcpCDdmUGfEFJcPUujvKxlXml-od5pY77JpNPsU&s=adzxGbNlkZfQ_qmKPnE2J-nNg-JMd0FJKfC37pwbpGc&e=
> >     > >>
> >     > >>On 03/03/2017 05:34 PM, Ulf Schiller wrote:
> >     > >>>Stuttgart, we (may) have a problem...
> >     > >>>
> >     > 
> > >>>https://urldefense.proofpoint.com/v2/url?u=https-3A__www.mirbsd.org_permalinks_wlog-2D10-5Fe20170301-2Dtg.htm&d=DwICaQ&c=Ngd-ta5yRYsqeUsEDgxhcqsYYY1Xs5ogLxWPA_2Wlc4&r=9HKBLELzR4cr8VDrFLGdom3-YW9ayOJQwH65Vt5eFQc&m=24y_szbgE4sBUEMNZ-RRUdRRc6VOxrgeS0BSjdf9QHY&s=cFwd9YLHoQmAIZxA3oAH3t7v3bOc_0uquQbGS9Vkdp8&e=
> >     > >>>
> >     > >
> >     >
> >     > --
> >     > Dr. Ulf D. Schiller
> >     > Assistant Professor, Department of Materials Science and Engineering
> >     > Faculty Scholar, School of Health Research
> >     > Clemson University
> >     > 161 Sirrine Hall
> >     > Clemson, SC 29634
> >     >
> >     > Office: 299c Sirrine Hall
> >     > Phone: 1-864-656-2669 <tel:(864)%20656-2669>
> >     > Fax: 1-864-656-5973 <tel:(864)%20656-5973>
> > 
>

pgpUBooUz8jMR.pgp
Description: PGP signature