Re: Licensing terms for data derived from ARB/SILVA data
Hi,
what aout putting barnap into main and barrnap-additional-data
into non-free? Packages in non-free do not gain the same QA means
which we do nit need for the non-free data.
Greetings from Hanoi
Andreas.
On Sat, Nov 29, 2014 at 01:14:20PM +0000, Sascha Steinbiss wrote:
> Hi Olivier and all,
>
> thanks for the feedback. I have just adapted the barrnap package to go
> into the non-free area and added the respective licenses to d/copyright.
> If there are no other suggestions, I would consider the package to be
> ready for upload then.
> I will also reply to Frank Gloeckner and inform Torsten Seemann about
> the results of the enquiry later.
>
> Thanks
> Sascha
>
> On 29/11/14 11:49, olivier.sallou@codeless.fr wrote:
> > On 11/29/2014 12:29 PM, Sascha Steinbiss wrote:
> >> Hi all,
> >>
> >>>> But we would like to ask you to clearly indicate that the profiles are
> >>>> only free for non-commercial usage, commercial users need a licence.
> >>>> The easiest approach is to add a reference to the SILVA terms of
> >>>> license assigned to the software/profile.
> >>> Argghh! this will prevent from being in free section. And I suppose
> >>> those profiles are mandary for software usage ?
> >> To ensure that the Debian version and the upstream version produce the
> >> same results, yes, they are necessary. I guess there are two choices now:
> >>
> >> 1. Put barrnap in non-free. What consequences would that have for a
> >> 'regular' academic user -- they would have to enable non-free on a
> >> default install, right?
> > Yeap, they need to add non-free to apt sources
> >>
> >> 2. Use an alternative set of pHMMs in the Debian version of barrnap,
> >> only including those which are built from Rfam alignments only (Rfam
> >> has a free license (CC0)). However, that would strongly limit the
> >> functionality of barrnap as whole rRNA families could end up
> >> being missing from the output of the dfsg compliant version.
> >>
> >> I guess to have a usable barrnap version, 1. is the only viable
> >> solution. Any comments?
> > I agree
> >> Thanks
> >> Sascha
> >
> >
>
>
> --
> To UNSUBSCRIBE, email to debian-med-REQUEST@lists.debian.org
> with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
> Archive: [🔎] 5479C6AC.2080800@steinbiss.name">https://lists.debian.org/[🔎] 5479C6AC.2080800@steinbiss.name
>
>
--
http://fam-tille.de
Reply to: