Mime-Version: 1.0
Content-Transfer-Encoding: quoted-printable
Content-Type: text/plain; charset=UTF-8
Date: Mon, 17 Oct 2022 22:40:29 +0200
Message-Id: <CNOHM83VUDDZ.2QMYUC8MIPF5F@sumire>
To: "Ariadne Conill" <ariadne@dereferenced.org>
Cc: "Riccardo Mottola" <riccardo.mottola@libero.it>, "Winston Weinert"
 <winston@ml1.net>, <~alpine/users@lists.alpinelinux.org>
Subject: Re: ipw2100 wifi card firmware
From: "alice" <alice@ayaya.dev>
References: <bdb05306-0f84-50d6-c71c-ff4b311d7ffb@libero.it>
 <005dc987-0461-9189-383e-d690b7686e96@ml1.net>
 <d03805ad-4f6e-ef23-db58-dde2bf32ae45@libero.it>
 <CNOFY00I1FSZ.294LCIZHOORKP@sumire>
 <eda4842d-5be7-5dfb-c9c6-23f927764d46@dereferenced.org>
 <CNOGTU7H8RZ5.3BK216FYLUEVO@sumire>
 <7eddf22-613c-7d48-9a78-12609bb8712@dereferenced.org>
 <CNOH9O1DWNVX.37UPC3Q0KL05E@sumire>
 <8a60ec67-6fe-c8ab-8c11-45423a5a02e@dereferenced.org>
In-Reply-To: <8a60ec67-6fe-c8ab-8c11-45423a5a02e@dereferenced.org>

On Mon Oct 17, 2022 at 10:34 PM CEST, Ariadne Conill wrote:
> Hi,
>
> On Mon, 17 Oct 2022, alice wrote:
>
> > On Mon Oct 17, 2022 at 10:20 PM CEST, Ariadne Conill wrote:
> >> Hi,
> >>
> >> On Mon, 17 Oct 2022, alice wrote:
> >>
> >>> On Mon Oct 17, 2022 at 9:28 PM CEST, Ariadne Conill wrote:
> >>>> Hi,
> >>>>
> >>>> On Mon, 17 Oct 2022, alice wrote:
> >>>>
> >>>>> On Mon Oct 17, 2022 at 9:07 PM CEST, Riccardo Mottola wrote:
> >>>>>> Hi Winston,
> >>>>>>
> >>>>>> I was unable to download the firmware directly from sourceforge.ne=
t, my
> >>>>>> browser(s) continue to fail the accept test!
> >>>>>> However, I did as you suggested: got the Debian package, "ar x" an=
d then
> >>>>>> tar the data file... put the firmware in /lib/firmware... and yes,=
 my
> >>>>>> ThinkPad connects to the wireless, the WiFi LED lights up.
> >>>>>>
> >>>>>> I wonder why Alpine cannot have such a package? I have seen other
> >>>>>> non-free firmware (how much I hate that... a reason why I skipped =
all
> >>>>>> those "libre" distributions: without binary blobs a laptop is almo=
st
> >>>>>> useless nowadays).
> >>>>> alpine doesn't ship anything non-free, with really the only excepti=
on
> >>>>> being firmware under the "linux-firmware" project umbrella (and any=
thing
> >>>>> overlooked, of course). that project (afaik) has the vendors adding
> >>>>> firmware to it by petitioning it directly, and it's more or less a
> >>>>> community consensus that it gets a pass since otherwise nothing is
> >>>>> really usable at all. that part is the distinction from "libre" dis=
tros
> >>>>> that ship nothing and let you use them on like 5 laptops (maybe a b=
it
> >>>>> harsh, but you know what i mean).
> >>>>>
> >>>>> but as for everything else, it is all free software. no shortcuts t=
here.
> >>>>>
> >>>>> as for the ipw2x00 firmware, the licence isn't really much differen=
t than a
> >>>>> lot of things in linux-firmware:
> >>>>>
> >>>>> ipw2100 firmware license For OEMs, IHVs, and ISVs:
> >>>>> =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D
> >>>>>
> >>>>> LICENSE. This Software is licensed for use only in conjunction with
> >>>>> Intel component products. Use of the Software in conjunction with
> >>>>> non-Intel component products is not licensed hereunder. Subject to =
the
> >>>>> terms of this Agreement, Intel grants to you a nonexclusive,
> >>>>> nontransferable, worldwide, fully paid-up license under Intel's
> >>>>> copyrights to: (i) copy the Software internally for your own develo=
pment
> >>>>> and maintenance purposes; (ii) copy and distribute the Software to =
your
> >>>>> end-users, but only under a license agreement with terms at least a=
s
> >>>>> restrictive as those contained in Intel's Final, Single User Licens=
e
> >>>>> Agreement, attached as Exhibit A; and (iii) modify, copy and distri=
bute
> >>>>> the end-user documentation which may accompany the Software, but on=
ly in
> >>>>> association with the Software.
> >>>>>
> >>>>> but it's still nonfree. i guess adding it to the repositories would
> >>>>> require the same exemption linux-firmware does.
> >>>>>
> >>>>> https://git.kernel.org/pub/scm/linux/kernel/git/firmware/linux-firm=
ware.git/
> >>>>> (read the various LICENSE.* files in there)
> >>>>
> >>>> I think it would be better if Intel just submitted it to linux-firmw=
are
> >>>> instead.
> >>> "just" is certainly doing a lot of heavy lifting there. i wouldn't mi=
nd
> >>> perhaps sending an email somewhere, but i don't really know where to
> >>> even start aside from generic-corporate-support-address or
> >>> very-specific-personal-email-to-intel-linux-engineer-via-lkml (bad). =
i
> >>> doubt they'd do it anyway, but it's worth a try if there's a non-gene=
ric
> >>> point of contact you can point me to.
> >>
> >> I will see what I can dig up through various channels.  I think that i=
t
> >> should not be a problem to get Intel to submit their firmware through =
that
> >> channel.
> > sounds good. hopefully they do :)
> >
> >> An alternative is that we include it in linux-firmware ourselves, but =
I
> >> don't necessarily like the idea of opening that pandora's box.
> > yeah, me neither. though it's an opened one, i think, as it's what
> > debian does. maybe they were lax here, or maybe they researched it a
> > bunch somewhere i haven't looked for:
> > https://packages.debian.org/bullseye/firmware-ipw2x00
> >
> > the .orig is the original firmware, and the .debian addition (verified
> > via the source packages) contains a few custom extras, this package
> > being one of them.
>
> Debian choosing to do it that way does not mean that we are also committe=
d=20
> to doing the same way, of course.
>
> My concern about going that way is that we probably need to have strict=
=20
> licensing requirements for any firmware we add there, otherwise everyone=
=20
> will want firmware added there, even if we do not actually have permissio=
n=20
> to redistribute it.
yeah, of course. the first thing i checked was the actual licence of the
thing in question. in that hypothetical future, everyone else would be
subject to the same checking.. which would mean someone would have to
specialise doing it. i don't think i'd even count, since i'm not a
lawyer (and i prefer to never read these things if i don't have to). a
pretty bad can of worms to open there..

>
> Ariadne