| Message ID | cover.1676213809.git.jlicht@fsfe.org |
|---|---|
| Headers | show |
| Series | Expose upstream linux sources | expand |
Am Sonntag, dem 12.02.2023 um 16:39 +0100 schrieb jlicht@fsfe.org:
> * Is this fundamentally going to be an issue with the FSDG?
A free system distribution must not steer users towards obtaining any
nonfree information for practical use, or encourage them to do so.
Cheers
Hi Jelle! jlicht@fsfe.org 写道： > * Is this fundamentally going to be an issue with the FSDG? I can't think of a reading of the FSDG where it is not against both the letter and the spirit. Guix already ventures close to the edge; this would push us clean over. > * This is the 'dumb' solution; alternatively, I was thinking of > introducing a > record to unify all the moving parts (the upstream sources, > the deblob > scripts) involved in building our linux-libre kernels. I'm not sure this will suit your purposes any better without amounting to the same thing. Kind regards, T G-R
Hi Tobias, Tobias Geerinckx-Rice <me@tobias.gr> writes: > Hi Jelle! > > jlicht@fsfe.org 写道： >> * Is this fundamentally going to be an issue with the FSDG? > > I can't think of a reading of the FSDG where it is not against > both the letter and the spirit. > > Guix already ventures close to the edge; this would push us clean > over. Fair enough. I'll go ahead and assume that any "workaround" using public bindings exposed by guix can be considered a bug, later to be addressed by guix in order to prevent similar situations. >> * This is the 'dumb' solution; alternatively, I was thinking of >> introducing a >> record to unify all the moving parts (the upstream sources, >> the deblob >> scripts) involved in building our linux-libre kernels. > > I'm not sure this will suit your purposes any better without > amounting to the same thing. It does, so never mind! Thanks for your input, - Jelle
From: Jelle Licht <jlicht@fsfe.org> Hey guix, This patch series should not lead to any rebuilds. The aim is to expose the used linux upstream sources, for use by custom local packages or external channels. Two main concerns I still have: * Is this fundamentally going to be an issue with the FSDG? * This is the 'dumb' solution; alternatively, I was thinking of introducing a record to unify all the moving parts (the upstream sources, the deblob scripts) involved in building our linux-libre kernels. That bigger change is ideally something we achieve consensus on before writing some code. Jelle Licht (5): gnu: linux-libre 4.14: Expose upstream sources. gnu: linux-libre 4.19: Expose upstream sources. gnu: linux-libre 5.10: Expose upstream sources. gnu: linux-libre 5.15: Expose upstream sources. gnu: linux-libre 6.1: Expose upstream sources. gnu/packages/linux.scm | 47 ++++++++++++++++++++++++------------------ 1 file changed, 27 insertions(+), 20 deletions(-) base-commit: fb9799ff5f1d90a443dc197535c48041ad6b3865