Hi Yves-Alexis
On 6/6/2020 10:43 AM, Yves-Alexis Perez wrote:
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA256
On Fri, 2020-06-05 at 11:58 -0400, Mark Pearson wrote:
That's where doing a Debian pre-load would be challenging (which
is really where this conversation started). If support for a
platform isn't there until 6 to 8 months after it's shipped (or
more) then really it's not worth doing a preload (note - I'm not
saying it's not worth supporting the platform - that is still
important0. Fedora have this somewhat solved by being on the latest
of everything, Ubuntu solve it by having their oem image model.
Hi Mark and thanks for the nice thread.
I have the feeling that support comes late to Lenovo machines because
porting efforts start when the machine are actually available and in
the hands of willing developpers. That was my experience when I
bought my trusty X250 back in 2015. I've documented part of the
process (https://www.corsac.net/X250/): because I bought the machine
early in its life, support was not perfect (although honestly it was
really good) and I had to do some backporting myself, poke some
Debian or upstream maintainers here and there.
For quite some years now volunteer do that for IBM then Lenovo
hardware (and others as well), but obviously they can only do it once
they bought it, received it and started playing with it. Lenovo
obviously has access to the hardware way earlier, and could thus
start the porting effort (which is mostly shared between
distributions anyway because work has to be done upstream) and then
make sure it propagates to the various supported distro in time for
the release.
What can be done in the Debian community to help you do that *before*
the hardware are in the hand of volunteers, because as you already
said that means the laptops work perfectly 6-8 months after the
release which is too late.
That's a really good question. I was thinking about it over the weekend
and I'm not sure what the right answer is here. If Lenovo were able to
make systems available earlier who would they go to? How does that work
with a community like Debian? If NDAs are involved (which would depend
on how early in the process you get HW) is that a problem?
My plan before this conversation came up was to keep an eye on what
fixes were needed to get things working on the Ubuntu/Fedora/RHEL front
and then once those were upstream work on getting those pulled into Debian.
Obviously having more competent people than myself do that process and
be able to test it directly would speed things up (a lot :)) but that's
potentially a bunch of work to place on a few people (due to limited
HW). On the plus side - my understanding is that whoever worked on it
would get to keep the HW...don't know if that is tempting or not :)
I think it's very important for Lenovo to become active and competent
contributors to the community. I don't think it is healthy for us to
just dump HW and say "please fix" - we really should be contributing to
the community for our HW. Maybe a hybrid model where some Debian folk
with some time and interest get HW and are willing to mentor/support?
I'm guessing the first couple of platforms would be more challenging but
in theory it would get easier and less demanding (hopefully :))?
Anyway - if there was interest we could explore what was involved with
choosing a couple of platforms, getting them to a Debian developer or
two and going from there. I don't know how early in the process we'd be
able to make HW available - it is *really* hard to get hold of these
early systems (based on personal experience). That's a challenge I'm
willing to take on if it's something that there is interest in.
Mark
