On Sun, 8/21/16, Daniel Reurich <daniel@???> wrote:

Subject: Re: [DNG] eudev [was: vdev]
To: dng@???
Date: Sunday, August 21, 2016, 3:08 PM
>>
>> He guys,
>>
>> I've been at work for a week or so and today I looked at the DNG list
>> for the latest activities around vdev, but there has almost been no
>> activity on vdev as far as I can see. OTOH, last week I tested eudev on
>> a separate partition and that seems to work quite well.
>
> You are mistaken, there has been lots of activity around vdev and making
> it installable.
>
>>
>> I think it might be a good idea to leave vdev for what it is and
>> to switch to eudev. It is moreorless maintained (the latest change is
>> two weeks ago) and it works well. We should not reinvent the wheel IMHO.
>> And as there has been no response from the original vdev author, I think
>> it's better to package eudev for Devuan and to make it available for
>> Jessie and Ascii. The latest version is 3.2.
>
> Well quite frankly you don't get to make that call. Eudev is just a
> hack that from what I gather is isolating the systemd-udev changes and
> bringing them in to eudev. IMHO that is less sustainable then vdev
> because it relies on developers from systemd to play nice with udev and
> not deprecate features that don't serve systemd's needs. At the end of
> the day, I consider eudev as at best marginally better the eudev, but
> still far to closely coupled with systemd to be useful in the medium to
> long term.
>
> With regards to vdev, I'm sure if Jude didn't come back, others would
> pick up his work and progress it, as is happening now around packaging
> it. I think it rather disingenuous of you to imply it's a dead project
> whilst claiming that eudev, the re-animated zombie of systemd-udev as a
> better and only option. It's not better, and it's not the only option
> either.
>
> Whilst I respect the work to package eudev and having it as an option in
> Devuan, I will personally very loudly push back on any attempt to derail
> alternatives such as vdev - unless those alternative are demonstrably
> built on the same flawed design principles as systemd.
>
> Daniel.
>

--------------------------------------------

I agree that relying on anything connected to udev will likely not be sustainable in the long term. I was reminded of this just today in a private discussion I'm having with someone over at FDN . . . yes, I still hang out there to advocate for non-systemd Linux. S/he posted this link which finally pushed them over the edge and away from the path that Debian has taken:

https://lists.freedesktop.org/archives/systemd-devel/2014-May/019657.html

I remember that from the time it was written and heated discussions were swirling on FDN and debian-user. Let it be a reminder to all of you also.

Now that I finally have qemu up and running (thanks to fsmithred), I'll be happy to test any vdev isos from aitor and hope that many of you will do the same. It's time to buckle down and get serious . . .

golinux