Re: Perhaps we're rehashed enough of the systemd discussions?

[Scott Kitterman]

On November 4, 2019 5:11:22 PM UTC, Russ Allbery  wrote:
>Jonas Smedegaard  writes:
>
>> Scott is refering to Wouter's "prepare a GR proposal in private".
>
>> So a secret _preparation_ announced publicly which triggering the 
>> thread.
>
>I frequently run ideas past a small number of people before posting
>them
>to a large audience just to make sure I don't make some stupid mistake.
>For highly political topics, that's even more important because opening
>a
>public discussion with something that doesn't get at the actual problem
>or
>misrepresents one side can cause a lot of noise and confusion and
>derail
>the attempt to solve the real problem.
>
>If you're trying to resolve a contentious dispute between (at least)
>two
>groups of people, giving them a heads-up first and making sure that
>you're
>being fair to their arguments is a pretty reasonable way to start.
>
>I think people are reading a bit too much into this.  If anything, the
>problem was that Sam was extremely transparent about what he was
>preparing
>to do next, which prematurely started the conversation (for which I'm
>as
>responsible as anyone else).  We were going to have the conversation
>anyway; I'm sure Sam wasn't planning on posting a GR and immediately
>calling for a vote.

No doubt.  I realize Sam is in a bit of a damned if you do/damned if you don't 
situation here.  If he'd waited until he was ready with GR text, including 
private consultation with some, then people (possibly including me) would be 
whining about a cabal.

OTOH, this is an important topic for the project and so I don't think a 
spirited debate should surprise anyone.  I don't think it's likely to stop 
until well after the GR vote is complete.  The best way, IMO, to end a long, 
draining thread on debian-devel is to publish the GR text so we can have 
another long, draining thread on debian-vote.

Scott K

Re: Perhaps we're rehashed enough of the systemd discussions?

[Russ Allbery]

Jonas Smedegaard  writes:

> Scott is refering to Wouter's "prepare a GR proposal in private".

> So a secret _preparation_ announced publicly which triggering the 
> thread.

I frequently run ideas past a small number of people before posting them
to a large audience just to make sure I don't make some stupid mistake.
For highly political topics, that's even more important because opening a
public discussion with something that doesn't get at the actual problem or
misrepresents one side can cause a lot of noise and confusion and derail
the attempt to solve the real problem.

If you're trying to resolve a contentious dispute between (at least) two
groups of people, giving them a heads-up first and making sure that you're
being fair to their arguments is a pretty reasonable way to start.

I think people are reading a bit too much into this.  If anything, the
problem was that Sam was extremely transparent about what he was preparing
to do next, which prematurely started the conversation (for which I'm as
responsible as anyone else).  We were going to have the conversation
anyway; I'm sure Sam wasn't planning on posting a GR and immediately
calling for a vote.

-- 
Russ Allbery (r...@debian.org)  

Re: Perhaps we're rehashed enough of the systemd discussions?

[Theodore Y. Ts'o]

On Mon, Nov 04, 2019 at 11:16:15AM +, Holger Levsen wrote:
> > I'm wrong, but even before it's been published, I feel like it's already 
> > caused people to become more firmly entrenched in their existing positions.
> 
> I was sceptical whether a GR on systemd at this point in time was
> useful, but Russ convinced me it is. (I was and am quite annoyed by yet
> another systemd thread on -devel the last weeks, and I now think that
> Russ is right that a GR will finally allow us to move forward on this
> topic.)

Same here.  I was initially skeptical, because I was hoping that the
status quo without a vote would be, "Debian maintainers should make
best effort attempts to accomodate Alternative Init Systems (AIS);
people who want to work on accomodating AIS should do so; Debian
Developers should try not to make things worse for AIS."  My hope was
that in a few years, it would become clear what the right outcome
should be, based on the actions of the overall Linux and Open Source
ecosystem.

But it's become clear that we don't have consensus that this is what
the status quo is, or should be.  And perhaps having a GR will allow
us to at least get a good sense of what the overall project believes
to be the best path forward.

- Ted

Re: Perhaps we're rehashed enough of the systemd discussions?

[Jonas Smedegaard]

Quoting Holger Levsen (2019-11-04 12:16:15)
> On Mon, Nov 04, 2019 at 03:27:29AM -0500, Scott Kitterman wrote:
> > Personally, I've no idea what enough is since I'm still seriously concerned 
> > this secret GR will make a bad situation worse.
> 
> I'm not sure why you call it a secret GR, there are no secret GRs.

Scott is refering to Wouter's "prepare a GR proposal in private".

So a secret _preparation_ announced publicly which triggering the 
thread.

 - Jonas

-- 
 * Jonas Smedegaard - idealist & Internet-arkitekt
 * Tlf.: +45 40843136  Website: http://dr.jones.dk/

 [x] quote me freely  [ ] ask before reusing  [ ] keep private


signature.asc
Description: signature

Re: Perhaps we're rehashed enough of the systemd discussions?

[Holger Levsen]

On Mon, Nov 04, 2019 at 03:27:29AM -0500, Scott Kitterman wrote:
> Personally, I've no idea what enough is since I'm still seriously concerned 
> this secret GR will make a bad situation worse.

I'm not sure why you call it a secret GR, there are no secret GRs.

> Personally, I think your GR is highly unlikely to make things better.  I hope 
> I'm wrong, but even before it's been published, I feel like it's already 
> caused people to become more firmly entrenched in their existing positions.

I was sceptical whether a GR on systemd at this point in time was
useful, but Russ convinced me it is. (I was and am quite annoyed by yet
another systemd thread on -devel the last weeks, and I now think that
Russ is right that a GR will finally allow us to move forward on this
topic.)

However this is also because I'm pretty sure what will happen once Sam
proposes his GR draft: there will be alternative resolutions and
amendments proposed, we will have more discussions and then we will have
a vote.

IOW: it doesnt really matter that much whether Sam will propose a good
or a bad GR, I'm confident that debian-vote@ will produce good options
to vote on and I also think we will get a good result out of such a GR.

Whatever a good result here means. ;)

(And I should maybe also say that I believe Sam will propose a good GR,
but thats mostly besides my point.)


-- 
cheers,
Holger

---
   holger@(debian|reproducible-builds|layer-acht).org
   PGP fingerprint: B8BF 5413 7B09 D35C F026 FE9D 091A B856 069A AA1C



signature.asc
Description: PGP signature

Re: Perhaps we're rehashed enough of the systemd discussions?

[Scott Kitterman]

On Sunday, November 3, 2019 3:26:40 PM EST Sam Hartman wrote:
> > "Wouter" == Wouter Verhelst  writes:
> Wouter> On Thu, Oct 31, 2019 at 03:45:47PM +0100, Marco d'Itri wrote:
> >> http://www.islinuxaboutchoice.com/
> 
> Wouter>
> https://grep.be/blog/en/computer/cluebat/Systemd__Devuan__and_Debian/
> 
> Wouter> -- To the thief who stole my anti-depressants: I hope you're
> Wouter> happy
> 
> Hi.  We've received a lot of feedback about the negative costs of long
> threads here.
> I think that some aspects of the discussion were necessary and
> inevitable.
> It sounds like the discussion has helped some people understand why a GR
> might be a good idea.
> It has helped me sanity check whether I was including issues important
> to the project.
> 
> I think that we'll need to discuss a few other things once again when
> draft GR text comes out (although that will be on debian-vote).
> 
> But as someone charged with helping try to facilitate discussions within
> the project, I think this particular set of related threads have reached
> the point of diminishing returns on debian-devel for now.
> If there's some last message you absolutely have to send, get that *one
> last message* out of your system.
> But let's be almost done, out of respect for everyone who has had this
> discussion before and out of the many people who have talked about the
> costs of relatively long threads like this.
> 
> Thanks for considering,

Personally, I've no idea what enough is since I'm still seriously concerned 
this secret GR will make a bad situation worse.  Wouter's last email really 
struck a chord with me.  

It's possible for us to have a reasonable init system thread on debian-devel, 
but this isn't way to do it.  I felt like the discussion that ensued from me 
asking about the utility of sysv init scripts in docker instances was very 
helpful and I learned some things without there being too much stress of flames 
being thrown.

Personally, I think your GR is highly unlikely to make things better.  I hope 
I'm wrong, but even before it's been published, I feel like it's already 
caused people to become more firmly entrenched in their existing positions.

Scott K

signature.asc
Description: This is a digitally signed message part.

Re: Perhaps we're rehashed enough of the systemd discussions?

[Wouter Verhelst]

On Sun, Nov 03, 2019 at 03:26:40PM -0500, Sam Hartman wrote:
> > "Wouter" == Wouter Verhelst  writes:
> 
> Wouter> On Thu, Oct 31, 2019 at 03:45:47PM +0100, Marco d'Itri wrote:
> >> http://www.islinuxaboutchoice.com/
> 
> Wouter> 
> https://grep.be/blog/en/computer/cluebat/Systemd__Devuan__and_Debian/
> 
> Wouter> -- To the thief who stole my anti-depressants: I hope you're
> Wouter> happy
> 
> Hi.  We've received a lot of feedback about the negative costs of long
> threads here.

Yes.

[...]
> But as someone charged with helping try to facilitate discussions within
> the project, I think this particular set of related threads have reached
> the point of diminishing returns on debian-devel for now.
> If there's some last message you absolutely have to send, get that *one
> last message* out of your system.
> But let's be almost done, out of respect for everyone who has had this
> discussion before and out of the many people who have talked about the
> costs of relatively long threads like this.

I wasn't going to reply to this at first, but...

I'm not sure if you replying to my particular email has any
significance. But if it does, I think that's pretty terribly bad of you.

I replied exactly twice to this whole thread: once to ask you to *not*
prepare a GR proposal in private, *precicely because* announcing you're
going to propose a GR but not actually (yet) doing so has in the past
led to long and very much not productive threads about the general
subject the almost-proposed GR is about. You declined, for reasons that
I didn't agree with (but didn't feel strongly enough about to argue to
death). A second reply was the above one, to point out a particularly
bad style of "argument", of pointing to a URL that kindof okayishly
defends the way one of our competing distributions works, but doesn't
(at least in my opinion) match up with Debian's traditions very well. I
happen to believe that using a bad style of argument in an already-long
thread is not very productive, and I wanted to point that out in a short
and simple way.

So now that you've seen that your actions have caused a huge and
inproductive thread, rather than owning up to that and just doing the
right thing, you're pointing fingers and telling *me* that I'm making a
bad situation worse?

Srsly. Yes, the DPL is supposed to "lead discussion", but no, you do
*not* do that by telling people to shut up when they're just reacting to
*your* mistakes.

Thanks for considering.

-- 
To the thief who stole my anti-depressants: I hope you're happy

  -- seen somewhere on the Internet on a photo of a billboard

Perhaps we're rehashed enough of the systemd discussions?

[Sam Hartman]

> "Wouter" == Wouter Verhelst  writes:

Wouter> On Thu, Oct 31, 2019 at 03:45:47PM +0100, Marco d'Itri wrote:
>> http://www.islinuxaboutchoice.com/

Wouter> 
https://grep.be/blog/en/computer/cluebat/Systemd__Devuan__and_Debian/

Wouter> -- To the thief who stole my anti-depressants: I hope you're
Wouter> happy

Hi.  We've received a lot of feedback about the negative costs of long
threads here.
I think that some aspects of the discussion were necessary and
inevitable.
It sounds like the discussion has helped some people understand why a GR
might be a good idea.
It has helped me sanity check whether I was including issues important
to the project.

I think that we'll need to discuss a few other things once again when
draft GR text comes out (although that will be on debian-vote).

But as someone charged with helping try to facilitate discussions within
the project, I think this particular set of related threads have reached
the point of diminishing returns on debian-devel for now.
If there's some last message you absolutely have to send, get that *one
last message* out of your system.
But let's be almost done, out of respect for everyone who has had this
discussion before and out of the many people who have talked about the
costs of relatively long threads like this.

Thanks for considering,

--Sam