Re: Bits from the DPL For December 2019

classic Classic list List threaded Threaded
5 messages Options
Reply | Threaded
Open this post in threaded view
|

Re: Bits from the DPL For December 2019

Wouter Verhelst
Hi Sam,

On Tue, Jan 21, 2020 at 02:29:31PM -0500, Sam Hartman wrote:

> Seeking Volunteers for Deescalation
> ===================================
>
> While discussing the role of the Community Team, I talked about my
> desire for the community team to help deescalate conflict.  I used the
> dread m-word (mediation) and confused us all.
> Russ and others expressed doubt that this sort of deescalation was
> something we could accomplish.
>
> I think it is really important.  We've demonstrated that we can make
> decisions; we can effect change.  But for it to be healthy to do so, I
> think we need to deescalate and heal.
>
> The first question is whether we can find volunteers to work on that.
> I don't know whether it will fit into the Community Team: I think we
> need to see who volunteers, get input from the Community Team, and get
> input from the project.
>
> If you would be willing to help with deescalation, please reach out.
>
> Sorts of things this might include:
>
> * Working with people to make sure they are heard--whether it is a
>   technical issue or a community issue.  The point is not to judge or
>   mediate, but to make sure that people and their opinions are not lost
>   in the noise.  And of course to raise the issue if they are getting
>   lost.
>
> * work with people to split meta-issues away from discussions.  So for
>   example during the conduct discussions this December, provide
>   reassurance that questions about our standards would eventually be
>   heard, but help separate that from a discussion where we were trying
>   to stand behind our transgender community.
>
> * work with people so they walk away from discussions feeling that they
>   were considered.  Try to figure out where frustration is festering and
>   respond to that.
>
> * Provide reassurance.  When we do need to take strong actions, help
>   people understand  how they can respond if they are nervous about
>   whether they are meeting our standards.  Help get them to a place
>   where they have confidence  that they could work constructively with
>   the community if there were a concern.
>
> * Help people let go of past disagreements.

I don't think any of the above is something that can be done by a team
of any sort. I believe our community has been slowly migrating to a
situation where these kinds of actions are seen as normal, and I think
we should work to encourage that going forward; but IMHO, delegating
that to a team is not going to help, on the contrary.

I specifically also disagree that trying to "split meta-issues away from
discussions" is in any way or form helpful. What may be a meta-issue to
you might be the core reason why someone else is upset about the
situation, and they may want to explain that to you to make you
understand *why* they are upset; being told in such a situation that
"we'll talk about it later" does not help deescalating things (on the
contrary). Additionally to that, telling people "we'll talk about it
later" can also be a way to (disingeniously) kill a discussion that
people don't want to have, with no intention to ever talk about it
later. In my experience, when tensions are running high, people get a
little bit paranoid about the "other side", and then being told things
like this will add oil to the fire.

I'm sorry to only shoot down suggestions and not have anything more
constructive to offer; but I can't think of anything better than
"continue to educate the community"...

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

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

Reply | Threaded
Open this post in threaded view
|

Re: Bits from the DPL For December 2019

Sam Hartman-3

TL;DR: I think we need a team to focus training and skill sharing even
if we try and drive community wide change.
I try to explore when splitting off meta issues is a good idea.

>>>>> "Wouter" == Wouter Verhelst <[hidden email]> writes:

    Wouter> I don't think any of the above is something that can be done by a team
    Wouter> of any sort. I believe our community has been slowly migrating to a
    Wouter> situation where these kinds of actions are seen as normal, and I think
    Wouter> we should work to encourage that going forward; but IMHO, delegating
    Wouter> that to a team is not going to help, on the contrary.

I hope the entire community grows at deescalation and that we all teach
each other.
However, deescalation is something you can get better at through
practice, study and investigation of techniques.
It's something that benefits from training and focus.
Thus it is something that benefits from having a group of people who
commit to spend time on the problem.
It probably benefits from the project committing money to help train
those people.

As such, I do think a team is essential.
I appreciate your reminder that eventually this is something that we can
all focus on and that any team should be leading by example and
fostering community wide change.



    Wouter> I specifically also disagree that trying to "split meta-issues away from
    Wouter> discussions" is in any way or form helpful. What may be a meta-issue to
    Wouter> you might be the core reason why someone else is upset about the
    Wouter> situation, and they may want to explain that to you to make you
    Wouter> understand *why* they are upset; being told in such a situation that
    Wouter> "we'll talk about it later" does not help deescalating things (on the
    Wouter> contrary).

I think I hear you.  Are you saying that you are concerned when we talk
about splitting off the meta issues because you worry people concerned
about these meta issues will view this as an escalation?

I think that will happen.
And so, I think that when we work with people to split off the meta
issues, we will need to actively work to deescalate the situation.
I think that splitting off the meta issues is necessary to reduce the
impact of even greater escalations that routinely make Debian a deeply
frustrating and painful place to work.


I'd ask you to consider whether you're coming at this from the viewpoint
of technical discussions and of the prevailing viewpoint from the time
the CoC was written.  I find that as the world has evolved, and as CoCs
have become more important for treating people (and not just
communications) with respect, the requirements for what we as a
community need to do have changed.  I don't know, but that difference
may be one of the ways in which we see this differently.



The importance of splitting off these issues is something I've only come
to gradually appreciate, and some of my most recent growth is based on
mail Steve wrote to a small group after the December incident.
I'll approach explaining my view from two directions.

First, in matters of behavior, the meta issues can explicitly create
situations where people do not feel welcome.
We had a thread in December where people argued that using people's
pronouns was optional.  As a community we needed to send a strong
message that was unacceptable.  We got feedback from trans members of
our community--the same people we're trying to protect--that the side
discussions diluted that message of support.

If we're going to come across as supportive when bad things happen, we
need to keep focused.

For some of the side threads I think it would have been sufficient to
just change the subject.

For others, I actually think handling them off-list and/or at a later
time with different tone would have been needed.


My second observation comes from exploration of empathy
frameworks--particularly in my case NVC.
Often when you are upset, those strong feelings get in the way of
connection--get in the way of empathy.  You cannot both go first in
building understanding.  You can give or receive empathy, but especially
when you don't have a connection now, it is difficult to impossible to
do both at the same time.

When people are upset, starting a thread is often about asking for
empathy (among other things).
Some of the side issues can come across as a refusal to give empathy
until the point of the person bringing up the side issue has been
considered.

For the technical discussions, what counts as a side issue is much more
complex and probably involves much more latitude on the person bringing
up the side issue.
I assure you it is frustrating when you bring up a current problem, and
the discussion is derailed by a blue-sky design that may solve today's
problem three years from now in an alternate universe where assumptions
are different.
In many of those cases, the person bringing up the "what if we
redesigned …" might well agree their point is a side issue.
Especially for the technical discussions, significant progress could be
made simply by changing subject lines appropriately.

--Sam

Reply | Threaded
Open this post in threaded view
|

Re: Bits from the DPL For December 2019

Wouter Verhelst
On Thu, Jan 23, 2020 at 08:26:45AM -0500, Sam Hartman wrote:

>
> TL;DR: I think we need a team to focus training and skill sharing even
> if we try and drive community wide change.
> I try to explore when splitting off meta issues is a good idea.
>
> >>>>> "Wouter" == Wouter Verhelst <[hidden email]> writes:
>
>     Wouter> I don't think any of the above is something that can be done by a team
>     Wouter> of any sort. I believe our community has been slowly migrating to a
>     Wouter> situation where these kinds of actions are seen as normal, and I think
>     Wouter> we should work to encourage that going forward; but IMHO, delegating
>     Wouter> that to a team is not going to help, on the contrary.
>
> I hope the entire community grows at deescalation and that we all teach
> each other.
> However, deescalation is something you can get better at through
> practice, study and investigation of techniques.
> It's something that benefits from training and focus.
> Thus it is something that benefits from having a group of people who
> commit to spend time on the problem.
> It probably benefits from the project committing money to help train
> those people.
>
> As such, I do think a team is essential.
> I appreciate your reminder that eventually this is something that we can
> all focus on and that any team should be leading by example and
> fostering community wide change.

OK, I see now where you're coming from. I guess that *could* work,
although it's certainly not the approach that I would take.

>     Wouter> I specifically also disagree that trying to "split meta-issues away from
>     Wouter> discussions" is in any way or form helpful. What may be a meta-issue to
>     Wouter> you might be the core reason why someone else is upset about the
>     Wouter> situation, and they may want to explain that to you to make you
>     Wouter> understand *why* they are upset; being told in such a situation that
>     Wouter> "we'll talk about it later" does not help deescalating things (on the
>     Wouter> contrary).
>
> I think I hear you.  Are you saying that you are concerned when we talk
> about splitting off the meta issues because you worry people concerned
> about these meta issues will view this as an escalation?

That's one way of putting it, yes.

> I think that will happen.
> And so, I think that when we work with people to split off the meta
> issues, we will need to actively work to deescalate the situation.
> I think that splitting off the meta issues is necessary to reduce the
> impact of even greater escalations that routinely make Debian a deeply
> frustrating and painful place to work.

So, I don't think Debian is "routinely [...] a deeply frustrating and
painful place to work". We have our disagreements from time to time, but
we do have ways of resolving them, even though some of those ways are
rather hardhanded. Having been on the "losing" side in a major change of
policies in Debian in the past[1], I can state from personal experience
that such a time is, indeed, not fun. But our conflict resolution
systems are set up so that all sides of the argument will have their
position heard *before* we make a decision. This creates such conflict
resolutions long and, yes, sometimes somewhat draining. But I am
absolutely of the opinion that it is necessary to do it that way.

It seems to me like what you're advocating here is to avoid conflict and
disagreement. I don't think that's the right thing to do.

Do correct me if I'm wrong, though ;-)

[1] I was heavily involved in the m68k port up to about half a decade
    ago (IIRC), and was strongly disappointed by the decision in 2005
    which essentially reduced the value of the m68k port for the rest of
    the project. It felt to me like the ground was removed from under
    me, when the decision happened. And while I would have preferred
    another outcome at the time, I do think today that the way in which
    the whole conflict was resolved was, ultimately, done very well, and
    it has been a very big part in allowing me to resign myself to the
    situation.

> I'd ask you to consider whether you're coming at this from the viewpoint
> of technical discussions and of the prevailing viewpoint from the time
> the CoC was written.  I find that as the world has evolved, and as CoCs
> have become more important for treating people (and not just
> communications) with respect, the requirements for what we as a
> community need to do have changed.  I don't know, but that difference
> may be one of the ways in which we see this differently.

Not necessarily.

I pushed for a Debian Code of Conduct, because I believed that it would
help the community move forward. I organized a "Community BOF" at
DebConf15 to discuss further ways in which the community could perhaps
move forward even more. Although not much came from that, I do think
it's never a good idea to rest on one's laurels in this area, and that
we may end up doing more at some point in the future.

But being told that the real issues you're trying to bring to the
discussion are non-issues for the original poster and that therefore
you're wrong for even trying to bring them up, can be extremely
distressing and demotivating. I've had that happen to me a few times
over the past year, and I think it has contributed to my being less
involved in discussions in Debian.

> The importance of splitting off these issues is something I've only come
> to gradually appreciate, and some of my most recent growth is based on
> mail Steve wrote to a small group after the December incident.
> I'll approach explaining my view from two directions.
>
> First, in matters of behavior, the meta issues can explicitly create
> situations where people do not feel welcome.
> We had a thread in December where people argued that using people's
> pronouns was optional.  As a community we needed to send a strong
> message that was unacceptable.  We got feedback from trans members of
> our community--the same people we're trying to protect--that the side
> discussions diluted that message of support.

I don't see it that way.

What I saw in that thread was people going off on a tangent. One person
said "don't do this". Another person said "I can do this because X".
Another person said "you can't do this, because Y". After three emails
of doing that you end up with a discussion that isn't about the original
point being made anymore, and I agree that such cases are not helpful.

However, such side discussions are not meta-discussions. They are
tangents, and tangents are almost never helpful.

A meta discussion is where one person says "we should come up with a way
so we can fix X", and another person says "I agree we should fix X, but
these Y and Z situations are really other cases of the same problem (and
in my opinion worse ones), and I think we should look at how we can make
sure we also fix Y and Z".

This can bring with it some level of frustration to the person
originally starting the discussion, but it is not necessarily a problem.
In fact, it could be providing another point of view that the person had
not even thought of, changing their minds as to what needs to be done.

[...don't disagree with the rest of what you said...]

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

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

Reply | Threaded
Open this post in threaded view
|

Re: Bits from the DPL For December 2019

Sam Hartman-3
>>>>> "Wouter" == Wouter Verhelst <[hidden email]> writes:
    >> First, in matters of behavior, the meta issues can explicitly create
    >> situations where people do not feel welcome.
    >> We had a thread in December where people argued that using people's
    >> pronouns was optional.  As a community we needed to send a strong
    >> message that was unacceptable.  We got feedback from trans members of
    >> our community--the same people we're trying to protect--that the side
    >> discussions diluted that message of support.

    Wouter> I don't see it that way.

    Wouter> What I saw in that thread was people going off on a tangent. One person
    Wouter> said "don't do this". Another person said "I can do this because X".
    Wouter> Another person said "you can't do this, because Y". After three emails
    Wouter> of doing that you end up with a discussion that isn't about the original
    Wouter> point being made anymore, and I agree that such cases are not helpful.

    Wouter> However, such side discussions are not meta-discussions.

Okay, so you disagree with my terminology.  I'm more focused on the harm
of the things you call tangents.  but I guess we're now having a meta
discussion about what words mean so we can go back to having our
discussion.  For myself, I'd rather focus on the effects and let other
people decide on the terminology.
It sounds like we have agreement that there are some direction shifts
that are harmful and could be separated and improve the community.
I absolutely agree we'll need to figure out when it is helpful to do
that and when it is harmful.
And choosing terminology should be part of that.

 

    >>
    >> However, deescalation is something you can get better at through
    >> practice, study and investigation of techniques.
    >> It's something that benefits from training and focus.
    >> Thus it is something that benefits from having a group of people who
    >> commit to spend time on the problem.
    >> It probably benefits from the project committing money to help train
    >> those people.
    >>
    >> As such, I do think a team is essential.
    >> I appreciate your reminder that eventually this is something that we can
    >> all focus on and that any team should be leading by example and
    >> fostering community wide change.

    Wouter> OK, I see now where you're coming from. I guess that *could* work,
    Wouter> although it's certainly not the approach that I would take.

What approach would you take?
My hope  is to improve Debian's ability to deescalate, and to do so
sooner rather than later.
    >> I think [escalations when we ask people to hold or split their
    tangents]  will happen.
    >> And so, I think that when we work with people to split off the meta
    >> issues, we will need to actively work to deescalate the situation.
    >> I think that splitting off the meta issues is necessary to reduce the
    >> impact of even greater escalations that routinely make Debian a deeply
    >> frustrating and painful place to work.

    Wouter> So, I don't think Debian is "routinely [...] a deeply frustrating and
    Wouter> painful place to work". We have our disagreements from time to time, but
    Wouter> we do have ways of resolving them, even though some of those ways are
    Wouter> rather hardhanded.

We disagree then.
I love Debian.
I invest a lot of myself here.
But there are times where I don't feel respected or valued or treated in
ways that are at all reasonable.  That's true in technical discussions
as well as community discussions.
I stick around because I hope we can improve and because I believe there
are a number of people committed to improving.

People don't bring up issues because they believe the cost of discussing
them is too high and painful.
So, no, we don't hear all sides.
We hear the sides from the people who are willing to face the pain of
our processes.
I know that we don't hear all sides because I've listened when people
have talked about times when they chose not to engage (both on-list and
in private).

I hear that you perceive things differently; I'm glad that you don't
face some of the negatives I do in Debian.

Again, I want to stress that Debian is something I value greatly,.  I do
recognize the positive aspects of our community, and I'm glad to be
here.


    Wouter> It seems to me like what you're advocating here is to avoid conflict and
    Wouter> disagreement. I don't think that's the right thing to do.

    Wouter> Do correct me if I'm wrong, though ;-)

No, not at all.
I actually have been creating more conflict during my term as DPL, and I
think that trend needs to continue.

Together we explored several tools for making decisions.  We used
consensus for the DH and Git discussions.  I facilitated those
discussions, but we've seen other developers start refining and
exploring those techniques in other contexts.

We just got done using a facilitated  GR process to make a decision
where consensus was impossible.

We've demonstrated that the people who claim Debian cannot make
decisions are *wrong*.  I'm really happy the project has been able to do
that; I'm proud of my role in facilitating that.

But oh, the cost!  We've generated more conflict.

If from where you sit, the pain and cost of that conflict isn't high,
perhaps this is not a big deal.

>From my viewpoint, the cost and pain of that conflict is high.
If we're going to continue this trend--continue to actually work on
long-standing issues, make decisions, and move forward, we will have
more conflict.
For those of us who are feeling this as painful, we're going to need to
get better at deescalating and healing for this to be healthy.

I would be disappointed to learn that yes, Debian *can* make decisions
and resolve issues, but that the price of doing so is too high, and
stagnation is preferable.
Without better skills at healing and deescalation, I think we may reach
that conclusion.

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



    Wouter> But being told that the real issues you're trying to bring to the
    Wouter> discussion are non-issues for the original poster and that therefore
    Wouter> you're wrong for even trying to bring them up, can be extremely
    Wouter> distressing and demotivating. I've had that happen to me a few times
    Wouter> over the past year, and I think it has contributed to my being less
    Wouter> involved in discussions in Debian.

I agree.
I'd rather us help people figure out appropriate ways and places for
bringing up concerns.

I don't think there are any concerns that are wrong to want to bring up.
There are some concerns that are out of scope for Debian.
As an example it's basically off topic to talk about replacing Linux
with the Windows kernel, at least unless Microsoft makes that free
software.



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

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

Reply | Threaded
Open this post in threaded view
|

Re: Bits from the DPL For December 2019

Felix Lechner-4
In reply to this post by Wouter Verhelst
Hi,

On Thu, Jan 23, 2020 at 08:26:45AM -0500, Sam Hartman wrote:
>
> I find that as the world has evolved, and as CoCs
> have become more important for treating people (and not just
> communications) with respect, the requirements for what we as a
> community need to do have changed.

Yes, the world has evolved. Respect and kindness are in short supply.

> >>>>> "Wouter" == Wouter Verhelst <[hidden email]> writes:
>
> I pushed for a Debian Code of Conduct, because I believed that it would
> help the community move forward. I organized a "Community BOF" at
> DebConf15 to discuss further ways in which the community could perhaps
> move forward even more. Although not much came from that, I do think
> it's never a good idea to rest on one's laurels in this area, and that
> we may end up doing more at some point in the future.

Indeed, more could be done.

Judgment and Compassion work in opposite directions.

The Code of Conduct stays mostly on the side of judgment (do/do not),
and the community responded in kind. Our interactions are heavy on
judgment and short of compassion.

If we had instead a document that employed grander language and
inspired a sense of exceptionalism, similar to the US Declaration of
Independence, people would forget their differences. Who wants to live
in a world without a free computing platform?

Without Debian, we might all fall prey to corporate surveillance or
government control. Access to the world's information would be in the
hands of a privileged few who can pay for a secure browser. Debian is
a gem, and all our contributors are special.

Let us write such a Declaration of Purpose. Debian unite!

Kind regards
Felix Lechner