Re: On having and using a Code of Conduct

[Ben Hutchings]

On Thu, 2019-01-03 at 11:26 -0700, Eldon Koyle wrote:
> Hi all,
> 
> On Wed, Jan 2, 2019 at 5:25 AM Steve McIntyre  wrote:
> 
> > For those trying to undermine it with statements like "I'm worried
> > I'll be thrown out of Debian if I make a single mistake", please give
> > it a rest already. These are basic principles on how we want all
> > people to interact.
> 
> 
> I think there are many who are concerned about the process, not the CoC
> itself.  Here are the main concerns as I see them (at least from the few
> who have come forward), and I believe these are the reasons that people
> are worrying:
> 
>   1. The process itself is not well documented (it's new, so expected)
> 
>   2. The accused isn't allowed to address the claims against them
> 
>   3. The a-h team is acting as both prosecution and judge/jury (usually
> separated to reduce confirmation bias)

There is a separation of roles.  The Debian Account Managers (DAMs)
have the delegated power to decide on expulsions and additions to the
project members.  (Latest delegation is at
.)
The anti-harassment team is the usual contact point for complaints and
can recommend actions to the DAMs (or other teams) but doesn't have
delegated powers (as I understand it).

>   4. The proceedings are closed, so claims of unfairness aren't refuted
> 
>   5. There doesn't appear to be an appeals process (contact DAM?)
[...]

There is, since any decision by the DPL or a delegate can be overridden
by General Resolution.

Ben.

-- 
Ben Hutchings
Absolutum obsoletum. (If it works, it's out of date.) - Stafford Beer



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

Re: On having and using a Code of Conduct

[Eldon Koyle]

Hi all,

On Wed, Jan 2, 2019 at 5:25 AM Steve McIntyre  wrote:
>

> For those trying to undermine it with statements like "I'm worried
> I'll be thrown out of Debian if I make a single mistake", please give
> it a rest already. These are basic principles on how we want all
> people to interact.


I think there are many who are concerned about the process, not the CoC
itself.  Here are the main concerns as I see them (at least from the few
who have come forward), and I believe these are the reasons that people
are worrying:

  1. The process itself is not well documented (it's new, so expected)

  2. The accused isn't allowed to address the claims against them

  3. The a-h team is acting as both prosecution and judge/jury (usually
separated to reduce confirmation bias)

  4. The proceedings are closed, so claims of unfairness aren't refuted

  5. There doesn't appear to be an appeals process (contact DAM?)

I believe that the a-h team gives people warnings and tries to help them
understand why what they are saying is unacceptable and how they might
have been able to express their opinion more appropriately before
starting this process, but again I have no evidence of this, and they
cannot provide it.

IANADD, but the limited information available about the process and the
outcomes is difficult in a community that is typically as transparent as
possible, and I think it is reasonable for people to have concerns about a
closed process.  Any information that can be provided about the process
would probably help with these concerns -- and it should be published
somewhere other than mailing list archives.

This situation is especially difficult since the interpretation of the
CoC can be highly subjective, and there is no real feedback on how the
a-h team is interpreting it.  Maybe writing a more in-depth document on
the a-h team's interpretation of the CoC would help (examples of bad
behavior, examples of behavior that although someone might be offended,
is not forbidden)?

In Norbert's case, I get the impression that the bar was raised for him
after his first offense, and he may have actually been removed from the
project for insubordination (ie. re-adding his blog to planet, which
although ill-advised, may have been an honest mistake as he removed the
offending post before doing so).  However, I only have half of the story.

Finally, due to 2 and 3, there is going to be a lot more bias (toward
guilt) in this process than in a typical legal proceeding (this is about
the process, not the a-h team; it is just the nature of searching for
evidence of a crime or breach of the CoC in this case -- it is the
reason we have a hopefully impartial judge hearing both sides in legal
proceedings).

This is especially difficult since the interpretation of the CoC can be
highly subjective, and there is no real feedback on how the a-h team is
interpreting it.  Maybe writing a more in-depth document on what the a-h
team expects and what kind of behavior is the most common or most
disruptive would help?

(As an aside: I have noticed a tendency on the lists for people to pick
out the point they think is the worst in an email, beat on it
repeatedly, and ignore the rest of the argument -- this has the very
negative side-effect of making people feel like their main point has not
been and will not be heard.)

--
Eldon Koyle

Re: Planet Debian revisions

[Ulrike Uhlig]

Hi!

Roberto C. Sánchez:
> On Wed, Jan 02, 2019 at 05:39:58PM +0200, Jonathan Carter wrote:
>>
>> On #6 I was tempted to add "When a blog is removed, the committer should
>> add a comment listing the posts that resulted in it being removed", but
>> not sure if that's overloading it a bit too much.
>>
> Probably better to say something like, "When a blog is removed, the
> committer should send a direct email message to the author of the
> removed content explaining the reason for the removal."

Please, no. A commit message ensures that everybody is aware of the
removal reason, including planet admins. Resorting to email? I don't
think emails are encoded in the feeds and we cannot reasonably expect
people to search for them...

Cheers
u.

Re: Planet Debian revisions

[Ulrike Uhlig]

Hi,

Sean Whitton:
> Hello,
> 
> On Thu 03 Jan 2019 at 10:19am -0500, Roberto C. Sánchez wrote:
> 
>> On Wed, Jan 02, 2019 at 05:39:58PM +0200, Jonathan Carter wrote:
>>>
>>> On #6 I was tempted to add "When a blog is removed, the committer should
>>> add a comment listing the posts that resulted in it being removed", but
>>> not sure if that's overloading it a bit too much.
>>>
>> Probably better to say something like, "When a blog is removed, the
>> committer should send a direct email message to the author of the
>> removed content explaining the reason for the removal."
>>
>> That keeps potentially loaded statements from being recorded in commit
>> message forever.
> 
> Yes, let's avoid this.

As currently phrased, if people are unsure, they should contact the
planet administrator's team. I think this should solve such concerns.

Ulrike

Re: Planet Debian revisions

[Roberto C . Sánchez]

On Thu, Jan 03, 2019 at 05:50:03PM +0100, Joerg Jaspert wrote:
> On 15271 March 1977, Roberto C. Sánchez wrote:
> 
> > > And I sometimes remove blogs for them just going 5xx. A commit msg
> > > is fine.
> > I still think an email to the author would be a good thing in that case.
> > I have had parts of my site stop functioning and known of it for some
> > time.  An email from someone telling me that it is broken is something I
> > consider to be helpful.
> 
> In principle I agree. Now tell me, for a good chunk of the planet blogs,
> which email? Without investing lots of time to find out.
> 
I see your point.  My invalid assumption and my ignorance regarding the
implementation of Planet did not allow me to see that there was a
potential obstacle there.

> > > And who says a commi message is short? Write a novel, if you want.
> > > :)
> > I think we have enough flamewars ongoing at the moment that I am not
> > going to take the bait to start a philosophical/religious discussion on
> > the merits of short/concise commit messages :-)
> 
> But but, I was short, I only used 4242 words why I added a comma at that
> position!
> 
Just be sure to keep the first line to a maximum of 72 characters
followed by a hard line break and a blank line so 'git log --oneline'
looks sane.

Regards,

-Roberto

-- 
Roberto C. Sánchez

Re: Planet Debian revisions

[Joerg Jaspert]

On 15271 March 1977, Roberto C. Sánchez wrote: 

And I sometimes remove blogs for them just going 5xx. A commit 
msg is fine.
I still think an email to the author would be a good thing in 
that case.  I have had parts of my site stop functioning and 
known of it for some time.  An email from someone telling me 
that it is broken is something I consider to be helpful.


In principle I agree. Now tell me, for a good chunk of the planet 
blogs, which email? Without investing lots of time to find out.


And who says a commi message is short? Write a novel, if you 
want. :)
I think we have enough flamewars ongoing at the moment that I am 
not going to take the bait to start a philosophical/religious 
discussion on the merits of short/concise commit messages :-) 


But but, I was short, I only used 4242 words why I added a comma 
at that position!



--
bye, Joerg

Re: Planet Debian revisions

[Jonas Smedegaard]

Quoting Roberto C. Sánchez (2019-01-03 16:33:51)
> I have built up quite a backlog of email, so I did not see that the
> discussion had effectively concluded when I wrote my message.
> 
> On Thu, Jan 03, 2019 at 04:25:14PM +0100, Joerg Jaspert wrote:
> > On 15271 March 1977, Roberto C. Sánchez wrote:
> > 
> > > Probably better to say something like, "When a blog is removed, 
> > > the committer should send a direct email message to the author of 
> > > the removed content explaining the reason for the removal."
> > 
> > Ah please not.
> > 
> > > That keeps potentially loaded statements from being recorded in 
> > > commit message forever.  It also allows the author something 
> > > perhaps more complete than a short sentence fragment in a commit 
> > > message upon which to base a decision on how to proceed.
> > 
> > And I sometimes remove blogs for them just going 5xx. A commit msg 
> > is fine.
> > 
> I still think an email to the author would be a good thing in that 
> case. I have had parts of my site stop functioning and known of it for 
> some time.  An email from someone telling me that it is broken is 
> something I consider to be helpful.

I find it problematic to _promise_ an email notice upon removal.

Sure it is a nice gesture to notify, but I see no reason to codify that.


 - 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: Planet Debian revisions

[Michael Stone]

On Thu, Jan 03, 2019 at 03:25:07PM +, Sean Whitton wrote:

On Thu 03 Jan 2019 at 02:47pm GMT, Ulrike Uhlig wrote:

Looks good! I like it.

One tiny thingy based on a remark: I've looked up 'slur' in the
dictionary and 'slander' and 'libel' seem to be synonyms that might be
more widely known. Maybe a native speaker could confirm this.


'slander' seems fine but 'libel' implies you are doing something
illegal.  'slander' and 'slurs' need not be illegal.


Slander and libel are equally "illegal" under common law: the former is 
spoken while the latter is written, and both are civil rather than 
criminal matters. This varies by jurisdiction, as does protection of 
true statements, so it's important to be clear about the "where". (And 
historically, slander was punishable by the removal of the tongue--so it 
hardly seems a lesser matter!)


Mike Stone

Re: Planet Debian revisions

[Roberto C . Sánchez]

I have built up quite a backlog of email, so I did not see that the
discussion had effectively concluded when I wrote my message.

On Thu, Jan 03, 2019 at 04:25:14PM +0100, Joerg Jaspert wrote:
> On 15271 March 1977, Roberto C. Sánchez wrote:
> 
> > Probably better to say something like, "When a blog is removed, the
> > committer should send a direct email message to the author of the
> > removed content explaining the reason for the removal."
> 
> Ah please not.
> 
> > That keeps potentially loaded statements from being recorded in commit
> > message forever.  It also allows the author something perhaps more
> > complete than a short sentence fragment in a commit message upon which
> > to base a decision on how to proceed.
> 
> And I sometimes remove blogs for them just going 5xx. A commit msg is fine.
> 
I still think an email to the author would be a good thing in that case.
I have had parts of my site stop functioning and known of it for some
time.  An email from someone telling me that it is broken is something I
consider to be helpful.

In any event, I don't think it is particularly important enough to
warrant changing something for which consensus has already been
established.

> And who says a commi message is short? Write a novel, if you want. :)
> 
I think we have enough flamewars ongoing at the moment that I am not
going to take the bait to start a philosophical/religious discussion on
the merits of short/concise commit messages :-)

Regards,

-Roberto

-- 
Roberto C. Sánchez

Re: Planet Debian revisions

[Sean Whitton]

Hello,

On Thu 03 Jan 2019 at 02:47pm GMT, Ulrike Uhlig wrote:

> Looks good! I like it.
>
> One tiny thingy based on a remark: I've looked up 'slur' in the
> dictionary and 'slander' and 'libel' seem to be synonyms that might be
> more widely known. Maybe a native speaker could confirm this.

'slander' seems fine but 'libel' implies you are doing something
illegal.  'slander' and 'slurs' need not be illegal.

-- 
Sean Whitton


signature.asc
Description: PGP signature

Re: Planet Debian revisions

[Sean Whitton]

Hello,

On Thu 03 Jan 2019 at 10:19am -0500, Roberto C. Sánchez wrote:

> On Wed, Jan 02, 2019 at 05:39:58PM +0200, Jonathan Carter wrote:
>>
>> On #6 I was tempted to add "When a blog is removed, the committer should
>> add a comment listing the posts that resulted in it being removed", but
>> not sure if that's overloading it a bit too much.
>>
> Probably better to say something like, "When a blog is removed, the
> committer should send a direct email message to the author of the
> removed content explaining the reason for the removal."
>
> That keeps potentially loaded statements from being recorded in commit
> message forever.

Yes, let's avoid this.

-- 
Sean Whitton


signature.asc
Description: PGP signature

Re: Planet Debian revisions

[Roberto C . Sánchez]

On Thu, Jan 03, 2019 at 02:47:00PM +, Ulrike Uhlig wrote:
> Hi!
> 
> Jonathan Carter:
> > On 2019/01/03 00:26, Joerg Jaspert wrote:
> 
> > Full text: https://wiki.debian.org/PlanetDebian/ProposedChanges
> 
> Looks good! I like it.
> 
> One tiny thingy based on a remark: I've looked up 'slur' in the
> dictionary and 'slander' and 'libel' seem to be synonyms that might be
> more widely known. Maybe a native speaker could confirm this.
> 
A slander or libel is something that attacks an individual's character,
like falsely accusing someone of corruption.  A slur, on the other hand,
might attack someone's race, ethnicity, gender, etc.

"Disparaging statement" might work better, as it would cover both.

Regards,

-Roberto

-- 
Roberto C. Sánchez

Re: Planet Debian revisions

[Joerg Jaspert]

On 15271 March 1977, Roberto C. Sánchez wrote: 

Probably better to say something like, "When a blog is removed, 
the committer should send a direct email message to the author 
of the removed content explaining the reason for the removal." 


Ah please not.

That keeps potentially loaded statements from being recorded in 
commit message forever.  It also allows the author something 
perhaps more complete than a short sentence fragment in a commit 
message upon which to base a decision on how to proceed. 


And I sometimes remove blogs for them just going 5xx. A commit msg 
is fine.


And who says a commi message is short? Write a novel, if you want. 
:)


--
bye, Joerg

Re: Planet Debian revisions

[Roberto C . Sánchez]

On Wed, Jan 02, 2019 at 05:39:58PM +0200, Jonathan Carter wrote:
> 
> On #6 I was tempted to add "When a blog is removed, the committer should
> add a comment listing the posts that resulted in it being removed", but
> not sure if that's overloading it a bit too much.
> 
Probably better to say something like, "When a blog is removed, the
committer should send a direct email message to the author of the
removed content explaining the reason for the removal."

That keeps potentially loaded statements from being recorded in commit
message forever.  It also allows the author something perhaps more
complete than a short sentence fragment in a commit message upon which
to base a decision on how to proceed.

Regards,

-Roberto

-- 
Roberto C. Sánchez

Re: Planet Debian revisions

[Ulrike Uhlig]

Hi!

Jonathan Carter:
> On 2019/01/03 00:26, Joerg Jaspert wrote:

> Full text: https://wiki.debian.org/PlanetDebian/ProposedChanges

Looks good! I like it.

One tiny thingy based on a remark: I've looked up 'slur' in the
dictionary and 'slander' and 'libel' seem to be synonyms that might be
more widely known. Maybe a native speaker could confirm this.

Cheers!
Ulrike