[gentoo-council] Extent of Code of Conduct enforcement

[gentoo-council] Extent of Code of Conduct enforcement

[Donnie Berkholz]

[-- Attachment #1: Type: text/plain, Size: 837 bytes --]

Can people be entirely banned from Gentoo?

- What would such a ban include? Some ideas -- the person could not:
  - Post to any gentoo mailing list;
  - Post to gentoo bugzilla;
  - Participate in #gentoo- IRC channels; 
  - Contribute to gentoo (hence my corner case of a security fix) except 
    perhaps through a proxy;

- Why would we do it?

- Under whose authority would it happen?

- Would it be reversible? What conditions would cause this?

  Since the banned person couldn't participate in Gentoo, we'd never 
  know whether anything changed.

- How would one appeal this? Would there be a chance to respond before
  the ban?

- Would moderating the gentoo-dev mailing list obsolete this concept?

-- 
Thanks,
Donnie

Donnie Berkholz
Developer, Gentoo Linux
Blog: http://dberkholz.wordpress.com

[-- Attachment #2: Type: application/pgp-signature, Size: 197 bytes --]

Re: [gentoo-council] Extent of Code of Conduct enforcement

[Alec Warner]

On Mon, Jul 14, 2008 at 6:35 AM, Donnie Berkholz <dberkholz@gentoo.org> wrote:
> Can people be entirely banned from Gentoo?

no

>
> - What would such a ban include? Some ideas -- the person could not:
>  - Post to any gentoo mailing list;

technically feasible

>  - Post to gentoo bugzilla;

technically feasible

>  - Participate in #gentoo- IRC channels;

technically infeasible.  Also a hard sell; traditionally gentoo-*
channels that are not #gentoo-dev and #gentoo are owned and operated
by gentoo subprojects with the permission of the gentoo reps.
Enforcing a ban in all channels would be difficult.

>  - Contribute to gentoo (hence my corner case of a security fix) except
>    perhaps through a proxy;

nothing stops them from contributing to the community; it is not as if
Gentoo controls all outlets anyway.

>
> - Why would we do it?

To prevent gentoo from being messed with by people who have routinely
proven that they are unfit to assist the distribution.

>
> - Under whose authority would it happen?

Userrel.

>
> - Would it be reversible? What conditions would cause this?

I assume the person would ask to return or have someone vouch for them.

>
>  Since the banned person couldn't participate in Gentoo, we'd never
>  know whether anything changed.

I would think that if the person wanted to come back they would:

  Make an effort to contact Gentoo; It is not as if developers would
not talk to this individual.
  Gentoo itself would take this person back provisionally to ensure
things were different.  This is
  a case by case deal and I think is difficult to pin down.

>
> - How would one appeal this? Would there be a chance to respond before
>  the ban?

Since the ban would require some amount of history I don't see any
particular reason not to solicit feedback from said person.

>
> - Would moderating the gentoo-dev mailing list obsolete this concept?

If moderating gentoo-dev obsoletes this concept then I think the
concept itself is flawed (gentoo is more than gentoo-dev)

>
> --
> Thanks,
> Donnie
>
> Donnie Berkholz
> Developer, Gentoo Linux
> Blog: http://dberkholz.wordpress.com
>
-- 
gentoo-council@lists.gentoo.org mailing list

Re: [gentoo-council] Extent of Code of Conduct enforcement

[Ferris McCormick]

[-- Attachment #1: Type: text/plain, Size: 4029 bytes --]


On Mon, 2008-07-14 at 08:23 +0000, Alec Warner wrote:
> On Mon, Jul 14, 2008 at 6:35 AM, Donnie Berkholz <dberkholz@gentoo.org> wrote:
> > Can people be entirely banned from Gentoo?
> 
> no
> 
> >
> > - What would such a ban include? Some ideas -- the person could not:
> >  - Post to any gentoo mailing list;
> 
> technically feasible
> 
> >  - Post to gentoo bugzilla;
> 
> technically feasible
> 
> >  - Participate in #gentoo- IRC channels;
> 
> technically infeasible.  Also a hard sell; traditionally gentoo-*
> channels that are not #gentoo-dev and #gentoo are owned and operated
> by gentoo subprojects with the permission of the gentoo reps.
> Enforcing a ban in all channels would be difficult.
> 
> >  - Contribute to gentoo (hence my corner case of a security fix) except
> >    perhaps through a proxy;
> 
> nothing stops them from contributing to the community; it is not as if
> Gentoo controls all outlets anyway.
> 
Although why would they want to?

> >
> > - Why would we do it?
> 
> To prevent gentoo from being messed with by people who have routinely
> proven that they are unfit to assist the distribution.
> 

That's a very strong statement.  I personally have never seen anyone I'd
categorize as "unfit to assist."

> >
> > - Under whose authority would it happen?
> 
> Userrel.
> 
> >
> > - Would it be reversible? What conditions would cause this?
> 
> I assume the person would ask to return or have someone vouch for them.
> 
> >
> >  Since the banned person couldn't participate in Gentoo, we'd never
> >  know whether anything changed.
> 
> I would think that if the person wanted to come back they would:
> 
>   Make an effort to contact Gentoo; It is not as if developers would
> not talk to this individual.
>   Gentoo itself would take this person back provisionally to ensure
> things were different.  This is
>   a case by case deal and I think is difficult to pin down.
> 
> >
> > - How would one appeal this? Would there be a chance to respond before
> >  the ban?
> 
> Since the ban would require some amount of history I don't see any
> particular reason not to solicit feedback from said person.
> 

I disagree.  I think everyone deserves to be heard.  After all, one of
Gentoo's guiding principles reads:

========================================================
Gentoo is open

Every aspect of Gentoo is and remains open. Gentoo does not benefit from
hiding any of its development processes (whether it is source code or
documentation, decisions or discussions, coordination or management).
=========================================================

This sort of thing certainly falls within that, I think (management if
nothing else, and certainly some aspect of Gentoo).



> >
> > - Would moderating the gentoo-dev mailing list obsolete this concept?
> 
> If moderating gentoo-dev obsoletes this concept then I think the
> concept itself is flawed (gentoo is more than gentoo-dev)
> 

That's why I keep saying we should at least give moderating gentoo-dev a
chance first.  Most "problems" are flame wars there, and I believe that
most flames involve more developers than non-developers.  #gentoo-dev is
sometimes a problem, but mostly brief outbursts resulting from frayed
tempers, and largely from developers.  Bugzilla is sometimes a problem,
but not all that often, and usually resulting from strongly held
technical disagreements or disagreements on who can make changes to a
some package (perceived poaching).

I really don't think it is in our interests to find a user or some users
to "make examples" of if less extreme measures can achieve what we
actually want (calmer gentoo-dev mostly).

Regards,
Ferris 
> >
> > --
> > Thanks,
> > Donnie
> >
> > Donnie Berkholz
> > Developer, Gentoo Linux
> > Blog: http://dberkholz.wordpress.com
> >
-- 
Ferris McCormick (P44646, MI) <fmccor@gentoo.org>
Developer, Gentoo Linux (Devrel, Sparc, Userrel, Trustees)

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 197 bytes --]

Re: [gentoo-council] Extent of Code of Conduct enforcement

[Roy Bamford]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On 2008.07.14 07:35, Donnie Berkholz wrote:
> Can people be entirely banned from Gentoo?
> 
> - What would such a ban include? Some ideas -- the person could not:
>   - Post to any gentoo mailing list;
>   - Post to gentoo bugzilla;
>   - Participate in #gentoo- IRC channels; 
>   - Contribute to gentoo (hence my corner case of a security fix)
> except 
>     perhaps through a proxy;

All of these things require the co-operation of the banee.
email addresses to subscribe to lists are unlimited  and easy to 
change.

IRC nicks can be registered for the asking.
Does #gentoo- include non-English channels?

We should not make rules we cannot enforce, the best we can do it to 
make it difficult for the bannee to do these things without pushing 
them to the trouble of swapping email addresses, IRC nicks etc.

> 
> - Why would we do it?
We wouldn't - its more trouble than its worth to enforce.

> 
> - Under whose authority would it happen?
*rel against 'contrubutors' with a history.

> 
> - Would it be reversible? What conditions would cause this?
N/A
> 
>   Since the banned person couldn't participate in Gentoo, we'd never 
>   know whether anything changed.
> 
> - How would one appeal this? Would there be a chance to respond 
> before
>   the ban?
Why would there be any right of appeal at all?
Its not like the bannee would have had no warning.
While I don't see this being implemented at all, its really a desparate 
last measure. 

> 
> - Would moderating the gentoo-dev mailing list obsolete this concept?
No.

It would make it less likely to be used as potential banees would get 
gentle tugs on their leashes more frequently, as they currently do in 
#gentoo and on the forums. Even there, some contributors are 
permanently banned and a some are known to have come back with new 
identities. 

Following on that model, I could go with a permanent ban on an online 
identity without a permanent ban on the individual holding it, so that 
if they come back with a new identity, after a suitable period and 
behave, they are permitted to remain.

That also gets around all the 'mole' hunting.

> 
> -- 
> Thanks,
> Donnie
> 
> Donnie Berkholz
> Developer, Gentoo Linux
> Blog: http://dberkholz.wordpress.com
> 

- -- 
Regards,

Roy Bamford
(NeddySeagoon) a member of
gentoo-ops
forum-mods
treecleaners
trustees

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.9 (GNU/Linux)

iEYEARECAAYFAkh7mbgACgkQTE4/y7nJvas+ggCcC3QfJLRrXt2NiM3U4E/5uhfK
P7oAnREQU5IVGmFeU2G1jfYGcedgu6mE
=FCE4
-----END PGP SIGNATURE-----

--
gentoo-council@lists.gentoo.org mailing list

Re: [gentoo-council] Extent of Code of Conduct enforcement

[Tobias Scherbaum]

[-- Attachment #1: Type: text/plain, Size: 2871 bytes --]

Donnie Berkholz wrote:
> Can people be entirely banned from Gentoo?

At least from a technical pov I tend to say "no". Implementing a
"feature" we (as in Gentoo) cannot technically enforce is useless, as
enforcing it would require lots of manpower and manual interaction which
we need more urgently in lots of other areas of Gentoo.

> - What would such a ban include? Some ideas -- the person could not:
>   - Post to any gentoo mailing list;
>   - Post to gentoo bugzilla;
>   - Participate in #gentoo- IRC channels; 
>   - Contribute to gentoo (hence my corner case of a security fix) except 
>     perhaps through a proxy;
> 
> - Why would we do it?

don't know, I don't see the need. People play wanker on #gentoo -> they
get banned from that channel. People play wanker in the forums -> they
get a warning and finally their account will get locked. I think these
mechanisms are quite effective and proved to be good (tm), creating a
next step of a "full Gentoo ban" isn't needed (nor doable) from my pov.

> - Under whose authority would it happen?

As people who would be banned are no developers any more this clearly
falls under Userrels authority.

> - Would it be reversible? What conditions would cause this?

It needs to be reversible, people change, their attitude changes.
Therefore we would need to implement a process which allows every banned
user (after a fixed timeframe following the ban) to let userrel re-check
the ban.

>   Since the banned person couldn't participate in Gentoo, we'd never 
>   know whether anything changed.

They could still talk to people on IRC or via mail - or request to
re-check if their ban is still necessary or if they deserve a second
chance as described above.

> - How would one appeal this? Would there be a chance to respond before
>   the ban?

As such a ban would require fast intervention to just stop people
playing wankers we would need to have different steps of bans, temporary
bans followed by a longer ban and permanent bans as the last resort.
Having several steps (i.e. short bans for a few days or a week at last)
before someone gets banned permanently there's no need to be able to
appeal these decisions - except a permanent ban would require such a
process being in place.

> - Would moderating the gentoo-dev mailing list obsolete this concept?

It wouldn't obsolete this concept, but for now I see no need to ban
people from interacting with our (developer) community - besides that I
question if such a ban would be technically doable.
As we had the most problems with our dev-ml in the past (and we have
other working mechanisms like operators on #gentoo or mods in forums
already in place) putting the ml on moderation would help and *might*
obsolete the need for bans if the implementation works and will be
accepted.

  Tobias

[-- Attachment #2: Dies ist ein digital signierter Nachrichtenteil --]
[-- Type: application/pgp-signature, Size: 197 bytes --]

Re: [gentoo-council] Extent of Code of Conduct enforcement

[Ferris McCormick]

[-- Attachment #1: Type: text/plain, Size: 3379 bytes --]

Top posting because it's brief.

This reflects my own views pretty much exactly, and states them better
than I've been able to.  So I'll try to refrain from posting further.


On Mon, 2008-07-14 at 21:20 +0200, Tobias Scherbaum wrote:
> Donnie Berkholz wrote:
> > Can people be entirely banned from Gentoo?
> 
> At least from a technical pov I tend to say "no". Implementing a
> "feature" we (as in Gentoo) cannot technically enforce is useless, as
> enforcing it would require lots of manpower and manual interaction which
> we need more urgently in lots of other areas of Gentoo.
> 
> > - What would such a ban include? Some ideas -- the person could not:
> >   - Post to any gentoo mailing list;
> >   - Post to gentoo bugzilla;
> >   - Participate in #gentoo- IRC channels; 
> >   - Contribute to gentoo (hence my corner case of a security fix) except 
> >     perhaps through a proxy;
> > 
> > - Why would we do it?
> 
> don't know, I don't see the need. People play wanker on #gentoo -> they
> get banned from that channel. People play wanker in the forums -> they
> get a warning and finally their account will get locked. I think these
> mechanisms are quite effective and proved to be good (tm), creating a
> next step of a "full Gentoo ban" isn't needed (nor doable) from my pov.
> 
> > - Under whose authority would it happen?
> 
> As people who would be banned are no developers any more this clearly
> falls under Userrels authority.
> 
> > - Would it be reversible? What conditions would cause this?
> 
> It needs to be reversible, people change, their attitude changes.
> Therefore we would need to implement a process which allows every banned
> user (after a fixed timeframe following the ban) to let userrel re-check
> the ban.
> 
> >   Since the banned person couldn't participate in Gentoo, we'd never 
> >   know whether anything changed.
> 
> They could still talk to people on IRC or via mail - or request to
> re-check if their ban is still necessary or if they deserve a second
> chance as described above.
> 
> > - How would one appeal this? Would there be a chance to respond before
> >   the ban?
> 
> As such a ban would require fast intervention to just stop people
> playing wankers we would need to have different steps of bans, temporary
> bans followed by a longer ban and permanent bans as the last resort.
> Having several steps (i.e. short bans for a few days or a week at last)
> before someone gets banned permanently there's no need to be able to
> appeal these decisions - except a permanent ban would require such a
> process being in place.
> 
> > - Would moderating the gentoo-dev mailing list obsolete this concept?
> 
> It wouldn't obsolete this concept, but for now I see no need to ban
> people from interacting with our (developer) community - besides that I
> question if such a ban would be technically doable.
> As we had the most problems with our dev-ml in the past (and we have
> other working mechanisms like operators on #gentoo or mods in forums
> already in place) putting the ml on moderation would help and *might*
> obsolete the need for bans if the implementation works and will be
> accepted.
> 
>   Tobias

Regards,
Ferris
-- 
Ferris McCormick (P44646, MI) <fmccor@gentoo.org>
Developer, Gentoo Linux (Devrel, Sparc, Userrel, Trustees)

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 197 bytes --]

RE: [gentoo-council] Extent of Code of Conduct enforcement

[Chrissy Fullam]

> Can people be entirely banned from Gentoo?

By technical merit I'd not say that anything is entirely outside of
technical ability if one has that ability to circumvent the system. Doesn't
mean we should throw the towel in and just sit back to take abuse though.

> - What would such a ban include? Some ideas -- the person could not:
>   - Post to any gentoo mailing list;
>   - Post to gentoo bugzilla;
>   - Participate in #gentoo- IRC channels;
>   - Contribute to gentoo (hence my corner case of a security fix) except
>     perhaps through a proxy;

Mailing list rights and bugzilla rights for that email address would be
revoked; the argument is if they subscribed with a different email address
what then? If they are confirmed to be the original person User Rel should
talk to them and quite possibly extend the ban to the new accounts - not my
decision just an idea; argument being when do we stop because they could
keep trying with new accounts? My response is situational, likely I'd
consider continuing doing it so long as it was verified to be that person
but that is my opinion. Changing your name doesn't make you less of an ass,
only you can make yourself less of an ass.
Regarding IRC, we should only remove their nick from #gentoo-dev as other
channels are moderated and operated by the relevant team. I for one would
remove anyone from #gentoo-devrel, but that is my right to do so.
Contributions... I'd vote to only accept via a proxy so long as said proxy
understands they are vouching 100% for the quality of the code as well. This
should be common practice today I suspect.

> - Why would we do it?

Because we do our developers and users a disservice to allow such repeat
offenders the luxury of participating in such a manner while ruining the
experience for others. Simply put people quit and drop Gentoo solely as a
result of one or several vocal negative and abusive influences. Let's do our
jobs and do something about it.

> - Under whose authority would it happen?

Dev Rel would have removed such a person from being a developer in which
case this round of elimination would fall upon User Rel, whom Dev Rel would
gladly assist if assistance were requested.

> - Would it be reversible? What conditions would cause this?
> 
>   Since the banned person couldn't participate in Gentoo, we'd never
>   know whether anything changed.

Life happens, people change. No ban should be 100% forever so let's be
realistic here. If User Rel banned someone from participating they could
follow the same rule set that Dev Rel uses: a forcibly removed developer may
return upon approval of the current Dev Rel lead. That said, if a forcibly
removed contributor/participator wanted to come back that person would need
to convince the User Rel lead of a darn good reason and such re-joining
should be done under a probationary period subject to review.

> - How would one appeal this? Would there be a chance to respond before
>   the ban?

Appeal a ban, no. I think if someone is being considered for the ban there
should be an exchange of dialog. That said, no person eligible for such an
extreme course of action could honestly say they were surprised and caught
off guard. Such a person would have a repeated history of conflicts and
discussions warning them that they are out of line, be it in a bug or via
emails.

> - Would moderating the gentoo-dev mailing list obsolete this concept?

Nope, just make it less likely to happen as hopefully we'd be able to show
people that we are taking this seriously and it must stop or the
consequences continue to get more severe. Much like the legal system fmccor
references, a jury or judge is less sympathetic to the protests of a repeat
offender.


Kind regards,
Christina Fullam
Gentoo Developer Relations Lead | Gentoo Public Relations 




-- 
gentoo-council@lists.gentoo.org mailing list

Re: [gentoo-council] Extent of Code of Conduct enforcement

[Mike Doty]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Chrissy Fullam wrote:
|> Can people be entirely banned from Gentoo?
|
| By technical merit I'd not say that anything is entirely outside of
| technical ability if one has that ability to circumvent the system.
Doesn't
| mean we should throw the towel in and just sit back to take abuse though.
|
|> - What would such a ban include? Some ideas -- the person could not:
|>   - Post to any gentoo mailing list;
|>   - Post to gentoo bugzilla;
|>   - Participate in #gentoo- IRC channels;
|>   - Contribute to gentoo (hence my corner case of a security fix) except
|>     perhaps through a proxy;
|
| Mailing list rights and bugzilla rights for that email address would be
[snip]
Not to pick on you musikc, but can we all stop using the term 'rights'
please.  Regarding anything the infrastructure team provides, the proper
term is privilege.

TIA

- --
=======================================================
Mike Doty                      kingtaco -at- gentoo.org
Gentoo Infrastructure
Gentoo/AMD64 Strategic Lead
GPG: E1A5 1C9C 93FE F430 C1D6  F2AF 806B A2E4 19F4 AE05
=======================================================
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.9 (GNU/Linux)

iJwEAQECAAYFAkh9YtAACgkQgGui5Bn0rgUR6AP/XQGPbymLZhZ53qmK8oU3SauG
BI3OC7jQ1CMUr2HLUNvSYDO8q/PIu9PKO9/AjBA7EEEtRtmt2ic6lWAS/+JfcPsE
TCYWs7kV+xpPmQRPziTR1Vz6lsV+3t+M6V/0Kl2UcWp2M/tvc+yLIHN+TfOiq4Xj
oI8+r3Hu012kD1a/yOQ=
=+kRN
-----END PGP SIGNATURE-----
-- 
gentoo-council@lists.gentoo.org mailing list

RE: [gentoo-council] Extent of Code of Conduct enforcement

[Chrissy Fullam]

> | Mailing list rights and bugzilla rights for that email address would be
> [snip]
> Not to pick on you musikc, but can we all stop using the term 'rights'
> please.  Regarding anything the infrastructure team provides, the proper
> term is privilege.

rights rights rights rights rights rights
rights rights rights rights rights rights

Seriously, upon making requests from infrastructure I'll do my darnedest to
try to remember your preferred wording and that your POV is that rights in
this sense is not synonymous with privilege. 

However this email was to a larger crowd, not infra specific, and the
subject of which is not correct wording but rather is regarding whether we
should extend the CoC enforcement to include Gentoo wide bans. IT grammar
police not required at this time but undoubtedly would prove invaluable if
this suggestion were agreed upon and documents required revising, in which
case you'd make an excellent editor and would have my vote! :)


Kind regards,
Christina Fullam
Gentoo Developer Relations Lead | Gentoo Public Relations 




-- 
gentoo-council@lists.gentoo.org mailing list

Re: [gentoo-council] Extent of Code of Conduct enforcement

[Mark Loeser]

[-- Attachment #1: Type: text/plain, Size: 1951 bytes --]

Donnie Berkholz <dberkholz@gentoo.org> said:
> Can people be entirely banned from Gentoo?

It would be ideal, but not technically feasible.

> - What would such a ban include? Some ideas -- the person could not:
>   - Post to any gentoo mailing list;
>   - Post to gentoo bugzilla;
>   - Participate in #gentoo- IRC channels; 

All of these seem to be possible to an extent and would be valid if we
would like to limit the damage an individual does to the community.

>   - Contribute to gentoo (hence my corner case of a security fix) except 
>     perhaps through a proxy;

This would be difficult to control, and I don't think we should care in
this case.  If they are contributing through a proxy (say Linus), should
we reject their fix?

> - Why would we do it?

Because they are damaging the community and driving possible
contributors aways.

> - Under whose authority would it happen?

Devrel and/or userrel

> - Would it be reversible? What conditions would cause this?

This would be something I would like to hear opinions from userrel and
devrel on.  Do they think someone that we would want to ban permanently
would turn around enough to unban?

>   Since the banned person couldn't participate in Gentoo, we'd never 
>   know whether anything changed.

Refer to my comment above.

> - How would one appeal this? Would there be a chance to respond before
>   the ban?

If it got to the point of us considering a permanent ban, I don't think
there is any reason to even consider listening to an appeal at that
point.  With that being said, I would expect a permanent ban to be a
majority vote from devrel or userrel to put such a ban in place.

> - Would moderating the gentoo-dev mailing list obsolete this concept?

I don't think so.


-- 
Mark Loeser
email         -   halcy0n AT gentoo DOT org
email         -   mark AT halcy0n DOT com
web           -   http://www.halcy0n.com

[-- Attachment #2: Type: application/pgp-signature, Size: 197 bytes --]

Re: [gentoo-council] Extent of Code of Conduct enforcement

[Ferris McCormick]

[-- Attachment #1: Type: text/plain, Size: 3599 bytes --]


On Tue, 2008-07-22 at 02:34 -0400, Mark Loeser wrote:
> Donnie Berkholz <dberkholz@gentoo.org> said:
> > Can people be entirely banned from Gentoo?
> 
> It would be ideal, but not technically feasible.
> 
> > - What would such a ban include? Some ideas -- the person could not:
> >   - Post to any gentoo mailing list;
> >   - Post to gentoo bugzilla;
> >   - Participate in #gentoo- IRC channels; 
> 
> All of these seem to be possible to an extent and would be valid if we
> would like to limit the damage an individual does to the community.
> 

Perhaps you could ban a name (like fmccor), But if I reappeared just as
Ferris@guaranteed-notguilty.pro, that would be harder.  Ferris is not
all that unusual last name, and not unique as a first name (it's Celtic)
--- think of "Ferris Buehler's day off."  So, you might be convinced
it's really me, but if by mistake you banned Ferris Buehler, you'd have
a problem.  (Just an unethical lawyer, not the poisonous fmccor. :) )

> >   - Contribute to gentoo (hence my corner case of a security fix) except 
> >     perhaps through a proxy;
> 
> This would be difficult to control, and I don't think we should care in
> this case.  If they are contributing through a proxy (say Linus), should
> we reject their fix?
> 
> > - Why would we do it?
> 
> Because they are damaging the community and driving possible
> contributors aways.
> 
> > - Under whose authority would it happen?
> 
> Devrel and/or userrel
> 
> > - Would it be reversible? What conditions would cause this?
> 
> This would be something I would like to hear opinions from userrel and
> devrel on.  Do they think someone that we would want to ban permanently
> would turn around enough to unban?
> 

Sure.  Everyone changes.

> >   Since the banned person couldn't participate in Gentoo, we'd never 
> >   know whether anything changed.
> 
> Refer to my comment above.
> 
> > - How would one appeal this? Would there be a chance to respond before
> >   the ban?
> 
> If it got to the point of us considering a permanent ban, I don't think
> there is any reason to even consider listening to an appeal at that
> point.  With that being said, I would expect a permanent ban to be a
> majority vote from devrel or userrel to put such a ban in place.
> 

Let's see, devrel population is about 8 or 9; userrel is smaller, I
think.  That's a tremendous amount of power in the hands of very few
developers, most of whom participate in these groups for reasons
unrelated to discipline as such (most of devrel is taken up with
recruiting people, retiring people, or documentation).  Further more,
unless we are considering imposing permanent bans on developers, it's
hard for me to see devrel's interest in such a process.

And you are saying that Council hears appeals of disciplinary actions
(GLEP 39), except in the case of the most severe action possible?

> > - Would moderating the gentoo-dev mailing list obsolete this concept?
> 
> I don't think so.
> 
Why not?  No one has explained that in a way I can comprehend.  Perhaps
I'm slow, but I just don't see it.  Having watched gentoo-dev@ over the
last few weeks, I conclude that usually the moderators would have
nothing to do in fact.  Almost all the time everything is pretty calm,
in fact.  We have problems when someone says something provocative and
someone else way overreacts, and things spiral out of control.

Regards,
Ferris
-- 
Ferris McCormick (P44646, MI) <fmccor@gentoo.org>
Developer, Gentoo Linux (Devrel, Sparc, Userrel, Trustees)

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 197 bytes --]

Re: [gentoo-council] Extent of Code of Conduct enforcement

[Ferris McCormick]

[-- Attachment #1: Type: text/plain, Size: 2546 bytes --]


On Tue, 2008-07-22 at 02:34 -0400, Mark Loeser wrote:
> Donnie Berkholz <dberkholz@gentoo.org> said:
> > Can people be entirely banned from Gentoo?
> 
> It would be ideal, but not technically feasible.
> 
> > - 
 ....................... <snip
> ..........................................
> 
> > - Why would we do it?
> 
> Because they are damaging the community and driving possible
> contributors aways.
> 
Let me respond to this specifically.  I have pretty strong views on
this, and I suspect they might reflect a minority opinion.

I'm going to divide this into two cases, because I think the first one
is easy.  First case is developers who leave or threaten to leave giving
the reason that "XXX (a developer or user) drove them away because
of ...".  Second case is a sponsor who threatens to withdraw support
"unless something is done about XXX."

In the first case, my reaction is absolute.  The developer who threatens
to leave because of someone else is (1) making the judgment call that we
care if he leaves; (2) Is resorting to extortion to get rid of someone
else (or reign someone else in or whatever).  At that point, I'd wish
him well in his future endeavors and start retirement process.  I view
giving in to such a threat as at least as harmful as whatever or whoever
triggered it in the first place.  This is based on my own background and
experiences, and others no doubt react differently.

Now, there is a variation on this:  The developer who resigns, citing
abuse as the reason.  Here, the process has broken down.  Believe it or
not, devrel and userrel will work with problems like this if we know
there are such problems to address.  For example, if you want me
involved, best is to contact me personally or open a bug assigned to me.
If you want someone else, do whatever that person prefers.  It's a
matter of posture --- we should all be prepared to help work out
conflicts, but we should not give in to "He goes or I go." 

The second case is more delicate.  It is still a form of extortion, but
conceivably with merit.  I think the resolution requires negotiation
with the sponsor and the "problem child".  If we can reach no agreement,
I suppose we have to do what seems best for the community.  That will
always be a decision depending on each circumstance.

Probably nothing new here, but I wanted to clarify my opinion a bit.
Regards,
Ferris
-- 
Ferris McCormick (P44646, MI) <fmccor@gentoo.org>
Developer, Gentoo Linux (Devrel, Sparc, Userrel, Trustees)

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 197 bytes --]

RE: [gentoo-council] Extent of Code of Conduct enforcement

[Chrissy Fullam]

> Ferris McCormick wrote:
> > On Tue, 2008-07-22 at 02:34 -0400, Mark Loeser wrote:
> > > Donnie Berkholz <dberkholz@gentoo.org> said:
> > > Can people be entirely banned from Gentoo?
> >
> > > - Why would we do it?
> > Because they are damaging the community and driving possible
> > contributors aways.
> >
> Let me respond to this specifically.  I have pretty strong views on this,
> and I suspect they might reflect a minority opinion.
> 
> I'm going to divide this into two cases, because I think the first one is
> easy.  First case is developers who leave or threaten to leave giving the
> reason that "XXX (a developer or user) drove them away because of ...".
> Second case is a sponsor who threatens to withdraw support "unless
> something is done about XXX."
> 
> In the first case, my reaction is absolute.  The developer who threatens
> to leave because of someone else is (1) making the judgment call that we
> care if he leaves; (2) Is resorting to extortion to get rid of someone
> else (or reign someone else in or whatever).  At that point, I'd wish him
> well in his future endeavors and start retirement process.  I view giving
> in to such a threat as at least as harmful as whatever or whoever
> triggered it in the first place.  This is based on my own background and
> experiences, and others no doubt react differently.

(1) I am bothered that as a member of devrel you state something that
implies devrel doesn't care if a developer leaves. While we have personal
opinions of people, devrel is expected to leave them at the door to our
jobs. I care when people leave, I care to know why they leave (this gives us
grounds for improvement which is what we should be seeking), and I care that
people do what is really best for them.
(2) You are making assumptions that people are resorting to extortion. I've
known people who simply said I'm through and leaving, here is why, and they
did so not to have someone try to win them back but rather so that the
appropriate people knew the areas that may require investigation. Not
everyone is a malicious ass trying to take advantage of people or ruin other
people.

> Now, there is a variation on this:  The developer who resigns, citing
> abuse as the reason.  Here, the process has broken down.  Believe it or
> not, devrel and userrel will work with problems like this if we know there
> are such problems to address.  

We are here to help, this is so very true, but devrel also recognizes the
stereo type that we are labeled with as a result of the past actions of a
variety of people within devrel, that being ineffective and not desiring to
do anything. This is not the case of devrel today. We will do what it takes
to appropriately address issues and if someone doesn't desire to participate
in devrel then they are welcome to seek out other areas of Gentoo in which
to participate.

> For example, if you want me involved, best
> is to contact me personally or open a bug assigned to me.
> If you want someone else, do whatever that person prefers.

If you contact a member of devrel directly then such a person will not be
acting as a devrel official but rather as a peer seeking to help. This is
quite alright and an option that anyone can exercise. The official means of
seeking devrel assistance is quoted from our policy as follows:

"To involve Developer Relations in your issue please send an email to
devrel@gentoo.org or open a Bug and assign it to Developer Relations; either
is acceptable. Please note that opening a bug is not necessary for
mediation, however the developer may open a bug if he/she wishes to do so;
opening a bug is mandatory if mediation efforts fail."

> The second case is more delicate.  It is still a form of extortion, but
> conceivably with merit.  I think the resolution requires negotiation with
> the sponsor and the "problem child".  If we can reach no agreement, I
> suppose we have to do what seems best for the community.  That will always
> be a decision depending on each circumstance.

A sponsor saying that they will withdraw unless we address XYZ is welcome to
do just that. They are sponsors and not legally bound to us for a specified
period of time. If the mutually beneficial relationship is no longer
mutually beneficial then it indeed should end if there is no agreeable
resolution. For example, I cannot condone firing a developer who we feel has
not done something worthy of such punishment just to please a sponsor.
However I have no objection to a sponsor stating something that causes us to
review the situation and determine whether we do in fact agree that such a
statement is in the best interest of Gentoo, just such a decision should not
be made solely to keep a sponsor. I realize my opinion may not be popular
there, but it's my own. ;-)


Kind regards,
Christina Fullam
Gentoo Developer Relations Lead | Gentoo Public Relations 

RE: [gentoo-council] Extent of Code of Conduct enforcement

[Chrissy Fullam]

> Mark Loeser wrote:
> > Donnie Berkholz <dberkholz@gentoo.org> said:
> > Can people be entirely banned from Gentoo?
> > - Why would we do it?
> 
> Because they are damaging the community and driving possible contributors
> aways.
> > - Under whose authority would it happen?
> 
> Devrel and/or userrel
> > - Would it be reversible? What conditions would cause this?
> 
> This would be something I would like to hear opinions from userrel and
> devrel on.  Do they think someone that we would want to ban permanently
> would turn around enough to unban?
>
> >   Since the banned person couldn't participate in Gentoo, we'd never
> >   know whether anything changed.
> Refer to my comment above.


I'm more than happy to state again that I am not opposed to allowing someone
back, as a user or a developer, after a period of supervision to ensure they
truly have turned things around for the better. People can change, no doubt
there, but I am also not foolish or naïve enough to think it happens over
night or with every person.


Kind regards,
Christina Fullam
Gentoo Developer Relations Lead | Gentoo Public Relations 

RE: [gentoo-council] Extent of Code of Conduct enforcement

[Ferris McCormick]

[-- Attachment #1: Type: text/plain, Size: 4870 bytes --]

After yesterday's Council meeting, I asked a question about just where
the current Code of Conduct was, or the status of the posted version.  I
asked this in the context of (1) I was surprised to see that the posted
version still talked about proctors, even though that project was
disbanded about a year ago, and (2) I remembered several discussions in
the previous council about amending CoC to explain its scope and how we
actually *did* intend to enforce it.

I thought this was a simple question the answer to which would be either
(1) Actual (proposed for revision?) Code of Conduct is <here>, or (2)
Council still has pending action on proposed revisions from last
council.  All of this is relevant, because this thread must apply to
current (proposed for revision?) code of conduct; it does not fit well
with a Code of Conduct enforced by proctors.  And I don't think anyone
can make much of an informed decision without some sort of Code of
Conduct which reflects reality (such as we don't have proctors) to work
from.

The reaction to my question seemed to suggest I was visiting from some
other universe or something, Hold over Council member(s) did not seem to
recall what if anything old council had done with Code of Conduct, and
everything here should be read in context of posted Code of Conduct.
Now, that can't be correct because it erases several months of
discussion and decisions from the history of previous council.

Anyway, people asked me to post my questions here, so that is what I am
doing.  Everything needs to be read in the context of the summaries of
the Council meetings from 20071011 -- 20080214 (five months).  Also,
there seem to be two proposals for Code of Conduct revision, both from
Donnie.  The first is discussed in a thread here:
http://thread.gmane.org/gmane.linux.gentoo.council/82
In fact, at the time I supported that idea enthusiastically, but others
pointed out difficulties.

Eventually, this was revised here:
http://archives.gentoo.org/gentoo-council/msg_ba125098c929ea31f34051dfb009d436.xml
advantage that it did not require any approval at all, but it most
certainly *did* imply revisions to the Code of Conduct itself, because
effectively it stripped out any authority from the (non-existent)
proctors and instead identify a small group of people who would take
aside CoC violators (in private) and suggest they quit doing whatever
they were doing wrong.  For the record here, my response to this was:

[quote]
'Nice idea and worth a try.  I have one concern.  Since we are talking
CoC here, I'd like to emphasise that "assholeness" should be determined
within the guidelines of the CoC.  Not by some person's own conception
of "assholeness" --- I'd hate to see a flame war about just who is being
the asshole in any particular instance.'
[/quote]

As best as I can tell, this proposal was *APPROVED*, and at 20080214, was left like this:

================================
Code of Conduct enforcement
---------------------------
	Promote individual devs responding to people who are being jerks.
	Keep responses private, unless that person gets out of hand.

	dberkholz will get things going.
	To help or get advice, contact him.
===================================

I think the final intent was that Council expected the Code of Conduct
to be pretty much self-enforcing, driven by members of the community who
cared enough to take violators aside and calm them down, beat them over
the head, or whatever.

So, my question remains:  Did this resolution ever make it into a
revision of the Code of Conduct or not.  I thought it did, but can't
find it.  If it didn't, it probably should, and this entire discussion
should be interpreted with that intent.

In passing, I'll note something else.  The underlying assumption of the
entire Code of Conduct threads over five months last year was that:

[quote from Donnie]

A primary focus of CoC enforcement is deterrence from continued 
violation, so permanent action is unnecessary. Thus, what seems 
necessary is a way to take rapid, private, temporary action.

[/quote from Donnie]

The focus here was on errant developers, but by validating userrel's role in
all of this, we know it now explicitly applies to the entire community.
However, the underlying "immediate and temporary" assumptions still apply,
I would think.  Anything else would be a fundamental change as best as I
can tell, and discussion should be framed and clearly understood on that
context.

I guess the answer to my question is that Council did approve changes to the Code
of Conduct, but if they ever got incorporated into the official document, I
can't find it.

Regards,
Ferris

-- 
Ferris McCormick (P44646, MI) <fmccor@gentoo.org>
Developer, Gentoo Linux (Devrel, Sparc, Userrel, Trustees)

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 197 bytes --]

Re: [gentoo-council] Extent of Code of Conduct enforcement

[Donnie Berkholz]

[-- Attachment #1: Type: text/plain, Size: 3585 bytes --]

On 13:24 Fri 25 Jul     , Ferris McCormick wrote:
> Eventually, this was revised here:
> http://archives.gentoo.org/gentoo-council/msg_ba125098c929ea31f34051dfb009d436.xml
> advantage that it did not require any approval at all, but it most
> certainly *did* imply revisions to the Code of Conduct itself, because
> effectively it stripped out any authority from the (non-existent)
> proctors and instead identify a small group of people who would take
> aside CoC violators (in private) and suggest they quit doing whatever
> they were doing wrong.  For the record here, my response to this was:
> 
> [quote]
> 'Nice idea and worth a try.  I have one concern.  Since we are talking
> CoC here, I'd like to emphasise that "assholeness" should be determined
> within the guidelines of the CoC.  Not by some person's own conception
> of "assholeness" --- I'd hate to see a flame war about just who is being
> the asshole in any particular instance.'
> [/quote]
> 
> As best as I can tell, this proposal was *APPROVED*, and at 20080214, was left like this:
> 
> ================================
> Code of Conduct enforcement
> ---------------------------
> 	Promote individual devs responding to people who are being jerks.
> 	Keep responses private, unless that person gets out of hand.
> 
> 	dberkholz will get things going.
> 	To help or get advice, contact him.
> ===================================
> 
> I think the final intent was that Council expected the Code of Conduct
> to be pretty much self-enforcing, driven by members of the community who
> cared enough to take violators aside and calm them down, beat them over
> the head, or whatever.
> 
> So, my question remains:  Did this resolution ever make it into a
> revision of the Code of Conduct or not.  I thought it did, but can't
> find it.  If it didn't, it probably should, and this entire discussion
> should be interpreted with that intent.

Let me share my interpretation to make sure we're on the same page. The 
whole idea here is that there is no new official global response team 
like the proctors.

Nothing changed about the abilities & authorities of individual groups 
that were already in charge of their specific areas.

> In passing, I'll note something else.  The underlying assumption of the
> entire Code of Conduct threads over five months last year was that:
> 
> [quote from Donnie]
> 
> A primary focus of CoC enforcement is deterrence from continued 
> violation, so permanent action is unnecessary. Thus, what seems 
> necessary is a way to take rapid, private, temporary action.
> 
> [/quote from Donnie]
> 
> The focus here was on errant developers, but by validating userrel's role in
> all of this, we know it now explicitly applies to the entire community.
> However, the underlying "immediate and temporary" assumptions still apply,
> I would think.  Anything else would be a fundamental change as best as I
> can tell, and discussion should be framed and clearly understood on that
> context.

It's always desirable for people to change their behavior, but 
unfortunately it doesn't always happen.

Here, the first sentence you quoted is the important one. The second 
sentence only deals with the deterrence part of the first sentence. 
Nothing is addressed about the permanent action, because that email was 
about creating a new proctors-like group, and devrel/userrel already 
existed for permanent action.

-- 
Thanks,
Donnie

Donnie Berkholz
Developer, Gentoo Linux
Blog: http://dberkholz.wordpress.com

[-- Attachment #2: Type: application/pgp-signature, Size: 197 bytes --]