Re: Should mailing list bans be published?

[Holger Levsen]

Hi,

I'm quite very late to this party, but I want to express that I do agree with 
the general idea of publishing mailinglist bans. (In public, with a duration 
attached to it, probably with other restraints I've forgotten since I've read 
this thread.)

What's the "official status" of the listmasters on this now? (And how can we 
get "there"?)


cheers,
Holger


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

Re: Should mailing list bans be published?

[Lisandro Damián Nicanor Pérez Meyer]

On Sunday 27 October 2013 08:54:30 Enrico Zini wrote:
> On Sat, Oct 26, 2013 at 05:27:25PM -0400, Joey Hess wrote:
> > Simply obfuscating the name on the list of banned users (or not posting
> > any names at all, only links to the posts that led to the ban) would
> 
> I'm in favour of not posting names but links to the posts that led to
> the ban. That it solves the problem of google-shaming is welcome, but
> marginal to me. The most important thing to me in doing that, is that we
> make it clear that "we banned you from the list for what you did",
> instead of "we banned you from the list for what you are".
> 
> That subtle distinction in the message is very important to me, because
> it makes the distinction from a community where I'm judged for what I
> am, to a community where I'm judged for what I do.
> 
> In the former, the incentive is to do as little public activity as
> possible, because it might turn into stigma. Places that work in that
> way see very little involvement from me.
> 
> In the latter, I'm happy to participate and take my chances, because I
> know that if I screw up I can learn something from my mistakes, grow up
> and do better next time.

+1.

-- 

Lisandro Damián Nicanor Pérez Meyer
http://perezmeyer.com.ar/
http://perezmeyer.blogspot.com/


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

Re: Should mailing list bans be published?

[Tollef Fog Heen]

]] Stefano Zacchiroli 

> On Mon, Nov 04, 2013 at 02:51:52PM +0100, Tollef Fog Heen wrote:
> > > So, what would be the beneficial social effects of publishing the ban
> > > *duration*?
> > 
> > The ban duration is an indication of how severe we think the violation
> > is.  You don't get a lifetime ban for a minor transgression and you
> > don't get a one-day ban for serious harassments.
> 
> Of course. The question is: do you think disclosing ban duration will
> discourage bad behavior on our mailing lists more than just disclosing
> the bans currently in effect? (I don't.)

To the extent that publishing the bans in the first place is going to
help: yes, I think so.

Another important point is it will enable more oversight so you can see
that a ban was instated at $time and is due to be removed at $time+$x,
but now > $time+$x and it's still not removed and it's the possible to
poke the relevant people to get it fixed.  If the expiry is unknown, you
have no idea if the ban is still there on purpose or not.

-- 
Tollef Fog Heen
UNIX is user friendly, it's just picky about who its friends are


-- 
To UNSUBSCRIBE, email to debian-project-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/87ppqfqktr@qurzaw.varnish-software.com

Re: Should mailing list bans be published?

[Uoti Urpala]

Russ Allbery wrote:
> Camaleón  writes:
> 
> > The mailing list managers/admins have the right to ban whoever they
> > decide, but in the aim of "fair play", the user should also have the
> > right to defend him/herself from the accusations, expose his/her
> > reasoning and be able to restore him/her reputation or recognize the
> > error, say sorry and come back to the list again.
> 
> I don't agree with this part of your message.  Debian is under no
> obligation to play fair.  We are not a legal system and have no obligation
> to give people any sort of due process.  Debian's mailing lists are
> instruments of the project, and the project can decide how they are used.
> 
> This sort of quasi-judicial process eats up *vast* quantities of time and
> effort, and if we go down this path, the listmasters will end up spending

> It's not like we're depriving people of life and limb, or even property.
> They just can't send mail to one of our mailing lists for a while.  If

I think these arguments work much worse in the context of _publishing_
bans as examples or even as a punishment. You can ban anyone whose
face/opinions/attitude you don't like, and there is no particular
obligation of "fairness". However, it's another thing to make public
claims (especially as an official project stance) that someone's posts
are examples of particularly unacceptable behavior or contain false
claims, and give no opportunity to respond to such accusations. I think
there *are* obligations of fairness in this case.



-- 
To UNSUBSCRIBE, email to debian-project-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/1383606635.1856.311.camel@glyph.nonexistent.invalid

Re: Should mailing list bans be published?

[Charles Plessy]

Le Mon, Nov 04, 2013 at 02:33:34PM -0500, Brian Gupta a écrit :
> 
> I don't know the answer but perhaps, we can try experimenting with a system
> where the first action is a polite public warning by listmaster, pointing to
> code of conduct. (Assuming that the code of conduct is updated to cover
> this.)

Hello everybody,

I think that we should also encourage anybody (not just the listmasters) to
send complain messages in private first.  Public reminders should be the last
ressort, not because of considerations on the person receiving the reminder,
but because the very high probability that it triggers the sending of more and
more off-topic messages.

To take a concrete example, in last month's thread about systemd, I sent a
public comment about "assuming good faith", and I hesitated a lot about making
it private.  I sent it in public because at that point, I thought that it was
important to compensate the harsh message by an appreciation to the Gnome
developers, but the cost of it was an extra 8 messages that probably would
never have been posted otherwise: "censorship !" -> "no it isn't" -> "angriness
is not surprising" -> "no it is" -> "no it isn't, go figure yourself" -> "you
are contradicting yourself" -> "what do you mean ?" -> "please stop off-topic
discussion" (which thankfully was listened).  Altogether, I am not sure that I
made the right choice.

Cheers,

-- 
Charles Plessy
Tsurumi, Kanagawa, Japan


-- 
To UNSUBSCRIBE, email to debian-project-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20131104210524.ga1...@falafel.plessy.net

Re: Should mailing list bans be published?

[Brian Gupta]

On Mon, Nov 4, 2013 at 12:17 PM, Russ Allbery  wrote:
> Camaleón  writes:
>
>> The mailing list managers/admins have the right to ban whoever they
>> decide, but in the aim of "fair play", the user should also have the
>> right to defend him/herself from the accusations, expose his/her
>> reasoning and be able to restore him/her reputation or recognize the
>> error, say sorry and come back to the list again.
>
> I don't agree with this part of your message.  Debian is under no
> obligation to play fair.  We are not a legal system and have no obligation
> to give people any sort of due process.  Debian's mailing lists are
> instruments of the project, and the project can decide how they are used.
>
> This sort of quasi-judicial process eats up *vast* quantities of time and
> effort, and if we go down this path, the listmasters will end up spending
> all their time responding to people who milk every possible process step
> out of the judicial process just to be irritating.  I've seen this happen
> repeatedly in the IETF, which has this sort of judicial process for
> mailing list bans, and it's a disaster.  It also strongly discourages ever
> banning anyone, since the process is such a nightmare.
>
> It's not like we're depriving people of life and limb, or even property.
> They just can't send mail to one of our mailing lists for a while.  If
> there's an occasional mistake, oh well, life goes on.  If I get banned for
> a month or two from a mailing list for something I said that was
> ill-advised, I'm not going to argue about it; I'm going to realize that,
> whatever I thought about what I said, other people were quite upset about
> it, and I should take that into account in the future.  It gives me some
> time to think about it.
>
> Review of the decisions *by other project members* is fine, but generally
> the messages themselves are self-explanatory and I neither care nor want
> to know what justifications the person banned is going to dig up.  If
> they're sorry and won't do it again, great!  After the ban expires, they
> can demonstrate that.  But let's keep this process simple.
>
> If some decision seems egregiously wrong, other developers who are worried
> about it can always approach the banned person in private and ask for
> their side of the story, but we don't need to make anything formal.
>
>> So I have to agree with Alexander's POV that these things need to be
>> done in the background to preserve the privacy right of the user,
>> despite if he/she is using a real name or a nickname.
>
>> In brief: IMO there's no need to make a public list and Debian Project
>> has nothing to demonstrate nobody because being effectively banned is
>> the only "proof of action" worth doing.
>
>> Here in Spain we have a saying ("hacer leña del árbol caído") which can
>> resemble into "kick a man when he's down" and that's IMO what we should
>> avoid here.
>
> I'm still somewhat inclined to agree with this, though, and prefer
> debian-private as the venue for advertising these, although the arguments
> about making it publicly clear that we're doing something about bad
> behavior on our lists are fairly compelling.

I don't know the answer but perhaps, we can try experimenting with a system
where the first action is a polite public warning by listmaster,
pointing to code of
conduct. (Assuming that the code of conduct is updated to cover this.)

This has a few advantages:

1) We point out bad behavior as examples of what not to do as they happen,
where they happen
2) We potentially change the behavior before it gets out of hand.
  (I suspect that in many cases an official warning would be enough.)
3) For the majority of incidents where people forget to treat each other with
  respect, we give them a chance to correct the behavior, and apologize,
  without some sort of "this is going on your permanent record".

In the end though I do feel that having our listmasters officially and publicly
call out bad behavior could be an improvement over the status quo.

I guess that leaves open:
1) what do when people don't heed the warnings? In this case, I'd say a public
notification is ok as well, since a warning was given.
2) Action was too far our of line to warrant a simple warning. In this
case I also
think a public notice is appropriate, as any action that is so bad
we can't give
a warning, needs to be addressed publicly.

As far as tracking on a wiki, I don't have a strong feeling either way, as to me
the most important place to make this public is on the list where the
action took
place.

In any case, whatever we decide, I do think we should document our policy in
the code of conduct and make sure it is well advertised. We needn't make the
policy too complex, and could have a simple cardinal rule: "Treat other people
with respect. Failure to do so could result in a public banning from Debian
mailing lists."

Thanks,
Brian

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

Re: Should mailing list bans be published?

[Camaleón]

On Mon, 04 Nov 2013 09:17:29 -0800, Russ Allbery wrote:

> Camaleón  writes:
> 
>> The mailing list managers/admins have the right to ban whoever they
>> decide, but in the aim of "fair play", the user should also have the
>> right to defend him/herself from the accusations, expose his/her
>> reasoning and be able to restore him/her reputation or recognize the
>> error, say sorry and come back to the list again.
> 
> I don't agree with this part of your message.  Debian is under no
> obligation to play fair.  We are not a legal system and have no
> obligation to give people any sort of due process.  Debian's mailing
> lists are instruments of the project, and the project can decide how
> they are used.

(...)

Well, being "fair" does not have to relate to any legal concept so please 
disregard any thought about it; to be fair is just a moral issue like 
being transparent. And yes, I would expect Debian (as a project) embraces 
that way.

> It's not like we're depriving people of life and limb, or even property.
> They just can't send mail to one of our mailing lists for a while.  If
> there's an occasional mistake, oh well, life goes on.  If I get banned
> for a month or two from a mailing list for something I said that was
> ill-advised, I'm not going to argue about it; I'm going to realize that,
> whatever I thought about what I said, other people were quite upset
> about it, and I should take that into account in the future.  It gives
> me some time to think about it.

The problem I see here is not "being banned" (which I agree) but "being 
listed" as rejected for posting in a list that can be accessible. 
Sincerely, I don't see any gain in exposing those facts (the facts are 
that a user has been banned from posting to certain mailing list and also 
that Debian has acted in this regard).

> Review of the decisions *by other project members* is fine, but
> generally the messages themselves are self-explanatory and I neither
> care nor want to know what justifications the person banned is going to
> dig up.  If they're sorry and won't do it again, great!  After the ban
> expires, they can demonstrate that.  But let's keep this process simple.

Simple, I agree but also fair (for both parts) and transparent.

> If some decision seems egregiously wrong, other developers who are
> worried about it can always approach the banned person in private and
> ask for their side of the story, but we don't need to make anything
> formal.

"Developers"? You mean only "developers" can lead an action to ask for 
someone to be banned? :-?

>> So I have to agree with Alexander's POV that these things need to be
>> done in the background to preserve the privacy right of the user,
>> despite if he/she is using a real name or a nickname.
> 
>> In brief: IMO there's no need to make a public list and Debian Project
>> has nothing to demonstrate nobody because being effectively banned is
>> the only "proof of action" worth doing.
> 
>> Here in Spain we have a saying ("hacer leña del árbol caído") which can
>> resemble into "kick a man when he's down" and that's IMO what we should
>> avoid here.
> 
> I'm still somewhat inclined to agree with this, though, and prefer
> debian-private as the venue for advertising these, although the
> arguments about making it publicly clear that we're doing something
> about bad behavior on our lists are fairly compelling.

I only wish that, whatever the final decision, it gets properly explained 
at the wiki so any user can be aware of this.

Greetings,

-- 
Camaleón


-- 
To UNSUBSCRIBE, email to debian-project-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/pan.2013.11.04.18.31...@gmail.com

Re: Should mailing list bans be published?

[Russ Allbery]

Camaleón  writes:

> The mailing list managers/admins have the right to ban whoever they
> decide, but in the aim of "fair play", the user should also have the
> right to defend him/herself from the accusations, expose his/her
> reasoning and be able to restore him/her reputation or recognize the
> error, say sorry and come back to the list again.

I don't agree with this part of your message.  Debian is under no
obligation to play fair.  We are not a legal system and have no obligation
to give people any sort of due process.  Debian's mailing lists are
instruments of the project, and the project can decide how they are used.

This sort of quasi-judicial process eats up *vast* quantities of time and
effort, and if we go down this path, the listmasters will end up spending
all their time responding to people who milk every possible process step
out of the judicial process just to be irritating.  I've seen this happen
repeatedly in the IETF, which has this sort of judicial process for
mailing list bans, and it's a disaster.  It also strongly discourages ever
banning anyone, since the process is such a nightmare.

It's not like we're depriving people of life and limb, or even property.
They just can't send mail to one of our mailing lists for a while.  If
there's an occasional mistake, oh well, life goes on.  If I get banned for
a month or two from a mailing list for something I said that was
ill-advised, I'm not going to argue about it; I'm going to realize that,
whatever I thought about what I said, other people were quite upset about
it, and I should take that into account in the future.  It gives me some
time to think about it.

Review of the decisions *by other project members* is fine, but generally
the messages themselves are self-explanatory and I neither care nor want
to know what justifications the person banned is going to dig up.  If
they're sorry and won't do it again, great!  After the ban expires, they
can demonstrate that.  But let's keep this process simple.

If some decision seems egregiously wrong, other developers who are worried
about it can always approach the banned person in private and ask for
their side of the story, but we don't need to make anything formal.

> So I have to agree with Alexander's POV that these things need to be
> done in the background to preserve the privacy right of the user,
> despite if he/she is using a real name or a nickname.

> In brief: IMO there's no need to make a public list and Debian Project
> has nothing to demonstrate nobody because being effectively banned is
> the only "proof of action" worth doing.

> Here in Spain we have a saying ("hacer leña del árbol caído") which can
> resemble into "kick a man when he's down" and that's IMO what we should
> avoid here.

I'm still somewhat inclined to agree with this, though, and prefer
debian-private as the venue for advertising these, although the arguments
about making it publicly clear that we're doing something about bad
behavior on our lists are fairly compelling.

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


--
To UNSUBSCRIBE, email to debian-project-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/87mwlkhys6@windlord.stanford.edu

Re: Should mailing list bans be published?

[Camaleón]

On Sat, 26 Oct 2013 10:46:41 -0700, Steve Langasek wrote:

> Hi folks,
> 
> Was discussing with one of the listmasters (Alexander Wirt) on IRC today
> about mailing list bans, because it turns out that someone I was just
> about to ask the listmasters to ban from debian-devel had just been
> blocked in response to a request from someone else.
> 
> This led to a philosophical debate about whether bans should be made
> public.
> Alexander expressed concern that having them published could be harmful
> to a person's reputation, since employers will google your name and see
> that you've been banned from a large project such as Debian.

(...)

> What do the rest of you think?

After carefully reading the exposed reasons and with my mailing list user
hat on, I have to disagree with making such a list publicly available.

If a user has been already banned from a list, what kind of additional
punishment do you want he/she suffers? It should be enough by just
privately informing the user that he/she has been banned, what mailing
list is the ban applied to and what's the reason (and length) for the ban.

The mailing list managers/admins have the right to ban whoever they
decide, but in the aim of "fair play", the user should also have the right
to defend him/herself from the accusations, expose his/her reasoning and
be able to restore him/her reputation or recognize the error, say sorry
and come back to the list again.

So I have to agree with Alexander's POV that these things need to be done
in the background to preserve the privacy right of the user, despite if
he/she is using a real name or a nickname.

In brief: IMO there's no need to make a public list and Debian Project has
nothing to demonstrate nobody because being effectively banned is the only
"proof of action" worth doing.

Here in Spain we have a saying ("hacer leña del árbol caído") which can
resemble into "kick a man when he's down" and that's IMO what we should
avoid here.

Greetings,

-- 
Camaleón


-- 
To UNSUBSCRIBE, email to debian-project-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/pan.2013.11.04.16.29...@gmail.com

Re: Should mailing list bans be published?

[Didier 'OdyX' Raboud]

Le lundi, 4 novembre 2013 15.08:05 Stefano Zacchiroli a écrit :
> On Mon, Nov 04, 2013 at 02:51:52PM +0100, Tollef Fog Heen wrote:
> > > So, what would be the beneficial social effects of publishing the
> > > ban *duration*?
> > 
> > The ban duration is an indication of how severe we think the
> > violation is.  You don't get a lifetime ban for a minor
> > transgression and you don't get a one-day ban for serious
> > harassments.
> 
> Of course. The question is: do you think disclosing ban duration will
> discourage bad behavior on our mailing lists more than just disclosing
> the bans currently in effect? (I don't.)

That's a good point.

I think (and apparently you agree) that the banned person must get 
notified of the duration of the ban; I think that this is absolutely 
essential for the avoidance of blanket punishments. "Infinite ban" 
should be used as very last resort [0] and subject to re-examination 
anyway.

I do see some value in having the ban durations published along with the 
rationales, as both these constitute the listmasters' decision for which 
review should be possible. Where rationales (offending posts) carry 
value for everyone because they show what behaviours are not OKay, I can 
agree that durations mostly matter to those exercising the review.

Cheers,
OdyX

[0] I don't doubt it is.

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

Re: Should mailing list bans be published?

[Stefano Zacchiroli]

On Mon, Nov 04, 2013 at 02:51:52PM +0100, Tollef Fog Heen wrote:
> > So, what would be the beneficial social effects of publishing the ban
> > *duration*?
> 
> The ban duration is an indication of how severe we think the violation
> is.  You don't get a lifetime ban for a minor transgression and you
> don't get a one-day ban for serious harassments.

Of course. The question is: do you think disclosing ban duration will
discourage bad behavior on our mailing lists more than just disclosing
the bans currently in effect? (I don't.)

-- 
Stefano Zacchiroli  . . . . . . .  z...@upsilon.cc . . . . o . . . o . o
Maître de conférences . . . . . http://upsilon.cc/zack . . . o . . . o o
Former Debian Project Leader  . . @zack on identi.ca . . o o o . . . o .
« the first rule of tautology club is the first rule of tautology club »


signature.asc
Description: Digital signature

Re: Should mailing list bans be published?

[Tollef Fog Heen]

]] Stefano Zacchiroli 

> So, what would be the beneficial social effects of publishing the ban
> *duration*?

The ban duration is an indication of how severe we think the violation
is.  You don't get a lifetime ban for a minor transgression and you
don't get a one-day ban for serious harassments.

-- 
Tollef Fog Heen
UNIX is user friendly, it's just picky about who its friends are


-- 
To UNSUBSCRIBE, email to debian-project-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/877gcos29z@qurzaw.varnish-software.com

Re: Should mailing list bans be published?

[Stefano Zacchiroli]

On Mon, Nov 04, 2013 at 02:21:17PM +0100, Didier 'OdyX' Raboud wrote:
> I disagree on the point of not making the ban durations public. Although 
> I understand the effect you're afraid of, I think that the benefits of 
> having the durations public outweigh the downsides: even if the banned 
> persons could try to trick the system, that would be easily detected I 
> think. Also, most of the effects of the ban are social, not technical.

Right.

> The ban durations should be made public (and very much communicated to 
> the banned person) because that gives a dimension to the punishment, you 
> can then get "short" or "long" bans, depending on how far you crossed 
> the line. An offender could first get a short ban, and when coming back, 
> if crossing the line again, a longer ban (exponentially). The only limit 
> to that would be the listmasters' patience.

So, what would be the beneficial social effects of publishing the ban
*duration*? I can't see any of that. The main beneficial effect we're
looking for is sending the message that bad behavior on Debian mailing
lists is not tolerated "here, see what happens when you post nasty
messages like those?". To have this effect you don't need to disclose
ban duration. (Of course, the banned person should be made aware of the
ban duration, but I'm sure that's already the case.)

-- 
Stefano Zacchiroli  . . . . . . .  z...@upsilon.cc . . . . o . . . o . o
Maître de conférences . . . . . http://upsilon.cc/zack . . . o . . . o o
Former Debian Project Leader  . . @zack on identi.ca . . o o o . . . o .
« the first rule of tautology club is the first rule of tautology club »


signature.asc
Description: Digital signature

Re: Should mailing list bans be published?

[Didier 'OdyX' Raboud]

Hi Jonathan,

Le dimanche, 3 novembre 2013 14.06:33 Jonathan Dowland a écrit :
> I think bans should be time-limited in
> almost all cases, with perma-bans being very rare indeed. I don't
> think that ban durations should be disclosed publically or to the
> person banned (otherwise badly behaved people will just count down to
> the ban lifting before posting.)

I disagree on the point of not making the ban durations public. Although 
I understand the effect you're afraid of, I think that the benefits of 
having the durations public outweigh the downsides: even if the banned 
persons could try to trick the system, that would be easily detected I 
think. Also, most of the effects of the ban are social, not technical.

The ban durations should be made public (and very much communicated to 
the banned person) because that gives a dimension to the punishment, you 
can then get "short" or "long" bans, depending on how far you crossed 
the line. An offender could first get a short ban, and when coming back, 
if crossing the line again, a longer ban (exponentially). The only limit 
to that would be the listmasters' patience.

Cheers,
OdyX

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

Re: Should mailing list bans be published?

[Jonathan Dowland]

My feeling before reading this thread was that bans should be accessible
to DDs somewhere but this thread has convinced me that they should be
made public. I think bans should be time-limited in almost all cases,
with perma-bans being very rare indeed. I don't think that ban durations
should be disclosed publically or to the person banned (otherwise badly
behaved people will just count down to the ban lifting before posting.)

I think if we did this we may be more inclined to suggest bans more
often, or softer solutions for less egregiously bad behaviour (periods
of time-delayed posting for example) which might mean manpower issues
for the listmasters. In which case a seperate moderator team should be
considered (superlist of listmasters?), or teams, or perhaps per-list
moderator teams. I'd trust the listmasters to decide if and when they
needed help.

I found Charles' post about self-limiting post frequency very
interesting. One of the disadvantages of the threaded model of
communication is that various facets of a discussion get scattered to
all the leaves which means if you have several points to make, the
logical place to make them is at the leaves. This has advantages and
disadvantages (not least, many duplicate leaves as the same points
are raised and re-raised in sub-threads). I do like the experiments to
use the wiki to maintain "position statements".


-- 
To UNSUBSCRIBE, email to debian-project-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20131103140633.gc9...@bryant.redmars.org

Re: Should mailing list bans be published?

[Charles Plessy]

Le Sun, Oct 27, 2013 at 07:58:03AM +, Lars Wirzenius a écrit :
> On Sun, Oct 27, 2013 at 09:00:20AM +0900, Charles Plessy wrote:
> > 
> > Given how arbitrarly other bans have been proposed, I think that the
> > outcome should stay private unless the banned person wishes so.
> 
> I don't understand this at all. Are you saying Debian listmasters, who
> decide on bans, have been making arbitrary, and therefore badly
> justified, bans?

Dear Lars and listmasters,

no, this is not what I wrote, I am not saying that listmasters are making
arbitrary bans (after checking an on-line dictionary, let me clarify that I
read "arbitrary" in the sense of "Determined by chance, whim, or impulse, and
not by necessity, reason, or principle", not "Established by a court or judge
rather than by a specific law or statute").

But I am saying that everybody, me, you, the listmasters, and anybody else who
do some work will eventually take a decision that is not the best.  Only the
ones who does nothing does no mistake.

More in particular, I have seen on another list this year a public call for
banning a contributor, that was written in a precise authoritative style and
looked well argumented, but was quickly contradicted by at least four debian
developers, who unerlined that the contributor was polite, constructive and
respectful.

The point I want to make is that decisions are hard to take and the listmasters
will eventually be presented contradicting informations, that might not even
come on the same day.

So why don't we mitigate in advance any possibility of error on our side ?

Anyway, the tone of this debate gives a strange taste that we are trying to
decide for the listmasters.  On my side, if they are reluctant to publish a
list of banned people, I say: I would be as well if I were in your position,
and I would certainely not blame you if you would keep this information
private.  (For the reasonning, it was written by me and others before,
and I will save everybody's time by not repeating it).

Have a nice day,

-- 
Charles Plessy
Tsurumi, Kanagawa, Japan


-- 
To UNSUBSCRIBE, email to debian-project-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20131028223043.gc14...@falafel.plessy.net

Re: Should mailing list bans be published?

[Gunnar Wolf]

Ingo Jürgensmann dijo [Sat, Oct 26, 2013 at 08:56:59PM +0200]:
> > This led to a philosophical debate about whether bans should be made public.
> > Alexander expressed concern that having them published could be harmful to a
> > person's reputation, since employers will google your name and see that
> > you've been banned from a large project such as Debian.
> 
> 
> I agree with Alexanders concern here. Publishing other peoples
> personal data without prior allowance might even violate privacy
> legislation in some countries.

I side with Steve's view here. Now, we *could* obscure the personal
data in a way that it won't show on general web searches — Say,
something as trivial as omitting the person's name, and publishing the
file with just sha256sum(email). This still allows us to make an easy
querying interface (even allowing for historical information on a
given mail address).

Of course, this would omit the fact we are dealing with people and not
with mail addresses. Am I "gw...@gwolf.org" or "gw...@debian.org"?
(But OTOH, am I "Gunnar Wolf", "Gunnar Eyal Wolf Iszaevich" or "Big
Bearded Troll"?)


signature.asc
Description: Digital signature

Re: Should mailing list bans be published?

[Ian Jackson]

Steve Langasek writes ("Should mailing list bans be published?"):
> I think we should publish them, for several reasons:

I agree wholeheartedly.

Also, to expand on this:

>  - It casts sunlight on the kinds of decisions that the listmasters are
>making WRT bans, so that we collectively have oversight of these
>decisions and can ensure our principles are being applied fairly and
>consistently.

Publishing the bans gives us the opportunity to congratulate or
castigate the listmasters (in private!) for their action.  This will
give the listmasters much better feedback on their thresholds and
criteria for disciplinary action.

For example, it would be much easier to write to the listmasters and
say "thanks for acting; I would have preferred it if you could have
stepped in sooner".

Ian.


-- 
To UNSUBSCRIBE, email to debian-project-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/21102.30750.96740.265...@chiark.greenend.org.uk

Re: Should mailing list bans be published?

[Ian Jackson]

Russ Allbery writes ("Re: Should mailing list bans be published?"):
> But, against that, I would say that the point of mailing list bans is not
> to humiliate or expose someone who behaved poorly, or even to call further
> attention to their poor behavior.  Rather, the goal is to *stop* calling
> attention to their behavior and to defend the usability of the list
> against further interference.  Actions that tend to produce *more*
> discussion seem directly contrary to that goal.  Ideally, everyone would
> simply forget the banned person ever participated in the list and proceed
> as if they had never said anything.

You're missing that one goal of publishing such bans is to provide an
example to other people, clearly showing what Debian will not
tolerate.

This works both to show potential good contributors that Debian does
take this seriously and will protect them if they are attacked, and to
show potential wrongdoers what they won't get away with.

Without publishing the bans it just looks like the trolls got bored,
or the victims went into hiding.

I agree with those who say that, for Debian's reptutation etc., a
series of bad messages is worse than a series of bad messages followed
by a disciplinary notice from the listmasters.

Before reading this thread I was frustrated by what appears looks to
an outsider (which includes a mere DD) like lack of action, by those
in authority, against bad behaviour.  I'm surprised to discover that
there are unpublished bans.

Ian.


-- 
To UNSUBSCRIBE, email to debian-project-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/21102.29336.22477.957...@chiark.greenend.org.uk

Re: Should mailing list bans be published?

[Didier 'OdyX' Raboud]

Le samedi, 26 octobre 2013 10.46:41 Steve Langasek a écrit :
> This led to a philosophical debate about whether bans should be made
> public. Alexander expressed concern that having them published could
> be harmful to a person's reputation, since employers will google your
> name and see that you've been banned from a large project such as
> Debian.
> 
> I think we should publish them, for several reasons:
> (…)

While I do agree that we should publish the bans, I think we should 
really do it carefully and that these publications should satisfy the 
following criterias:

* forgiveness. This goes along with the concerns about harming one's 
reputation. As we all change during our lives, I think it shouldn't be 
of Debian's role to keep an eternal track record of past bans. I would 
therefore propose to only list active bans on a non-indexed webpage. 
Notifications of new bans could be posted to a private non-archived list 
such as -private (because posting them to an archived list would defeat 
the purpose). This would ensure people that later get unbanned also get 
their names removed from that list. They would start afresh (at least 
publically, not for the listmasters of course) and I think that's a good 
thing.

* least surprise and non-retroactivity. I think that the legal concerns 
can be circumvented by warning offenders in advance that in the case of 
a ban their names would be published. As Alexander puts it [0], "Banning 
is usually the last action we take and we only use it if we really have 
to.". This implies that the offenders get mails from the listmasters 
before the actual ban is put in place. If they get informed of the 
consequences, they can simply stop posting and don't have to face the 
consequences. With this in mind, I think we should also only publish new 
bans and not the ones already in force (non-retroactivity).

[0] <20131027164642.ga15...@smithers.snow-crash.org>

* transparency. This is very much along the lines of Steve's original 
words:

> So I don't think bans need to be posted anywhere prominent like
> debian-devel-announce, but I do think basic facts like who is banned,
> for how long, and the rationale (with links to specific mailing list
> posts as reference) should be made public.

It is IMHO very important that published bans go along with a list of 
links to offending mails, a rationale (short-ish explanation of why the 
mails are a problem) and the duration of the ban. This makes the 
listmaster's job public, defendable and challengeable, which are good 
things™ too.

Cheers,

OdyX

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

Re: Should mailing list bans be published?

[Philip Hands]

Hi Steve,

Thanks for starting this thread.

Steve Langasek  writes:

> On Sun, Oct 27, 2013 at 10:33:42PM +0100, Florian Weimer wrote:
>> * Joey Hess:
>
>> > Simply obfuscating the name on the list of banned users (or not posting
>> > any names at all, only links to the posts that led to the ban) would
>> > eliminate most reputational damage. Ie, random searches for that
>> > person would not turn up a high pagerank debian.org page listing their
>> > youthful indiscretions.
>
>> > Using eg "J. Hess" would probably be fine in most cases.
>
>> I recommend to use a web page, and not announce bans on public mailing
>> lists because such announcements invite subsequent discussion, likely
>> decloaking the banned poster.
>
> Reducing subsequent discussion is inseparable from reducing both oversight
> and the closure given to other list participants.  I don't consider posting
> such content on a web page to suitably address the concerns.

I think it would be fair enough to have a fully public (but not very
well linked) web page that lists mails that were considered sufficient
to provoke a ban, and the duration and conditions of each ban.

I do not think such a page is liable to violate any rights because it
would not list names, and it would not end up being top hit for the
abuser's name in later years -- the mails they sent that provoked the
ban might well end up being their top hit, but that would be without the
help of the "Last straw" page.

The page could perhaps also be a place to collect resources that might
encourage people to express themselves more constructively, and so could
be referred to by the listmasters when issuing a first warning.

To address the need for oversight/closure, would you consider a
simultaneous post to debian-private sufficient?  I don't think it's
enough without the public list, but the combination allows future
abusers to be refereed to the list as an indication that they might want
to moderate their behaviour because we do actually ban people

The post to debian-private does fail to provide closure for non DDs but
otherwise does the job, and I would think that the readership of
debian-private is diverse enough that the spectrum of opinion should be
wide enough to ensure good oversight.

Also, if we're going to make these changes, I think we should publicise
them very widely, possibly going as far as a mail sent to every mailing
list where the policy is going to be implemented, and then any bans that
are then published in this way should be justifiable by reference only
to mails sent after that announcement -- it would not be fair to spring
this on a troll for sins committed before the announcement.

Cheers, Phil.

P.S. in case it's not obvious, I fully support publication, as long as we
can do it without putting a blight on the futures of people that might now
be committing childhood sins, and also without getting our listmasters sued.
-- 
|)|  Philip Hands [+44 (0)20 8530 9560]http://www.hands.com/
|-|  HANDS.COM Ltd.http://ftp.uk.debian.org/
|(|  10 Onslow Gardens, South Woodford, London  E18 1NE  ENGLAND


pgp9dkiFoUIl8.pgp
Description: PGP signature

Re: Should mailing list bans be published?

[Ingo Jürgensmann]

Am 27.10.2013 um 20:46 schrieb Steve Langasek :

> On Sat, Oct 26, 2013 at 05:27:25PM -0400, Joey Hess wrote:
>> Bart Martens wrote:
>>> I suggest we keep things civil, with respect for the persons involved.  It's
>>> really not up to Debian to harm someone's reputation, and that could reflect
>>> bad on Debian's reputation.
> 
>>> Approaches I could support :
>>> - post the bans with reasons on debian-private
>>> - or maintain a list of bans with reasons in a text file on a Debian machine
>>>  where DDs can read this info.
> 
>> Simply obfuscating the name on the list of banned users (or not posting
>> any names at all, only links to the posts that led to the ban) would
>> eliminate most reputational damage. Ie, random searches for that
>> person would not turn up a high pagerank debian.org page listing their
>> youthful indiscretions.
> 
>> Using eg "J. Hess" would probably be fine in most cases.
> 
> This also seems like a good compromise to me.  Do the other folks who object
> to publishing information that could damage the poster's reputation (e.g.,
> Bart, Ingo) think this is ok?


Hmmm... difficult. 

I can understand the reasons to make bans public. It's a way of checks and 
balances for the listmasters and the public, which is good. 

On the other hand, making such a ban publically known is like a pillory for 
that person and might violate the persons fundamental rights, especially the 
right of informational self-determination (EN: 
http://en.wikipedia.org/wiki/Informational_self-determination, DE: 
http://de.wikipedia.org/wiki/Informationelle_Selbstbestimmung) in Germany, 
which has its counter part in Article 8, 1 European Convention on Human Rights 
(http://en.wikipedia.org/wiki/Article_8_of_the_European_Convention_on_Human_Rights).

The name of a person is, of course, a personal data and thus proteced under 
privacy rights. When that person writes to a mailing list, s/he is granting the 
right to publish that mail to others - because in most cases people are aware 
of the fact that the willingly write to a public and open mailing list. But 
listing the name in a different context that it was intended by the author 
itself, namely on list of ML bans, might violate this basic right. Also keep in 
mind that a General Data Protection Regulation 
(http://en.wikipedia.org/wiki/General_Data_Protection_Regulation) in the 
Eruopean Union is on its way, where a "Right to be Forgotten" is mentioned in 
Article 17, which could be applied in this case as well. 

Obfuscating the name by shorten it will not work for rare family names like 
mine for example, so I don't think this could be an option at all. 
Conducting a webpage with links to the posts in question that led to the ban 
seems to break the direct connection between the ban and the name when doing 
some easy websearchs for the name, but I don't know the detials of search 
algorithms, so maybe the ban page will pop up quickly when searching for the 
name. 

Currently I see no good way of making the ban public - except to get the 
consent of the banned person. 
For example when a person gets banned s/he can be asked whether s/he is ok with 
taking this ban to the mailing list to get input from other users and start a 
discussion whether this ban should stand or get removed again. When the person 
agrees the issue can be made publically known and discussed. 

PS:
I'm aware of the fact that privacy standards are lower or seen in a more 
relaxed way that do not match the high standards we have in the EU and 
especially in Germany and that our high german standard causes shaking the head 
in most other countries. But, well, I'm a German and can't ignore basic rights 
and when asked what's more important to me, Debians reputation or peoples 
fundamental rights, I'll always stand and fight for the fundamental rights. 
Sorry. :)

-- 
Ciao...//  Fon: 0381-2744150
  Ingo   \X/   http://blog.windfluechter.net


gpg pubkey:  http://www.juergensmann.de/ij_public_key.asc



signature.asc
Description: Message signed with OpenPGP using GPGMail

Re: Should mailing list bans be published?

[Bart Martens]

On Sun, Oct 27, 2013 at 12:46:07PM -0700, Steve Langasek wrote:
> On Sat, Oct 26, 2013 at 05:27:25PM -0400, Joey Hess wrote:
> > Bart Martens wrote:
> > > I suggest we keep things civil, with respect for the persons involved.  
> > > It's
> > > really not up to Debian to harm someone's reputation, and that could 
> > > reflect
> > > bad on Debian's reputation.
> 
> > > Approaches I could support :
> > > - post the bans with reasons on debian-private
> > > - or maintain a list of bans with reasons in a text file on a Debian 
> > > machine
> > >   where DDs can read this info.
> 
> > Simply obfuscating the name on the list of banned users (or not posting
> > any names at all, only links to the posts that led to the ban) would
> > eliminate most reputational damage. Ie, random searches for that
> > person would not turn up a high pagerank debian.org page listing their
> > youthful indiscretions.
> 
> > Using eg "J. Hess" would probably be fine in most cases.
> 
> This also seems like a good compromise to me.  Do the other folks who object
> to publishing information that could damage the poster's reputation (e.g.,
> Bart, Ingo) think this is ok?

Publishing the bans with links to the posts that led to the bans, means that
the names are published with the bans, because the names are on the posts.

Regards,

Bart Martens


-- 
To UNSUBSCRIBE, email to debian-project-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20131028044235.ga26...@master.debian.org

Re: Should mailing list bans be published?

[Steve Langasek]

On Sun, Oct 27, 2013 at 10:33:42PM +0100, Florian Weimer wrote:
> * Joey Hess:

> > Simply obfuscating the name on the list of banned users (or not posting
> > any names at all, only links to the posts that led to the ban) would
> > eliminate most reputational damage. Ie, random searches for that
> > person would not turn up a high pagerank debian.org page listing their
> > youthful indiscretions.

> > Using eg "J. Hess" would probably be fine in most cases.

> I recommend to use a web page, and not announce bans on public mailing
> lists because such announcements invite subsequent discussion, likely
> decloaking the banned poster.

Reducing subsequent discussion is inseparable from reducing both oversight
and the closure given to other list participants.  I don't consider posting
such content on a web page to suitably address the concerns.

-- 
Steve Langasek   Give me a lever long enough and a Free OS
Debian Developer   to set it on, and I can move the world.
Ubuntu Developerhttp://www.debian.org/
slanga...@ubuntu.com vor...@debian.org


signature.asc
Description: Digital signature

Re: Should mailing list bans be published?

[Charles Plessy]

> On Sun, 27 Oct 2013, Charles Plessy wrote:
> 
> > In parallel, I think that we need some technical or social pressure for
> > limiting to 1 or 2 messages a day each individual contribution to long 
> > threads.

Le Sun, Oct 27, 2013 at 07:35:56AM +0100, Alexander Wirt a écrit :
>
> That is nonsense. There will always be people that have to write more mails
> to a thread. For example the maintainer of a discussed software, dpl, or the
> ctte member. And so on. Such a general limitation just won't work.

Hi Alex,

Just to clarify: I am not asking for new inflexible rules.

I refrain myself from sending more than 1 or 2 messages per thread and per day,
and I would be grateful to others if they would do so.  If this means that
somebody else will be the one posting an opinion or proposing an idea that is
same to mine, I think that it is a gain for the project in terms of diversity,
and it is not much of a loss for me as I hope that my contribution to Debian is
far more than just telling what I think on mailing lists.

I also make exceptions, and expect others to do so when it makes sense.  But in
many large discussions, there are a large number of messages that are not
particuarly urgent.  Not even judging on the contents, these messages are
driving out of the discussion a lot of people who will not have time to read a
dozens of them in a day.

To implement technical measures is quite far-fetched; I should probably have
not mentionned it.  But the social pressure is simply to see the main
contributors in term of achievements (not posting) moderating the pace of
discussions by the rythm of their answers.

(PS: still about reducing traffic, I am not sure if I would have answered
if your reply had not started by "this is nonsense"…)

PPS: Reducing the traffic is also why I refrain from posting "+1" messages.
Let's not making it important, otherwise, our mailboxes will explode.

Cheers,

-- 
Charles Plessy
Tsurumi, Kanagawa, Japan


-- 
To UNSUBSCRIBE, email to debian-project-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20131028002048.gb7...@falafel.plessy.net

Re: Should mailing list bans be published?

[Russ Allbery]

Christoph Anton Mitterer  writes:

> Perhaps one should think whether such publishing might have legal
> consequences...

> In some countries (like the US) it seems not be so uncommon to publicly
> name offenders or criminals on webpages... in Europe though, you might
> get into legal troubles.

In the United States, fact is an absolute defense against libel, so a web
page or email message stating simply that "this email address has been
banned from this mailing list" or "the author of the message at URL has
been banned from this mailing list" is not actionable.  I don't know to
what extent this applies in other countries, however.

Regardless of the legalities, I think that's the sort of notification that
we should present.  There's not much gained in trying to characterize the
reasons for the ban beyond pointing to an applicable sample, and quite a
bit to be lost (particularly since it's an invitation to continue the
argument via other means).

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


-- 
To UNSUBSCRIBE, email to debian-project-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/87eh76mi7t@windlord.stanford.edu

Re: Should mailing list bans be published?

[Christoph Anton Mitterer]

Hi.

Perhaps one should think whether such publishing might have legal
consequences...

In some countries (like the US) it seems not be so uncommon to publicly
name offenders or criminals on webpages... in Europe though, you might
get into legal troubles.


Cheers,
Chris.


smime.p7s
Description: S/MIME cryptographic signature

Re: Should mailing list bans be published?

[Florian Weimer]

* Joey Hess:

> Simply obfuscating the name on the list of banned users (or not posting
> any names at all, only links to the posts that led to the ban) would
> eliminate most reputational damage. Ie, random searches for that
> person would not turn up a high pagerank debian.org page listing their
> youthful indiscretions.
>
> Using eg "J. Hess" would probably be fine in most cases.

I recommend to use a web page, and not announce bans on public mailing
lists because such announcements invite subsequent discussion, likely
decloaking the banned poster.

We could also add metatags/robots.txt exclusions to prevent indexing
of the page.


-- 
To UNSUBSCRIBE, email to debian-project-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/87iowipfex@mid.deneb.enyo.de

Re: Should mailing list bans be published?

[Stefano Zacchiroli]

On Sat, Oct 26, 2013 at 10:46:41AM -0700, Steve Langasek wrote:
> I think we should publish them, for several reasons:

I'm in favor of this.

There is clearly a trade-off between the interests of the individual
being banned and those of the community which suffers the consequences
of not knowing about bans, for the reasons you've mentioned. As the
community is the most important value of any volunteer project, and also
considering how conservative we are in banning people, I've no doubt
where to place the cue in that trade-off.

This is also why I do not think that disclosing bans to -private is good
enough. Non-DDs (or simply DDs who don't read every -private mail) would
have no way of knowing who is banned, and therefore might be tricked
into thinking that specific bad behavior patterns visible in the mailing
list archives have been tolerated.

Once the principle of disclosing ban is established, we should adopt
various clever communication techniques to minimize reputation damages
to the involved individuals. In that respect, both Joey's and Enrico's
suggestions in this thread seem very sane to me.

Thanks for raising this topic, Steve.
Cheers.
-- 
Stefano Zacchiroli  . . . . . . .  z...@upsilon.cc . . . . o . . . o . o
Maître de conférences . . . . . http://upsilon.cc/zack . . . o . . . o o
Former Debian Project Leader  . . @zack on identi.ca . . o o o . . . o .
« the first rule of tautology club is the first rule of tautology club »


signature.asc
Description: Digital signature

Re: Should mailing list bans be published?

[Russ Allbery]

Steve Langasek  writes:

> This also seems like a good compromise to me.  Do the other folks who
> object to publishing information that could damage the poster's
> reputation (e.g., Bart, Ingo) think this is ok?

The problem that I have with publicly posting mailing list bans is that I
think it constitutes picking a fight.  This is backed up by some of the
messages that have been posted in defense of the idea (they destroyed
their own reputation, it's worthwhile to make an example of them, etc.).

I would instead take a step back and ask what the goal here is.

I agree with Steve's point that it's useful to know when someone has been
banned so that one can stop worrying about their messages.  There is a
drawback to publishing that information privately in that other public
participants in the newsgroup won't have access to that information.  We
also have a general rule that says that we always publish everything in
public unless there's some compelling reason not to, and that's valuable.

But, against that, I would say that the point of mailing list bans is not
to humiliate or expose someone who behaved poorly, or even to call further
attention to their poor behavior.  Rather, the goal is to *stop* calling
attention to their behavior and to defend the usability of the list
against further interference.  Actions that tend to produce *more*
discussion seem directly contrary to that goal.  Ideally, everyone would
simply forget the banned person ever participated in the list and proceed
as if they had never said anything.

I like either the idea of publishing the bans on debian-private or
publishing pointers to specific mailing list messages that resulted in a
ban rather than mentioning the name or email address directly because I
think both are less likely to result in further discussion and further
attention drawn to the contributions that we're specifically trying to get
rid of.  That additional hop of having to follow a link to the message in
question will, I think, reduce the quantity of discussion.  Both still
allow oversight.

If we didn't have debian-private, I would be in favor of publicly posting
all bans, and that's what other technical mailing lists do.  There's no
other good way to let people know what happened and let them know that
they should stop responding to that person.  But we do have
debian-private, which reaches most of the contributors and (hopefully) all
of the core project contributors, without making it a public issue.

However, I will say that if the choice is between not posting the bans at
all and posting them publicly, I'd lean towards the latter.

I'm definitely open to changing my mind if the listmasters have a
different opinion, since they're the ones who will have to deal with most
of the fallout.

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


-- 
To UNSUBSCRIBE, email to debian-project-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/874n82o532@windlord.stanford.edu

Re: Should mailing list bans be published?

[Steve Langasek]

On Sat, Oct 26, 2013 at 05:27:25PM -0400, Joey Hess wrote:
> Bart Martens wrote:
> > I suggest we keep things civil, with respect for the persons involved.  It's
> > really not up to Debian to harm someone's reputation, and that could reflect
> > bad on Debian's reputation.

> > Approaches I could support :
> > - post the bans with reasons on debian-private
> > - or maintain a list of bans with reasons in a text file on a Debian machine
> >   where DDs can read this info.

> Simply obfuscating the name on the list of banned users (or not posting
> any names at all, only links to the posts that led to the ban) would
> eliminate most reputational damage. Ie, random searches for that
> person would not turn up a high pagerank debian.org page listing their
> youthful indiscretions.

> Using eg "J. Hess" would probably be fine in most cases.

This also seems like a good compromise to me.  Do the other folks who object
to publishing information that could damage the poster's reputation (e.g.,
Bart, Ingo) think this is ok?

-- 
Steve Langasek   Give me a lever long enough and a Free OS
Debian Developer   to set it on, and I can move the world.
Ubuntu Developerhttp://www.debian.org/
slanga...@ubuntu.com vor...@debian.org


signature.asc
Description: Digital signature

Re: Should mailing list bans be published?

[Alexander Wirt]

On Sun, 27 Oct 2013, Boris Pek wrote:

> Hi,
> 
> > What do the rest of you think?
> 
> +1 for publishing the facts of bans and their reasons in a public mailing 
> list.
> Only with one correction which have been well described by Rhonda:
> 
> >>   - It provides a reference point for newcomers to the Debian community to
> >> judge their actions by, to understand what kinds of things will get 
> >> them
> >> banned from participation (although I expect few of the people who need
> >> such guidance will actually take advantage of it...)
> >
> >  People who goe by judging what action might get them banned are usually
> > those who try to *abuse* the rules and avoid banning instead of those
> > who would really like to go well with the community.  Taking "advantage
> > of it" already transports somehow a feeling of it might get abused.
> 
> By the way, how many bans we are talking about? Just few digits for estimating
> the scale of problem... For example, how many people (or email addresses) were
> banned this year or last year? How many mailing lists were affected?
Maybe ~2-3 bans a year. Banning is usually the last action I/we take and we
only use it if we really have to. 

Alex
 


-- 
To UNSUBSCRIBE, email to debian-project-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20131027164642.ga15...@smithers.snow-crash.org

Re: Should mailing list bans be published?

[Boris Pek]

Hi,

> What do the rest of you think?

+1 for publishing the facts of bans and their reasons in a public mailing list.
Only with one correction which have been well described by Rhonda:

>>   - It provides a reference point for newcomers to the Debian community to
>> judge their actions by, to understand what kinds of things will get them
>> banned from participation (although I expect few of the people who need
>> such guidance will actually take advantage of it...)
>
>  People who goe by judging what action might get them banned are usually
> those who try to *abuse* the rules and avoid banning instead of those
> who would really like to go well with the community.  Taking "advantage
> of it" already transports somehow a feeling of it might get abused.

By the way, how many bans we are talking about? Just few digits for estimating
the scale of problem... For example, how many people (or email addresses) were
banned this year or last year? How many mailing lists were affected?

Best wishes,
Boris


-- 
To UNSUBSCRIBE, email to debian-project-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/82411382887...@web19g.yandex.ru

Re: Should mailing list bans be published?

[Ana Guerrero Lopez]

On Sun, Oct 27, 2013 at 08:54:30AM +0100, Enrico Zini wrote:
> I'm in favour of not posting names but links to the posts that led to
> the ban. That it solves the problem of google-shaming is welcome, but
> marginal to me. The most important thing to me in doing that, is that we
> make it clear that "we banned you from the list for what you did",
> instead of "we banned you from the list for what you are".
> 
> That subtle distinction in the message is very important to me, because
> it makes the distinction from a community where I'm judged for what I
> am, to a community where I'm judged for what I do.
> 
> In the former, the incentive is to do as little public activity as
> possible, because it might turn into stigma. Places that work in that
> way see very little involvement from me.
> 
> In the latter, I'm happy to participate and take my chances, because I
> know that if I screw up I can learn something from my mistakes, grow up
> and do better next time.

I like this idea.

Ana
 


-- 
To UNSUBSCRIBE, email to debian-project-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20131027112816.ga2...@pryan.ekaia.org

Re: Should mailing list bans be published?

[Martin Steigerwald]

Am Sonntag, 27. Oktober 2013, 08:54:30 schrieb Enrico Zini:
> On Sat, Oct 26, 2013 at 05:27:25PM -0400, Joey Hess wrote:
> > Simply obfuscating the name on the list of banned users (or not posting
> > any names at all, only links to the posts that led to the ban) would
> 
> I'm in favour of not posting names but links to the posts that led to
> the ban. That it solves the problem of google-shaming is welcome, but
> marginal to me. The most important thing to me in doing that, is that we
> make it clear that "we banned you from the list for what you did",
> instead of "we banned you from the list for what you are".
> 
> That subtle distinction in the message is very important to me, because
> it makes the distinction from a community where I'm judged for what I
> am, to a community where I'm judged for what I do.
> 
> In the former, the incentive is to do as little public activity as
> possible, because it might turn into stigma. Places that work in that
> way see very little involvement from me.
> 
> In the latter, I'm happy to participate and take my chances, because I
> know that if I screw up I can learn something from my mistakes, grow up
> and do better next time.

I like this.

Thanks,
-- 
Martin 'Helios' Steigerwald - http://www.Lichtvoll.de
GPG: 03B0 0D6C 0040 0710 4AFA  B82F 991B EAAC A599 84C7

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

Re: Should mailing list bans be published?

[Martin Steigerwald]

Am Sonntag, 27. Oktober 2013, 07:58:03 schrieb Lars Wirzenius:
> On Sun, Oct 27, 2013 at 09:00:20AM +0900, Charles Plessy wrote:
> > Le Sat, Oct 26, 2013 at 10:46:41AM -0700, Steve Langasek a écrit :
> > > What do the rest of you think?
> > 
> > Given how arbitrarly other bans have been proposed, I think that the
> > outcome should stay private unless the banned person wishes so.
> 
> I don't understand this at all. Are you saying Debian listmasters, who
> decide on bans, have been making arbitrary, and therefore badly
> justified, bans? In other words, are you saying that they have
> prevented, without sufficient justification, people from posting to
> Debian mailing lists? If that's what you're saying, could you give
> examples of this?
> 
> I doubt it has happened, but if it has, that's an excellent reason
> make decisions about banning people from lists public.
> 
> What's better?
> 
> a) Someone gets banned from the lists, and nobody is told they have
> been or at least nobody is told publically. The person might deduce it
> from the fact that their mails never show up in the list archives, or
> that people stop responding to them, but that's not always very
> likely. Other people probably won't realise it at all. It's nearly
> impossible to argue against a ban you don't agree with, or the reasons
> for the ban, if everything is kept secret.

I always thought that if a listmaster bans someone from a list, he notifies 
that someone via mail personally? Isn´t that the case?

So it is not just a decision between a and b or black and white.

> b) Someone gets banned from the lists, and an explanation of why this
> happened is posted in public. Anyone can review the decision, and if
> it seems inappropriate, action can be taken. There is no room for
> insinuating that the listmasters are abusing their powers. If the ban
> was, in fact, inappropriate, it can be overturned, and the banned
> person's reputation is cleared.
> 
> To me, b) is obviously the better choice.

Ciao,
-- 
Martin 'Helios' Steigerwald - http://www.Lichtvoll.de
GPG: 03B0 0D6C 0040 0710 4AFA  B82F 991B EAAC A599 84C7


--
To UNSUBSCRIBE, email to debian-project-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/2721239.niUskzxTTu@merkaba

Re: Should mailing list bans be published?

[Enrico Zini]

On Sat, Oct 26, 2013 at 05:27:25PM -0400, Joey Hess wrote:

> Simply obfuscating the name on the list of banned users (or not posting
> any names at all, only links to the posts that led to the ban) would

I'm in favour of not posting names but links to the posts that led to
the ban. That it solves the problem of google-shaming is welcome, but
marginal to me. The most important thing to me in doing that, is that we
make it clear that "we banned you from the list for what you did",
instead of "we banned you from the list for what you are".

That subtle distinction in the message is very important to me, because
it makes the distinction from a community where I'm judged for what I
am, to a community where I'm judged for what I do.

In the former, the incentive is to do as little public activity as
possible, because it might turn into stigma. Places that work in that
way see very little involvement from me.

In the latter, I'm happy to participate and take my chances, because I
know that if I screw up I can learn something from my mistakes, grow up
and do better next time.


Ciao,

Enrico

-- 
GPG key: 4096R/E7AD5568 2009-05-08 Enrico Zini 


signature.asc
Description: Digital signature

Re: Should mailing list bans be published?

[Lars Wirzenius]

On Sun, Oct 27, 2013 at 09:00:20AM +0900, Charles Plessy wrote:
> Le Sat, Oct 26, 2013 at 10:46:41AM -0700, Steve Langasek a écrit :
> > 
> > What do the rest of you think?
> 
> Given how arbitrarly other bans have been proposed, I think that the
> outcome should stay private unless the banned person wishes so.

I don't understand this at all. Are you saying Debian listmasters, who
decide on bans, have been making arbitrary, and therefore badly
justified, bans? In other words, are you saying that they have
prevented, without sufficient justification, people from posting to
Debian mailing lists? If that's what you're saying, could you give
examples of this?

I doubt it has happened, but if it has, that's an excellent reason
make decisions about banning people from lists public.

What's better?

a) Someone gets banned from the lists, and nobody is told they have
been or at least nobody is told publically. The person might deduce it
from the fact that their mails never show up in the list archives, or
that people stop responding to them, but that's not always very
likely. Other people probably won't realise it at all. It's nearly
impossible to argue against a ban you don't agree with, or the reasons
for the ban, if everything is kept secret.

b) Someone gets banned from the lists, and an explanation of why this
happened is posted in public. Anyone can review the decision, and if
it seems inappropriate, action can be taken. There is no room for
insinuating that the listmasters are abusing their powers. If the ban
was, in fact, inappropriate, it can be overturned, and the banned
person's reputation is cleared.

To me, b) is obviously the better choice.

PS. I realise you phrased your objection such that a literal
interpretation makes it be about the _proposed_ bans only. I
understand that even less. People can, and do, propose bans
willy-nilly regardless of how public the bans are.

-- 
http://www.cafepress.com/trunktees -- geeky funny T-shirts
http://gtdfh.branchable.com/ -- GTD for hackers


-- 
To UNSUBSCRIBE, email to debian-project-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20131027075803.GV4353@holywood

Re: Should mailing list bans be published?

[Gerfried Fuchs]

Hi!

 In general I agree, but one reason can fire back:

* Steve Langasek  [2013-10-26 19:46:41 CEST]:
>  - It provides a reference point for newcomers to the Debian community to
>judge their actions by, to understand what kinds of things will get them
>banned from participation (although I expect few of the people who need
>such guidance will actually take advantage of it...)

 People who goe by judging what action might get them banned are usually
those who try to *abuse* the rules and avoid banning instead of those
who would really like to go well with the community.  Taking "advantage
of it" already transports somehow a feeling of it might get abused.

 But yes, I totally agree with the other points.
Rhonda
-- 
Fühlst du dich mutlos, fass endlich Mut, los  |
Fühlst du dich hilflos, geh raus und hilf, los| Wir sind Helden
Fühlst du dich machtlos, geh raus und mach, los   | 23.55: Alles auf Anfang
Fühlst du dich haltlos, such Halt und lass los|


-- 
To UNSUBSCRIBE, email to debian-project-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20131027075754.ga30...@anguilla.debian.or.at

Re: Should mailing list bans be published?

[Alexander Wirt]

On Sun, 27 Oct 2013, Charles Plessy wrote:

> Le Sat, Oct 26, 2013 at 10:46:41AM -0700, Steve Langasek a écrit :
> > 
> > What do the rest of you think?
> 
> Given how arbitrarly other bans have been proposed, I think that the outcome
> should stay private unless the banned person wishes so.  This will also reduce
> the pressure on the listmasters, by reducing the consequences of giving 
> unequal
> treatments to people.  Why not making the list readable on a machine open to
> the Debian Developers only ?
> 
> In parallel, I think that we need some technical or social pressure for
> limiting to 1 or 2 messages a day each individual contribution to long 
> threads.
That is nonsense. There will always be people that have to write more mails
to a thread. For example the maintainer of a discussed software, dpl, or the
ctte member. And so on. Such a general limitation just won't work.

Alex


-- 
To UNSUBSCRIBE, email to debian-project-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20131027063556.gf8...@smithers.snow-crash.org

Re: Should mailing list bans be published?

[Steve McIntyre]

On Sat, Oct 26, 2013 at 10:46:41AM -0700, Steve Langasek wrote:
>
>So I don't think bans need to be posted anywhere prominent like
>debian-devel-announce, but I do think basic facts like who is banned, for
>how long, and the rationale (with links to specific mailing list posts as
>reference) should be made public.
>
>What do the rest of you think?

Sounds perfectly reasonable to me.

-- 
Steve McIntyre, Cambridge, UK.st...@einval.com
< sladen> I actually stayed in a hotel and arrived to find a post-it
  note stuck to the mini-bar saying "Paul: This fridge and
  fittings are the correct way around and do not need altering"


-- 
To UNSUBSCRIBE, email to debian-project-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20131027062914.gf11...@einval.com

Re: Should mailing list bans be published?

[Yaroslav Halchenko]

On Sat, 26 Oct 2013, Joey Hess wrote:

> Bart Martens wrote:
> > I suggest we keep things civil, with respect for the persons involved.  It's
> > really not up to Debian to harm someone's reputation, and that could reflect
> > bad on Debian's reputation.

> > Approaches I could support :
> > - post the bans with reasons on debian-private
> > - or maintain a list of bans with reasons in a text file on a Debian machine
> >   where DDs can read this info.

> Simply obfuscating the name on the list of banned users (or not posting
> any names at all, only links to the posts that led to the ban) would
> eliminate most reputational damage. Ie, random searches for that
> person would not turn up a high pagerank debian.org page listing their
> youthful indiscretions.

> Using eg "J. Hess" would probably be fine in most cases.

+1 for a elegance and gesture, but -1 for efficiency: F.Last is not
sufficiently obfuscated when Last name is not very common.   but
probably it would still be better indeed than publishing names publicly
complete unobfuscated.

-- 
Yaroslav O. Halchenko, Ph.D.
http://neuro.debian.net http://www.pymvpa.org http://www.fail2ban.org
Senior Research Associate, Psychological and Brain Sciences Dept.
Dartmouth College, 419 Moore Hall, Hinman Box 6207, Hanover, NH 03755
Phone: +1 (603) 646-9834   Fax: +1 (603) 646-1419
WWW:   http://www.linkedin.com/in/yarik


-- 
To UNSUBSCRIBE, email to debian-project-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20131027013949.gs18...@onerussian.com

Re: Should mailing list bans be published?

[Charles Plessy]

Le Sat, Oct 26, 2013 at 10:46:41AM -0700, Steve Langasek a écrit :
> 
> What do the rest of you think?

Given how arbitrarly other bans have been proposed, I think that the outcome
should stay private unless the banned person wishes so.  This will also reduce
the pressure on the listmasters, by reducing the consequences of giving unequal
treatments to people.  Why not making the list readable on a machine open to
the Debian Developers only ?

In parallel, I think that we need some technical or social pressure for
limiting to 1 or 2 messages a day each individual contribution to long threads.

Cheers,

-- 
Charles Plessy
Tsurumi, Kanagawa, Japan


-- 
To UNSUBSCRIBE, email to debian-project-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/2013102720.gb14...@falafel.plessy.net

Re: Should mailing list bans be published?

[Luca Filipozzi]

Bart wrote:
> > > The harm that could come to Debian's reputation is that Debian could be
> > > perceived as an organization that harms people's reputation by judging 
> > > them in
> > > public about their behavior on the mailing lists.

Steve replied:
> > Ok, thanks for explaining.  This isn't something that concerns me at all,
> > but I understand that it concerns you.

Paul agreed:
> Nor I. The fact of the matter is that forcing folks to think twice
> before posting complete garbage to the mailing lists is nothing but
> good. If we get the reputation for harming the reputations of folks
> who harass and abuse others, well, fine by me -- just don't troll the
> MLs.

I think that our reputation is harmed more by mailing list archives containing
argumentative / vitriolic emails than by a reasoned (!) declaration of a ban.

The public publication of the ban cuts both ways.  If the reasoning behind the
ban is sound, then it will enhance our reputation for all the reasons already
mentioned by Steve and others; if it is not, then our reputation is damaged,
and appropriately so.  Public publication keeps both discussion participants
AND listmasters in check.

-- 
Luca Filipozzi
http://www.crowdrise.com/SupportDebian


signature.asc
Description: Digital signature

Re: Should mailing list bans be published?

[Paul Tagliamonte]

On Sat, Oct 26, 2013 at 04:05:05PM -0700, Steve Langasek wrote:
> On Sat, Oct 26, 2013 at 10:08:42PM +, Bart Martens wrote:
> 
> > > > > What do the rest of you think?
> 
> > > > I suggest we keep things civil, with respect for the persons involved. 
> > > > It's really not up to Debian to harm someone's reputation, and that 
> > > > could
> > > > reflect bad on Debian's reputation.
> 
> > > I don't understand this argument.  What harm comes to Debian's reputation
> > > from showing publically that we do not tolerate abusive behavior on our
> > > mailing list?
> 
> > The harm that could come to Debian's reputation is that Debian could be
> > perceived as an organization that harms people's reputation by judging them 
> > in
> > public about their behavior on the mailing lists.
> 
> Ok, thanks for explaining.  This isn't something that concerns me at all,
> but I understand that it concerns you.

Nor I. The fact of the matter is that forcing folks to think twice
before posting complete garbage to the mailing lists is nothing but
good. If we get the reputation for harming the reputations of folks
who harass and abuse others, well, fine by me -- just don't troll the
MLs.

> > > > Approaches I could support :
> > > > - post the bans with reasons on debian-private
> > > > - or maintain a list of bans with reasons in a text file on a Debian 
> > > > machine
> > > >   where DDs can read this info.
> > > 
> > > I think posting this on debian-private is not as good as posting it
> > > publically, for some of the reasons mentioned in my original mail.  (E.g.,
> > > making it clear to outsiders that certain behavior will not be tolerated.)
> 
> > That can be made clear without harming individuals' reputations.
> 
> How do you think it can be made clear?  We do have a list code of conduct
> already (), but the rules
> are vague; past attempts to make them more explicit have foundered.  So
> while in theory there are other ways to make this clear, in practice it
> seems to be quite difficult.

Indeed. The "I know it when I see it" method isn't very good to explain
the rules; a list of behavior that caused bans would be pretty nice in
this regard.

> > > I don't think maintaining a list "somewhere" is sufficient; there should 
> > > be
> > > some notification to the project when the bans take place.
> 
> > I can imagine that some DDs prefer to receive notifications, which can be
> > obtained by simply using diff in crontab.
> 
> That would fail to provide any of the benefits outlined in my original mail.

+1 for publishing bans.

There's a line between privacy and transparency; and this isn't a
privacy issue (indeed, the lists are public) - bad behavior in public
almost warents discipline that's public, otherwise folks (such as
myself, who didn't even know there was a ban) might continue to think
that listmasters would turn a blind eye so such emails.


Happy to hear of the ban, happy to hear of this discussion; +1

Cheers,
  T

-- 
 .''`.  Paul Tagliamonte 
: :'  : Proud Debian Developer
`. `'`  4096R / 8F04 9AD8 2C92 066C 7352  D28A 7B58 5B30 807C 2A87
 `- http://people.debian.org/~paultag


signature.asc
Description: Digital signature

Re: Should mailing list bans be published?

[Steve Langasek]

On Sat, Oct 26, 2013 at 10:08:42PM +, Bart Martens wrote:

> > > > What do the rest of you think?

> > > I suggest we keep things civil, with respect for the persons involved. 
> > > It's really not up to Debian to harm someone's reputation, and that could
> > > reflect bad on Debian's reputation.

> > I don't understand this argument.  What harm comes to Debian's reputation
> > from showing publically that we do not tolerate abusive behavior on our
> > mailing list?

> The harm that could come to Debian's reputation is that Debian could be
> perceived as an organization that harms people's reputation by judging them in
> public about their behavior on the mailing lists.

Ok, thanks for explaining.  This isn't something that concerns me at all,
but I understand that it concerns you.

> > > Approaches I could support :
> > > - post the bans with reasons on debian-private
> > > - or maintain a list of bans with reasons in a text file on a Debian 
> > > machine
> > >   where DDs can read this info.
> > 
> > I think posting this on debian-private is not as good as posting it
> > publically, for some of the reasons mentioned in my original mail.  (E.g.,
> > making it clear to outsiders that certain behavior will not be tolerated.)

> That can be made clear without harming individuals' reputations.

How do you think it can be made clear?  We do have a list code of conduct
already (), but the rules
are vague; past attempts to make them more explicit have foundered.  So
while in theory there are other ways to make this clear, in practice it
seems to be quite difficult.

> > I don't think maintaining a list "somewhere" is sufficient; there should be
> > some notification to the project when the bans take place.

> I can imagine that some DDs prefer to receive notifications, which can be
> obtained by simply using diff in crontab.

That would fail to provide any of the benefits outlined in my original mail.

-- 
Steve Langasek   Give me a lever long enough and a Free OS
Debian Developer   to set it on, and I can move the world.
Ubuntu Developerhttp://www.debian.org/
slanga...@ubuntu.com vor...@debian.org


signature.asc
Description: Digital signature

Re: Should mailing list bans be published?

[Bart Martens]

On Sat, Oct 26, 2013 at 12:58:34PM -0700, Steve Langasek wrote:
> Hi Bart,
> 
> On Sat, Oct 26, 2013 at 07:33:34PM +, Bart Martens wrote:
> > On Sat, Oct 26, 2013 at 10:46:41AM -0700, Steve Langasek wrote:
> > > This led to a philosophical debate about whether bans should be made 
> > > public.
> > > Alexander expressed concern that having them published could be harmful 
> > > to a
> > > person's reputation, since employers will google your name and see that
> > > you've been banned from a large project such as Debian.
> 
> > I join Alexander on the above.
> 
> > > What do the rest of you think?
> 
> > I suggest we keep things civil, with respect for the persons involved. 
> > It's really not up to Debian to harm someone's reputation, and that could
> > reflect bad on Debian's reputation.
> 
> I don't understand this argument.  What harm comes to Debian's reputation
> from showing publically that we do not tolerate abusive behavior on our
> mailing list?

The harm that could come to Debian's reputation is that Debian could be
perceived as an organization that harms people's reputation by judging them in
public about their behavior on the mailing lists.

> > Approaches I could support :
> > - post the bans with reasons on debian-private
> > - or maintain a list of bans with reasons in a text file on a Debian machine
> >   where DDs can read this info.
> 
> I think posting this on debian-private is not as good as posting it
> publically, for some of the reasons mentioned in my original mail.  (E.g.,
> making it clear to outsiders that certain behavior will not be tolerated.)

That can be made clear without harming individuals' reputations.

> But it's a compromise I could support, if that's the consensus in the
> project.

I appreciate that you are open for this compromise.  Let's see if it becomes a
consensus.

> I don't think maintaining a list "somewhere" is sufficient; there should be
> some notification to the project when the bans take place.

I can imagine that some DDs prefer to receive notifications, which can be
obtained by simply using diff in crontab.

Regards,

Bart Martens


-- 
To UNSUBSCRIBE, email to debian-project-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20131026220842.ga26...@master.debian.org

Re: Should mailing list bans be published?

[Joey Hess]

Bart Martens wrote:
> I suggest we keep things civil, with respect for the persons involved.  It's
> really not up to Debian to harm someone's reputation, and that could reflect
> bad on Debian's reputation.
> 
> Approaches I could support :
> - post the bans with reasons on debian-private
> - or maintain a list of bans with reasons in a text file on a Debian machine
>   where DDs can read this info.

Simply obfuscating the name on the list of banned users (or not posting
any names at all, only links to the posts that led to the ban) would
eliminate most reputational damage. Ie, random searches for that
person would not turn up a high pagerank debian.org page listing their
youthful indiscretions.

Using eg "J. Hess" would probably be fine in most cases.

-- 
see shy jo, not banned yet


signature.asc
Description: Digital signature

Re: Should mailing list bans be published?

[davidson]

expanding on this point of the OP,

On Sat, Oct 26, 2013 at 10:46:41AM -0700, Steve Langasek wrote:

- It provides a reference point for newcomers to the Debian
  community to judge their actions by, to understand what kinds of
  things will get them banned from participation (although I expect
  few of the people who need such guidance will actually take
  advantage of it...)


and regarding the counterproposal below,

On Sat, 26 Oct 2013, Bart Martens wrote:

On Sat, Oct 26, 2013 at 10:46:41AM -0700, Steve Langasek wrote:

What do the rest of you think?


I suggest we keep things civil, with respect for the persons involved.  It's
really not up to Debian to harm someone's reputation, and that could reflect
bad on Debian's reputation.

Approaches I could support :
- post the bans with reasons on debian-private
- or maintain a list of bans with reasons in a text file on a Debian machine
 where DDs can read this info.


if bans and their rationales are not posted publicly, potential
ban-candidates cannot read them, and cannot learn from others'
experiences.

so withholding this information from ban-candidates weakens their
ability to make informed decisions, regarding what is out of bounds on
debian's mailing lists.

and so, withholding this information from ban-candidates *decreases*
their own share of responsibility for getting banned, when they do.

and, arguably, it is the withholder of that information (i.e. debian)
which shoulders that relieved responsibility, instead.

i have no argument against the proposition that it is not up to debian
to manage others' reputations.  but making bans and their rationales
public, as proposed by OP, might make it easier for a ban-candidate to
manage their own.

thinking out loud,
wes


--
To UNSUBSCRIBE, email to debian-project-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/alpine.deb.2.02.1310261607290.25...@brutus.ling.ohio-state.edu

Re: Should mailing list bans be published?

[Yves-Alexis Perez]

On Sat, 2013-10-26 at 20:24 +, Bart Martens wrote:
> Cover up ? I did suggest approaches with full transparency among DDs.

I don't think that's the meaning of “public” Steve (And Lars) initially
thought about…

Regards,
-- 
Yves-Alexis


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

Re: Should mailing list bans be published?

[Bart Martens]

On Sat, Oct 26, 2013 at 09:20:27PM +0100, Ben Hutchings wrote:
> On Sat, 2013-10-26 at 19:33 +, Bart Martens wrote:
> > On Sat, Oct 26, 2013 at 10:46:41AM -0700, Steve Langasek wrote:
> > > This led to a philosophical debate about whether bans should be made 
> > > public.
> > > Alexander expressed concern that having them published could be harmful 
> > > to a
> > > person's reputation, since employers will google your name and see that
> > > you've been banned from a large project such as Debian.
> > 
> > I join Alexander on the above.
> > 
> > > What do the rest of you think?
> > 
> > I suggest we keep things civil, with respect for the persons involved.  It's
> > really not up to Debian to harm someone's reputation, and that could reflect
> > bad on Debian's reputation.
> [...]
> 
> This is the same argument used to cover up all kinds of abuses.  Maybe
> in the case of mailing list bans the infraction is minor enough that we
> should not make a public record of it, but I am very sceptical of the
> argument in general.

Cover up ? I did suggest approaches with full transparency among DDs.

Regards,

Bart Martens


-- 
To UNSUBSCRIBE, email to debian-project-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20131026202406.gb16...@master.debian.org

Re: Should mailing list bans be published?

[Ben Hutchings]

On Sat, 2013-10-26 at 10:46 -0700, Steve Langasek wrote:
> Hi folks,
> 
> Was discussing with one of the listmasters (Alexander Wirt) on IRC today
> about mailing list bans, because it turns out that someone I was just about
> to ask the listmasters to ban from debian-devel had just been blocked in
> response to a request from someone else.
> 
> This led to a philosophical debate about whether bans should be made public.
> Alexander expressed concern that having them published could be harmful to a
> person's reputation, since employers will google your name and see that
> you've been banned from a large project such as Debian.
> 
> I think we should publish them, for several reasons:
[...]

I agree with your reasons.

Ben.

-- 
Ben Hutchings
Editing code like this is akin to sticking plasters on the bleeding stump
of a severed limb. - me, 29 June 1999


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

Re: Should mailing list bans be published?

[Ben Hutchings]

On Sat, 2013-10-26 at 19:33 +, Bart Martens wrote:
> On Sat, Oct 26, 2013 at 10:46:41AM -0700, Steve Langasek wrote:
> > This led to a philosophical debate about whether bans should be made public.
> > Alexander expressed concern that having them published could be harmful to a
> > person's reputation, since employers will google your name and see that
> > you've been banned from a large project such as Debian.
> 
> I join Alexander on the above.
> 
> > What do the rest of you think?
> 
> I suggest we keep things civil, with respect for the persons involved.  It's
> really not up to Debian to harm someone's reputation, and that could reflect
> bad on Debian's reputation.
[...]

This is the same argument used to cover up all kinds of abuses.  Maybe
in the case of mailing list bans the infraction is minor enough that we
should not make a public record of it, but I am very sceptical of the
argument in general.

Ben.

-- 
Ben Hutchings
Editing code like this is akin to sticking plasters on the bleeding stump
of a severed limb. - me, 29 June 1999


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

Re: Should mailing list bans be published?

[Steve Langasek]

Hi Bart,

On Sat, Oct 26, 2013 at 07:33:34PM +, Bart Martens wrote:
> On Sat, Oct 26, 2013 at 10:46:41AM -0700, Steve Langasek wrote:
> > This led to a philosophical debate about whether bans should be made public.
> > Alexander expressed concern that having them published could be harmful to a
> > person's reputation, since employers will google your name and see that
> > you've been banned from a large project such as Debian.

> I join Alexander on the above.

> > What do the rest of you think?

> I suggest we keep things civil, with respect for the persons involved. 
> It's really not up to Debian to harm someone's reputation, and that could
> reflect bad on Debian's reputation.

I don't understand this argument.  What harm comes to Debian's reputation
from showing publically that we do not tolerate abusive behavior on our
mailing list?

If the world knowing about a ban would harm our reputation, then maybe we
should pause to think whether the ban itself is correct.  But I see no harm
to Debian's reputation from banning people for the kinds of mailing list
behavior that they are actually getting banned for.  It can only improve
Debian's current bad reputation for having a take-no-prisoners mailing list
culture!

As for "respect for the persons involved": I don't believe the project owes
anything to someone who can't behave with the minimum of civility required
on our mailing lists to avoid being banned.  We should be guided by what's
best for the Debian project, not worry about hurting the feelings of someone
whose behavior is so far beyond the pale that we find it necessary to
ostracize them.

> Approaches I could support :
> - post the bans with reasons on debian-private
> - or maintain a list of bans with reasons in a text file on a Debian machine
>   where DDs can read this info.

I think posting this on debian-private is not as good as posting it
publically, for some of the reasons mentioned in my original mail.  (E.g.,
making it clear to outsiders that certain behavior will not be tolerated.)
But it's a compromise I could support, if that's the consensus in the
project.

I don't think maintaining a list "somewhere" is sufficient; there should be
some notification to the project when the bans take place.

Thanks,
-- 
Steve Langasek   Give me a lever long enough and a Free OS
Debian Developer   to set it on, and I can move the world.
Ubuntu Developerhttp://www.debian.org/
slanga...@ubuntu.com vor...@debian.org


signature.asc
Description: Digital signature

Re: Should mailing list bans be published?

[Russ Allbery]

Bart Martens  writes:

> Approaches I could support :
> - post the bans with reasons on debian-private

+1.  I think this provides most of the benefits that Steve names (albeit
in a reduced form) and allows oversight without getting into a public
fight with that person.  (Or getting into weird issues in countries with
aggressive libel protection, etc.)

This seems like exactly the right use of debian-private to me.

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


-- 
To UNSUBSCRIBE, email to debian-project-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/87r4b7yfz5@windlord.stanford.edu

Re: Should mailing list bans be published?

[Bart Martens]

On Sat, Oct 26, 2013 at 10:46:41AM -0700, Steve Langasek wrote:
> This led to a philosophical debate about whether bans should be made public.
> Alexander expressed concern that having them published could be harmful to a
> person's reputation, since employers will google your name and see that
> you've been banned from a large project such as Debian.

I join Alexander on the above.

> What do the rest of you think?

I suggest we keep things civil, with respect for the persons involved.  It's
really not up to Debian to harm someone's reputation, and that could reflect
bad on Debian's reputation.

Approaches I could support :
- post the bans with reasons on debian-private
- or maintain a list of bans with reasons in a text file on a Debian machine
  where DDs can read this info.

Regards,

Bart Martens


-- 
To UNSUBSCRIBE, email to debian-project-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20131026193334.ga1...@master.debian.org

Re: Should mailing list bans be published?

[Steve Langasek]

On Sat, Oct 26, 2013 at 08:56:59PM +0200, Ingo Jürgensmann wrote:
> Am 26.10.2013 um 19:46 schrieb Steve Langasek :

> > This led to a philosophical debate about whether bans should be made public.
> > Alexander expressed concern that having them published could be harmful to a
> > person's reputation, since employers will google your name and see that
> > you've been banned from a large project such as Debian.

> I agree with Alexanders concern here.

I understand Alexander's concern, but why should it outweigh all of the
benefits to the project that I listed?  Particularly given that anyone can
already see the behavior that got them banned, why is it bad to also
disclose publically that they've been banned?

> Publishing other peoples personal data without prior allowance might even
> violate privacy legislation in some countries.

That's a red herring.  They've already posted their name and email address
publically by engaging in the behavior that got them banned in the first
place.  Posting that same name and email address with a statement that
they've been banned is not a violation of "privacy" under any reasonable
interpretation.

-- 
Steve Langasek   Give me a lever long enough and a Free OS
Debian Developer   to set it on, and I can move the world.
Ubuntu Developerhttp://www.debian.org/
slanga...@ubuntu.com vor...@debian.org


signature.asc
Description: Digital signature

Re: Should mailing list bans be published?

[Jose Luis Rivas]

On 10/26/2013 02:26 PM, Ingo Jürgensmann wrote:
> Am 26.10.2013 um 19:46 schrieb Steve Langasek :
> 
>> This led to a philosophical debate about whether bans should be made public.
>> Alexander expressed concern that having them published could be harmful to a
>> person's reputation, since employers will google your name and see that
>> you've been banned from a large project such as Debian.
> 
> 
> I agree with Alexanders concern here. Publishing other peoples personal data 
> without prior allowance might even violate privacy legislation in some 
> countries. 
> 

It's data that's already being published by themselves when they publish
anything into the mailing lists. Their name and email. This will not
violate any law.

And making it public is just following Debian's Social Contract: not
hidding our problems.

Transparency in all our procedures is crucial for a project like Debian
and that's why I support Steve's request.

This will allow to know why people gets banned too, so we can make the
administrator accountable for their decisions.

I think this is a win-win situation.

Kind Regards.
-- 
The Debian Project - http://debian.org/
Jose Luis Rivas - http://joseluisrivas.net/#ghostbar



signature.asc
Description: OpenPGP digital signature

Re: Should mailing list bans be published?

[Ingo Jürgensmann]

Am 26.10.2013 um 19:46 schrieb Steve Langasek :

> This led to a philosophical debate about whether bans should be made public.
> Alexander expressed concern that having them published could be harmful to a
> person's reputation, since employers will google your name and see that
> you've been banned from a large project such as Debian.


I agree with Alexanders concern here. Publishing other peoples personal data 
without prior allowance might even violate privacy legislation in some 
countries. 

-- 
Ciao...//  Fon: 0381-2744150
  Ingo   \X/   http://blog.windfluechter.net


gpg pubkey:  http://www.juergensmann.de/ij_public_key.asc



signature.asc
Description: Message signed with OpenPGP using GPGMail

Re: Should mailing list bans be published?

[Luca Filipozzi]

On Sat, Oct 26, 2013 at 10:46:41AM -0700, Steve Langasek wrote:
> Hi folks,
> 
> Was discussing with one of the listmasters (Alexander Wirt) on IRC today
> about mailing list bans, because it turns out that someone I was just about
> to ask the listmasters to ban from debian-devel had just been blocked in
> response to a request from someone else.
> 
> This led to a philosophical debate about whether bans should be made public.
> Alexander expressed concern that having them published could be harmful to a
> person's reputation, since employers will google your name and see that
> you've been banned from a large project such as Debian.
> 
> I think we should publish them, for several reasons:
> 
>  - Debian is not responsible for the reputation of someone who has gotten
>themselves banned for their behavior; their reputation is already in the
>mud if employers read their actual posts to the Debian lists.
>  - It brings closure to the rest of our community to know that action has
>been taken against an abuser, showing that we've stood up for the
>principle of civil discourse and that the problem hasn't just gone away
>on its own because a troll got bored.
>  - It gives Debian contributors confidence that bad behavior doesn't have to
>be silently endured as a cost of participating in Debian lists.
>  - It improves *Debian's* reputation to the rest of the world, by showing
>that our mailing lists are not "anything goes".
>  - It provides a reference point for newcomers to the Debian community to
>judge their actions by, to understand what kinds of things will get them
>banned from participation (although I expect few of the people who need
>such guidance will actually take advantage of it...)
>  - It casts sunlight on the kinds of decisions that the listmasters are
>making WRT bans, so that we collectively have oversight of these
>decisions and can ensure our principles are being applied fairly and
>consistently.
> 
> So I don't think bans need to be posted anywhere prominent like
> debian-devel-announce, but I do think basic facts like who is banned, for
> how long, and the rationale (with links to specific mailing list posts as
> reference) should be made public.
> 
> What do the rest of you think?

The counterargument would be that disclosing our reasons for a ban might show
us as capricious ... which is yet another reason to publish the bans so that we
are also held to account for our decisions.

If the above is unclear: I'm in favour of publishing our decisions to ban.

-- 
Luca Filipozzi
http://www.crowdrise.com/SupportDebian


-- 
To UNSUBSCRIBE, email to debian-project-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20131026180913.ga27...@emyr.net

Re: Should mailing list bans be published?

[Lars Wirzenius]

On Sat, Oct 26, 2013 at 10:46:41AM -0700, Steve Langasek wrote:
> This led to a philosophical debate about whether bans should be made public.
> Alexander expressed concern that having them published could be harmful to a
> person's reputation, since employers will google your name and see that
> you've been banned from a large project such as Debian.
> 
> I think we should publish them, for several reasons:
...
> What do the rest of you think?

I agree with you, Steve, on this issue, and for the reasons you
express.

-- 
http://www.cafepress.com/trunktees -- geeky funny T-shirts
http://gtdfh.branchable.com/ -- GTD for hackers


-- 
To UNSUBSCRIBE, email to debian-project-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20131026175724.GS4353@holywood

Re: Should mailing list bans be published?

[Cyril Brulebois]

Hi Steve,

Steve Langasek  (2013-10-26):
> Was discussing with one of the listmasters (Alexander Wirt) on IRC today
> about mailing list bans, because it turns out that someone I was just about
> to ask the listmasters to ban from debian-devel had just been blocked in
> response to a request from someone else.
> 
> This led to a philosophical debate about whether bans should be made public.
> Alexander expressed concern that having them published could be harmful to a
> person's reputation, since employers will google your name and see that
> you've been banned from a large project such as Debian.
> 
> I think we should publish them, for several reasons:
> 
>  - Debian is not responsible for the reputation of someone who has gotten
>themselves banned for their behavior; their reputation is already in the
>mud if employers read their actual posts to the Debian lists.
>  - It brings closure to the rest of our community to know that action has
>been taken against an abuser, showing that we've stood up for the
>principle of civil discourse and that the problem hasn't just gone away
>on its own because a troll got bored.
>  - It gives Debian contributors confidence that bad behavior doesn't have to
>be silently endured as a cost of participating in Debian lists.
>  - It improves *Debian's* reputation to the rest of the world, by showing
>that our mailing lists are not "anything goes".
>  - It provides a reference point for newcomers to the Debian community to
>judge their actions by, to understand what kinds of things will get them
>banned from participation (although I expect few of the people who need
>such guidance will actually take advantage of it...)
>  - It casts sunlight on the kinds of decisions that the listmasters are
>making WRT bans, so that we collectively have oversight of these
>decisions and can ensure our principles are being applied fairly and
>consistently.
> 
> So I don't think bans need to be posted anywhere prominent like
> debian-devel-announce, but I do think basic facts like who is banned, for
> how long, and the rationale (with links to specific mailing list posts as
> reference) should be made public.
> 
> What do the rest of you think?

looks very reasonable to me, thanks for writing that down.

Mraw,
KiBi.


signature.asc
Description: Digital signature

Should mailing list bans be published?

[Steve Langasek]

Hi folks,

Was discussing with one of the listmasters (Alexander Wirt) on IRC today
about mailing list bans, because it turns out that someone I was just about
to ask the listmasters to ban from debian-devel had just been blocked in
response to a request from someone else.

This led to a philosophical debate about whether bans should be made public.
Alexander expressed concern that having them published could be harmful to a
person's reputation, since employers will google your name and see that
you've been banned from a large project such as Debian.

I think we should publish them, for several reasons:

 - Debian is not responsible for the reputation of someone who has gotten
   themselves banned for their behavior; their reputation is already in the
   mud if employers read their actual posts to the Debian lists.
 - It brings closure to the rest of our community to know that action has
   been taken against an abuser, showing that we've stood up for the
   principle of civil discourse and that the problem hasn't just gone away
   on its own because a troll got bored.
 - It gives Debian contributors confidence that bad behavior doesn't have to
   be silently endured as a cost of participating in Debian lists.
 - It improves *Debian's* reputation to the rest of the world, by showing
   that our mailing lists are not "anything goes".
 - It provides a reference point for newcomers to the Debian community to
   judge their actions by, to understand what kinds of things will get them
   banned from participation (although I expect few of the people who need
   such guidance will actually take advantage of it...)
 - It casts sunlight on the kinds of decisions that the listmasters are
   making WRT bans, so that we collectively have oversight of these
   decisions and can ensure our principles are being applied fairly and
   consistently.

So I don't think bans need to be posted anywhere prominent like
debian-devel-announce, but I do think basic facts like who is banned, for
how long, and the rationale (with links to specific mailing list posts as
reference) should be made public.

What do the rest of you think?

Cheers,
-- 
Steve Langasek   Give me a lever long enough and a Free OS
Debian Developer   to set it on, and I can move the world.
Ubuntu Developerhttp://www.debian.org/
slanga...@ubuntu.com vor...@debian.org


signature.asc
Description: Digital signature