Re: [gentoo-dev] Mailing list moderation and community openness

[Tom Wijsman]

Hey!!!


I'm not going to open that bug, read all these related mailing list 
discussions or waste time on whatever!


Instead, I think it's important some of you read this message:

I hope that you choose to stand still for some time, or even sit or lie 
down for once.
Take a deep breath and count to ten, then think about what the goal of 
Gentoo is and what your goal in this context is.


Don't let these goals confuse others into random directions, but make it 
clear to yourself and everyone what they are.
And with those thoughts, as well as second guesses; decide what you 
really want to do with it, for yourself and for others...


Live your life; live it together <3


P.S: Not responding to you in particular, I'm spending my last time to 
collectively answer multiple threads from now and history


On 3/20/2018 1:17 PM, Michael Palimaka wrote:

I see that in bug #650964[1] Council is pushing forward again with
implementing user whitelisting on this mailing list (ie. anyone that is
not "approved" will have their mail rejected).

Could someone please explain how this doesn't directly contradict the
core tenets of an open and inclusive community?

1: https://bugs.gentoo.org/650964

Re: [gentoo-dev] Mailing list moderation and community openness

[Rich Freeman]

On Mon, Mar 26, 2018 at 10:38 PM, kuzetsa  wrote:
>
> I think this may be a misunderstanding? no? there might be some mailing
> list jargon term: "moderation" which I was unaware of:
>

Historically moderation meant having list traffic held prior to
distribution for approval from a moderator.

> I've never used mailing list software which has that feature (I think
> that may be what you're referring to) - I mostly meant someone (or a
> team) with the specific duty to hold people accountable for their posts
> (since the list is public-facing, this should include @gentoo.org devs
> too because it sets a weird precedent to have disparate enforcement)

Well, ultimately the question is whether unverified members of the
community can post or not.  If they can, then there is no way to hold
anybody accountable for anything, because they can just create a new
email address to continue posting.

If you require verification prior to posting it gives everybody a
reputation to have to be concerned about.

> the "require whitelist / default deny" version of having a closed list
> seems the same - expecting users to contact a dev to relay messages, or
> go through the dubiously [un]documented process of getting whitelisted.

The process is simple, and certainly could be documented on the wiki
(it was already described in emails).  Get a dev to whitelist you.  It
can be any dev, and it is up to that dev to agree to the request or
not.

> unless that process has a standardized format, it seems worse than the
> greylist because individual developers have the autonomy to [not]
> sponsor people for whitelist, or approve posting on a user's behalf.

I'd consider that a feature, not a bug.  Gentoo has well over 100
developers.  All it takes is the approval of any one of them to be
whitelisted.  That is a very permissive system.  If every single one
of them is unwilling to whitelist somebody, is it really necessary to
have every single one of them make some kind of case for their
individual decisions?  Who would even judge such a case, considering
that all of comrel and the council (and even the current Trustees) are
all developers who presumably could have done the whitelisting
themselves?

You could still layer something like the proctors or comrel on top of
this, and they would presumably be a bit more formalized in how they
operate.  (The typical conception is that Proctors would have a lot of
discretion but would generally only enforce short-term "punishments"
like bans of a few days, warnings, and so on.  On the other hand
comrel would be much more formalized but would be able to take
long-term action.  The goal of course would be for Proctors to defuse
situations before they ever get to Comrel.)

-- 
Rich

Re: [gentoo-dev] Mailing list moderation and community openness

[kuzetsa]

On 03/26/2018 09:26 PM, Rich Freeman wrote:
> On Mon, Mar 26, 2018 at 9:19 PM, kuzetsa  wrote:
>> On 03/20/2018 08:08 PM, Rich Freeman wrote:
>>
>>>
>>> Actually, I think it is more of a technical constraint.  It is
>>> basically impossible to blacklist somebody on a mailing list, since
>>> all they need to do is roll up a new email address.
>>>
>>> I can think of various arguments for whitelisting or not whitelisting,
>>> but it seems silly to blacklist.
>>>
>>
>> require active stewardship (moderation, blacklisting, etc.)
>>
>> entry barriers to participation (default deny / require whitelist)
>>
>> if there are limitations on free speech, someone has to bear the burden.
>> for gentoo to have list moderation (blacklist approach) which isn't
>> dysfunctional, the main barrier to resources will be the human resources
>> end of things, not technical constraints. The tools themselves are easy
>> enough to use, but people who are willing and able to use them, and with
>> a clear guideline for how it needs done is a comrel issue which the
>> foundation needs to sort out.
>>
> 
> List moderation isn't the same as blacklisting.  With moderation
> you're effectively whitelisting because the first post anybody makes
> would be held for moderation, and depending on the approach you could
> moderate everything.
> 
> If you allowed new subscribers to post without being held for
> moderation, then the issues I spoke of would still apply, no matter
> how much manpower you threw at it.
> 

I think this may be a misunderstanding? no? there might be some mailing
list jargon term: "moderation" which I was unaware of:

I was more referring to how IRC chatrooms have an op, forums have
moderators which DO NOT screen individual posts, etc. I think I know of
the other version, and it might be analogous to the mechanism you meant?

for example: websites which hold back all comments which are posted
anonymously (non-trusted users) until a moderator can approve it.

I've never used mailing list software which has that feature (I think
that may be what you're referring to) - I mostly meant someone (or a
team) with the specific duty to hold people accountable for their posts
(since the list is public-facing, this should include @gentoo.org devs
too because it sets a weird precedent to have disparate enforcement)

specifically - I was referring to persons (staff) who are moderators.

(active stewardship to check for problems which need addressed)

I think the mechanism you describes sounds like some sort of greylist /
tiered version of default deny or something like that. Interesting.

the "require whitelist / default deny" version of having a closed list
seems the same - expecting users to contact a dev to relay messages, or
go through the dubiously [un]documented process of getting whitelisted.

unless that process has a standardized format, it seems worse than the
greylist because individual developers have the autonomy to [not]
sponsor people for whitelist, or approve posting on a user's behalf. the
lack of transparency for the process is a concern, I mean.

Re: [gentoo-dev] Mailing list moderation and community openness

[Rich Freeman]

On Mon, Mar 26, 2018 at 9:19 PM, kuzetsa  wrote:
> On 03/20/2018 08:08 PM, Rich Freeman wrote:
>
>>
>> Actually, I think it is more of a technical constraint.  It is
>> basically impossible to blacklist somebody on a mailing list, since
>> all they need to do is roll up a new email address.
>>
>> I can think of various arguments for whitelisting or not whitelisting,
>> but it seems silly to blacklist.
>>
>
> require active stewardship (moderation, blacklisting, etc.)
>
> entry barriers to participation (default deny / require whitelist)
>
> if there are limitations on free speech, someone has to bear the burden.
> for gentoo to have list moderation (blacklist approach) which isn't
> dysfunctional, the main barrier to resources will be the human resources
> end of things, not technical constraints. The tools themselves are easy
> enough to use, but people who are willing and able to use them, and with
> a clear guideline for how it needs done is a comrel issue which the
> foundation needs to sort out.
>

List moderation isn't the same as blacklisting.  With moderation
you're effectively whitelisting because the first post anybody makes
would be held for moderation, and depending on the approach you could
moderate everything.

If you allowed new subscribers to post without being held for
moderation, then the issues I spoke of would still apply, no matter
how much manpower you threw at it.

-- 
Rich

Re: [gentoo-dev] Mailing list moderation and community openness

[kuzetsa]

On 03/20/2018 08:08 PM, Rich Freeman wrote:
> On Tue, Mar 20, 2018 at 7:54 PM, Benda Xu  wrote:
>> William Hubbs  writes:
>>
>>> I do feel that this decision reflects badly on us as a community and
>>> should be reversed immediately. The proper way to deal with people who
>>> have bad behavior is to deal with them individually and not put a
>>> restriction on the community that is not necessary.
>>
>> I agree with William.  Dealing with individuals makes more sense.
>>
>> It boils down to an attitude of assuming outsiders are good (blacklist
>> to ML) or bad (whitelist to ML) by default.
> 
> Actually, I think it is more of a technical constraint.  It is
> basically impossible to blacklist somebody on a mailing list, since
> all they need to do is roll up a new email address.
> 
> I can think of various arguments for whitelisting or not whitelisting,
> but it seems silly to blacklist.
> 

require active stewardship (moderation, blacklisting, etc.)

entry barriers to participation (default deny / require whitelist)

if there are limitations on free speech, someone has to bear the burden.
for gentoo to have list moderation (blacklist approach) which isn't
dysfunctional, the main barrier to resources will be the human resources
end of things, not technical constraints. The tools themselves are easy
enough to use, but people who are willing and able to use them, and with
a clear guideline for how it needs done is a comrel issue which the
foundation needs to sort out.

Re: [gentoo-dev] Mailing list moderation and community openness

[Kristian Fiskerstrand]

On 03/22/2018 12:38 PM, Rich Freeman wrote:
> On Thu, Mar 22, 2018 at 4:30 AM, Alexander Berntsen  
> wrote:
>> On 22/03/18 07:31, Benda Xu wrote:
>>> We might be able to require GPG signed email to make a post.
>> Almost definitely.
>>
>> But before bikeshedding that, it would be advisable to find out whether
>> it would be a good idea in the first place. Unless you want only
>> prospective developers to be able to contribute to the ML (maybe you do
>> want that?), it seems like a poor idea to unnecessarily exclude anyone
>> who doesn't care (nor want to care) about OpenPGP.
> 
> That, and getting yourself whitelisted by a dev is gong to be a lower
> barrier than having to meet one in-person to have a key signed.  That
> is unless devs just start signing keys for people they've never met
> (which honestly doesn't really bother me much as I don't put much
> faith in the WoT anyway), in which case it turns into a whitelist that
> only comrel can un-whitelist since I don't think you can revoke a
> signature.

The one issuing the signature can also revoke it (see revsig in --edit-key).

That said, I'd rather focus on our own devs having WoT and requiring it
to become a developer long before we require it to be part of a mailing
list. If anything the technical complexity of verifying it doesn't make
much sense to me vs a simple whitelist.

> 
> Plus signing emails is a pain if you don't use an MUA that has this
> feature, and the web-based ones which do aren't very good.
> 


-- 
Kristian Fiskerstrand
OpenPGP keyblock reachable at hkp://pool.sks-keyservers.net
fpr:94CB AFDD 3034 5109 5618 35AA 0B7F 8B60 E3ED FAE3



signature.asc
Description: OpenPGP digital signature

Re: [gentoo-dev] Mailing list moderation and community openness

[Rich Freeman]

On Thu, Mar 22, 2018 at 4:30 AM, Alexander Berntsen  wrote:
> On 22/03/18 07:31, Benda Xu wrote:
>> We might be able to require GPG signed email to make a post.
> Almost definitely.
>
> But before bikeshedding that, it would be advisable to find out whether
> it would be a good idea in the first place. Unless you want only
> prospective developers to be able to contribute to the ML (maybe you do
> want that?), it seems like a poor idea to unnecessarily exclude anyone
> who doesn't care (nor want to care) about OpenPGP.

That, and getting yourself whitelisted by a dev is gong to be a lower
barrier than having to meet one in-person to have a key signed.  That
is unless devs just start signing keys for people they've never met
(which honestly doesn't really bother me much as I don't put much
faith in the WoT anyway), in which case it turns into a whitelist that
only comrel can un-whitelist since I don't think you can revoke a
signature.

Plus signing emails is a pain if you don't use an MUA that has this
feature, and the web-based ones which do aren't very good.

-- 
Rich

Re: [gentoo-dev] Mailing list moderation and community openness

[Alexander Berntsen]

On 22/03/18 07:31, Benda Xu wrote:
> We might be able to require GPG signed email to make a post.
Almost definitely.

But before bikeshedding that, it would be advisable to find out whether
it would be a good idea in the first place. Unless you want only
prospective developers to be able to contribute to the ML (maybe you do
want that?), it seems like a poor idea to unnecessarily exclude anyone
who doesn't care (nor want to care) about OpenPGP.
-- 
Alexander
berna...@gentoo.org
https://secure.plaimi.net/~alexander



signature.asc
Description: OpenPGP digital signature

Re: [gentoo-dev] Mailing list moderation and community openness

[Benda Xu]

Hi Rich,

Rich Freeman  writes:

> Actually, I think it is more of a technical constraint.  It is
> basically impossible to blacklist somebody on a mailing list, since
> all they need to do is roll up a new email address.

> I can think of various arguments for whitelisting or not whitelisting,
> but it seems silly to blacklist.

Okay, I see your argument.


Just a random bikeshedding.  We might be able to require GPG signed
email to make a post.  Then we can blacklist the GPG identity?

To think of it further, the web of trust is basically a whitelist.  But
it has the flexibility to chain the trust from a Gentoo dev by several
'hops'.

Benda

Re: [gentoo-dev] Mailing list moderation and community openness

[M. J. Everitt]

On 22/03/18 00:33, Kristian Fiskerstrand wrote:
> 
> Most contributions should happen via patches on b.g.o
>
Who was lamenting about the every-increasing bug queue on this Very list
recently?

And what about those 5+ year old bugs that are rotting for packages long
last-rited from the tree ?



signature.asc
Description: OpenPGP digital signature

Re: [gentoo-dev] Mailing list moderation and community openness

[Kristian Fiskerstrand]

On 03/22/2018 12:56 AM, Chí-Thanh Christopher Nguyễn wrote:
> Rich Freeman schrieb:
> 
>> Actually, I think it is more of a technical constraint.  It is
>> basically impossible to blacklist somebody on a mailing list, since
>> all they need to do is roll up a new email address.
>>
>> I can think of various arguments for whitelisting or not whitelisting,
>> but it seems silly to blacklist.
> 
> And how often did it actually happen that blacklisting was evaded on -dev
> mailing list?

we are aware of at least one case of evasion like this within the
relatively near past, but it is more a matter of principle as we don't
know if there are sock-puppets etc around.

The main things is really, the bar for whiltelisting is very low, you
just need a current dev to vouch for you, which is similiar to editbugs
and the #-dev channel on IRC. Most discussion should anyways happen on
-project ML, whereby the -dev ML is technical in nature. So there isn't
any real restriction being imposed here.

Most contributions should happen via patches on b.g.o

-- 
Kristian Fiskerstrand
OpenPGP keyblock reachable at hkp://pool.sks-keyservers.net
fpr:94CB AFDD 3034 5109 5618 35AA 0B7F 8B60 E3ED FAE3



signature.asc
Description: OpenPGP digital signature

Re: [gentoo-dev] Mailing list moderation and community openness

[Kristian Fiskerstrand]

On 03/22/2018 12:56 AM, Chí-Thanh Christopher Nguyễn wrote:
> Michael Palimaka schrieb:
>> I see that in bug #650964[1] Council is pushing forward again with
>> implementing user whitelisting on this mailing list (ie. anyone that is
>> not "approved" will have their mail rejected).
>>
>> Could someone please explain how this doesn't directly contradict the
>> core tenets of an open and inclusive community?
> 
> (I do not intend to single out your post, just replying to the thread in
> general here)
> 
> I would like to ask people to stay respectful in their disagreement towards
> the Council and their decision here. You might not agree with the decision,
> but the Council is an elected body and was given these powers by the
> developer community which they represent. Also I have no doubt that Council
> members who voted for -dev moderation are aware of the counter arguments and
> honestly expect a net positive effect from this.
> 
> If you dislike mailing list moderation, campaign and/or vote in the next
> period for candidates who want to reverse this decision.

+1 for this, and also note that the whitelisting approach allows for
those opposing to start a project for -dev ML moderation that is similar
to editbugs and proxy maintenance as we currently have in the project.


-- 
Kristian Fiskerstrand
OpenPGP keyblock reachable at hkp://pool.sks-keyservers.net
fpr:94CB AFDD 3034 5109 5618 35AA 0B7F 8B60 E3ED FAE3



signature.asc
Description: OpenPGP digital signature

Re: [gentoo-dev] Mailing list moderation and community openness

[Kristian Fiskerstrand]

On 03/22/2018 12:56 AM, Chí-Thanh Christopher Nguyễn wrote:
> Kristian Fiskerstrand schrieb:
> 
>> Switching to mailman might have some good merits on its own, but as I
>> understand it it isn't necessary for the proposal at hand, that can be
>> solved using access control lists in mlmmj-process?
> 
> I would advise caution that Council better not try to micro-manage Infra here.

By all means, infra should do what they think is right, but this
particular decision doesn't force infra to switch mailing list
infrastructure. If they believe there are reasons to do so, by all
means, but the decision doesn't result in

On 03/20/2018 04:28 PM, Matthew Thode wrote:
> There are still some issues with it infra side (archiving will still
> have to use the old system) and moving mailing lists is going to be fun,
> but them the breaks.


-- 
Kristian Fiskerstrand
OpenPGP keyblock reachable at hkp://pool.sks-keyservers.net
fpr:94CB AFDD 3034 5109 5618 35AA 0B7F 8B60 E3ED FAE3



signature.asc
Description: OpenPGP digital signature

Re: [gentoo-dev] Mailing list moderation and community openness

[Chí-Thanh Christopher Nguyễn]

Rich Freeman schrieb:

> Actually, I think it is more of a technical constraint.  It is
> basically impossible to blacklist somebody on a mailing list, since
> all they need to do is roll up a new email address.
> 
> I can think of various arguments for whitelisting or not whitelisting,
> but it seems silly to blacklist.

And how often did it actually happen that blacklisting was evaded on -dev
mailing list?


Best regards,
Chí-Thanh Christopher Nguyễn

Re: [gentoo-dev] Mailing list moderation and community openness

[Chí-Thanh Christopher Nguyễn]

Michael Palimaka schrieb:
> I see that in bug #650964[1] Council is pushing forward again with
> implementing user whitelisting on this mailing list (ie. anyone that is
> not "approved" will have their mail rejected).
> 
> Could someone please explain how this doesn't directly contradict the
> core tenets of an open and inclusive community?

(I do not intend to single out your post, just replying to the thread in
general here)

I would like to ask people to stay respectful in their disagreement towards
the Council and their decision here. You might not agree with the decision,
but the Council is an elected body and was given these powers by the
developer community which they represent. Also I have no doubt that Council
members who voted for -dev moderation are aware of the counter arguments and
honestly expect a net positive effect from this.

If you dislike mailing list moderation, campaign and/or vote in the next
period for candidates who want to reverse this decision.


Best regards,
Chí-Thanh Christopher Nguyễn

Re: [gentoo-dev] Mailing list moderation and community openness

[Chí-Thanh Christopher Nguyễn]

Kristian Fiskerstrand schrieb:

> Switching to mailman might have some good merits on its own, but as I
> understand it it isn't necessary for the proposal at hand, that can be
> solved using access control lists in mlmmj-process?

I would advise caution that Council better not try to micro-manage Infra here.


Best regards,
Chí-Thanh Christopher Nguyễn

Re: [gentoo-dev] Mailing list moderation and community openness

[Gregory Woodbury]

John Levine, author of "The Internet For Dummies," once set up a robo-moderation
process for the Usenet newsgroup soc.religion.unitarian-univ
(Unitarian Universalists).
The group, along with most of Usenet, ultimately "died" due to lack of
attention from
the moderators, who failed to curb one of their own.

However, the robo-moderator worked quite well, and still is
technically in-place. The
first post by a person generated an email to the poster to verify the
email addres,
and required the poster to reply with a confirmation. The posts then
went through
without anyone having to approve or whilelist things.  If a poster subsequently
became a "problem" their postings could be placed in a moderation
required status,
and some human would evelute the posts and handle the quelling of off-topic or
flame generating posts. In extreme cases, posters could be banned for
varying periods
of time.

The programs where quite powerful, and amazingly simple and elegant to implent.
The source is available, and should be easily adapted for practically
any system with
bash shell hook capabilities. The infra team might want to look at
that code and try
something like it.  Some addresses can be injected at setup time requiring human
action before posts are approved (Rejected posts would be sent back to the perp
requesting re-writing or abandoning.

The moderators did not have to login anywhere to work with the bot,
all interactions
were done via email.  The system is/was quite nice, and my mangled memories
should not be the deciding factors when looking at it.

Such a system might well serve as a means of allowing fully free entry into the
list, while still providing the ability to control things if it gets
out of hand.

On Wed, Mar 21, 2018 at 1:19 PM, Rich Freeman  wrote:
> On Wed, Mar 21, 2018 at 12:55 PM, R0b0t1  wrote:
>>
>> I can't tell, and I suspect other people can't either.
>>
>
> This is the crux of the issue.  Decisions involving people issues are
> made behind closed doors, which means that others are not free to
> confirm for themselves whether those actions are correct.  This tends
> to lead to ongoing debate over whether those decisions were
> appropriate, with everybody arguing from their own knowledge, and the
> only ones who know the information used to make the decision are
> barred from talking about it.  This is basically a debate where
> participation is limited to the ignorant, at least as far as the
> particular details go (the general principles are debated by all).
>
> That said, even if the decisions were made in the open I wouldn't
> expect all to agree with them.
>
> Ultimately though there are pros and cons to making these kinds of
> decisions in the open, and there is not universal agreement regarding
> how these situations ought to be handled.  We can either fight about
> it until the end of time, or we can agree on some way to determine
> what approach we are going to take and then support it (perhaps
> begrudgingly).  Right now the mechanism that we have in place is the
> Council.  The only other mechanism I could see that would make any
> sense would be a referendum on the issue.  That gets unwieldy if we
> try to apply it to every little decision, but maybe for the big
> picture issues it would make sense.
>
> However, I think a lot of people would be surprised at the outcome.
> We all assume that we're all here for the same reasons, but as I
> commented on my blog Gentoo is a bit unique among distros and many of
> us are here for very different reasons, and have different priorities.
> Also, there is sometimes a tendency to assume that all FOSS projects
> work the same way.  When I was listening to a talk about how one of
> the BSDs dealt with these kinds of issues I was shocked to discover
> that much of their dev communications happens on completely closed
> lists (not just closed to posting, but to reading as well).  Gentoo
> has the gentoo-core list but it is very low traffic and it tends to be
> used for things like swapping cell phone numbers before conferences.
> When anything substantive comes up there are usually several people
> who chime in to rightly point out that this talk belongs on a public
> list.
>
> Bottom line is that there are a lot of different ways projects can
> run, and they all have their pros and cons.  A lot of the FOSS we
> depend on actually gets built or discussed behind closed doors.  I
> doubt many of us want Gentoo to go that far, but I suspect there is a
> lot of interest in taking smaller steps in that general direction.
>
> --
> Rich
>



-- 
G.Wolfe Woodbury
redwo...@gmail.com

Re: [gentoo-dev] Mailing list moderation and community openness

[Rich Freeman]

On Wed, Mar 21, 2018 at 12:55 PM, R0b0t1  wrote:
>
> I can't tell, and I suspect other people can't either.
>

This is the crux of the issue.  Decisions involving people issues are
made behind closed doors, which means that others are not free to
confirm for themselves whether those actions are correct.  This tends
to lead to ongoing debate over whether those decisions were
appropriate, with everybody arguing from their own knowledge, and the
only ones who know the information used to make the decision are
barred from talking about it.  This is basically a debate where
participation is limited to the ignorant, at least as far as the
particular details go (the general principles are debated by all).

That said, even if the decisions were made in the open I wouldn't
expect all to agree with them.

Ultimately though there are pros and cons to making these kinds of
decisions in the open, and there is not universal agreement regarding
how these situations ought to be handled.  We can either fight about
it until the end of time, or we can agree on some way to determine
what approach we are going to take and then support it (perhaps
begrudgingly).  Right now the mechanism that we have in place is the
Council.  The only other mechanism I could see that would make any
sense would be a referendum on the issue.  That gets unwieldy if we
try to apply it to every little decision, but maybe for the big
picture issues it would make sense.

However, I think a lot of people would be surprised at the outcome.
We all assume that we're all here for the same reasons, but as I
commented on my blog Gentoo is a bit unique among distros and many of
us are here for very different reasons, and have different priorities.
Also, there is sometimes a tendency to assume that all FOSS projects
work the same way.  When I was listening to a talk about how one of
the BSDs dealt with these kinds of issues I was shocked to discover
that much of their dev communications happens on completely closed
lists (not just closed to posting, but to reading as well).  Gentoo
has the gentoo-core list but it is very low traffic and it tends to be
used for things like swapping cell phone numbers before conferences.
When anything substantive comes up there are usually several people
who chime in to rightly point out that this talk belongs on a public
list.

Bottom line is that there are a lot of different ways projects can
run, and they all have their pros and cons.  A lot of the FOSS we
depend on actually gets built or discussed behind closed doors.  I
doubt many of us want Gentoo to go that far, but I suspect there is a
lot of interest in taking smaller steps in that general direction.

-- 
Rich

Re: [gentoo-dev] Mailing list moderation and community openness

[R0b0t1]

On Wed, Mar 21, 2018 at 9:44 AM, Alec Warner  wrote:
> The community has a 'toxic people problem'

Maybe certain people who feel they are being attacked are idiots and
don't like hearing it? I can't tell, and I suspect other people can't
either.

Respectfully,
 R0b0t1

Re: [gentoo-dev] Mailing list moderation and community openness

[Alec Warner]

On Wed, Mar 21, 2018 at 12:31 PM, Eray Aslan  wrote:

> On Wed, Mar 21, 2018 at 10:44:48AM -0400, Alec Warner wrote:
> > [1] Which isn't to say that I would accept 'orders' to commit crimes, or
> > other obviously bad things.
>
> This is the crux of the problem.  There are certain lines you will not
> cross.  I am saying that my line is different and by voicing that,
> hopefully, making you re-consider yours.
>
> > I'm again asserting that this idea is not
> > fundamentally bad. The community has a 'toxic people problem' and our
> > previous attempts at resolution have not really produced great results.
> > Will this also produce great results? Not sure. But willing to try it.
>
> Openness, transparency, inclusiveness.  Those are some pretty
> fundemental values.  Reconsider.  But if you decide to go ahead, I am
> not going to judge you.  You (or the council members who voted yes) are
> not bad persons.  Just somewhat different values - which is surprising
> in a sad way.
>

I think of my aim is just playing a longer field here. I've been a part of
Gentoo for a long time. I've considered leaving numerous
times for a variety of reasons; yet I remain.

I don't disagree that the issue is important, but leaving an organization
really changes the velocity and direction of influence one
can have on it. Traditionally I have not seen external contributors have a
strong influence in Gentoo; so leaving to me implies a
loss of influence. If my goal is to have a good outcome; I'm not convinced
leaving accomplishes it. If I leave, will the council change their mind?
Why would they?

Perhaps you think myself (and other developers) should do more and I think
that is a reasonable thing to advocate for; but I'm also fairly happy with
a timeline
of:

1) We add moderation in ~April.
2) Council election happens in summer (I expect something of a strong
reckoning here, in terms of council makeup.)
3) Council` repeals the previous decision and we undo the moderation[1].

I tend to like this approach because I feel like its how the organization
was designed to work. I think alternatives involve essentially
'protesting'. E.g. I could propose the council discuss this topic at every
meeting. I could try to use my developer-ship to force extra council
meetings (emergency meetings perhaps.) I could collect signatures. I'm
still not convinced these things would be vehicles for change though.

[1] There is of course the risk that this doesn't come about, either
because the same council is re-elected or because the new council chooses
not to repeal. But I accept this risk willingly.


> --
> Eray
>
>

Re: [gentoo-dev] Mailing list moderation and community openness

[Eray Aslan]

On Wed, Mar 21, 2018 at 10:44:48AM -0400, Alec Warner wrote:
> [1] Which isn't to say that I would accept 'orders' to commit crimes, or
> other obviously bad things.

This is the crux of the problem.  There are certain lines you will not
cross.  I am saying that my line is different and by voicing that,
hopefully, making you re-consider yours.

> I'm again asserting that this idea is not
> fundamentally bad. The community has a 'toxic people problem' and our
> previous attempts at resolution have not really produced great results.
> Will this also produce great results? Not sure. But willing to try it.

Openness, transparency, inclusiveness.  Those are some pretty
fundemental values.  Reconsider.  But if you decide to go ahead, I am
not going to judge you.  You (or the council members who voted yes) are
not bad persons.  Just somewhat different values - which is surprising
in a sad way.

-- 
Eray

Re: [gentoo-dev] Mailing list moderation and community openness

[Alec Warner]

On Wed, Mar 21, 2018 at 1:36 AM, Eray Aslan  wrote:

> On Tue, Mar 20, 2018 at 10:28:48AM -0500, Matthew Thode wrote:
> > While I personally do no agree with mailing list moderation infra has
> > been tasked with moving forward on it.
>
> You can always resign from infra.
>

> That was a somewhat tongue-in-cheek comment but not wholly.  You cant
> cop out by saying it was an order from council.  I understand if you
> dont but do consider it.  Fight the good fight.
>

So when there is conflict its pretty often that you have 3 options.

1) Accept
2) Leave
3) Escalate

I'm not sure 3 is possible (the council is already the highest body). I
also think that as a organization this is how we
arranged it to be. Speaking for myself, this is not the worst issue I've
seen in Gentoo and so I thing doing 2 is probably
not very effective. Its also likely I can only do 2 once (because maybe I
would not be welcome'd back or want to contribute anymore.)

That leaves 1 and one interests me for many reasons.

a) as noted earlier, decisions are not set in stone. Its possible we could
turn on this whitelisting solution for a brief period and the decision is
overturned at the next council meeting, or perhaps at the next council
election once the existing council is replaced.
b) I am never afraid of making mistakes. I too think this is a mistake; but
I don't think its a critical mistake for the organization. Maybe I'm wrong
though.
c) I have a selfish interest to migrate off of mmlmj because I have an
intense dislike (of the software) and I think we need a "modernized" list
setup. So this effort is a driver to get some infra work done.
d) Infra as a organization wields a lot of power in Gentoo and I think its
organizationally dangerous to wield that power in this way. For example, if
the entire infra team retired rather than implement this solution; or even
worse, refused to retire but just didn't implement it. Ultimately
Infrastructure is here to meet the needs of the distribution and if we are
not doing that then we have failed as an organization.[1]
e) In the past, infra *has* wielded its power in a fashion that had
negative impacts on the distribution (e.g. arbitrarily removing commit
rights for developers with no warning, process, or oversight). I think
there is an additional focus in the the Infra team to avoid that sort of
activity and "inaction is still action" and I think it results in similar
repercussions.

[1] Which isn't to say that I would accept 'orders' to commit crimes, or
other obviously bad things. I'm again asserting that this idea is not
fundamentally bad. The community has a 'toxic people problem' and our
previous attempts at resolution have not really produced great results.
Will this also produce great results? Not sure. But willing to try it.

-A

>
> --
> Eray
>
>

Re: [gentoo-dev] Mailing list moderation and community openness

[Rich Freeman]

On Wed, Mar 21, 2018 at 1:36 AM, Eray Aslan  wrote:
> On Tue, Mar 20, 2018 at 10:28:48AM -0500, Matthew Thode wrote:
>> While I personally do no agree with mailing list moderation infra has
>> been tasked with moving forward on it.
>
> That was a somewhat tongue-in-cheek comment but not wholly.  You cant
> cop out by saying it was an order from council.  I understand if you
> dont but do consider it.  Fight the good fight.

Interesting.  When exactly should we all start ignoring the Council,
and when should we do what they say?  And what is the likely result of
that?

For all the complaining of "cabals" in Gentoo it seems odd to suggest
putting the final decisions of the one group that is about the least
democratic in the organization.

(That isn't really intended as a criticism: there are a lot of
practical reasons why infra operates as it does and I've yet to come
up with any better approach.  With the council/trustees the authority
comes from the collective, and nobody would pay attention to a
directive that didn't have a majority backing or the appearance of due
process.  With any other project the decisions are appealable to
council.  With infra one guy with the root password can cause a lot of
havoc, and the computer isn't going to stop and question what they're
doing.  That creates a lot of incentive to minimize the number of
people who are trusted.  In any case, I think it makes the most sense
to do the decision-making in more open/democratic processes, and then
minimize the execution footprint that requires "cabals.")

As I've commented elsewhere [1] I think an issue here is that we just
don't have enough of a critical mass to be able to afford to split
along ideological lines.  The set of developers interested in a
source-based distro is barely sufficient to create a viable
source-based distro.  If you split it into the subsets who prefer open
vs closed mailing lists on top of this then the individual groups lack
critical mass.  And so we're forced to co-exist, and agree on one or
the other, or some kind of compromise.

1 - 
https://rich0gentoo.wordpress.com/2016/02/27/gentoo-ought-to-be-about-choice/

-- 
Rich

Re: [gentoo-dev] Mailing list moderation and community openness

[Eray Aslan]

On Tue, Mar 20, 2018 at 10:28:48AM -0500, Matthew Thode wrote:
> While I personally do no agree with mailing list moderation infra has
> been tasked with moving forward on it.

You can always resign from infra.

That was a somewhat tongue-in-cheek comment but not wholly.  You cant
cop out by saying it was an order from council.  I understand if you
dont but do consider it.  Fight the good fight.

-- 
Eray

Re: [gentoo-dev] Mailing list moderation and community openness

[Paweł Hajdan , Jr .]

On 20/03/2018 05:17, Michael Palimaka wrote:
> I see that in bug #650964[1] Council is pushing forward again with
> implementing user whitelisting on this mailing list (ie. anyone that is
> not "approved" will have their mail rejected).
> 
> Could someone please explain how this doesn't directly contradict the
> core tenets of an open and inclusive community?
> 
> 1: https://bugs.gentoo.org/650964

This is a controversial topic which continues to be rehashed.

I think it'd be good for people opposing it (I share at least some of
your concern) to make sure they read the following resources and suggest
the best means to keep our community a nice place.





Paweł



signature.asc
Description: OpenPGP digital signature

Re: [gentoo-dev] Mailing list moderation and community openness

[Rich Freeman]

On Tue, Mar 20, 2018 at 7:54 PM, Benda Xu  wrote:
> William Hubbs  writes:
>
>> I do feel that this decision reflects badly on us as a community and
>> should be reversed immediately. The proper way to deal with people who
>> have bad behavior is to deal with them individually and not put a
>> restriction on the community that is not necessary.
>
> I agree with William.  Dealing with individuals makes more sense.
>
> It boils down to an attitude of assuming outsiders are good (blacklist
> to ML) or bad (whitelist to ML) by default.

Actually, I think it is more of a technical constraint.  It is
basically impossible to blacklist somebody on a mailing list, since
all they need to do is roll up a new email address.

I can think of various arguments for whitelisting or not whitelisting,
but it seems silly to blacklist.

-- 
Rich

Re: [gentoo-dev] Mailing list moderation and community openness

[Benda Xu]

William Hubbs  writes:

> I do feel that this decision reflects badly on us as a community and
> should be reversed immediately. The proper way to deal with people who
> have bad behavior is to deal with them individually and not put a
> restriction on the community that is not necessary.

I agree with William.  Dealing with individuals makes more sense.

It boils down to an attitude of assuming outsiders are good (blacklist
to ML) or bad (whitelist to ML) by default.

Benda


signature.asc
Description: PGP signature

Re: [gentoo-dev] Mailing list moderation and community openness

[Kristian Fiskerstrand]

On 03/20/2018 04:28 PM, Matthew Thode wrote:
> On 18-03-20 23:17:52, Michael Palimaka wrote:
>> I see that in bug #650964[1] Council is pushing forward again with
>> implementing user whitelisting on this mailing list (ie. anyone that is
>> not "approved" will have their mail rejected).
>>
>> Could someone please explain how this doesn't directly contradict the
>> core tenets of an open and inclusive community?
>>
>> 1: https://bugs.gentoo.org/650964
>>
> While I personally do no agree with mailing list moderation infra has
> been tasked with moving forward on it.  In that vein, this is what we
> are proposing.
> 
> Install and configure mailman3/hyperkitty/postorius once they all
> support python3.  Specifically we wish to use docker-mailman for this so
> we can easilly redeploy this on diferent machines as needed.
> 
> mailman3 gives us two good things, it has support for moderation (for
> better or worse) and it handles senders using dmarc.
> 
> There are still some issues with it infra side (archiving will still
> have to use the old system) and moving mailing lists is going to be fun,
> but them the breaks.

Switching to mailman might have some good merits on its own, but as I
understand it it isn't necessary for the proposal at hand, that can be
solved using access control lists in mlmmj-process?

-- 
Kristian Fiskerstrand
OpenPGP keyblock reachable at hkp://pool.sks-keyservers.net
fpr:94CB AFDD 3034 5109 5618 35AA 0B7F 8B60 E3ED FAE3



signature.asc
Description: OpenPGP digital signature

Re: [gentoo-project] Re: [gentoo-dev] Mailing list moderation and community openness

[Rich Freeman]

On Tue, Mar 20, 2018 at 9:41 AM, Gregory Woodbury  wrote:
> On gentoo-dev list: k_f
> points out that this should have been talked about during previous
> discussion periods...
>
> It was discussed "to death" over and over, and many argued against it
> till they were blue in the face.

Indeed, it will probably still be discussed over and over up until the
point where those who disagree are either unable to post on the lists,
or told that it is off-topic and will result in them losing access to
post on the lists.

Seriously, everything that has been said today in this thread was said
in the last thread on this topic.  The whole reason we have GLEP 39 is
that there are simply topics that not everybody will agree on...

-- 
Rich

Re: [gentoo-dev] Mailing list moderation and community openness

[William Hubbs]

On Tue, Mar 20, 2018 at 04:44:26PM +0100, Alexander Berntsen wrote:
> On 20/03/18 13:17, Michael Palimaka wrote:
> > Could someone please explain how this doesn't directly contradict the
> > core tenets of an open and inclusive community?
> It's fairly simple to produce a justification of the decision. I can
> think of several ways of doing so. One is through an appeal to some
> notion of community health improvement from impeding toxic contributors.
> In this strategy, the argument would be something pertaining to how
> allowing these toxic posters free rein on the mailing list would
> contradict the core tenet of an open and inclusive community. There are
> several more ways to rationalise the decision.
> 
> But you won't buy into either of those purported vindications of this
> decision. (I won't either.) So don't bother requesting them. Another
> aimless (and thus endless) back and forth in Jackal language isn't
> likely to achieve anything worthwhile beyond what the initial exchange
> achieved.

As the council member who voted against this decision, I am going to
express my opinion, even though it will be unpopular with the majority of
the council and probably others as well.

I do feel that this decision reflects badly on us as a community and
should be reversed immediately. The proper way to deal with people who
have bad behavior is to deal with them individually and not put a
restriction on the community that is not necessary.

William



signature.asc
Description: Digital signature

Re: [gentoo-dev] Mailing list moderation and community openness

[Pengcheng Xu]

I can understand the need to reduce meaningless spams on the dev list,
but seems like general rejection of posts from non-developers would
distract the idea of this being an open mailing list: a list that one can’t
post to effectively decays to something like a bulletin board, and obviously
the developing process shouldn’t be kept in a showcase, which would greatly
discourage people who are not part of the dev team, yet still wanting to
get involved in the discussing, maybe even decision-making.

Pengcheng Xu
i...@jsteward.moe



> H30/03/20 20:17、Michael Palimaka のメール:
> 
> I see that in bug #650964[1] Council is pushing forward again with
> implementing user whitelisting on this mailing list (ie. anyone that is
> not "approved" will have their mail rejected).
> 
> Could someone please explain how this doesn't directly contradict the
> core tenets of an open and inclusive community?
> 
> 1: https://bugs.gentoo.org/650964
> 



signature.asc
Description: Message signed with OpenPGP

Re: [gentoo-dev] Mailing list moderation and community openness

[Alexander Berntsen]

On 20/03/18 13:17, Michael Palimaka wrote:
> Could someone please explain how this doesn't directly contradict the
> core tenets of an open and inclusive community?
It's fairly simple to produce a justification of the decision. I can
think of several ways of doing so. One is through an appeal to some
notion of community health improvement from impeding toxic contributors.
In this strategy, the argument would be something pertaining to how
allowing these toxic posters free rein on the mailing list would
contradict the core tenet of an open and inclusive community. There are
several more ways to rationalise the decision.

But you won't buy into either of those purported vindications of this
decision. (I won't either.) So don't bother requesting them. Another
aimless (and thus endless) back and forth in Jackal language isn't
likely to achieve anything worthwhile beyond what the initial exchange
achieved.
-- 
Alexander
berna...@gentoo.org
https://secure.plaimi.net/~alexander



signature.asc
Description: OpenPGP digital signature

Re: [gentoo-dev] Mailing list moderation and community openness

[Matthew Thode]

On 18-03-20 23:17:52, Michael Palimaka wrote:
> I see that in bug #650964[1] Council is pushing forward again with
> implementing user whitelisting on this mailing list (ie. anyone that is
> not "approved" will have their mail rejected).
> 
> Could someone please explain how this doesn't directly contradict the
> core tenets of an open and inclusive community?
> 
> 1: https://bugs.gentoo.org/650964
> 

While I personally do no agree with mailing list moderation infra has
been tasked with moving forward on it.  In that vein, this is what we
are proposing.

Install and configure mailman3/hyperkitty/postorius once they all
support python3.  Specifically we wish to use docker-mailman for this so
we can easilly redeploy this on diferent machines as needed.

mailman3 gives us two good things, it has support for moderation (for
better or worse) and it handles senders using dmarc.

There are still some issues with it infra side (archiving will still
have to use the old system) and moving mailing lists is going to be fun,
but them the breaks.

-- 
Matthew Thode (prometheanfire)


signature.asc
Description: PGP signature

Re: [gentoo-dev] Mailing list moderation and community openness

[Gregory Woodbury]

On gentoo-dev list: k_f
points out that this should have been talked about during previous
discussion periods...

It was discussed "to death" over and over, and many argued against it
till they were blue in the face.
Their concerns were ignored, and Gentoo lost a lot more of the "Free
and Open" reputation it theoretically
prides itself on.

-- 
G.Wolfe Woodbury
redwo...@gmail.com

Re: [gentoo-dev] Mailing list moderation and community openness

[Lars Wendler]

On Tue, 20 Mar 2018 23:17:52 +1100 Michael Palimaka wrote:

>I see that in bug #650964[1] Council is pushing forward again with
>implementing user whitelisting on this mailing list (ie. anyone that is
>not "approved" will have their mail rejected).
>
>Could someone please explain how this doesn't directly contradict the
>core tenets of an open and inclusive community?
>
>1: https://bugs.gentoo.org/650964
>

+1

This is ridiculous and council should be ashamed of this decision.

-- 
Lars Wendler
Gentoo package maintainer
GPG: 21CC CF02 4586 0A07 ED93  9F68 498F E765 960E 9B39


pgp8dhy4aanHN.pgp
Description: Digitale Signatur von OpenPGP

Re: [gentoo-dev] Mailing list moderation and community openness

[Kristian Fiskerstrand]

On 03/20/2018 01:17 PM, Michael Palimaka wrote:
> I see that in bug #650964[1] Council is pushing forward again with
> implementing user whitelisting on this mailing list (ie. anyone that is
> not "approved" will have their mail rejected).
> 
> Could someone please explain how this doesn't directly contradict the
> core tenets of an open and inclusive community?
> 
> 1: https://bugs.gentoo.org/650964
> 

The correct place to have pointed this out would have been during the
previous ML discussions, and in particular ahead of either of the two
council meetings on the matter where it was clearly put on the agenda.
The bug in question is just a technical matter of implementing a final
decision.

-- 
Kristian Fiskerstrand
OpenPGP keyblock reachable at hkp://pool.sks-keyservers.net
fpr:94CB AFDD 3034 5109 5618 35AA 0B7F 8B60 E3ED FAE3



signature.asc
Description: OpenPGP digital signature

[gentoo-dev] Mailing list moderation and community openness

[Michael Palimaka]

I see that in bug #650964[1] Council is pushing forward again with
implementing user whitelisting on this mailing list (ie. anyone that is
not "approved" will have their mail rejected).

Could someone please explain how this doesn't directly contradict the
core tenets of an open and inclusive community?

1: https://bugs.gentoo.org/650964