Re: Amendment to Proposed GR: Declassifying parts of -private of historical interest

[Ian Jackson]

Bart Martens writes ("Re: Amendment to Proposed GR: Declassifying parts of 
-private of historical interest"):
> For example, your two points quoted above could easily be included
> in a GR text using these phrases:
> 
> - "The scope is limited to messages posted on debian-private before
>   debian-project was introduced." (And I have no strong opinion on
>   whether this should be included.)
> 
> - "The consent of the original author of the message on debian-private is
>   required before declassification." (I think this should have been in.)

It is a shame that you didn't raise these concerns during the
extensive discussion phase, during which the wording of this proposal
was carefully considered by various people, and modified.  Even if you
hadn't got consensus for your suggested restrictions, they could have
been on the ballot, if you had enough supporters.

Personally I don't read the current GR text as an authorisation or
encouragement for listmasters to go and do something nonconsensual and
inappropriate with old messages.

Ian.

-- 
Ian JacksonThese opinions are my own.

If I emailed you from an address @fyvzl.net or @evade.org.uk, that is
a private address which bypasses my fierce spamfilter.

more GRs to come (Re: Current GRs clarification)

[Holger Levsen]

Hi Kurt,

thanks for these clarifications!

On Sun, Aug 07, 2016 at 11:06:03PM +0200, Debian Project Secretary - Kurt 
Roeckx wrote:
> GR about declassifying debian-private
> =
> 
> There are 2 options on the ballot:
> - Choice 1: Allow declassifying parts of debian-private
> - Choice 2: Further Discussion
> 
> If the first option wins, the GR of 2005 is repealed and replaced
> by this GR.  If the second option wins, nothing changes and the GR
> of 2005 will stay in effect.

I expressed my surprise about a missing third option ("depeal the GR
of 2005 and burry the idea of systematically declassifying
debian-private") on #debian-private and have learned there that this
seems to have been an oversight / others agree that there should have
been this third option.

So, I hereby announce that I'll propose another GR to "depeal the GR
of 2005 and burry the idea of systematically declassifying debian-private"
if *this* GR turns out to result in "further discussion".

(Because I think if choice 1 does *not* win, the project doesnt really
want further discussion but rather this idea to be burried for good.)

Obviously, if choice 1 does win, I will *not* propose a GR to overcome this.
But if choice 1 does *not* win, I don't think the projects want "further
discussion" but rather "choice 3".


If you reply, please respect the reply-to: headers.

-- 
cheers,
Holger


signature.asc
Description: Digital signature

Re: more GRs to come (Re: Current GRs clarification)

[John Paul Adrian Glaubitz]

On 08/08/2016 03:34 PM, Holger Levsen wrote:
> I expressed my surprise about a missing third option ("depeal the GR
> of 2005 and burry the idea of systematically declassifying
> debian-private") on #debian-private and have learned there that this
> seems to have been an oversight / others agree that there should have
> been this third option.

I agree. I found it rather alienating that we are having a vote which basically
has the options "Yes" and "Continue arguing" but not a "No" which is I think
is as much as a legitimate answer as "Yes" is. I feel the same about the other,
currently pending GR, independent of what it's actually about.

If we set up a vote to ask a closed-ended question, we should always provide
both options, i.e. agreement or disagreement. Because if you're having a vote
and ask a large audience for their opinion on a certain topic you should
always be prepared to accept the opposite view.

> Obviously, if choice 1 does win, I will *not* propose a GR to overcome this.
> But if choice 1 does *not* win, I don't think the projects want "further
> discussion" but rather "choice 3".

I fully agree.

> If you reply, please respect the reply-to: headers.

Done! Thanks for reminding me :).

Cheers,
Adrian

-- 
 .''`.  John Paul Adrian Glaubitz
: :' :  Debian Developer - glaub...@debian.org
`. `'   Freie Universitaet Berlin - glaub...@physik.fu-berlin.de
  `-GPG: 62FF 8A75 84E0 2956 9546  0006 7426 3B37 F5B5 F913



signature.asc
Description: OpenPGP digital signature

Re: Amendment to Proposed GR: Declassifying parts of -private of historical interest

[Don Armstrong]

On Sun, 07 Aug 2016, Micha Lenk wrote:
> That would establishing some kind of "ex post facto" law (which by the
> way is prohibited in many constitutions for good reasons). I really
> don't want to leave the decision whether past messages will be
> affected or not up to the list masters.

This is why the GR text requires that at minimum DDs can object via GR.

-- 
Don Armstrong  https://www.donarmstrong.com

Your absence has gone through me
Like thread through a needle.
Everything I do is stitched with its color.
 -- W. S. Merwin "Poetry in Motion" p107

Re: Amendment to Proposed GR: Declassifying parts of -private of historical interest

[Bart Martens]

On Mon, Aug 08, 2016 at 09:58:45AM -0500, Don Armstrong wrote:
> On Sun, 07 Aug 2016, Micha Lenk wrote:
> > That would establishing some kind of "ex post facto" law (which by the
> > way is prohibited in many constitutions for good reasons). I really
> > don't want to leave the decision whether past messages will be
> > affected or not up to the list masters.
> 
> This is why the GR text requires that at minimum DDs can object via GR.

I don't see how that covers Micha's concern.

DDs can always initiate a GR, so the text in GR 2016/vote_002 "which at minimum
provides sufficient time and opportunity for Debian Developers to object by GR
prior to declassification" only gives the impression that one can prevent
declassification. In fact there is no way of preventing declassification since
the outcome of a GR is unknown in advance.

I think it's plain wrong that we're now about to give a permission for
declassifying debian-private possibly against the authors' will. GR
2005/vote_002 did include "requests by the author of a post for that post not
to be published will be honored" while 2016/vote_002 does not include any means
for the original authors to prevent declassification.

Regards,

Bart Martens

Re: Amendment to Proposed GR: Declassifying parts of -private of historical interest

[Don Armstrong]

On Mon, 08 Aug 2016, Bart Martens wrote:
> On Mon, Aug 08, 2016 at 09:58:45AM -0500, Don Armstrong wrote:
> > On Sun, 07 Aug 2016, Micha Lenk wrote:
> > > That would establishing some kind of "ex post facto" law (which by the
> > > way is prohibited in many constitutions for good reasons). I really
> > > don't want to leave the decision whether past messages will be
> > > affected or not up to the list masters.
> > 
> > This is why the GR text requires that at minimum DDs can object via GR.
> 
> I don't see how that covers Micha's concern.

It makes it so that the decision whether a particular set of messages or
quotes is released publicly isn't solely up to listmaster@ or another
DPL delegate. At minimum, DDs can object by GR.

> In fact there is no way of preventing declassification since the
> outcome of a GR is unknown in advance.

I don't follow you here; that a GR could potentially go against
you doesn't mean that the outcome of the GR is meaningless.

> I think it's plain wrong that we're now about to give a permission for
> declassifying debian-private possibly against the authors' will. GR
> 2005/vote_002 did include "requests by the author of a post for that
> post not to be published will be honored" while 2016/vote_002 does not
> include any means for the original authors to prevent
> declassification.

I envision that anyone who is delegated to do the declassification will
include something along those lines. But they are in the best position
to decide how to do that, if that ever happens.

-- 
Don Armstrong  https://www.donarmstrong.com

America was far better suited to be the World's Movie Star. The
world's tequila-addled pro-league bowler. The world's acerbic bi-polar
stand-up comedian. Anything but a somber and tedious nation of
socially responsible centurions.
 -- Bruce Sterling, _Distraction_ p122

Re: Amendment to Proposed GR: Declassifying parts of -private of historical interest

[Colin Tuckley]

On 08/08/16 17:53, Don Armstrong wrote:

> I envision that anyone who is delegated to do the declassification will
> include something along those lines. But they are in the best position
> to decide how to do that, if that ever happens.

Indeed, and that means that a message written to debian-private back
when all messages there were private *forever* might now be made public
against the wishes of the author, since a GR may go against the authors
wishes.

Colin

-- 
Colin Tuckley| +44(0)1223 830814 |  PGP/GnuPG Key Id
Debian Developer | +44(0)7799 143369 | 0xFA0C410738C9D903

Re: Amendment to Proposed GR: Declassifying parts of -private of historical interest

[David Kalnischkies]

On Sun, Aug 07, 2016 at 03:45:12PM +, Bart Martens wrote:
> debian-private might make. You've proven that point today on debian-private.

Which is either a leak of information, an argument that such a message
(if it exists) shouldn't have been on d-private in the first place or
you providing disinformation to the public by implying there exists
a message in a channel they can't access which doesn't exist in reality
but nobody can repel as saying that such a message doesn't exist is
leaking information by itself – all of which is in the end an argument
for declassification of messages to remove the mystery surrounding it.


Or as an "Explain like I'm Five" question: Why is the idea that
a process could be proposed by listmasters for declassification which
would be subject to review and with objection opportunity so frightening
given that declassification happens every day by individuals – without
review and without the possibility to object – via (accidental) leakage?


> > > I hope that everyone fully realizes that before voting.
> > 
> > And I hope that, at one point, we as a project will learn to trust one 
> > another
> > and stop micro-managing people that actually want to get things done.
> 
> It's really not about trust and micro-management, but about "what" we want do
> decide with this GR.
> 
> For example, your two points quoted above could easily be included in a GR 
> text
> using these phrases:
> 
> - "The scope is limited to messages posted on debian-private before
>   debian-project was introduced." (And I have no strong opinion on whether 
> this
>   should be included.)
> - "The consent of the original author of the message on debian-private is
>   required before declassification." (I think this should have been in.)

No, it couldn't because these are part of the "how" which will probably
be a mile long IF anyone is ever attempting to do declassification.

For example: "How" are you going to handle mails from people who have by
now passed away – or an author who is currently that ill that (s)he has
a guardian appointed to handle matters for him/her. And how is that
consent even given, verified and recorded… you would need to define all
these things or otherwise someone will end up asking a medium who can
talk to the dead for proxying the consent OR we need a GR to repel yours
with a no-mediums GR because the process outlined in your GR requires
consent, but doesn't forbid (specific) proxies giving it.

If on the other hand we say: Listmaster can come up with a proposal
which can be discussed and as ultima ratio vetoed by GR (or by DPL via
delegation revocation) we can adept as we go[0] with input from anyone
arriving eventually at a process which can a) work and b) nobody objects
to rather than after twenty-something GRs at a process which might work
and nobody objects to *enough* to have yet another GR about it.


[0] I think, but that is just personal opinion and guessing, to have
a realistic stab at declassifying at the very least the first few if not
all ever attempted declassifications will be around a single thread done
by human beings and the process hence tailored to the real needs of the
thread rather than a giant all-knowing fully-automated AI doing it for
all threads in a single pass, which is why the previous GR didn't work,
but just IMHO.


Besides, we talk about d-private. Nobody can honestly believe that an
unencrypted mail sent to ~1000 mail addresses is in any sense of the
word a secret that the public doesn't and will never know. The only
thing it real is: It is non-public archived (on Debian infrastructure…
I would actually be surprised if it couldn't be found somewhere else if
you look really hard for it as if it were really as private as suggested
it would be Debians version of revenge porn to publish it) and that only
because listmasters have chosen to make it so.

That isn't to say that anything posted there should be automatically
public – thinking mostly about VAC messages, expulsions or generic
issues of the day like DDs forming a fellowship (with a ring), forking,
resigning or any other real-world-leaks-into-my-Debian-time announcement
with the appropriated responses – but the idea that it is a safe haven
you can say anything on without risking that it will ever become public
knowledge is at the very least equally wrong. aka: d-private is Debians
archived online version of water-cooler talk. Sometimes, historically
significant things happen around the water-cooler, but most of the time…


Best regards

David Kalnischkies

"Three may keep a Secret, if two of them are dead."
 -- Benjamin Franklin


signature.asc
Description: PGP signature

Re: Amendment to Proposed GR: Declassifying parts of -private of historical interest

[Russ Allbery]

Colin Tuckley  writes:
> On 08/08/16 17:53, Don Armstrong wrote:

>> I envision that anyone who is delegated to do the declassification will
>> include something along those lines. But they are in the best position
>> to decide how to do that, if that ever happens.

> Indeed, and that means that a message written to debian-private back
> when all messages there were private *forever* might now be made public
> against the wishes of the author, since a GR may go against the authors
> wishes.

But that's always been the case, with or without this proposal, since a GR
is the highest decision-making mechanism we have.  Whether or not this GR
passes, I could propose a GR tomorrow that says "all of the debian-private
archives will immediately be released on the web," and if that passes, we
would release messages against their authors' wishes.

The backstop against this happening has always been the good-faith
assumption that Debian as a whole will not vote for such a GR.  I don't
think this proposal materially changes that.

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

Re: Amendment to Proposed GR: Declassifying parts of -private of historical interest

[Micha Lenk]

Hi everybody,

Am 08.08.2016 um 16:58 schrieb Don Armstrong:
> On Sun, 07 Aug 2016, Micha Lenk wrote:
>> That would establishing some kind of "ex post facto" law (which by the
>> way is prohibited in many constitutions for good reasons). I really
>> don't want to leave the decision whether past messages will be
>> affected or not up to the list masters.
> 
> This is why the GR text requires that at minimum DDs can object via GR.

That -- and also the listmasters or another DPL delegate deciding about
what/how to release publicly -- would be perfectly fine with me, but
only for any mails posted after the GR is accepted. I just don't want
this GR to change the rules of the past in the same go.

On a related note, I would also vote against any GR to "repeal the GR of
2005 and burry the idea of systematically declassifying debian-private",
as suggested by Holger Levsen. This is because despite my opposition to
the current GR, I do support the rationale behind the 2005 GR:

>  In accordance with principles of openness and transparency, Debian
>  will seek to declassify and publish posts of historical or ongoing
>  significance made to the Debian Private Mailing List

IMHO the currently proposed GR tries to solve too many issues at once.
Changing the rules for future mails on -private is one goal that is hard
enough to accomplish but certainly worth it. Cleaning up the backlog of
declassifying the -private archive is another one, but this does not
necessarily need to follow the same rules. So let's decide them
separately, I would say.

For this reason "further discussion" is correctly representing what I
want to vote for.

Best regards,
Micha

Re: Amendment to Proposed GR: Declassifying parts of -private of historical interest

[Nick Phillips]

On Mon, 2016-08-08 at 19:56 +0200, David Kalnischkies wrote:
> On Sun, Aug 07, 2016 at 03:45:12PM +, Bart Martens wrote:
> > 
> > debian-private might make. You've proven that point today on
> > debian-private.
> Which is either a leak of information, an argument that such a
> message
> (if it exists) shouldn't have been on d-private in the first place or
> you providing disinformation to the public by implying there exists
> a message in a channel they can't access which doesn't exist in
> reality
> but nobody can repel as saying that such a message doesn't exist is
> leaking information by itself – all of which is in the end an
> argument
> for declassification of messages to remove the mystery surrounding
> it.
> 
> 
> Or as an "Explain like I'm Five" question: Why is the idea that
> a process could be proposed by listmasters for declassification which
> would be subject to review and with objection opportunity so
> frightening
> given that declassification happens every day by individuals –
> without
> review and without the possibility to object – via (accidental)
> leakage?
> 


It might be that a hypothetical process would ensure that it did not
make public any message without approval of the author. I believe that
this would be fine. It might however be that the hypothetical process
would make public any message from any author who did not respond to an
invitation to object, posted perhaps in a filing cabinet in a locked
closet with a sign on the door saying "Beware of the Leopard"; since
those authors did not object via GR, they would be assumed to be OK
with the process. This would be Wrong. 

And it is "frightening" because it would be deliberate, and on the part
of the Project. For the Project to be morally/ethically lacking would
be far more disturbing than for an individual to be so.

To be clear - I do not believe that it would be acceptable for any message to 
be made public without explicit approval of the author. A mere lack of 
objection is not enough - however it does seem to me that this is a road that 
some are keen to travel down.


Cheers,


Nick
-- 
Nick Phillips / nick.phill...@otago.ac.nz / 03 479 4195
# These statements are mine, not those of the University of Otago