Re: Re: Secret ballot and RMS Resolution

[Davide Lombardo]

How you can credibly blame the FSF to have kept secret the process of RMS 
reinstatement, when you are the first asking to keep secret the votes in regard 
 
of this  stupid GR  ? You should select better the DPL the next time, pushing 
a such divisive GR in freeze time was a very bad decision for the overall 
ending quality of the project.  Please don't do I such stupid thing again,  
learn from the past. Maybe you should make a rule, no highly divisive GR in 
freeze period again.



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

Re: Secret ballot and RMS Resolution

[Wouter Verhelst]

Hi,

On Thu, Apr 01, 2021 at 03:06:07PM +, Didier 'OdyX' Raboud wrote:
> Could we get a Constitution Amendment GR passed along the lines of the
> following?
> 
> Provided that 2*Q developers demand it, votes are kept secret after
> the vote ended.

I would actually prefer it to be easier:

If one developer demands it, the votes will be kept secret, *unless* N
developers oppose the secrecy of the vote (with the definition of N
being up for debate).

I think it may be more important to make people feel safe to express
their vote than to keep the process open.

-- 
To the thief who stole my anti-depressants: I hope you're happy

  -- seen somewhere on the Internet on a photo of a billboard

Re: Secret ballot and RMS Resolution

[Ulrike Uhlig]

Hello,

On 01.04.21 20:08, Bart Martens wrote:

On Thu, Apr 01, 2021 at 05:00:47PM +0200, Kurt Roeckx wrote:

The constitution says:
 3. Votes are taken by the Project Secretary. Votes, tallies, and
results are not revealed during the voting period; after the vote
the Project Secretary lists all the votes cast. The voting period
is 2 weeks, but may be varied by up to 1 week by the Project
Leader.



You could say that "all the votes cast" could mean what was voted,
now who voted what, but I think that conflicts with the intention
of the text.


Could one use hashes and publish those, while keeping the original votes 
in a place that's accessible only to the secretary? The constitution 
does not talk about _how_ the cast votes need to be listed. Also it uses 
the word "listed" not "published". But I might be playing on words here.

But again, that option would place the secretary at the center of trust.

That said, i don't feel very strongly about having that particular vote 
be private, specifically since the initial email came from a pseudonym 
referring to https://en.wikipedia.org/wiki/Jean_Duprat_(Avignon) and 
since I could not find that email having interacted with any Debian list 
before [1], nor find it anywhere else referred to on the internet.


Ulrike

[1] https://lists.debian.org/cgi-bin/search?P=je.duprat%40protonmail.com

Re: Secret ballot and RMS Resolution

[Bart Martens]

On Thu, Apr 01, 2021 at 05:00:47PM +0200, Kurt Roeckx wrote:
> The constitution says:
> 3. Votes are taken by the Project Secretary. Votes, tallies, and
>results are not revealed during the voting period; after the vote
>the Project Secretary lists all the votes cast. The voting period
>is 2 weeks, but may be varied by up to 1 week by the Project
>Leader.
> 
> While for DPL election it says:
> 5. The next two weeks are the polling period during which Developers
>may cast their votes. Votes in leadership elections are kept
>secret, even after the election is finished.
> 
> You could say that "all the votes cast" could mean what was voted,
> now who voted what, but I think that conflicts with the intention
> of the text.

Kurt, one could argue that publishing the names is always forbidden. Article 5
already plainly forbids it. And article 3 does not specify it, so with
privacy... I'm just sharing another way of looking at this, without suggesting
to change commonly accepted practice now in the middle of two votings. - B.



signature.asc
Description: PGP signature

Re: Secret ballot and RMS Resolution

[Philip Hands]

Kurt Roeckx  writes:

...
>after the vote the Project Secretary lists all the votes cast.
...
> You could say that "all the votes cast" could mean what was voted,

You could also note that there is no stipulation of how soon after the
vote that publication must occur, and decide that in this case that an
appropriate interval would be e.g. no less than ten years.

Cheers, Phil.
-- 
|)|  Philip Hands  [+44 (0)20 8530 9560]  HANDS.COM Ltd.
|-|  http://www.hands.com/http://ftp.uk.debian.org/
|(|  Hugo-Klemm-Strasse 34,   21075 Hamburg,GERMANY


signature.asc
Description: PGP signature

Re: Secret ballot and RMS Resolution

[Russ Allbery]

Felix Lechner  writes:

> Moving to new software without preparation or a chance to practice

We would have to pass a constitutional amendment to allow for secret votes
for anything other than DPL elections, at least by my understanding of the
constitution and this thread, so there will be plenty of preparation and
warning and opportunity to practice.  No one has even proposed the
constitutional GR yet.

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

Re: Secret ballot and RMS Resolution

[Kurt Roeckx]

On Thu, Apr 01, 2021 at 10:18:08AM -0700, Felix Lechner wrote:
> Hi,
> 
> On Thu, Apr 1, 2021 at 9:59 AM Kurt Roeckx  wrote:
> >
> > I could move to voting software like Belenios
> 
> Moving to new software without preparation or a chance to practice
> could discourage centrist voters—i.e. those who care least but provide
> the gravity to hold the project together.
> 
> It could perhaps harm the vote's legitimacy, especially in view of the
> perceived rush and the already abbreviated discussion period.

I'm not suggesting that for either the current DPL vote or the GR
that's being discussed.


Kurt

Re: Secret ballot and RMS Resolution

[Stefano Zacchiroli]

On Thu, Apr 01, 2021 at 09:47:51AM -0700, Russ Allbery wrote:
> I didn't quite parse that, so I'm not sure if this is what you were
> already proposing, but I wonder if we should just have secret ballots for
> all votes.

My bad for my previous non-parsable paragraph. I was proposing something
less radical (namely: secret vote for elections + decision by the
secretary for all other votes). That was in the spirit of minimizing
changes to the Constitution. But upon reflection I think your proposal
is just better.

Non-secret voting is useful and important for the votes cast by elected
representatives (e.g., in parliaments), because it provides
accountability. But all GRs in Debian are referendums, where everyone
votes for themselves, without having to be accountable to anyone else.

So, yeah, I (now) see no point any more in having non-secret votes in
Debian.

(There's the marginal advantage of verifiability that you point out, but
if either DSA or the Secretary are not trusted, we already have a major
problem anyway. And both bodies have ways to interfere with voting even
today with non-secret votes: ballot stuffing on behalf of inactive
accounts, dropping ballots of people who will not verify, etc.)

Cheers
-- 
Stefano Zacchiroli . z...@upsilon.cc . upsilon.cc/zack . . o . . . o . o
Computer Science Professor . CTO Software Heritage . . . . . o . . . o o
Former Debian Project Leader & OSI Board Director  . . . o o o . . . o .
« the first rule of tautology club is the first rule of tautology club »

Re: Secret ballot and RMS Resolution

[Pierre-Elliott Bécue]

Le jeudi 01 avril 2021 à 10:18:08-0700, Felix Lechner a écrit :
> Hi,
> 
> On Thu, Apr 1, 2021 at 9:59 AM Kurt Roeckx  wrote:
> >
> > I could move to voting software like Belenios
> 
> Moving to new software without preparation or a chance to practice
> could discourage centrist voters—i.e. those who care least but provide
> the gravity to hold the project together.
> 
> It could perhaps harm the vote's legitimacy, especially in view of the
> perceived rush and the already abbreviated discussion period.

I'm pretty sure the idea is not to move right now.

Apart from that, Belenios is starting to be considered as reliable.

-- 
Pierre-Elliott Bécue
GPG: 9AE0 4D98 6400 E3B6 7528  F493 0D44 2664 1949 74E2
It's far easier to fight for one's principles than to live up to them.


signature.asc
Description: PGP signature

Re: Secret ballot and RMS Resolution

[Felix Lechner]

Hi,

On Thu, Apr 1, 2021 at 9:59 AM Kurt Roeckx  wrote:
>
> I could move to voting software like Belenios

Moving to new software without preparation or a chance to practice
could discourage centrist voters—i.e. those who care least but provide
the gravity to hold the project together.

It could perhaps harm the vote's legitimacy, especially in view of the
perceived rush and the already abbreviated discussion period.

Kind regards
Felix Lechner

Re: Secret ballot and RMS Resolution

[Kurt Roeckx]

On Thu, Apr 01, 2021 at 09:47:51AM -0700, Russ Allbery wrote:
> Stefano Zacchiroli  writes:
> 
> > This is probably something that should be fixed in the Constitution, by
> > mandating secret voting for elections whereas living to the judgment of
> > the secretary whether other GR votes should be secret or not.
> 
> I didn't quite parse that, so I'm not sure if this is what you were
> already proposing, but I wonder if we should just have secret ballots for
> all votes.  That avoids the argument over whether a given vote should be a
> secret ballot.  I admit to being endlessly curious about the opinions of
> my fellow contributors, but that's just me being nosy and I don't have any
> real reason to need to know.  Secret ballots seem like they would be
> reassuring for the people who are the least willing to be vocal on mailing
> lists.
> 
> In the current on-line landscape, I think it's also incumbent on any
> organization to look for ways in which their procedures or tools could be
> used for harassment and to try to make them safer, and secret ballots seem
> like an obvious thing for us to do here.
> 
> The obvious difficulty with secret ballots is that it requires a lot of
> trust in the votetaker.  We have a verification process for people to
> check whether their vote was recorded correctly, but I don't think there's
> any good way with our style of secret ballot to verify that the votetaker
> didn't add new votes that no one actually sent.  But (a) personally I am
> not worried about the Project Secretary manipulating the vote, and (b) I
> can think of a few ways to deal with that should that ever be enough of a
> concern, such as having a small number of people trusted with confidential
> information jointly double-check the tally.
> 
> This obviously would have an impact on the Project Secretary, so I'd love
> to hear Kurt and Neil's opinion about the possible merits of going to
> secret ballots for all votes.

I could move to voting software like Belenios that only supports
secret votes and has a way that doesn't require trust in (just)
the Secretary.


Kurt

Re: Secret ballot and RMS Resolution

[Russ Allbery]

Stefano Zacchiroli  writes:

> This is probably something that should be fixed in the Constitution, by
> mandating secret voting for elections whereas living to the judgment of
> the secretary whether other GR votes should be secret or not.

I didn't quite parse that, so I'm not sure if this is what you were
already proposing, but I wonder if we should just have secret ballots for
all votes.  That avoids the argument over whether a given vote should be a
secret ballot.  I admit to being endlessly curious about the opinions of
my fellow contributors, but that's just me being nosy and I don't have any
real reason to need to know.  Secret ballots seem like they would be
reassuring for the people who are the least willing to be vocal on mailing
lists.

In the current on-line landscape, I think it's also incumbent on any
organization to look for ways in which their procedures or tools could be
used for harassment and to try to make them safer, and secret ballots seem
like an obvious thing for us to do here.

The obvious difficulty with secret ballots is that it requires a lot of
trust in the votetaker.  We have a verification process for people to
check whether their vote was recorded correctly, but I don't think there's
any good way with our style of secret ballot to verify that the votetaker
didn't add new votes that no one actually sent.  But (a) personally I am
not worried about the Project Secretary manipulating the vote, and (b) I
can think of a few ways to deal with that should that ever be enough of a
concern, such as having a small number of people trusted with confidential
information jointly double-check the tally.

This obviously would have an impact on the Project Secretary, so I'd love
to hear Kurt and Neil's opinion about the possible merits of going to
secret ballots for all votes.

> But that cannot be fixed in a timely manner. I'm not sure what other
> options we have here.

I can't think of a good one either.  There's proxy voting (find Debian
Developers who don't want to vote personally and have them cast votes for
other Debian Developers who aren't willing to vote a public ballot), but
gah, what a logistical mess.  I'm not sure we could pull it off.

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

Re: Secret ballot and RMS Resolution

[Pierre-Elliott Bécue]

Le jeudi 01 avril 2021 à 19:04:41+0300, Andrei POPESCU a écrit :
> On Jo, 01 apr 21, 17:00:47, Kurt Roeckx wrote:
> > 
> > This would be a vote I would also like to see as secret. The
> > constitution says:
> > 3. Votes are taken by the Project Secretary. Votes, tallies, and
> >results are not revealed during the voting period; after the vote
> >the Project Secretary lists all the votes cast. The voting period
> >is 2 weeks, but may be varied by up to 1 week by the Project
> >Leader.
> > 
> > While for DPL election it says:
> > 5. The next two weeks are the polling period during which Developers
> >may cast their votes. Votes in leadership elections are kept
> >secret, even after the election is finished.
> > 
> > You could say that "all the votes cast" could mean what was voted,
> > now who voted what, but I think that conflicts with the intention
> > of the text.
> 
> How about making the votes available to Debian Members only?
> 
> Kind regards,

One could also fear that Debian Members would harrass them because they
voted X.

-- 
Pierre-Elliott Bécue
GPG: 9AE0 4D98 6400 E3B6 7528  F493 0D44 2664 1949 74E2
It's far easier to fight for one's principles than to live up to them.


signature.asc
Description: PGP signature

Re: Secret ballot and RMS Resolution

[Andrei POPESCU]

On Jo, 01 apr 21, 17:00:47, Kurt Roeckx wrote:
> 
> This would be a vote I would also like to see as secret. The
> constitution says:
> 3. Votes are taken by the Project Secretary. Votes, tallies, and
>results are not revealed during the voting period; after the vote
>the Project Secretary lists all the votes cast. The voting period
>is 2 weeks, but may be varied by up to 1 week by the Project
>Leader.
> 
> While for DPL election it says:
> 5. The next two weeks are the polling period during which Developers
>may cast their votes. Votes in leadership elections are kept
>secret, even after the election is finished.
> 
> You could say that "all the votes cast" could mean what was voted,
> now who voted what, but I think that conflicts with the intention
> of the text.

How about making the votes available to Debian Members only?

Kind regards,
Andrei
-- 
http://wiki.debian.org/FAQsFromDebianUser


signature.asc
Description: PGP signature

Re: Secret ballot and RMS Resolution

[Felix Lechner]

Hi,

On Thu, Apr 1, 2021 at 8:01 AM Kurt Roeckx  wrote:
>
> This would be a vote I would also like to see as secret. The

Maybe a solution would be to offer two alternatives of each option on
the ballot—one as currently stated, and one that includes a secret
vote instruction to the secretary. It would require an explanation to
voters to clear up the confusion, but would result in the greatest
legitimacy.

As an alternative, all delegates and the DPL could perhaps vote to
stay the release of voter identities until the electorate voted on the
matter. As a stop-gap solution conceived to head off a constitutional
crisis, it might work because there is no explicit promise of
timeliness in the constitutional part that reads "after the vote the
Project Secretary lists all the votes cast."

Unfortunately, neither alternative will comfort voters afraid of
reprisals, because there is no guarantee of privacy.

Kind regards
Felix Lechner

Re: Secret ballot and RMS Resolution

[Taowa]

Stefano Zacchiroli, 2021-04-01 10:41 -0400:
> On Thu, Apr 01, 2021 at 12:42:01PM +, Jean Duprat (Avignon) wrote:
> > Votes in leadership elections are kept secret even after the end of
> > the voting period for obvious reasons: by knowing that the ballot is
> > secret, voters can feel free to express their opinion as they see
> > fit. This constitutional guarantee sadly does not apply to General
> > Resolutions.
> 
> I think we're at opposite ends of the voting spectrum here on this
> matter, but I share your concern.
> 
> In fact, I think that *any* position taken in this vote could result in
> targeted harassment of the voters.

Thanks Stefano. My thoughts exactly.

Further, I will be voting on this GR with the same set of worries I have
whenever I post to Debian lists.  I signed the letter with that same set
of worries. I would not be surprised in the least if I found out that a
large proportion of other women and non-binary Debian contributors feel
the same way.  Painting this as either novel or unique to those who
object to Debian making a statement seems disingenuous.

Taowa

-- 
Taowa (they)
LOC FN35EM

Re: Secret ballot and RMS Resolution

[Didier 'OdyX' Raboud]

1 avril 2021 17:01 "Kurt Roeckx"  a écrit:
> On Thu, Apr 01, 2021 at 04:40:59PM +0200, Stefano Zacchiroli wrote: 
>> On Thu, Apr 01, 2021 at 12:42:01PM +, Jean Duprat (Avignon) wrote:
>> Votes in leadership elections are kept secret even after the end of
>> the voting period for obvious reasons: by knowing that the ballot is
>> secret, voters can feel free to express their opinion as they see
>> fit. This constitutional guarantee sadly does not apply to General
>> Resolutions.
>> 
>> I think we're at opposite ends of the voting spectrum here on this
>> matter, but I share your concern.
>> 
>> In fact, I think that *any* position taken in this vote could result in
>> targeted harassment of the voters.
>> 
>> As this is a vote on a person (even if an indirect one, in the sense
>> that is not an election), most democratic systems will treat it as a
>> vote where ballots should be secret. But I fear that the Debian
>> Constitution ties the hands of the secretary here, or am I missing
>> something here Kurt?
> 
> This would be a vote I would also like to see as secret. The
> constitution says:
> 3. Votes are taken by the Project Secretary. Votes, tallies, and
> results are not revealed during the voting period; after the vote
> the Project Secretary lists all the votes cast. The voting period
> is 2 weeks, but may be varied by up to 1 week by the Project
> Leader.
> 
> While for DPL election it says:
> 5. The next two weeks are the polling period during which Developers
> may cast their votes. Votes in leadership elections are kept
> secret, even after the election is finished.
> 
> You could say that "all the votes cast" could mean what was voted,
> now who voted what, but I think that conflicts with the intention
> of the text.

Could we get a Constitution Amendment GR passed along the lines of the
following?

Provided that 2*Q developers demand it, votes are kept secret after
the vote ended.

(en_* advice welcome)

--
  OdyX

Re: Secret ballot and RMS Resolution

[Kurt Roeckx]

On Thu, Apr 01, 2021 at 04:40:59PM +0200, Stefano Zacchiroli wrote:
> On Thu, Apr 01, 2021 at 12:42:01PM +, Jean Duprat (Avignon) wrote:
> > Votes in leadership elections are kept secret even after the end of
> > the voting period for obvious reasons: by knowing that the ballot is
> > secret, voters can feel free to express their opinion as they see
> > fit. This constitutional guarantee sadly does not apply to General
> > Resolutions.
> 
> I think we're at opposite ends of the voting spectrum here on this
> matter, but I share your concern.
> 
> In fact, I think that *any* position taken in this vote could result in
> targeted harassment of the voters.
> 
> As this is a vote on a person (even if an indirect one, in the sense
> that is not an election), most democratic systems will treat it as a
> vote where ballots should be secret. But I fear that the Debian
> Constitution ties the hands of the secretary here, or am I missing
> something here Kurt?

This would be a vote I would also like to see as secret. The
constitution says:
3. Votes are taken by the Project Secretary. Votes, tallies, and
   results are not revealed during the voting period; after the vote
   the Project Secretary lists all the votes cast. The voting period
   is 2 weeks, but may be varied by up to 1 week by the Project
   Leader.

While for DPL election it says:
5. The next two weeks are the polling period during which Developers
   may cast their votes. Votes in leadership elections are kept
   secret, even after the election is finished.

You could say that "all the votes cast" could mean what was voted,
now who voted what, but I think that conflicts with the intention
of the text.


Kurt

Re: Secret ballot and RMS Resolution

[Stefano Zacchiroli]

On Thu, Apr 01, 2021 at 12:42:01PM +, Jean Duprat (Avignon) wrote:
> Votes in leadership elections are kept secret even after the end of
> the voting period for obvious reasons: by knowing that the ballot is
> secret, voters can feel free to express their opinion as they see
> fit. This constitutional guarantee sadly does not apply to General
> Resolutions.

I think we're at opposite ends of the voting spectrum here on this
matter, but I share your concern.

In fact, I think that *any* position taken in this vote could result in
targeted harassment of the voters.

As this is a vote on a person (even if an indirect one, in the sense
that is not an election), most democratic systems will treat it as a
vote where ballots should be secret. But I fear that the Debian
Constitution ties the hands of the secretary here, or am I missing
something here Kurt?

This is probably something that should be fixed in the Constitution, by
mandating secret voting for elections whereas living to the judgment of
the secretary whether other GR votes should be secret or not. But that
cannot be fixed in a timely manner. I'm not sure what other options we
have here.

Cheers
-- 
Stefano Zacchiroli . z...@upsilon.cc . upsilon.cc/zack . . o . . . o . o
Computer Science Professor . CTO Software Heritage . . . . . o . . . o o
Former Debian Project Leader & OSI Board Director  . . . o o o . . . o .
« the first rule of tautology club is the first rule of tautology club »


signature.asc
Description: PGP signature

Re: Secret ballot and RMS Resolution

[Felix Lechner]

Hi,

On Thu, Apr 1, 2021 at 6:17 AM Pasha  wrote:
>
> Could you please provide any historical example of  your "mob justice" ?

Sadly, a discussion over proof cannot solve the issue. The request of
a single eligible voter—motivated by fear or otherwise—should trigger
a secret procedure. (Although it is not so in a representative body,
where accountability is paramount.) That is what my high school did,
when I served as student leader.

In a positive light, secrecy requests are a great measure of the
environment in which a vote was conducted, and also show the health of
the community with respect to mutual trust.

Kind regards
Felix Lechner

Re: Secret ballot and RMS Resolution

[Pasha]

On Thu, 2021-04-01 at 12:42 +, Jean Duprat (Avignon) wrote:
> Votes in leadership elections are kept secret even after the end of
> the voting period for obvious reasons: by knowing that the ballot is
> secret, voters can feel free to express their opinion as they see
> fit. This constitutional guarantee sadly does not apply to General
> Resolutions.
> 
> The GR currently being discussed, "Statement regarding Richard
> Stallman's readmission to the FSF board", is significantly more
> polarizing and controversial than any of the DPL elections the
> project has ever had. Due to various examples of attacks based on
> guilt by association fallacies [1] and similar that took place during
> the recent debate about RMS, voters may not be able to freely express
> their opinion on this GR if the tally sheet is made public at the end
> of the voting period. It is unfortunate that the consequences of
> simply expressing an opinion may include personal attacks, ostracism
> and even losing one's job, but we have seen enough evidence of that
> to know that it is the case. [2]
> 
> The nuances of why someone might decide against a collective
> endorsement of the statement by the Debian Project are entirely lost
> by looking at the tally sheet alone. Maybe you are terrified by mob
> justice. Perhaps you disagree with some of the most extreme opinions
> expressed in the appendix to the statement, including that using
> singular gender-neutral pronouns instead of "they" is enough to be
> found guilty of transphobia. You could even simply think that Debian
> as a Project should not participate in the current culture of fear
> and intimidation resembling McCarthyism more than the type of society
> we all aspire to live in.
> 
> Whatever the case may be, it is not safe to vote choice 3: "do not,
> as the project itself, sign any letter regarding rms" unless the
> privacy of voters is guaranteed.
> 
> [1] https://twitter.com/sarahmei/status/1172744015885697025
> [2] https://en.wikipedia.org/wiki/Google%27s_Ideological_Echo_Chamber
> 

Hi Jean Duprat, 

Could you please provide any historical example of  your "mob justice"
?
Thank you.

Secret ballot and RMS Resolution

[Jean Duprat (Avignon)]

Votes in leadership elections are kept secret even after the end of the voting 
period for obvious reasons: by knowing that the ballot is secret, voters can 
feel free to express their opinion as they see fit. This constitutional 
guarantee sadly does not apply to General Resolutions.

The GR currently being discussed, "Statement regarding Richard Stallman's 
readmission to the FSF board", is significantly more polarizing and 
controversial than any of the DPL elections the project has ever had. Due to 
various examples of attacks based on guilt by association fallacies [1] and 
similar that took place during the recent debate about RMS, voters may not be 
able to freely express their opinion on this GR if the tally sheet is made 
public at the end of the voting period. It is unfortunate that the consequences 
of simply expressing an opinion may include personal attacks, ostracism and 
even losing one's job, but we have seen enough evidence of that to know that it 
is the case. [2]

The nuances of why someone might decide against a collective endorsement of the 
statement by the Debian Project are entirely lost by looking at the tally sheet 
alone. Maybe you are terrified by mob justice. Perhaps you disagree with some 
of the most extreme opinions expressed in the appendix to the statement, 
including that using singular gender-neutral pronouns instead of "they" is 
enough to be found guilty of transphobia. You could even simply think that 
Debian as a Project should not participate in the current culture of fear and 
intimidation resembling McCarthyism more than the type of society we all aspire 
to live in.

Whatever the case may be, it is not safe to vote choice 3: "do not, as the 
project itself, sign any letter regarding rms" unless the privacy of voters is 
guaranteed.

[1] https://twitter.com/sarahmei/status/1172744015885697025
[2] https://en.wikipedia.org/wiki/Google%27s_Ideological_Echo_Chamber