[GNU/consensus] A GNU Consensus for the GNU Year!

[hellekin (GNU Consensus)]

We're pleased to announce the launch of our activities on this new
year's eve.

Toward a GNU Consensus on Free Software for Social Networking[1]

Dear fellow hackers,

after almost a year working in the shadows, we're pleased to announce
the launch of the GNU/consensus project[2]. You're invited to join the
mailing list[3], and to register your Free Software project as a
stakeholder[4].

Happy GNU year!

==
hk

[1] https://gnu.org/consensus/manifesto
[2] https://gnu.org/consensus
[3] https://lists.gnu.org/mailman/listinfo/consensus
[4] https://gnu.org/consensus/stakeholders

[Melvin Carvalho]

Post by hellekin (GNU Consensus)
We're pleased to announce the launch of our activities on this new
year's eve.
Toward a GNU Consensus on Free Software for Social Networking[1]
Dear fellow hackers,
after almost a year working in the shadows, we're pleased to announce
the launch of the GNU/consensus project[2]. You're invited to join the
mailing list[3], and to register your Free Software project as a
stakeholder[4].

+1 looks like a great initiative

I very much look forward to testing interop. May I suggest three points
that can operate as a starting point.

1. Someone in Project A wishes to friend someone in Project B

2. Someone in Project B wishes to friend someone in Project A

3. The two people, having established friendship, should be able to send
each other a private message.

You will find that 90% of projects that claim to inteorp CANNOT do this.
Some will never be able to do it. Let's quickly try towork out which
projects are serious about interop, and for which, it is a marketing
vehicle.

Also I am concerned about GNU favouring proprietary protocols, over
existing web standards, or standards under creative commons.

Will it be a MUST that GNU consensus will preserve the concept of freedom
of thought, such that, anyone is free to inspect and free to MODIFY the
protocols used? Will the license of the protocols used reflect that?

Post by hellekin (GNU Consensus)
Happy GNU year!
==
hk
[1] https://gnu.org/consensus/manifesto
[2] https://gnu.org/consensus
[3] https://lists.gnu.org/mailman/listinfo/consensus
[4] https://gnu.org/consensus/stakeholders

[hellekin (GNU Consensus)]

Post by Melvin Carvalho
+1 looks like a great initiative

*** +12 so far, on the mailing list :)

Post by Melvin Carvalho
I very much look forward to testing interop. May I suggest three points
that can operate as a starting point.
1. Someone in Project A wishes to friend someone in Project B
2. Someone in Project B wishes to friend someone in Project A
3. The two people, having established friendship, should be able to send
each other a private message.
You will find that 90% of projects that claim to inteorp CANNOT do
this. Some will never be able to do it. Let's quickly try towork out
which projects are serious about interop, and for which, it is a
marketing vehicle.

*** Excellent points, Melvin. Welcome!

I put our conversations of the other day on the LibrePlanet wiki:
http://libreplanet.org/wiki/User_talk:Hellekin/A_Coder_Perspective_Of_GNU_Social

The plan is to register a GNU/consensus interest group on LibrePlanet.
as vector to distill the conversation of this list, and allow more
people to participate more easily. Then we can update the official
GNU/consensus website accordingly. What do you think?

Post by Melvin Carvalho
Also I am concerned about GNU favouring proprietary protocols, over
existing web standards, or standards under creative commons.

*** Of course, the GNU's mission is to foster user freedom. I contacted
the tent.io people earlier to start a discussion with them on that
topic, but didn't receive any response yet. I hope they're going to jump
on board, as well as Evan Prodromou, who's working on transition from
StatusNet to pump.io[1].

Post by Melvin Carvalho
free to MODIFY the protocols used?
Will the license of the protocols used reflect that?

*** I, nor the GNU project, don't have any authority to determine what
licenses the protocols are using. We can only issue recommendations, and
hope the community cooperates toward these goals. Obviously, being part
of the GNU project makes it clear what we're looking for!

Cheers,

==
hk

[Melvin Carvalho]

Post by hellekin (GNU Consensus)

Post by Melvin Carvalho
+1 looks like a great initiative

*** +12 so far, on the mailing list :)

Post by Melvin Carvalho
I very much look forward to testing interop. May I suggest three points
that can operate as a starting point.
1. Someone in Project A wishes to friend someone in Project B
2. Someone in Project B wishes to friend someone in Project A
3. The two people, having established friendship, should be able to send
each other a private message.
You will find that 90% of projects that claim to inteorp CANNOT do
this. Some will never be able to do it. Let's quickly try towork out
which projects are serious about interop, and for which, it is a
marketing vehicle.

*** Excellent points, Melvin. Welcome!
http://libreplanet.org/wiki/User_talk:Hellekin/A_Coder_Perspective_Of_GNU_Social
The plan is to register a GNU/consensus interest group on LibrePlanet.
as vector to distill the conversation of this list, and allow more
people to participate more easily. Then we can update the official
GNU/consensus website accordingly. What do you think?

Sounds good. If we can have a project that drives interop, and can record
and measure the results, that's going to be big plus. Because we've not
had the resources to do that before. If you've ever used the 'wine'
program they measure the comparability of many projects with GNU linux and
give it a rating on a wiki page bronze/silver/gold. I'm not sure that's
the perfect way, but if we can have something concrete, like badges, it
could mean something.

Post by hellekin (GNU Consensus)

Post by Melvin Carvalho
Also I am concerned about GNU favouring proprietary protocols, over
existing web standards, or standards under creative commons.

*** Of course, the GNU's mission is to foster user freedom. I contacted
the tent.io people earlier to start a discussion with them on that
topic, but didn't receive any response yet. I hope they're going to jump
on board, as well as Evan Prodromou, who's working on transition from
StatusNet to pump.io[1].

tent.io is a good project imho ... dont forget the open graph protocol,
everything you need is there, and it's the state of the art, im less
worried about web 2.0 stuff, they tend to do their own thing, let's see ...

Post by hellekin (GNU Consensus)

Post by Melvin Carvalho
free to MODIFY the protocols used?
Will the license of the protocols used reflect that?

*** I, nor the GNU project, don't have any authority to determine what
licenses the protocols are using. We can only issue recommendations, and
hope the community cooperates toward these goals. Obviously, being part
of the GNU project makes it clear what we're looking for!

We should encourage free software principles ie the freedom to inspect and
modify. for protocols just as software. I can understand some proprietary
protocols are useful, when they are the state of the art. But to use
proprietary stuff that's years behind leading edge needs a good
explanation. I would say we can learn a lot from the best proprietary
protocols, and have already. But we should also not be afraid to embrace
real standards like linked data that are starting to get a good track
record, and doubling every 9 months...

Post by hellekin (GNU Consensus)
Cheers,
==
hk

[Christian Grothoff]

Dear Hellekin,

While, as I'm sure you know, I'm happy about this initiative in general
and like to see it succeed, I'm not quite happy with some details in the
manifesto.

In particular, you write:
Anonymity

With interoperating free software social networking systems, no user
will be compelled to provide any particular kind of information, whether
it be her name, her age, or what country she lives in. It will be up to
those she communicates with to judge what information she chooses to
provide or withhold.
<<<

This is not _quite_ what I believe to be commonly understood by
'anonymity' by most users or the research community. What I think you
mean is more like a 'right to use pseudonyms'. The possibility to post
content anonymously is a separate issue (and somewhat incompatible with
mapping social relationships). Furthermore, the idea of making your
online identities unlinkable to real-world identities (which does relate
more to 'anonymity') is again a bit different (and much harder to
achieve, as having access to things like language, time and social
relationships can enable some adversaries to link pseudonyms to
real-world identities).

Given this, _strong_ pseudonymity (as in, user's pseudonyms being
fundamentally unlinkable to real-world identities) or _strong_ anonymity
are things we may strive for (at least for some applications), but
should not claim as major features (at this point) or fundamental
properties of GNU consensus. Thus, I'd urge you to change the section
title to something more like "Right to Pseudonymity".


My second nitpick is that it remains unclear to me how _exactly_ you
expect people (or projects) to contribute in the GNU consensus-context.
Is it supposed to be "just" a discussion platform? Or is it about
advocacy? Is there going to be code written under the "GNU consensus"
label? If so, what will that code do?


Finally, a little note: there is in GNUnet an API which is called the
"consensus" API (for set reconciliation). That API --- while possibly a
useful building block for social network applications --- should not be
confused with the "GNU Consensus" protocol. I wonder if we need to
rename it to avoid a possible confusion...


Happy hacking!

Christian

Post by hellekin (GNU Consensus)
We're pleased to announce the launch of our activities on this new
year's eve.
Toward a GNU Consensus on Free Software for Social Networking[1]
Dear fellow hackers,
after almost a year working in the shadows, we're pleased to announce
the launch of the GNU/consensus project[2]. You're invited to join the
mailing list[3], and to register your Free Software project as a
stakeholder[4].
Happy GNU year!
==
hk
[1] https://gnu.org/consensus/manifesto
[2] https://gnu.org/consensus
[3] https://lists.gnu.org/mailman/listinfo/consensus
[4] https://gnu.org/consensus/stakeholders

[hellekin (GNU Consensus)]

Post by Christian Grothoff
Dear Hellekin,
While, as I'm sure you know, I'm happy about this initiative in general
and like to see it succeed, I'm not quite happy with some details in the
manifesto.

*** Hi Christian! Nice to have you around. I want to make you happy :)

Post by Christian Grothoff
Anonymity
Thus, I'd urge you to change the section
title to something more like "Right to Pseudonymity".

*** You make an excellent point. Richard, do you have any objection to
that renaming, as you wrote this section?

Post by Christian Grothoff
My second nitpick is that it remains unclear to me how _exactly_ you
expect people (or projects) to contribute in the GNU consensus-context.

*** I have some ideas, such as providing a test suite that is more
modular than the original SWAT tests, as we've been discussing with
Melvin earlier. You can find a log of a conversation we had recently on
IRC at the LibrePlanet wiki[1].

In general, I want to engage in a conversation with the stakeholders, so
that we can reach a consensus--hence the name of the project, on what
has to be done in order to facilitate interoperability. I want to avoid
being too directive, and give a chance to everyone to voice their
opinions, concerns, and practical needs.

As we go, we'll refine the means and define more practical objectives,
but for now, I'd like to hear from every party involved.

I wrote a couple of articles two years ago, describing a generic user's
perspective[2], and a another describing a coder's perspective[3]. I
will send them both to the list for review and discussion.

In any case, I see the GNU/consensus project as a catalyst, and not an
authority. If it becomes an authority, it will be through consensus
within the community, not through imposing our views--we're looking for
consensus, not coercion.

Post by Christian Grothoff
Finally, a little note: there is in GNUnet an API which is called the
"consensus" API (for set reconciliation). That API --- while possibly a
useful building block for social network applications --- should not be
confused with the "GNU Consensus" protocol. I wonder if we need to
rename it to avoid a possible confusion...

*** We do not *own* the name "consensus", and I don't think we're going
to produce a "GNU/consensus API", so I think it's safe to keep it that
way. If we were to produce an API, we would choose another name for it,
because yours exists already, and that would be confusing to use the
same name. ;o)

==
hk

[1]
http://libreplanet.org/wiki/User_talk:Hellekin/A_Coder_Perspective_Of_GNU_Social
[2]
http://libreplanet.org/wiki/User:Hellekin/A_User_Perspective_Of_GNU_Social
[3]
http://libreplanet.org/wiki/User:Hellekin/A_Coder_Perspective_Of_GNU_Social

[Richard Stallman]

Post by Christian Grothoff
Thus, I'd urge you to change the section
title to something more like "Right to Pseudonymity".

*** You make an excellent point. Richard, do you have any objection to
that renaming, as you wrote this section?

Could someone show me that section, and explain the issue again?

[Hugo Roy]

Post by hellekin (GNU Consensus)

Post by Christian Grothoff
Thus, I'd urge you to change the section
title to something more like "Right to Pseudonymity".

*** You make an excellent point. Richard, do you have any objection to
that renaming, as you wrote this section?
Could someone show me that section, and explain the issue again?

The section from the current manifesto:

Anonymity

With interoperating free software social networking systems, no
user will be compelled to provide any particular kind of
information, whether it be her name, her age, or what country
she lives in. It will be up to those she communicates with to
judge what information she chooses to provide or withhold.

Post by hellekin (GNU Consensus)
This is not _quite_ what I believe to be commonly understood by
'anonymity' by most users or the research community. What I think you
mean is more like a 'right to use pseudonyms'. The possibility to
postcontent anonymously is a separate issue (and somewhat incompatible
with mapping social relationships). Furthermore, the idea of making
your online identities unlinkable to real-world identities (which does
relate more to 'anonymity') is again a bit different (and much harder
to achieve, as having access to things like language, time and social
relationships can enable some adversaries to link pseudonyms to
real-world identities).
Given this, _strong_ pseudonymity (as in, user's pseudonyms being
fundamentally unlinkable to real-world identities) or _strong_
anonymity are things we may strive for (at least for some
applications), but should not claim as major features (at this point)
or fundamental properties of GNU consensus. Thus, I'd urge you to
change the section title to something more like "Right to
Pseudonymity".

I think the problem is the title of the section: "Anonymity."
Christian proposes ``Right to Pseudonymity''

[Richard Stallman]

Anonymity

With interoperating free software social networking systems, no
user will be compelled to provide any particular kind of
information, whether it be her name, her age, or what country
she lives in. It will be up to those she communicates with to
judge what information she chooses to provide or withhold.

That's the meaning of anonymity as I understand it.
So I don't see any reason to change this.

Christian proposes ``Right to Pseudonymity''

"Pseudonymity" is not an everyday word. I am not convinced
this distinction is important, and I'd rather not make the text
harder to read.

[Christian Grothoff]

Post by Christian Grothoff
Anonymity
With interoperating free software social networking systems, no
user will be compelled to provide any particular kind of
information, whether it be her name, her age, or what country
she lives in. It will be up to those she communicates with to
judge what information she chooses to provide or withhold.
That's the meaning of anonymity as I understand it.
So I don't see any reason to change this.
Christian proposes ``Right to Pseudonymity''
"Pseudonymity" is not an everyday word. I am not convinced
this distinction is important, and I'd rather not make the text
harder to read.

Just checking google fight on this (how common is either word):

anonymity - 2.4 million results
pseudonym - 1.4 million results
pseudonymity - 16 thousand results

On 'wordcount.org', anonymity ranks 16393, pseudonym at 31330,
pseudonymity is not in the list.


So yes, "pseudonymity" is rarely used, but "pseudonym" is still quite
common. Clearly this is a trade-off between using simple language and
using precise language.

Your call.


-Christian