The pool is shrinking

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
91 messages Options
12345
Reply | Threaded
Open this post in threaded view
|

Re: The pool is shrinking

Robert J. Hansen-3
Mostly this is a response to Arnold, as for some reason his email never
showed up in my inbox:

> I thought SKS and PGP-keys is about one's ability to hide private
> data (by encryption).

Tools do not have intrinsic purposes.  There's the stuff they're
designed for and there's the stuff they actually wind up getting used
for, and very often the two are nothing alike.

The #1 use of OpenPGP today is for Linux distros to verify system
packages.  That accounts for 95% of all OpenPGP usage -- maybe more.

Tools are just tools.  We, we human beings, are the ones who have
purposes and ambitions and goals.

> GDPR is also about one's ability to hide private data

They are different far more than they are similar.

If I use OpenPGP to secure my communications, I'm not imposing anything
on people who acquire my communications.  If they can break the crypto,
go for it.  If they can't, tough luck.  But I'm not telling people who
already have the data, "oh sorry, you can't have it now."

The GDPR is completely different.  You can give me your personal
information.  I can give you complete up-front disclosure about what
you're getting into.  You can review it, you can decide that yes you
want to do this, you can give me your data... and then, ten years later,
you can force me "hey, I changed my mind, you've got to erase data now."

The OpenPGP model *compels absolutely no one*.  GDPR is built around the
idea *the EU has the right to compel people to delete data.*

I'm an American.  If the EU thinks it has the right to compel me to obey
a law I had no say in, well, good luck.

> To me, it is very strange to read one strongly supports one form of
> privacy, while totally ignoring other forms.

Then I think you really need to study ethics.

*How we do something* is just as important as *what it is we do*.  I
think there's a lot to be said about pursuing privacy in a way that
imposes no obligations on any other people.  And I think there's a lot
to be said against pursuing privacy in a way that imposes obligations on
people who don't even live in the EU.

> Remember, people in different parts of the world do have different
> values and different needs.

Yep.  And in America, we value our right to be left alone from the
government telling us that we're required to take certain acts just
because some people in Europe insist we follow their laws.

_______________________________________________
Sks-devel mailing list
[hidden email]
https://lists.nongnu.org/mailman/listinfo/sks-devel
Reply | Threaded
Open this post in threaded view
|

Re: The pool is shrinking

Steffen Kaiser
In reply to this post by Stefan Claas
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On Thu, 15 Aug 2019, Stefan Claas wrote:

> And has Mr. Rude then the right to freely distribute this data, without

"this data" => SKS stores private data, which are public by design and in
consens with the uploader, Art5 (1) a) and b)
Those dumps are used to full fill the purpose, intended by the SKS network
and intended to be processed by the uploader
The SKS servers fullfill the well-known purpose of making these data
available publically.

> protecting it, to the whole world? If that is the case then EU citizens
> having 'business' with the US can do the same with US citizens data.

Yes, you, personally, can dump the *SKS* database and make it available
yourself as well.

- --
Steffen Kaiser
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1

iQEVAwUBXVZQMiOjcd6avHZPAQIviQf/a9EjsJQUCZCILEdZTY+YuxFnjeJx+CDc
+9X3d52NLeYp8mBQFdRLSNMsBneDqHye+e7QFjcyE1R7aOgEe1/Cawzht7h8Fuu8
gs1ijA/l/Hdc0sy7uxBuEWA/mSrnyldwaxnNvInRz1GvDuxcmw48y74d20Gn/L8u
JpnemKYjeF2CssQRjN//kEJGweNMsVpuGjLSTSxJDigp0AFXXGBWsL4wyJv4BcPB
dpvsJ8tre7iyJoJVugT20oLs4V4EAAmKSCXDyJr1oJFtCdda6q8ii523QkEfb8hD
/aL3pJdAsxUz9WtHwTu0qrqEKPMkZqYaWPu7+hbrMlOTqj+4yfafBg==
=bwv8
-----END PGP SIGNATURE-----

_______________________________________________
Sks-devel mailing list
[hidden email]
https://lists.nongnu.org/mailman/listinfo/sks-devel
Reply | Threaded
Open this post in threaded view
|

Re: The pool is shrinking

Stefan Claas
Steffen Kaiser wrote:

> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> On Thu, 15 Aug 2019, Stefan Claas wrote:
>
> > And has Mr. Rude then the right to freely distribute this data, without
>
> "this data" => SKS stores private data, which are public by design and in
> consens with the uploader, Art5 (1) a) and b)

By what uploader? My current key wasn't uploaded by me and I gave no consent,
to whoever uploaded my key.

Regards
Stefan

--
box: 4a64758de9e8ceded2c481ee526440687fe2f3a828e3a813f87753ad30847b56
GPG: C93E252DFB3B4DB7EAEB846AD8D464B35E12AB77 (avail. on Hagrid, WKD)

_______________________________________________
Sks-devel mailing list
[hidden email]
https://lists.nongnu.org/mailman/listinfo/sks-devel
Reply | Threaded
Open this post in threaded view
|

Re: The pool is shrinking

stuff
In reply to this post by Steffen Kaiser
Once again pointing out the obvious that everyone is avoiding.

The keyservers don't have any mechanisms as required by the GDPR to remove data.

So once again if you load up someone else's personal data with out permission the servers instantly break the law due to the lack of those mechanisms. This is the simplest one to point out, Among many other issues. There is no exemption to this one at all!!!

Hansen its 2019 not 1990 and you need to evolve your thinking beyond your own personal interests! Do you think the GDPR is a bad thing? Do you think people having the right to better privacy is bad? from your resent responses you obviously do, strange attitude considering your interest in privacy????

-----------------------

>Its about pretty good privacy, not perfect privacy.. by design w/PGP and
>SKS, public keys are designed to be public, and not private.. in order to
>keep the private part secure, allowing people to arbitrary purge public
>data entirely undermines the entire thing.

And to Ryan, poor response! Also the world changes and laws change and peoples views of what is right and wrong change. And that's exactly what has happened especially in Europe! The sks keyservers where designed in the 1990s, its not 1990 any more. People think differently about privacy now. Hagrid or Keybase have solved issues for a majority of people.

It does not undermine it at all, this model is broken and its being laughed at by the entire tech community. Oh and it was never resilient to government interference that was just a fallacy which has been push right into the spot light. a single person or group just bitched slapped the sks keyservers recently with an attack, all it takes is someone to persist with a real attack and those are gone! NO RELIABLITIY, NO RESILIANCE....NO USE!!

-------------

The SKS Keyservers have brought a very bad light on GnuPG and other related projects, trust for most is low or gone in these projects, and people like Hansen and his approach to it has really not helped at all. Kristian meanwhile the maintainer remains quiet, not even making any attempts to suggest shutting down the servers or archiving the software.

The important Questions here for admins is :

Do you want to continue to:

1. Run broken and unreliable software?
2. Risk legal consequences?
3. be the laughing stock of modern security?

Kind regards

Yakamo


On Fri, 16 Aug 2019 08:41:53 +0200 (CEST)
Steffen Kaiser <[hidden email]> wrote:

> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> On Thu, 15 Aug 2019, Stefan Claas wrote:
>
> > And has Mr. Rude then the right to freely distribute this data, without
>
> "this data" => SKS stores private data, which are public by design and in
> consens with the uploader, Art5 (1) a) and b)
> Those dumps are used to full fill the purpose, intended by the SKS network
> and intended to be processed by the uploader
> The SKS servers fullfill the well-known purpose of making these data
> available publically.
>
> > protecting it, to the whole world? If that is the case then EU citizens
> > having 'business' with the US can do the same with US citizens data.
>
> Yes, you, personally, can dump the *SKS* database and make it available
> yourself as well.
>
> - --
> Steffen Kaiser
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1
>
> iQEVAwUBXVZQMiOjcd6avHZPAQIviQf/a9EjsJQUCZCILEdZTY+YuxFnjeJx+CDc
> +9X3d52NLeYp8mBQFdRLSNMsBneDqHye+e7QFjcyE1R7aOgEe1/Cawzht7h8Fuu8
> gs1ijA/l/Hdc0sy7uxBuEWA/mSrnyldwaxnNvInRz1GvDuxcmw48y74d20Gn/L8u
> JpnemKYjeF2CssQRjN//kEJGweNMsVpuGjLSTSxJDigp0AFXXGBWsL4wyJv4BcPB
> dpvsJ8tre7iyJoJVugT20oLs4V4EAAmKSCXDyJr1oJFtCdda6q8ii523QkEfb8hD
> /aL3pJdAsxUz9WtHwTu0qrqEKPMkZqYaWPu7+hbrMlOTqj+4yfafBg==
> =bwv8
> -----END PGP SIGNATURE-----
>
> _______________________________________________
> Sks-devel mailing list
> [hidden email]
> https://lists.nongnu.org/mailman/listinfo/sks-devel


--


_______________________________________________
Sks-devel mailing list
[hidden email]
https://lists.nongnu.org/mailman/listinfo/sks-devel
Reply | Threaded
Open this post in threaded view
|

Re: The pool is shrinking

Ryan Hunt-3
Yakamo, 
it still does its job of ensuring published keys are not tampered with, it was not designed to be resilient to denial attacks.. That does not interfere with the trust of PGP, its why there are local keystores.. and the SKS network is still around despite being unreliable/broken from a maintenance standpoint.. your poisoned keys are not altering other individuals keys in any way/shape/form, so its security has not been compromised.. availability of keyservers is not critical to the use of PGP, again by design.. there are many ways to distribute keys, it is resilient factually despite your opinions.. over the decades the need has not been lost.

You could not be more wrong about GnuPG, and it shows.. do you even work in the industry? Because where I sit, with over 54 million devices on my network.. PGP is one of the most trusted security tools we use, all of our software is signed by PGP, config files are signed by PGP, internal correspondence signed by PGP.. You are the only person in the world claiming GnuPG has lost its trust and you can write all the blog posts you want but your opinion means nothing to me, and the rest of the industry.. Snowden and all the other security industry's rock stars still fully advocate the use of PGP despite your feeble attacks. 

So to answer your questions:
1. Currently, its the only option until something better comes along.
2. There are absolutely none, but you seem to be beyond reason on this point so I digress. 
3. This is entirely arbitrary, not everyone has to share your perspective.. Most of the industry rallied against the GDPR, if anything the EU/Australia has become the laughing stock of the cryptography world.. you guys would give up master keys and implement backdoors to your government in exchange for a cookie and a pat on the back.

-Ryan

On Fri, Aug 16, 2019 at 8:08 AM <[hidden email]> wrote:
Once again pointing out the obvious that everyone is avoiding.

The keyservers don't have any mechanisms as required by the GDPR to remove data.

So once again if you load up someone else's personal data with out permission the servers instantly break the law due to the lack of those mechanisms. This is the simplest one to point out, Among many other issues. There is no exemption to this one at all!!!

Hansen its 2019 not 1990 and you need to evolve your thinking beyond your own personal interests! Do you think the GDPR is a bad thing? Do you think people having the right to better privacy is bad? from your resent responses you obviously do, strange attitude considering your interest in privacy????

-----------------------

>Its about pretty good privacy, not perfect privacy.. by design w/PGP and
>SKS, public keys are designed to be public, and not private.. in order to
>keep the private part secure, allowing people to arbitrary purge public
>data entirely undermines the entire thing.

And to Ryan, poor response! Also the world changes and laws change and peoples views of what is right and wrong change. And that's exactly what has happened especially in Europe! The sks keyservers where designed in the 1990s, its not 1990 any more. People think differently about privacy now. Hagrid or Keybase have solved issues for a majority of people.

It does not undermine it at all, this model is broken and its being laughed at by the entire tech community. Oh and it was never resilient to government interference that was just a fallacy which has been push right into the spot light. a single person or group just bitched slapped the sks keyservers recently with an attack, all it takes is someone to persist with a real attack and those are gone! NO RELIABLITIY, NO RESILIANCE....NO USE!!

-------------

The SKS Keyservers have brought a very bad light on GnuPG and other related projects, trust for most is low or gone in these projects, and people like Hansen and his approach to it has really not helped at all. Kristian meanwhile the maintainer remains quiet, not even making any attempts to suggest shutting down the servers or archiving the software.

The important Questions here for admins is :

Do you want to continue to:

1. Run broken and unreliable software?
2. Risk legal consequences?
3. be the laughing stock of modern security?

Kind regards

Yakamo


On Fri, 16 Aug 2019 08:41:53 +0200 (CEST)
Steffen Kaiser <[hidden email]> wrote:

> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> On Thu, 15 Aug 2019, Stefan Claas wrote:
>
> > And has Mr. Rude then the right to freely distribute this data, without
>
> "this data" => SKS stores private data, which are public by design and in
> consens with the uploader, Art5 (1) a) and b)
> Those dumps are used to full fill the purpose, intended by the SKS network
> and intended to be processed by the uploader
> The SKS servers fullfill the well-known purpose of making these data
> available publically.
>
> > protecting it, to the whole world? If that is the case then EU citizens
> > having 'business' with the US can do the same with US citizens data.
>
> Yes, you, personally, can dump the *SKS* database and make it available
> yourself as well.
>
> - --
> Steffen Kaiser
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1
>
> iQEVAwUBXVZQMiOjcd6avHZPAQIviQf/a9EjsJQUCZCILEdZTY+YuxFnjeJx+CDc
> +9X3d52NLeYp8mBQFdRLSNMsBneDqHye+e7QFjcyE1R7aOgEe1/Cawzht7h8Fuu8
> gs1ijA/l/Hdc0sy7uxBuEWA/mSrnyldwaxnNvInRz1GvDuxcmw48y74d20Gn/L8u
> JpnemKYjeF2CssQRjN//kEJGweNMsVpuGjLSTSxJDigp0AFXXGBWsL4wyJv4BcPB
> dpvsJ8tre7iyJoJVugT20oLs4V4EAAmKSCXDyJr1oJFtCdda6q8ii523QkEfb8hD
> /aL3pJdAsxUz9WtHwTu0qrqEKPMkZqYaWPu7+hbrMlOTqj+4yfafBg==
> =bwv8
> -----END PGP SIGNATURE-----
>
> _______________________________________________
> Sks-devel mailing list
> [hidden email]
> https://lists.nongnu.org/mailman/listinfo/sks-devel


--


_______________________________________________
Sks-devel mailing list
[hidden email]
https://lists.nongnu.org/mailman/listinfo/sks-devel

_______________________________________________
Sks-devel mailing list
[hidden email]
https://lists.nongnu.org/mailman/listinfo/sks-devel
Reply | Threaded
Open this post in threaded view
|

Re: The pool is shrinking

Stefan Claas
In reply to this post by stuff
[hidden email] wrote:

> Once again pointing out the obvious that everyone is avoiding.
>
> The keyservers don't have any mechanisms as required by the GDPR to remove
> data.
>
> So once again if you load up someone else's personal data with out permission
> the servers instantly break the law due to the lack of those mechanisms. This
> is the simplest one to point out, Among many other issues. There is no
> exemption to this one at all!!!
>
> Hansen its 2019 not 1990 and you need to evolve your thinking beyond your own
> personal interests! Do you think the GDPR is a bad thing? Do you think people
> having the right to better privacy is bad? from your resent responses you
> obviously do, strange attitude considering your interest in privacy????
>
> -----------------------
>
> >Its about pretty good privacy, not perfect privacy.. by design w/PGP and
> >SKS, public keys are designed to be public, and not private.. in order to
> >keep the private part secure, allowing people to arbitrary purge public
> >data entirely undermines the entire thing.
>
> And to Ryan, poor response! Also the world changes and laws change and
> peoples views of what is right and wrong change. And that's exactly what has
> happened especially in Europe! The sks keyservers where designed in the
> 1990s, its not 1990 any more. People think differently about privacy now.
> Hagrid or Keybase have solved issues for a majority of people.
>
> It does not undermine it at all, this model is broken and its being laughed
> at by the entire tech community. Oh and it was never resilient to government
> interference that was just a fallacy which has been push right into the spot
> light. a single person or group just bitched slapped the sks keyservers
> recently with an attack, all it takes is someone to persist with a real
> attack and those are gone! NO RELIABLITIY, NO RESILIANCE....NO USE!!
>
> -------------
>
> The SKS Keyservers have brought a very bad light on GnuPG and other related
> projects, trust for most is low or gone in these projects, and people like
> Hansen and his approach to it has really not helped at all. Kristian
> meanwhile the maintainer remains quiet, not even making any attempts to
> suggest shutting down the servers or archiving the software.
>
> The important Questions here for admins is :
>
> Do you want to continue to:
>
> 1. Run broken and unreliable software?
> 2. Risk legal consequences?
> 3. be the laughing stock of modern security?
>

1+

I would also like to make a little suggestion, to put a little bit
more trust in this broken SKS design, for people who still might
have a need for SKS usage.

How about to issue  monthly warrant canaries from SKS operators for
the pool, Kristian maintains?

It should tell users that SKS operators share no dumps with 3rd
parties for key analysis, i.e. social graph research etc. Those
who publish a warrant canary can stay in the pool, while others
who don't like to do so will be excluded from the pool.

Does this makes sense to honest operators? I think it would not
hurt and requires no additional work, except a monthly little
GnuPG signed statement.

Regards
Stefan

--
box: 4a64758de9e8ceded2c481ee526440687fe2f3a828e3a813f87753ad30847b56
GPG: C93E252DFB3B4DB7EAEB846AD8D464B35E12AB77 (avail. on Hagrid, WKD)

_______________________________________________
Sks-devel mailing list
[hidden email]
https://lists.nongnu.org/mailman/listinfo/sks-devel
Reply | Threaded
Open this post in threaded view
|

Re: The pool is shrinking

Andrew Gallagher
On 16/08/2019 16:13, Stefan Claas wrote:
> It should tell users that SKS operators share no dumps with 3rd
> parties for key analysis, i.e. social graph research etc. Those
> who publish a warrant canary can stay in the pool, while others
> who don't like to do so will be excluded from the pool.

That's an utterly worthless exercise, considering that keyserver
operators can't vouch for any other keyserver operators, and any or all
of them could be three-letter agencies in disguise. You don't need a
warrant to scrape publicly-available data, and you don't need to be in
the pool to sync with pool keyservers.

A

--
Andrew Gallagher


_______________________________________________
Sks-devel mailing list
[hidden email]
https://lists.nongnu.org/mailman/listinfo/sks-devel

signature.asc (849 bytes) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: The pool is shrinking

H Visage


> On 16 Aug 2019, at 18:01 , Andrew Gallagher <[hidden email]> wrote:
>
> Signed PGP part
> On 16/08/2019 16:13, Stefan Claas wrote:
>> It should tell users that SKS operators share no dumps with 3rd
>> parties for key analysis, i.e. social graph research etc. Those
>> who publish a warrant canary can stay in the pool, while others
>> who don't like to do so will be excluded from the pool.
>
> That's an utterly worthless exercise, considering that keyserver
> operators can't vouch for any other keyserver operators, and any or all
> of them could be three-letter agencies in disguise. You don't need a
> warrant to scrape publicly-available data, and you don't need to be in
> the pool to sync with pool keyservers.
Not to mention that the latest dumps are publicly available for syncing purposes...


_______________________________________________
Sks-devel mailing list
[hidden email]
https://lists.nongnu.org/mailman/listinfo/sks-devel

signature.asc (499 bytes) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: The pool is shrinking

stuff
In reply to this post by Ryan Hunt-3
On Fri, 16 Aug 2019 09:12:30 -0600
Ryan Hunt <[hidden email]> wrote:

> Yakamo,
> it still does its job of ensuring published keys are not tampered with, it
> was not designed to be resilient to denial attacks.. That does not
> interfere with the trust of PGP, its why there are local keystores.. and
> the SKS network is still around despite being unreliable/broken from a
> maintenance standpoint.. your poisoned keys are not altering other
> individuals keys in any way/shape/form, so its security has not been
> compromised.. availability of keyservers is not critical to the use of PGP,
> again by design.. there are many ways to distribute keys, it is resilient
> factually despite your opinions.. over the decades the need has not been
> lost.
>

That's correct its not designed to be resilient to denial attacks, making it unreliable as stated before! which means its not resilient to governments at all! This statement stands true. Now it barely fulfils its basic functions! the amount of posts littered over the internet about how people cant pull a key from the servers or unable to upload them. There are constant outages!

There are alternatives and they work! sks doesn't!

Its not the design or the attacks that's for me personally and others distrustful its the, closed minded approach to how vulnerabilities are handled, both people from the GnuPG community and SKS have attacked people for what's considered normal practice when it comes to disclosure of vulnerabilities and bugs. "stay quiet and hope nothing happens" or "your attacking us because you pointed out something wrong with our software" is not a good way to deal with things!

Also do you think its good Mr Hunt that data can be uploaded onto these servers such as peoples personal information without consent? This has happened to a lot of people. And yet no one is interested in addressing this!

> You could not be more wrong about GnuPG, and it shows.. do you even work in
> the industry? Because where I sit, with over 54 million devices on my
> network.. PGP is one of the most trusted security tools we use, all of our
> software is signed by PGP, config files are signed by PGP, internal
> correspondence signed by PGP.. You are the only person in the world
> claiming GnuPG has lost its trust and you can write all the blog posts you
> want but your opinion means nothing to me, and the rest of the industry..
> Snowden and all the other security industry's rock stars still fully
> advocate the use of PGP despite your feeble attacks.

Are we really comparing "network" size?

I didn't say it was not in demand or general use in the security community! or unpopular!
Although I come across very few people who actually use it these days and who are not middle aged. Even FreeBSD stopped using it who knows how long ago for signing packages.

Like wise your opinion holds no value to me either.


> So to answer your questions:
> 1. Currently, its the only option until something better comes along.

Keybase and Hagrid or self hosting your gpg key, plenty of options.

> 2. There are absolutely none, but you seem to be beyond reason on this
> point so I digress.

There's plenty why you claim none im not sure, maybe we should test this theory of yours?

> 3. This is entirely arbitrary, not everyone has to share your perspective..
> Most of the industry rallied against the GDPR, if anything the EU/Australia
> has become the laughing stock of the cryptography world.. you guys would
> give up master keys and implement backdoors to your government in exchange
> for a cookie and a pat on the back.

Of course big company's rallied against the GDPR, it gives users their privacy back again!
This messes with their business model!

And are you against the GDPR?
Do you even know what the GDPR covers?

what has Australia got to do with this?

and where are you from Mr Hunt? America?

Kind Regards

Yakamo


--


_______________________________________________
Sks-devel mailing list
[hidden email]
https://lists.nongnu.org/mailman/listinfo/sks-devel
Reply | Threaded
Open this post in threaded view
|

Re: The pool is shrinking

Stefan Claas
In reply to this post by Andrew Gallagher
Andrew Gallagher wrote:

> On 16/08/2019 16:13, Stefan Claas wrote:
> > It should tell users that SKS operators share no dumps with 3rd
> > parties for key analysis, i.e. social graph research etc. Those
> > who publish a warrant canary can stay in the pool, while others
> > who don't like to do so will be excluded from the pool.
>
> That's an utterly worthless exercise, considering that keyserver
> operators can't vouch for any other keyserver operators, and any or all
> of them could be three-letter agencies in disguise. You don't need a
> warrant to scrape publicly-available data, and you don't need to be in
> the pool to sync with pool keyservers.

They should not vouch for others only sign a monthly statement, which
Kristian can add to a column in his pool site. And like I said it
would *not* hurt! If however more key server operators  are against
this suggestion then IMHO at least people know that operators may
like to share dumps with 3rd parties, for whatever reasons they may
have.

And I think with all the discussions about the GDPR it would be also
a good sign from SKS operators doing so, while Kristian had it then
easier to maintain his pool, when getting asked questions in the future.

And addional feature could be that SKS operators implement a new
version of the .html search code, for their WWW interface, only allowing
fingerprint search, without revealing a long list of who signed whoms
pub key.

Regards
Stefan

--
box: 4a64758de9e8ceded2c481ee526440687fe2f3a828e3a813f87753ad30847b56
GPG: C93E252DFB3B4DB7EAEB846AD8D464B35E12AB77 (avail. on Hagrid, WKD)

_______________________________________________
Sks-devel mailing list
[hidden email]
https://lists.nongnu.org/mailman/listinfo/sks-devel
Reply | Threaded
Open this post in threaded view
|

Re: The pool is shrinking

Stefan Claas
In reply to this post by H Visage
Hendrik Visage wrote:

>
>
> > On 16 Aug 2019, at 18:01 , Andrew Gallagher <[hidden email]> wrote:
> >
> > Signed PGP part
> > On 16/08/2019 16:13, Stefan Claas wrote:
> >> It should tell users that SKS operators share no dumps with 3rd
> >> parties for key analysis, i.e. social graph research etc. Those
> >> who publish a warrant canary can stay in the pool, while others
> >> who don't like to do so will be excluded from the pool.
> >
> > That's an utterly worthless exercise, considering that keyserver
> > operators can't vouch for any other keyserver operators, and any or all
> > of them could be three-letter agencies in disguise. You don't need a
> > warrant to scrape publicly-available data, and you don't need to be in
> > the pool to sync with pool keyservers.
>
> Not to mention that the latest dumps are publicly available for syncing
> purposes...

This is well known and Kristian could exclude this person from the pool.

Regards
Stefan

--
box: 4a64758de9e8ceded2c481ee526440687fe2f3a828e3a813f87753ad30847b56
GPG: C93E252DFB3B4DB7EAEB846AD8D464B35E12AB77 (avail. on Hagrid, WKD)

_______________________________________________
Sks-devel mailing list
[hidden email]
https://lists.nongnu.org/mailman/listinfo/sks-devel
Reply | Threaded
Open this post in threaded view
|

Re: The pool is shrinking

Andrew Gallagher
In reply to this post by Stefan Claas
On 16/08/2019 17:23, Stefan Claas wrote:
> They should not vouch for others only sign a monthly statement, which
> Kristian can add to a column in his pool site. And like I said it
> would *not* hurt! If however more key server operators  are against
> this suggestion then IMHO at least people know that operators may
> like to share dumps with 3rd parties, for whatever reasons they may
> have.

What would that statement say? "I don't explicitly provide dumps to
third parties, but I sync my keyserver with random strangers on the
internet, and that's *totally* not *exactly* the same thing." ? :-P

The only way to prevent the SKS dataset getting into arbitrary people's
hands is for all the existing keyservers to refuse to sync with anyone
who's not vouched-for and trustworthy - which will turn the keyservers
into a closed network. That may or may not be *wrong*, but it would be a
fundamental change to the entire premise of the system.

And as Hendrik pointed out above, you can't bootstrap a new SKS
keyserver without a dump.

--
Andrew Gallagher


_______________________________________________
Sks-devel mailing list
[hidden email]
https://lists.nongnu.org/mailman/listinfo/sks-devel

signature.asc (849 bytes) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: The pool is shrinking

DevPGSV Pablo
In reply to this post by Stefan Claas
Please, if you consider the canary...
Don't centralize it.
SKS is supposed to be decentralized.
If the power to ban someone from the public pool or from peers, or somewhere else, falls on a single person, then you are centralizing an important part of SKS.
SKS is already broken due to the attacks... If you centralize it, then it loses the only thing the other alternatives do not have.


On Fri, Aug 16, 2019, 6:25 PM Stefan Claas <[hidden email]> wrote:
Hendrik Visage wrote:

>
>
> > On 16 Aug 2019, at 18:01 , Andrew Gallagher <[hidden email]> wrote:
> >
> > Signed PGP part
> > On 16/08/2019 16:13, Stefan Claas wrote:
> >> It should tell users that SKS operators share no dumps with 3rd
> >> parties for key analysis, i.e. social graph research etc. Those
> >> who publish a warrant canary can stay in the pool, while others
> >> who don't like to do so will be excluded from the pool.
> >
> > That's an utterly worthless exercise, considering that keyserver
> > operators can't vouch for any other keyserver operators, and any or all
> > of them could be three-letter agencies in disguise. You don't need a
> > warrant to scrape publicly-available data, and you don't need to be in
> > the pool to sync with pool keyservers.
>
> Not to mention that the latest dumps are publicly available for syncing
> purposes...

This is well known and Kristian could exclude this person from the pool.

Regards
Stefan

--
box: 4a64758de9e8ceded2c481ee526440687fe2f3a828e3a813f87753ad30847b56
GPG: C93E252DFB3B4DB7EAEB846AD8D464B35E12AB77 (avail. on Hagrid, WKD)

_______________________________________________
Sks-devel mailing list
[hidden email]
https://lists.nongnu.org/mailman/listinfo/sks-devel

_______________________________________________
Sks-devel mailing list
[hidden email]
https://lists.nongnu.org/mailman/listinfo/sks-devel
Reply | Threaded
Open this post in threaded view
|

Re: The pool is shrinking

Kiss Gabor (Bitman)
In reply to this post by Kiss Gabor (Bitman)
> So to answer your questions:

Ryan, have you ever seen this funny picture? :)
http://en.wikipedia.org/wiki/File:DoNotFeedTroll.svg

Gabor

_______________________________________________
Sks-devel mailing list
[hidden email]
https://lists.nongnu.org/mailman/listinfo/sks-devel
Reply | Threaded
Open this post in threaded view
|

Re: The pool is shrinking

Todd Fleisher
> On Aug 16, 2019, at 10:19 AM, Kiss Gabor (Bitman) <[hidden email]> wrote:
>
>> So to answer your questions:
>
> Ryan, have you ever seen this funny picture? :)
> http://en.wikipedia.org/wiki/File:DoNotFeedTroll.svg
>
> Gabor


+1 to this sentiment

If some really want to continue to debate particulars of the GDPR, I’d ask that they do it directly and off the list.

-T


_______________________________________________
Sks-devel mailing list
[hidden email]
https://lists.nongnu.org/mailman/listinfo/sks-devel

signature.asc (849 bytes) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: The pool is shrinking

Stefan Claas
In reply to this post by DevPGSV Pablo
DevPGSV Pablo wrote:

> Please, if you consider the canary...
> Don't centralize it.
> SKS is supposed to be decentralized.
> If the power to ban someone from the public pool or from peers, or
> somewhere else, falls on a single person, then you are centralizing an
> important part of SKS.
> SKS is already broken due to the attacks... If you centralize it, then it
> loses the only thing the other alternatives do not have.

O.k. I must admit I did not thought about the centralization issue,
people might have.

Well, then operators could put that on a link of their own WWW key
server interface and Kristian could add only a column in his pool
page, indicating that server x,y,z has a canary, without taking any
actions.

Even if Kristian does not like the idea than at least people could
see that operators are willing to support the idea.

Regards
Stefan

--
box: 4a64758de9e8ceded2c481ee526440687fe2f3a828e3a813f87753ad30847b56
GPG: C93E252DFB3B4DB7EAEB846AD8D464B35E12AB77 (avail. on Hagrid, WKD)

_______________________________________________
Sks-devel mailing list
[hidden email]
https://lists.nongnu.org/mailman/listinfo/sks-devel
Reply | Threaded
Open this post in threaded view
|

Re: The pool is shrinking

Ryan Hunt-3
In reply to this post by stuff
SKS is still resilient to anyone wiping out all references to my public key and replacing with their own for a man in the middle attack, you can go check multiple servers and compare keys against each other.. I can check keys in my local keystore or transmitted via other means against whats in SKS, its also resilient to keys being removed to prevent verifying data signed long ago.. none of that has changed, you can attack the whole network but its integrity is still intact when it comes back up.. 

Its role as a decentralized, tamper resistant key storage solution is still vital, and I would love it if we had the development going on to address the stability issues, but thats simply not the case at this point in time and until the actual integrity of the data the SKS network serves is compromised there is no need for its death.. yes there are alternatives, but those wont force enforcement of your precious GDPR, I can host all the same keys any way I want and ignore all your requests for removal just the same so your argument attacking SKS specifically is moot.

> Also do you think its good Mr Hunt that data can be uploaded onto these servers such as peoples personal information without consent? This has happened to a lot of people. And yet no one is interested in addressing this!
I've proposed solutions to simply add more sanity/validation checks to make sure keys are actual valid keys and limiting the overall size of keys to prevent abuse, but overall I'm not terribly concerned.. there's a billion places to make information public on the internet that is entirely out of reach of your local authorities, SKS is rather ineffective means of making information public since practically nobody is looking at the dataset as a whole and are only querying information directly, and almost always automated.. You are basically Gaslighting at this point.

> And are you against the GDPR?
Correct, the GDPR would be ruled unconstitutional in a heartbeat if someone tried to implement it here.

> Do you even know what the GDPR covers?
Yes, quite well.. I unfortunately work with many forms of Digital Compliance in my industry.

> what has Australia got to do with this?
Just another example of the road to hell is paved with good intentions.. Its a slippery slope you guys are already sliding down.. I can only think of one operator that was forced to shut down for being liable for data others posted publicly, and that was an Australian operator.. long before the GDPR was drafted.. and nothing was accomplished, the data they tried to take out of the public sphere still exists.. again SKS worked as designed, the government was unable to stop the distribution of that data.. and its still accessible, even within Australia. 

> and where are you from Mr Hunt? America?
Yes, Colorado to be precise if you need to figure out what court to waste your time with.

> There's plenty why you claim none im not sure, maybe we should test this theory of yours?
Go for it, I am completely willing to face any government and the resulting consequences to protect the integrity and availability of public cryptography, if my government were to ever insist on compromising it again in the future I would make it my mission to distribute the tools and spread awareness despite any legal ramifications or any moral perspective, yeah I might be assisting terrorists, child abusers, and other boogiemen; but thats the price of cryptographically secure communications. The EU can bring it on for all I care, this is a hill I'm fully prepared to die on, and have been for a while.. I advocated for and distributed the tools 30 years ago when strong crypto was illegal to export from the United States, and eventually we won that battle of attrition. 

-R



On Fri, Aug 16, 2019 at 10:12 AM <[hidden email]> wrote:
On Fri, 16 Aug 2019 09:12:30 -0600
Ryan Hunt <[hidden email]> wrote:

> Yakamo,
> it still does its job of ensuring published keys are not tampered with, it
> was not designed to be resilient to denial attacks.. That does not
> interfere with the trust of PGP, its why there are local keystores.. and
> the SKS network is still around despite being unreliable/broken from a
> maintenance standpoint.. your poisoned keys are not altering other
> individuals keys in any way/shape/form, so its security has not been
> compromised.. availability of keyservers is not critical to the use of PGP,
> again by design.. there are many ways to distribute keys, it is resilient
> factually despite your opinions.. over the decades the need has not been
> lost.
>

That's correct its not designed to be resilient to denial attacks, making it unreliable as stated before! which means its not resilient to governments at all! This statement stands true. Now it barely fulfils its basic functions! the amount of posts littered over the internet about how people cant pull a key from the servers or unable to upload them. There are constant outages!

There are alternatives and they work! sks doesn't!

Its not the design or the attacks that's for me personally and others distrustful its the, closed minded approach to how vulnerabilities are handled, both people from the GnuPG community and SKS have attacked people for what's considered normal practice when it comes to disclosure of vulnerabilities and bugs. "stay quiet and hope nothing happens" or "your attacking us because you pointed out something wrong with our software" is not a good way to deal with things!

Also do you think its good Mr Hunt that data can be uploaded onto these servers such as peoples personal information without consent? This has happened to a lot of people. And yet no one is interested in addressing this!

> You could not be more wrong about GnuPG, and it shows.. do you even work in
> the industry? Because where I sit, with over 54 million devices on my
> network.. PGP is one of the most trusted security tools we use, all of our
> software is signed by PGP, config files are signed by PGP, internal
> correspondence signed by PGP.. You are the only person in the world
> claiming GnuPG has lost its trust and you can write all the blog posts you
> want but your opinion means nothing to me, and the rest of the industry..
> Snowden and all the other security industry's rock stars still fully
> advocate the use of PGP despite your feeble attacks.

Are we really comparing "network" size?

I didn't say it was not in demand or general use in the security community! or unpopular!
Although I come across very few people who actually use it these days and who are not middle aged. Even FreeBSD stopped using it who knows how long ago for signing packages.

Like wise your opinion holds no value to me either.


> So to answer your questions:
> 1. Currently, its the only option until something better comes along.

Keybase and Hagrid or self hosting your gpg key, plenty of options.

> 2. There are absolutely none, but you seem to be beyond reason on this
> point so I digress.

There's plenty why you claim none im not sure, maybe we should test this theory of yours?

> 3. This is entirely arbitrary, not everyone has to share your perspective..
> Most of the industry rallied against the GDPR, if anything the EU/Australia
> has become the laughing stock of the cryptography world.. you guys would
> give up master keys and implement backdoors to your government in exchange
> for a cookie and a pat on the back.

Of course big company's rallied against the GDPR, it gives users their privacy back again!
This messes with their business model!

And are you against the GDPR?
Do you even know what the GDPR covers?

what has Australia got to do with this?

and where are you from Mr Hunt? America?

Kind Regards

Yakamo


--


_______________________________________________
Sks-devel mailing list
[hidden email]
https://lists.nongnu.org/mailman/listinfo/sks-devel
Reply | Threaded
Open this post in threaded view
|

Re: The pool is shrinking

Todd Fleisher
In reply to this post by Stefan Claas
> On Aug 16, 2019, at 10:24 AM, Stefan Claas <[hidden email]> wrote:

>
> DevPGSV Pablo wrote:
>
> O.k. I must admit I did not thought about the centralization issue,
> people might have.
>
> Well, then operators could put that on a link of their own WWW key
> server interface and Kristian could add only a column in his pool
> page, indicating that server x,y,z has a canary, without taking any
> actions.
>
> Even if Kristian does not like the idea than at least people could
> see that operators are willing to support the idea.
I don’t understand what benefit having operators post warrant canaries is supposed to provide in the context of the SKS network.

-T


_______________________________________________
Sks-devel mailing list
[hidden email]
https://lists.nongnu.org/mailman/listinfo/sks-devel

signature.asc (849 bytes) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: The pool is shrinking

stuff
In reply to this post by Ryan Hunt-3
I guess we will have to agree to disagree, i dont think we will get any further with this.
I fully support the GDPR as do alot of people in the EU.

You have made it clear where you stand as i have.

Still not sure what Austraila has to do with the EU and the GDPR though?

But i do know that the backdoor law implemented in Austraila is part of the 5 eyes progam which the US is part of and mostly spear heads these operations as it does in the UK as well. America is certainly not a budding example of privacy!


Just one correction keys.openpgp.org is fully GDPR compliant for those in the EU.

Kind Regards

Yakamo

On Fri, 16 Aug 2019 11:42:31 -0600
Ryan Hunt <[hidden email]> wrote:

> SKS is still resilient to anyone wiping out all references to my public key
> and replacing with their own for a man in the middle attack, you can go
> check multiple servers and compare keys against each other.. I can check
> keys in my local keystore or transmitted via other means against whats in
> SKS, its also resilient to keys being removed to prevent verifying data
> signed long ago.. none of that has changed, you can attack the whole
> network but its integrity is still intact when it comes back up..
>
> Its role as a decentralized, tamper resistant key storage solution is still
> vital, and I would love it if we had the development going on to address
> the stability issues, but thats simply not the case at this point in time
> and until the actual integrity of the data the SKS network serves is
> compromised there is no need for its death.. yes there are alternatives,
> but those wont force enforcement of your precious GDPR, I can host all the
> same keys any way I want and ignore all your requests for removal just the
> same so your argument attacking SKS specifically is moot.
>
> > Also do you think its good Mr Hunt that data can be uploaded onto these
> servers such as peoples personal information without consent? This has
> happened to a lot of people. And yet no one is interested in addressing
> this!
> I've proposed solutions to simply add more sanity/validation checks to make
> sure keys are actual valid keys and limiting the overall size of keys to
> prevent abuse, but overall I'm not terribly concerned.. there's a billion
> places to make information public on the internet that is entirely out of
> reach of your local authorities, SKS is rather ineffective means of making
> information public since practically nobody is looking at the dataset as a
> whole and are only querying information directly, and almost always
> automated.. You are basically Gaslighting at this point.
>
> > And are you against the GDPR?
> Correct, the GDPR would be ruled unconstitutional in a heartbeat if someone
> tried to implement it here.
>
> > Do you even know what the GDPR covers?
> Yes, quite well.. I unfortunately work with many forms of Digital
> Compliance in my industry.
>
> > what has Australia got to do with this?
> Just another example of the road to hell is paved with good intentions..
> Its a slippery slope you guys are already sliding down.. I can only think
> of one operator that was forced to shut down for being liable for data
> others posted publicly, and that was an Australian operator.. long before
> the GDPR was drafted.. and nothing was accomplished, the data they tried to
> take out of the public sphere still exists.. again SKS worked as designed,
> the government was unable to stop the distribution of that data.. and its
> still accessible, even within Australia.
>
> > and where are you from Mr Hunt? America?
> Yes, Colorado to be precise if you need to figure out what court to waste
> your time with.
>
> > There's plenty why you claim none im not sure, maybe we should test this
> theory of yours?
> Go for it, I am completely willing to face any government and the resulting
> consequences to protect the integrity and availability of public
> cryptography, if my government were to ever insist on compromising it again
> in the future I would make it my mission to distribute the tools and spread
> awareness despite any legal ramifications or any moral perspective, yeah I
> might be assisting terrorists, child abusers, and other boogiemen; but
> thats the price of cryptographically secure communications. The EU can
> bring it on for all I care, this is a hill I'm fully prepared to die on,
> and have been for a while.. I advocated for and distributed the tools 30
> years ago when strong crypto was illegal to export from the United States,
> and eventually we won that battle of attrition.
>
> -R
>
>
>
> On Fri, Aug 16, 2019 at 10:12 AM <[hidden email]> wrote:
>
> > On Fri, 16 Aug 2019 09:12:30 -0600
> > Ryan Hunt <[hidden email]> wrote:
> >
> > > Yakamo,
> > > it still does its job of ensuring published keys are not tampered with,
> > it
> > > was not designed to be resilient to denial attacks.. That does not
> > > interfere with the trust of PGP, its why there are local keystores.. and
> > > the SKS network is still around despite being unreliable/broken from a
> > > maintenance standpoint.. your poisoned keys are not altering other
> > > individuals keys in any way/shape/form, so its security has not been
> > > compromised.. availability of keyservers is not critical to the use of
> > PGP,
> > > again by design.. there are many ways to distribute keys, it is resilient
> > > factually despite your opinions.. over the decades the need has not been
> > > lost.
> > >
> >
> > That's correct its not designed to be resilient to denial attacks, making
> > it unreliable as stated before! which means its not resilient to
> > governments at all! This statement stands true. Now it barely fulfils its
> > basic functions! the amount of posts littered over the internet about how
> > people cant pull a key from the servers or unable to upload them. There are
> > constant outages!
> >
> > There are alternatives and they work! sks doesn't!
> >
> > Its not the design or the attacks that's for me personally and others
> > distrustful its the, closed minded approach to how vulnerabilities are
> > handled, both people from the GnuPG community and SKS have attacked people
> > for what's considered normal practice when it comes to disclosure of
> > vulnerabilities and bugs. "stay quiet and hope nothing happens" or "your
> > attacking us because you pointed out something wrong with our software" is
> > not a good way to deal with things!
> >
> > Also do you think its good Mr Hunt that data can be uploaded onto these
> > servers such as peoples personal information without consent? This has
> > happened to a lot of people. And yet no one is interested in addressing
> > this!
> >
> > > You could not be more wrong about GnuPG, and it shows.. do you even work
> > in
> > > the industry? Because where I sit, with over 54 million devices on my
> > > network.. PGP is one of the most trusted security tools we use, all of
> > our
> > > software is signed by PGP, config files are signed by PGP, internal
> > > correspondence signed by PGP.. You are the only person in the world
> > > claiming GnuPG has lost its trust and you can write all the blog posts
> > you
> > > want but your opinion means nothing to me, and the rest of the industry..
> > > Snowden and all the other security industry's rock stars still fully
> > > advocate the use of PGP despite your feeble attacks.
> >
> > Are we really comparing "network" size?
> >
> > I didn't say it was not in demand or general use in the security
> > community! or unpopular!
> > Although I come across very few people who actually use it these days and
> > who are not middle aged. Even FreeBSD stopped using it who knows how long
> > ago for signing packages.
> >
> > Like wise your opinion holds no value to me either.
> >
> >
> > > So to answer your questions:
> > > 1. Currently, its the only option until something better comes along.
> >
> > Keybase and Hagrid or self hosting your gpg key, plenty of options.
> >
> > > 2. There are absolutely none, but you seem to be beyond reason on this
> > > point so I digress.
> >
> > There's plenty why you claim none im not sure, maybe we should test this
> > theory of yours?
> >
> > > 3. This is entirely arbitrary, not everyone has to share your
> > perspective..
> > > Most of the industry rallied against the GDPR, if anything the
> > EU/Australia
> > > has become the laughing stock of the cryptography world.. you guys would
> > > give up master keys and implement backdoors to your government in
> > exchange
> > > for a cookie and a pat on the back.
> >
> > Of course big company's rallied against the GDPR, it gives users their
> > privacy back again!
> > This messes with their business model!
> >
> > And are you against the GDPR?
> > Do you even know what the GDPR covers?
> >
> > what has Australia got to do with this?
> >
> > and where are you from Mr Hunt? America?
> >
> > Kind Regards
> >
> > Yakamo
> >
> >
> > --
> >
> >


--


_______________________________________________
Sks-devel mailing list
[hidden email]
https://lists.nongnu.org/mailman/listinfo/sks-devel
Reply | Threaded
Open this post in threaded view
|

Re: The pool is shrinking

Todd Fleisher
In reply to this post by Ryan Hunt-3
On Aug 16, 2019, at 10:42 AM, Ryan Hunt <[hidden email]> wrote:

Its role as a decentralized, tamper resistant key storage solution is still vital, and I would love it if we had the development going on to address the stability issues, but thats simply not the case at this point in time and until the actual integrity of the data the SKS network serves is compromised there is no need for its death..

I think it would be much more constructive and on topic to this list if we could focus on this issue vs. what this thread has devolved into. There are very real operational issues with the SKS network and while I don’t agree it needs to die, I can attest to the fact that it has become a significant problem for some to rely on it for public key distribution because of the poison key issue. My key has already been targeted which means the public can no longer obtain it from the SKS network and I am not the only person this problem impacts.

I am personally not migrating to keys.openpgp.org because of the limitations it currently has over the SKS network:

- Cannot perform wildcard searches by domain
- Cannot discover keys that have not been submitted & verified
- Keys lack signatures which breaks the web of trust

I will also point out there is a movement amongst several major software distributions that bring PGP support to the masses (especially as it relates to email) that are migrating away from the SKS network in large part because of this very issue (https://keys.openpgp.org/about/usage). And while there are other use cases for the SKS network for sure, I believe the ongoing issue where keys can be rendered un-importable by malicious third parties without warning threatens its very existence and needs to be dealt with before it’s too late.

-T

+cc Kristian directly for higher visibility


_______________________________________________
Sks-devel mailing list
[hidden email]
https://lists.nongnu.org/mailman/listinfo/sks-devel

signature.asc (849 bytes) Download Attachment
12345