key question

MFPA expires2010 at
Fri Feb 26 21:14:47 CET 2010

Hash: SHA512

Hi Robert

On Friday 26 February 2010 at 6:05:56 PM, you wrote:

> On 2/26/10 12:38 PM, MFPA wrote:
>> I am *not* advocating the implementation of any form of
>> Digital Restrictions Malware (DRM).

> You can say you're not advocating DRM -- but if it looks like a duck,
> swims like a duck, flies like a duck and quacks like a duck, then it's a
> duck.

But if it bears only a slight resemblance to a duck, it is probably
*not* a duck.

> "Digital": yes, the public key is in a digital form.


> "Rights" : yes, you're advocating the owner possesses intrinsic rights.

I am simply advocating the owner's right to privacy.
Nothing spectacular, nor anything specific to PGP keys.

> "Management": yes, you're advocating the owner should be allowed to have
>    total control over how the key gets distributed.  That's pretty
>    extreme management.

I have not knowingly advocated anything so extreme.

The reasonable expectation that somebody will extend the common
courtesy of checking with the owner before publishing their key falls
somewhat short of the owner having total control over their key.

> But, hey.  If you don't like DRM on the honor system, I'm happy to call
> it ORCON ("Originator Controlled").

The term "ORCON" reminds me of a 1970s TV programme about an alien.

>  ORCON material doesn't get copied,
> shared, promulgated, forwarded on, without the originator's explicit
> permission.  It is the most extreme form of DRM imaginable.  I thought I
> was being generous by saying you were advocating DRM on the honor system
> instead of ORCON -- ORCON is much more onerous.

I am not advocating that at all.

I see the merit of a system that only allowed the key owner to publish
the key to a server. How this could reasonably be achieved is not
clear to me. And was not what I was discussing here.

> My exposure to ORCON material came from my work with electronic voting
> systems.  Government officials are sometimes willing to give electronic
> voting geeks a peek behind the curtain, so long as there's an ORCON
> agreement signed in blood with the Devil himself as an eyewitness.

Typical of a government to be ultra-secretive about the wrong things.
You would think trust in electronic voting systems would flow from
transparency, not secrecy. How can the voters have confidence that the
system cannot be manipulated by those running it?

> You're advocating public keys be treated like the inner secrets of how
> electronic voting machines work.  So am I.  It's just that you're
> advocating they all be kept secret by default and publication being an
> exception to the rule -- and I'm advocating they all be kept public by
> default and secrecy being the exception to the rule.

I think the inner secrets of how electronic voting machines work
should be open-source and available for peer-review.

I think personally-identifiable information, including an individual's
openPGP key, should not be made public without the consent of the

>> Uploading a somebody else's key without first checking it is OK by
>> them is a breach of their privacy

> You're claiming they have a reasonable expectation that, if they share
> data that is clearly marked *public*, the recipient should understand
> *public* means "clear it with me first"?

> I don't think that's a reasonable expectation.  The key says "public"
> right at the very top, and I think it's unreasonable to expect people to
> infer that it means "no, don't share it."

> This is why the burden is on the key provider: if you don't want the key
> shared, you have to explicitly tell someone about it.  If you don't tell
> someone about it, they are allowed to think the phrase "public" means
> just that.

I think it is reasonable to expect the recipient to know that it says
"PGP PUBLIC KEY BLOCK." I don't see any reason why they would split
the words and interpret each one as a standalone; if people do that,
I'm waiting to hear from those who think the key can't be used with
GPG, it will open a door or start a car, and that if they had a pile
of them they could build a wall. (-;

The use of the word "public" in the descriptor "public key" was an
unfortunate choice if people are going to interpret things in that

I think it is a reasonable expectation that the key owner would have
uploaded their key to the keyservers themselves if they wanted it to
be there. If the key is not already on the servers, that is a pretty
strong indicator that the key owner wants it that way.

>> and could well be illegal/unlawful
>> in jurisdictions with data protection legislation (for example, if a
>> company published a customer's key, showing their name and/or email
>> address, to a server).

> That's not the key sharer's problem.  That's the problem of the person
> who provided the key.  If you know it would be unlawful for you to share
> information, don't share it.

I don't understand your comment. It's not unlawful for the individual
to share their own information. It would be unlawful for the recipient
of that information to share it with others without consent from the
individual, or to keep it for longer than reasonably necessary, or to
use it for any purpose other than what the customer was told it would
be used for. So, the merchant told the customer he would communicate
by encrypted email if the customer supplied their public key. The
customer was not told the merchant would upload the key to a server;
if the merchant did upload it, the merchant would have acted

>> I don't see the connection between DRM and a perfectly proper respect
>> for individual privacy.

> By implication, then, I lack a proper respect for individual privacy.
> At this point this seems to be dropping straight into the ad-hominem range.

I was thinking maybe you might explain to me the connection you draw
between DRM and respecting individual privacy, since I do not see one.

It would appear I have offended you; for that I am very sorry.

- --
Best regards

MFPA                    mailto:expires2010 at

Vegetarian: Indian word for lousy hunter!!!


More information about the Gnupg-users mailing list