[gentoo-dev] gentoo's policy on sender id

[gentoo-dev] gentoo's policy on sender id

[John Davis]

[-- Attachment #1: Type: text/plain, Size: 587 bytes --]

Hi all -
Recent [1] announcements [2] from other large projects have prompted
this inquiry - should Gentoo adopt similar policies regarding Sender ID
and make a public statement or is it not worth our time?

1. http://it.slashdot.org/article.pl?sid=04/09/05/138257&tid=111&tid=1
2.
http://apache.slashdot.org/article.pl?sid=04/09/02/1446229&tid=148&tid=155&tid=109&tid=95
-- 
John Davis
Gentoo Linux Developer
<http://dev.gentoo.org/~zhen>

----
GnuPG Public Key: <http://dev.gentoo.org/~zhen/zhen_pub.asc>
Fingerprint: 4F9E 41F6 D072 5C1A 636C  2D46 B92C 4823 E281 41BB

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: [gentoo-dev] gentoo's policy on sender id

[Tom Gall]

On Sep 5, 2004, at 9:00 AM, John Davis wrote:

> Hi all -
> Recent [1] announcements [2] from other large projects have prompted
> this inquiry - should Gentoo adopt similar policies regarding Sender ID
> and make a public statement or is it not worth our time?

Personally I wonder how much of the attention with regards to Sender ID  
is more of a dig against micro$oft.  O wait that's the "in" thing to  
do. :-)

But what does saying something really mean?  Would it be somewhat sorta  
like the  xfree86 license change situation?

IE we won't have any package in portage that implements Sender ID?

Regards,

Tom

> 1. http://it.slashdot.org/article.pl?sid=04/09/05/138257&tid=111&tid=1
> 2.
> http://apache.slashdot.org/article.pl?sid=04/09/02/ 
> 1446229&tid=148&tid=155&tid=109&tid=95
> -- 
> John Davis
> Gentoo Linux Developer
> <http://dev.gentoo.org/~zhen>
>
> ----
> GnuPG Public Key: <http://dev.gentoo.org/~zhen/zhen_pub.asc>
> Fingerprint: 4F9E 41F6 D072 5C1A 636C  2D46 B92C 4823 E281 41BB
>
Tom Gall
Gentoo/PPC64 Team Leader
tgall@gentoo.org


--
gentoo-dev@gentoo.org mailing list

Re: [gentoo-dev] gentoo's policy on sender id

[Chris Bainbridge]

On Sunday 05 September 2004 15:00, John Davis wrote:
> Recent [1] announcements [2] from other large projects have prompted
> this inquiry - should Gentoo adopt similar policies regarding Sender ID
> and make a public statement or is it not worth our time?

One of the things that I like about gentoo is the lack of deep politics. If 
someone is willing to maintain ebuild for packages that support sender id 
then why shouldn't gentoo host them? We already have stuff like vmware that 
is both patented and proprietary binary-only software.

--
gentoo-dev@gentoo.org mailing list

Re: [gentoo-dev] gentoo's policy on sender id

[Stephen P. Becker]

Chris Bainbridge wrote:
> On Sunday 05 September 2004 15:00, John Davis wrote:
> 
>>Recent [1] announcements [2] from other large projects have prompted
>>this inquiry - should Gentoo adopt similar policies regarding Sender ID
>>and make a public statement or is it not worth our time?
> 
> 
> One of the things that I like about gentoo is the lack of deep politics. If 
> someone is willing to maintain ebuild for packages that support sender id 
> then why shouldn't gentoo host them? We already have stuff like vmware that 
> is both patented and proprietary binary-only software.
> 

Not that anything like this hasn't been done before (e.g. xorg-x11 
replacing xfree soon).

Steve

--
gentoo-dev@gentoo.org mailing list

Re: [gentoo-dev] gentoo's policy on sender id

[Nicholas Jones]

[-- Attachment #1: Type: text/plain, Size: 1552 bytes --]


Read this one:
http://www.apache.org/foundation/docs/sender-id-position.html

> One of the things that I like about gentoo is the lack of
> deep politics.

There are reasons you don't see deep politics. Most of us don't
like them, hence they are avoided. This is probably part of it.

> If someone is willing to maintain ebuild for packages that
> support sender id then why shouldn't gentoo host them? We
> already have stuff like vmware that is both patented and
> proprietary binary-only software.

Because we then have to deal with non-compliant interaction
with the liceneses. There is a great deal of bad-mojo here
if you read the full Apache position.

Recall that every distro has dropped XFree because of the
'logo adjacency' issue? Violation of the GPL... Well, that
is merely _one_ problem with Sender-ID.

We would not have the infrastructure to manage compliance
with such an annoying licence. I am not certain on this
point here, but it's entirely possible that arbitrary linking
of applications with sender-id may be inducing a violation
of the agreement. It's possible that we would be liable.

There are also points regarding termination of the licence
and the inability to transfer it. So there is no guarentee
that software using sender-id could be passed on to another
developer/team. Suddenly finding yourself in violation of
a license is probably not a good idea if your income is
zero, especially facing a prosecution with several billion
in the bank.

Someone get a law degree from somewhere and argue with me,
please.

--NJ



[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]

Re: [gentoo-dev] gentoo's policy on sender id

[Chris Bainbridge]

On Sunday 05 September 2004 22:33, Nicholas Jones wrote:
> Recall that every distro has dropped XFree because of the
> 'logo adjacency' issue? Violation of the GPL... Well, that
> is merely _one_ problem with Sender-ID.
>
> We would not have the infrastructure to manage compliance
> with such an annoying licence. I am not certain on this
> point here, but it's entirely possible that arbitrary linking
> of applications with sender-id may be inducing a violation
> of the agreement. It's possible that we would be liable.
>
> There are also points regarding termination of the licence
> and the inability to transfer it. So there is no guarentee
> that software using sender-id could be passed on to another
> developer/team. Suddenly finding yourself in violation of
> a license is probably not a good idea if your income is
> zero, especially facing a prosecution with several billion
> in the bank.

From the gentoo point of view all of these problems are restrictions on 
redistribution. At worst, RESTRICT="nomirror" solves them. Lets at least have 
a consistent policy - we already have software (particularly games) that we 
aren't allowed to redistribute. Sender-id software is no different. Having 
said that, I doubt that this patent actually prohibits 3rd party mirrors like 
cnet.

Theres no doubt that this is a bad patent/license from a software developer 
point of view, but gentoo is not in the business of developing mail server 
software. The decision of whether to support sender-id should be left in the 
hands of people who are (or are we going to start patching postfix to remove 
code?).

It's a slippery slope to reject ebuilds because we don't agree with the 
licenses imposed on the developers of those packages, or because we believe 
that they violate software patents. Where do you draw the line? There is no 
doubt that the linux kernel violates some (bad) patents. Should it 
potentially be removed? How about quake3 - I'm not allowed to rebrand and 
redistribute it without paying a lot of money for a license. Should it also 
be excluded from gentoo?

> Someone get a law degree from somewhere and argue with me,
> please.

You can email Microsofts licensing people and they will forward any questions 
they can't answer to their lawyers. It might be useful to do this anyway, for 
future protection.

--
gentoo-dev@gentoo.org mailing list

[gentoo-dev] Re: [gentoo-core] Re: [gentoo-dev] gentoo's policy on sender id

[Nicholas Jones]

[-- Attachment #1: Type: text/plain, Size: 1513 bytes --]

> It's a slippery slope to reject ebuilds because we don't agree with the 
> licenses imposed on the developers of those packages, or because we believe 
> that they violate software patents. Where do you draw the line? There is no 
> doubt that the linux kernel violates some (bad) patents. Should it 
> potentially be removed? How about quake3 - I'm not allowed to rebrand and 
> redistribute it without paying a lot of money for a license. Should it also 
> be excluded from gentoo?

The difference is optional component versus a (required) standard.
This licence applies to an Internet Standard. Something that is to
be implemented Internationally on a critical infrastructure. If you
wanted to modify this standard, you would be subject to Microsoft's
licensing and approval _prior_ to the work, or you'd be subject to
suit.

You don't have to have quake3 to receive you're email from your bank,
but it's entirely possible that the only way you'll be able to
communicate with utilities and other businesses is via a Standard
protocol that subjects you to a non-public license.

What happens if you ISP is using postfix and for some reason
Microsoft terminates the agreement allowing Postfix to user
sender-id? Would your mail be dropped?

The problem with Sender-ID is almost purely the license. If MS has
no malintent with this creation, then they can provide all parts
under an agreement that is suitable for GLOBAL acceptance and
without termination clauses, et al. 

--NJ


[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]

Re: [gentoo-dev] gentoo's policy on sender id

[Rob Cakebread]

REUTERS[ WEDNESDAY, SEPTEMBER 15, 2004 11:08:34 PM ]
SEATTLE: Microsoft Corp was dealt a setback on Tuesday after the 
Internet Engineering Task Force decided not to adopt Microsoft's e-mail 
sender ID standard that would make it easier for Internet providers to 
block unwanted junk e-mail.

http://timesofindia.indiatimes.com/cms.dll/html/uncomp/articleshow/852127.cms

--
gentoo-dev@gentoo.org mailing list

Re: [gentoo-dev] gentoo's policy on sender id

[Malte S. Stretz]

On Thursday 16 September 2004 06:38 CET Rob Cakebread wrote:
> REUTERS[ WEDNESDAY, SEPTEMBER 15, 2004 11:08:34 PM ]
> SEATTLE: Microsoft Corp was dealt a setback on Tuesday after the
> Internet Engineering Task Force decided not to adopt Microsoft's e-mail
> sender ID standard that would make it easier for Internet providers to
> block unwanted junk e-mail.

That's not true in that form; call it an uninformed reporter or a too short 
summary.  I'd suggest to read this [1] thread for the current state of 
things.

Cheers,
Malte

[1] http://thread.gmane.org/gmane.ietf.mxcomp/4937

-- 
[SGT] Simon G. Tatham: "How to Report Bugs Effectively"
      <http://www.chiark.greenend.org.uk/~sgtatham/bugs.html>
[ESR] Eric S. Raymond: "How To Ask Questions The Smart Way"
      <http://www.catb.org/~esr/faqs/smart-questions.html>

--
gentoo-dev@gentoo.org mailing list

Re: [gentoo-dev] gentoo's policy on sender id

[Kurt Lieber]

[-- Attachment #1: Type: text/plain, Size: 659 bytes --]

On Sun, Sep 05, 2004 at 10:00:45AM -0400 or thereabouts, John Davis wrote:
> Hi all -
> Recent [1] announcements [2] from other large projects have prompted
> this inquiry - should Gentoo adopt similar policies regarding Sender ID
> and make a public statement or is it not worth our time?


I left Debian in large part because of the stupid politics that went on in
that project.  I want a distribution that bases its decisions on technical
reasons, rather than political ones.

Also, for the record, regardless of what decisions are made regarding
including support for SenderID in portage, we *will* be implementing
SenderID on our infrastructure.

--kurt

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]

Re: [gentoo-dev] gentoo's policy on sender id

[Michiel de Bruijne]

On Monday 06 September 2004 01:44, Kurt Lieber wrote:
> On Sun, Sep 05, 2004 at 10:00:45AM -0400 or thereabouts, John Davis wrote:
> > Hi all -
> > Recent [1] announcements [2] from other large projects have prompted
> > this inquiry - should Gentoo adopt similar policies regarding Sender ID
> > and make a public statement or is it not worth our time?
>
> I left Debian in large part because of the stupid politics that went on in
> that project.  I want a distribution that bases its decisions on technical
> reasons, rather than political ones.
>
> Also, for the record, regardless of what decisions are made regarding
> including support for SenderID in portage, we *will* be implementing
> SenderID on our infrastructure.
>
> --kurt

Just out of curiosity; what functionality does Sender ID provide that Sender 
Policy Framework (SPF) doesn't?

--
gentoo-dev@gentoo.org mailing list