Re: [backstage] RTMP stream URL resolving script

2008-01-20 Thread Sean DALY 
I stand corrected. Concerning Corporation X, I should have said
"without attribution and without source code".

Sean


On Jan 19, 2008 2:22 PM, Dave Crossland <[EMAIL PROTECTED]> wrote:
> On 19/01/2008, Sean DALY <[EMAIL PROTECTED]> wrote:
> > Well, it's "public domain" then, which is fine
>
> The public domain exists in the UK only for works that have expired
> from copyright; its only in the USA that you can legally assign a work
> into the public domain before its term expires. Thus Creative Commons
> recently retracted its "PD" license in favor of "CC 0".
>
> > as long as you don't
> > mind Corporation X incorporating and selling your code.
>
> The GPL doesn't prohibit Corporation X incorporating and selling your code.
>
> --
> Regards,
> Dave
>
> -
> Sent via the backstage.bbc.co.uk discussion group.  To unsubscribe, please 
> visit http://backstage.bbc.co.uk/archives/2005/01/mailing_list.html.  
> Unofficial list archive: 
> http://www.mail-archive.com/backstage@lists.bbc.co.uk/
>
-
Sent via the backstage.bbc.co.uk discussion group.  To unsubscribe, please 
visit http://backstage.bbc.co.uk/archives/2005/01/mailing_list.html.  
Unofficial list archive: http://www.mail-archive.com/backstage@lists.bbc.co.uk/

Re: [backstage] RTMP stream URL resolving script

2008-01-20 Thread Iain Wallace 
Maybe we need a discussion on the pros and cons of the various OSS
licenses. Recommend me one!
-
Sent via the backstage.bbc.co.uk discussion group.  To unsubscribe, please 
visit http://backstage.bbc.co.uk/archives/2005/01/mailing_list.html.  
Unofficial list archive: http://www.mail-archive.com/backstage@lists.bbc.co.uk/

Re: [backstage] RTMP stream URL resolving script

2008-01-20 Thread Sean DALY 
I have found David Wheeler's analysis of the GPL vs. BSD-style
licences very helpful:

GPL, BSD, and NetBSD - why the GPL rocketed Linux to success
http://www.dwheeler.com/blog/2006/09/01/

Debates on this topic can be endless, in particular since the arrival
of the GPLv3 which had to be updated 17 years after its introduction
in 1991 and which is usually criticised for being "too complicated"
compared to GPLv2 or BSD. That said, I agree with Mr. Wheeler that the
GPL is a stronger incentive for companies to fight fair.

A longstanding rumor, for which I have no proof, is that parts of
Microsoft's network code was simply copied from BSD code, which if
true would naturally explain why Microsoft is so hesitant to documents
its protocols not to mention its code.

Sean



On Jan 20, 2008 4:35 PM, Iain Wallace <[EMAIL PROTECTED]> wrote:
> Maybe we need a discussion on the pros and cons of the various OSS
> licenses. Recommend me one!
>
> -
> Sent via the backstage.bbc.co.uk discussion group.  To unsubscribe, please 
> visit http://backstage.bbc.co.uk/archives/2005/01/mailing_list.html.  
> Unofficial list archive: 
> http://www.mail-archive.com/backstage@lists.bbc.co.uk/
>
-
Sent via the backstage.bbc.co.uk discussion group.  To unsubscribe, please 
visit http://backstage.bbc.co.uk/archives/2005/01/mailing_list.html.  
Unofficial list archive: http://www.mail-archive.com/backstage@lists.bbc.co.uk/

Re: [backstage] RTMP stream URL resolving script

2008-01-20 Thread Dave Crossland 
On 20/01/2008, Iain Wallace <[EMAIL PROTECTED]> wrote:
>
> Maybe we need a discussion on the pros and cons of the various OSS
> licenses. Recommend me one!

Using any free software license is good, and I hope you'll consider
which is best based on how they promote and protect software freedoms
for _all_ users of your program.

How would you feel if some developer who receive your program can
improve it and then tell people, even you as the original author, that
you can't share that version with your friends, or see how their
improvement works, or build upon their work as they built upon yours?

I would feel annoyed if that happened to me - even angry, if I had put
a lot of effort into the program. Peter Ferne has suggested that using
a free software license that permits that kind of antisocial behavior
would "keep everyone happy," but I think by "everyone" he means only a
few proprietary developers, not all the users of your program.

The GPL ensures this won't happen for software that runs on everyone's
own computers, and the Affero GPL ensures this won't happen for
software that runs on network servers.

http://www.gnu.org/licenses/quick-guide-gplv3.html explains why the
GPL is the best license for making a program free and keeping it that
way.

Because this might program may be run on a server and used though a
network, the Affero GPLv3 is better than the plain GPLv3. AGPLv3 is a
cutting edge license, just released in November last year, and a good
example of an Affero webapp you may have used is
http://petitions.pm.gov.uk/ :-)

Sean mentions GPLv3 may be criticised for being "too complicated" but
that seems like a sham to me; the GPL isn't longer than an average
sunday newspaper article and is written for a software developer
audience in mind.

-- 
Regards,
Dave
-
Sent via the backstage.bbc.co.uk discussion group.  To unsubscribe, please 
visit http://backstage.bbc.co.uk/archives/2005/01/mailing_list.html.  
Unofficial list archive: http://www.mail-archive.com/backstage@lists.bbc.co.uk/

Re: [backstage] RTMP stream URL resolving script

2008-01-20 Thread Rob Myers 

Dave Crossland wrote:

Sean mentions GPLv3 may be criticised for being "too complicated" 


The GPL2 was "too complicated" until GPL3 was released. Now GPL2 is 
perfect. ;-) The FSF should release a GPL4, let that be criticized for 
being too complex, withdraw it, and then everyone will be happy. ;-)


- Rob.
-
Sent via the backstage.bbc.co.uk discussion group.  To unsubscribe, please 
visit http://backstage.bbc.co.uk/archives/2005/01/mailing_list.html.  
Unofficial list archive: http://www.mail-archive.com/backstage@lists.bbc.co.uk/

Re: [backstage] RTMP stream URL resolving script

2008-01-20 Thread Noah Slater 
On Sun, Jan 20, 2008 at 06:23:31PM +, Dave Crossland wrote:
> How would you feel if some developer who receive your program can
> improve it and then tell people, even you as the original author, that
> you can't share that version with your friends, or see how their
> improvement works, or build upon their work as they built upon yours?

I am a CouchDB developer. 

We recently switched from GPL to the Apache licence. Now, I would like
to get it out of the way that I am unhappy with the switch, but as I
am not the primary developer, I decided to go along with the move.

The reason I am not happy is because Amazon could come along one day
and take all the code I had slaved over for so many hours, put a team
of 10 developers on improving it full time, and then release as a
competing product, be that closed source or via a web service.

Suddenly, all that time I had put into the project is being used
against me to compete. Not only that, but the competing product is
completely non-free, so it's not like /anyone/ benifits.

> Sean mentions GPLv3 may be criticised for being "too complicated" but
> that seems like a sham to me; the GPL isn't longer than an average
> sunday newspaper article and is written for a software developer
> audience in mind.

Lets remember that we are talking about legal documents here, not
poems. That the GPL is so long is a testiment to how complicated
copyright law is any how many precautions need to be taken to prevent
things like I just described from happening.

-- 
Noah Slater 

"Creativity can be a social contribution, but only in so far as
society is free to use the results." - R. Stallman
-
Sent via the backstage.bbc.co.uk discussion group.  To unsubscribe, please 
visit http://backstage.bbc.co.uk/archives/2005/01/mailing_list.html.  
Unofficial list archive: http://www.mail-archive.com/backstage@lists.bbc.co.uk/

Re: [backstage] RTMP stream URL resolving script

2008-01-20 Thread Michael Sparks 
On Sunday 20 January 2008 15:35:12 Iain Wallace wrote:
> Maybe we need a discussion on the pros and cons of the various OSS
> licenses. Recommend me one!

Summary:
   1 Copyright notice applied & enforced.
   2 No license - falls back to copyright law.
   3 Implied license (what you've done) - tentatively allowed to do stuff, but
  not very strong license as to what can be done with it. (technically
  falls back to copyright law meaning can't do anything, but your email is
  an implied license))
   4 License with "reciprocity clause". This means "if you do things (eg
  create a derivative work) with my code under this license and you
  redistribute this, you have to redistribute your code under the same
  license". Examples: GPL, MPL
   5 License without reciprocity clause. Not quite but close to "do what you
  like with this, but attribute me & don't remove this notice", doesn't
  constrain your choice of license for derivative work.
  Examples: BSD, MIT, Apache licenses
   6 Notice saying "released to the public domain". Unfortunately, as I
  understand things, not actually doable in the UK. 

Since 6 can't be done in the UK, many people who would choose 6 use 5. 

It's worth noting that license 5 is the weakest level of control a developer
can exert. Someone can take your work and either restrict your ability to take
changes (that you can release as 5) by either re-releasing your work in a
derivative licensed under 4) or 1). People who are don't like this (for 
various reasons) with this often choose either 4 or 1 up-front. 

Both the 4) & 5) camps can produce evidence as to why each is better than the 
other for various purposes, so I'm going to refrain from that :) Both camps 
can give ideological as well as pragmatic reasons in favour of their views as 
well.

Lawrence Rosen's book is also very readable: 
   * Open Source Licensing - ISBN: 0131487876

Different licenses have different benefits for differing groups. One creates
an artificial scarce resource that can be exploited for financial gain, one
can be used with an intent to allow all users to be able to modify and extend
the code by aiming to protect this ability from being taken away, and one 
allows the developer to wash their hands of this and make it someone else's 
problem :-)

Regards,


Michael.

 *This is not legal advice, and any opinions there not really intended, and
   certainly not anyone else's or any company's either :-) *
-
Sent via the backstage.bbc.co.uk discussion group.  To unsubscribe, please 
visit http://backstage.bbc.co.uk/archives/2005/01/mailing_list.html.  
Unofficial list archive: http://www.mail-archive.com/backstage@lists.bbc.co.uk/

Re: [backstage] RTMP stream URL resolving script

2008-01-20 Thread Michael Sparks 
On Sunday 20 January 2008 17:01:43 Sean DALY wrote:
> A longstanding rumor, for which I have no proof, is that parts of
> Microsoft's network code was simply copied from BSD code, which if
> true would naturally explain why Microsoft is so hesitant to documents
> its protocols not to mention its code.

That's misleading (I'm sure non-intentionally). Microsoft have indeed used BSD 
code in their systems in the past and as I recall it was the TCP/IP stack - 
or portions thereof. This isn't exactly uncommon and if you're choosing a 
TCP/IP stack to use, there are worse choices :-)

However they *have* complied with the BSD license - if you look in the manuals 
distributed with windows you will find the appropriate statements.

It is however not exactly a secret (or even a rumour!) - eg it's trivial to
find here:
* http://support.microsoft.com/kb/306819/en-us

(you'll see the various notices they're required to include)

I *believe* (but have no evidence beyond "I've been told") that they've been
reported to have rewritten that code since then, so I'd guess they no longer
need those statements. (I don't have a copy of Vista, so can't (and have no
inclination to) check :)

The reason for Microsoft not documenting it protocols and code in the way
demanded by some is IMO likely to be for some other reason. I'm going to 
refrain from speculating why. I will note that documenting protocols allows 
for multiple implementations - enabling competition. I suspect therefore 
their decision is based on whether they can see value in competition in that
space or not. (if it grows the market, then everyone benefits including them - 
since although their share shrinks the pie grows increasing their income. If 
the market is at peak size, it shrinks their market share whilst not growing 
the size of the pie, reducing their income)

Beyond speculating that their decision is based on cold hard money, I'm not 
speculating further :-)


Michael.

*personal opinions only*
-
Sent via the backstage.bbc.co.uk discussion group.  To unsubscribe, please 
visit http://backstage.bbc.co.uk/archives/2005/01/mailing_list.html.  
Unofficial list archive: http://www.mail-archive.com/backstage@lists.bbc.co.uk/

Re: [backstage] RTMP stream URL resolving script

2008-01-20 Thread Iain Wallace 
On Jan 20, 2008 9:10 PM, Michael Sparks <[EMAIL PROTECTED]> wrote:
> On Sunday 20 January 2008 15:35:12 Iain Wallace wrote:
> > Maybe we need a discussion on the pros and cons of the various OSS
> > licenses. Recommend me one!
>
> Summary: 

That's really useful, thanks! I think I'll go for a GPL license. I
can't really imagine why someone wouldn't want reciprocity in their
license. If someone makes something cool from my code I want to read
the source and see how they did it. I'll read up on GPLv3 and AGPLv3.

Iain
-
Sent via the backstage.bbc.co.uk discussion group.  To unsubscribe, please 
visit http://backstage.bbc.co.uk/archives/2005/01/mailing_list.html.  
Unofficial list archive: http://www.mail-archive.com/backstage@lists.bbc.co.uk/

Re: [backstage] RTMP stream URL resolving script

2008-01-20 Thread Dave Crossland 
On 20/01/2008, Michael Sparks <[EMAIL PROTECTED]> wrote:
>
> It's worth noting that license 5 is the weakest level of control a developer
> can exert. Someone can take your work and either restrict your ability to take
> changes (that you can release as 5) by either re-releasing your work in a
> derivative licensed under 4) or 1).

This is a common misconception.

With most "5" style licenses, such as X11 or BSD licenses, you can not
relicense (technically, "sublicense") the source code under "4" style
licenses, but you can combine sourcecode files with mixed licenses
into a single program.

http://www.softwarefreedom.org/resources/2007/gpl-non-gpl-collaboration.html
explains how to do this in depth.

The Expat license, that I linked to earlier in the thread, permits
sublicensing, which is why I recommend it.

-- 
Regards,
Dave
(Personal opinion only!)
-
Sent via the backstage.bbc.co.uk discussion group.  To unsubscribe, please 
visit http://backstage.bbc.co.uk/archives/2005/01/mailing_list.html.  
Unofficial list archive: http://www.mail-archive.com/backstage@lists.bbc.co.uk/

Re: [backstage] RTMP stream URL resolving script

2008-01-20 Thread Sean DALY 
> That's misleading (I'm sure non-intentionally). Microsoft have indeed used BSD
> code in their systems in the past and as I recall it was the TCP/IP stack -
> or portions thereof

Hmmm I meant aside from the TCP/IP stack -- after all, David Wheeler
mentions that in the article I linked to -- I should have been more
explicit. But, again, I have no proof, so l will call it just a rumor
:-)

However I am not at all sure Microsoft respects its licensing
obligations regarding reused code. The Services for Unix package
contains licence texts (including GPLv1, GPLv2, and BSD) but no source
code, nor any indication of where to find it.

Sean


On Jan 20, 2008 10:35 PM, Michael Sparks <[EMAIL PROTECTED]> wrote:
> On Sunday 20 January 2008 17:01:43 Sean DALY wrote:
> > A longstanding rumor, for which I have no proof, is that parts of
> > Microsoft's network code was simply copied from BSD code, which if
> > true would naturally explain why Microsoft is so hesitant to documents
> > its protocols not to mention its code.
>
> That's misleading (I'm sure non-intentionally). Microsoft have indeed used BSD
> code in their systems in the past and as I recall it was the TCP/IP stack -
> or portions thereof. This isn't exactly uncommon and if you're choosing a
> TCP/IP stack to use, there are worse choices :-)
>
> However they *have* complied with the BSD license - if you look in the manuals
> distributed with windows you will find the appropriate statements.
>
> It is however not exactly a secret (or even a rumour!) - eg it's trivial to
> find here:
> * http://support.microsoft.com/kb/306819/en-us
>
> (you'll see the various notices they're required to include)
>
> I *believe* (but have no evidence beyond "I've been told") that they've been
> reported to have rewritten that code since then, so I'd guess they no longer
> need those statements. (I don't have a copy of Vista, so can't (and have no
> inclination to) check :)
>
> The reason for Microsoft not documenting it protocols and code in the way
> demanded by some is IMO likely to be for some other reason. I'm going to
> refrain from speculating why. I will note that documenting protocols allows
> for multiple implementations - enabling competition. I suspect therefore
> their decision is based on whether they can see value in competition in that
> space or not. (if it grows the market, then everyone benefits including them -
> since although their share shrinks the pie grows increasing their income. If
> the market is at peak size, it shrinks their market share whilst not growing
> the size of the pie, reducing their income)
>
> Beyond speculating that their decision is based on cold hard money, I'm not
> speculating further :-)
>
>
> Michael.
>
> *personal opinions only*
>
> -
> Sent via the backstage.bbc.co.uk discussion group.  To unsubscribe, please 
> visit http://backstage.bbc.co.uk/archives/2005/01/mailing_list.html.  
> Unofficial list archive: 
> http://www.mail-archive.com/backstage@lists.bbc.co.uk/
>
-
Sent via the backstage.bbc.co.uk discussion group.  To unsubscribe, please 
visit http://backstage.bbc.co.uk/archives/2005/01/mailing_list.html.  
Unofficial list archive: http://www.mail-archive.com/backstage@lists.bbc.co.uk/

Re: [backstage] RTMP stream URL resolving script

2008-01-20 Thread Dave Crossland 
On 20/01/2008, Matthew Somerville <[EMAIL PROTECTED]> wrote:
>
> I can't see any difference between the Expat licence and the X11 license
> which you include above

You are right, there is none, I stand corrected; apologies and thank
you for pointing this out :-)

I remembered X11 being under BSD but it is under the Expat license.
I'll drop "expat" since its obscure and refer to X11 from now on :-)

> I can't see your earlier linking either,

Ah, it was in an offlist mail; apologies

-- 
Regards,
Dave
-
Sent via the backstage.bbc.co.uk discussion group.  To unsubscribe, please 
visit http://backstage.bbc.co.uk/archives/2005/01/mailing_list.html.  
Unofficial list archive: http://www.mail-archive.com/backstage@lists.bbc.co.uk/

Re: [backstage] RTMP stream URL resolving script

2008-01-20 Thread Matthew Somerville 

Dave Crossland wrote:

With most "5" style licenses, such as X11 or BSD licenses, you can not
relicense (technically, "sublicense") the source code under "4" style
licenses, but you can combine sourcecode files with mixed licenses
into a single program.


[...]


The Expat license, that I linked to earlier in the thread, permits
sublicensing, which is why I recommend it.


I can't see any difference between the Expat licence and the X11 license 
which you include above (I can't see your earlier linking either, I'm 
looking at the one at http://www.jclark.com/xml/copying.txt as compared to 
http://www.xfree86.org/3.3.6/COPYRIGHT2.html#3 ).


ATB,
Matthew  |  http://www.dracos.co.uk/
-
Sent via the backstage.bbc.co.uk discussion group.  To unsubscribe, please 
visit http://backstage.bbc.co.uk/archives/2005/01/mailing_list.html.  
Unofficial list archive: http://www.mail-archive.com/backstage@lists.bbc.co.uk/

Re: [backstage] RTMP stream URL resolving script

2008-01-20 Thread Michael Sparks 
On Sunday 20 January 2008 21:33:18 Dave Crossland wrote:
> On 20/01/2008, Michael Sparks <[EMAIL PROTECTED]> wrote:
> > It's worth noting that license 5 is the weakest level of control a
> > developer can exert. Someone can take your work and either restrict your
> > ability to take changes (that you can release as 5) by either
> > re-releasing your work in a derivative licensed under 4) or 1).
>
> This is a common misconception.

No, it's not a misconception. It's trivial to demonstrate with a useless toy
program. You can easily extend this to any larger work.

Suppose for example I create a program X:

(none of this is intended/likely to compile btw - it's an illustration :-)

X.c
/* licensed under the BSD license */
#include 

void gameover() {
printf("gameover\n");
}

int main(int argc, char *argv) {
   gameover()
   return 0;
}
/* - */

Recipient A refactors this:

gameover.c
/* licensed under the BSD license */
void gameover() {
printf("gameover\n");
}

X.c
/* licensed under the BSD license */
#include 
#include "gameover.c"

int main(int argc, char *argv) {
   gameover();
   return 0;
}
/* - */

That's a compliant change.

They create a test suite as well which tests that the file:
   * gameover.c contatins a function gameover that performs in a particular
  way. (Not unusual to add tests when refactoring)

Recipient B then creates API compatible reimplementation of gameover.c
based on this test, not based on the code. This is again, a compliant
change (throwing away code and writing your own).

gameover.c - from B
/* licensed under the Affero GPL V3 */
#include 
void
gameover() {
fprint(stdout, "gameover\n");
}

X.c - from B
/* licensed under the BSD license */
#include 
#include "gameover.c"

int main(int argc, char *argv) {
   gameover();
   return 0;
}
/* - */

The resulting work (as a whole) has to be under the Affero GPL V3. (The
individual files are under separate licenses at this point though). The new
improved gameover.c cannot be recombined back into the original work
without the original developer changing their license or without the
recipient granting them a BSD license on the new gameover.c .

(this is indeed, part of the point of the GPL - if the original developer
could do recombine the new GPL only code without an explicit BSD grant or 
without changing their license, the GPL wouldn't be doing it's job of
ensuring that new code stayed GPL - whatever version you pick :)

> With most "5" style licenses, such as X11 or BSD licenses, you can not
> relicense (technically, "sublicense") the source code under "4" style
> licenses, 

That is not what I said. I said:

> > re-releasing your work in a derivative licensed under 4) or 1).

That's not the same as what you said. (Take a BSD program, add in a call to 
GNU readline without which the program won't function & the derivative work 
has to be licensed as a whole as GPL - ie "a derivative licensed under 4)" )

> but you can combine sourcecode files with mixed licenses 
> into a single program.

Absolutely, and you also have to remember that if a project that was BSD
licensed only, if you add GPL'd portions in, those GPL'd extensions and
modifications cannot be reincorporated back into the BSD mainline (without
changing the license of the mainline or without a BSD license being granted
back to (at minimum) the mainline) - as demonstrated above.

So, whilst the code is open & free software, the improvements are denied
to the original developer (unless they change their license of their system
as a whole, or cease redistributing), in a /similar/ way that the code being
released as proprietary software denies the same author access to code.

That's what I meant by :

> > It's worth noting that license 5 is the weakest level of control a
> > developer can exert. Someone can take your work and either restrict your
> > ability to take changes (that you can release as 5) by either
> > re-releasing your work in a derivative licensed under 4) or 1).

As a specific example, the originator of the work (X.c) would not be able to
take the changes (improvements) in the GPL'd derivative (gameover.c, X.c
from B) to incorporate back into their BSD (only) version. If they could, the
GPL would be bust.

Note: I'm not passing a judgement on this being good or bad, just expanding
on what I said.

At that point I'm bowing out of that discussion, since otherwise I'll be 
breaking a new year's resolution :)

:-)


Michael.
--
*personal opinions only & not a lawyer :) *
-
Sent via the backstage.bbc.co.uk discussion group.  To unsubscribe, please 
visit http://backstage.bbc.co.uk/archives/2005/01/mailing_list.html.  
Unofficial list archive: http://www.mail-archive.com/backstage@lists.bbc.co.uk/

FW: [backstage] BBC recruits Kazaa's Rose

2008-01-20 Thread Christopher Woods 
> -Original Message-
> From: [EMAIL PROTECTED] 
> [mailto:[EMAIL PROTECTED] On Behalf Of Ian Forrester
> Sent: 21 September 2007 12:39
> To: backstage@lists.bbc.co.uk
> Subject: [backstage] BBC recruits Kazaa's Rose
> 
> Quote from Ashley Highfield's monthly 
> 
> "As part of these changes, we're very fortunate to have 
> Anthony Rose on the team, who is joining the BBC as our new 
> head of the Digital Media team. He come to us from Kazaa with 
> a formidable background in p2p networks, DRM based content 
> publishing and 3D animation."
> 
> http://media.guardian.co.uk/newmedia/story/0,,2171162,00.html


I was going through the archives and saw that, which made me wonder - almost
five months have passed, and aside from the Flash streaming - which did
indeed make its debut just before the end of last year - has the acquisition
of Mr. Rose actually brought about anything truly groundbreaking or
productive for the Beeb? What with him joining the Digital Media team, and
with his vast experience in content distribution... Certainly, the core
strategy and architecture of the iPlayer doesn't seem to have changed (an
albeit more streamlined Kontiki experience, but still with the same MSDRM
and still transmitted via the same means) - I've certainly not seen any
particular innovations past what's already there.

Or are there even better things planned in the pipeline for public trialing
/ release this year?

Just curious, tis all... (We've heard precious little if anything at all
from or about Mr. Rose on here, I know he's not mandated to discuss the
iPlayer on officially-unrelated lists, but you'd think he'd at least pop his
head in once in a while - a lot of the other knowledgeable Beeb bods
(beebods?) participate on Backstage, and the BBC made enough effort to
publically announce his acquisition last year...)

-
Sent via the backstage.bbc.co.uk discussion group.  To unsubscribe, please 
visit http://backstage.bbc.co.uk/archives/2005/01/mailing_list.html.  
Unofficial list archive: http://www.mail-archive.com/backstage@lists.bbc.co.uk/