Re: [Gendispatch] Updated draft: Policy experts are IETF stakeholders

[Phillip Hallam-Baker <[e-mail]>]

On Mon, Jul 10, 2023 at 7:45 AM Tony Rutkowski <[e-mail]>
wrote:

> What/who is a "policy maker"?
>

More importantly, 'how do policy makers differ from other participants?'.

The real heart of the issue is that everyone participates in IETF as
individuals. We do not represent constituencies. Government policy experts
do represent a constituency and very often, an agenda.

And the problem for me here is that while in theory, the constituency HMG
employees are answerable to is the British people, in practice it is much
more likely to be the editors of the London tabloid press. And here I speak
with exceptionally personal knowledge. I knew a half dozen people at Oxford
who went on to be cabinet ministers pretty well. These are not people who
are desperately well versed in technology issues and certainly not on the
impact technology has on society.

Take the latest 'Dangerous Dogs Bill': Client side scanning of CSAM.

It is not at all hard to see where this comes from: CEO of company with
CSAM technology has lunch with Murdoch editor, lurid stories appear in
Tabloid press, eventually a junior minister is faced with the traditional
civil service soliloquy:

SOMETHING must be done!
This is SOMETHING!
Therefore we must do it!

And so a solution that isn't actually a solution must be implemented by
every service offering messaging services in the UK irrespective of the
fact it will enable further hack-frame-and-leak attacks against the
democratic process.


That said, there are some technology interests that are critical to the
functioning of democratic society. In particular questions such as:

* Who controls access to the public square?
  * Is there a public square?
  * Is it dominated by the loud mouthed thug with an amplifier?
* Who owns the data?
  * The individual to which it relates or some corporation?
* Who has agency?
  * Can people choose their service provider?
  * Can people change their service provider?

Rather than writing up documents describing input on how to raise such
issues, why not just write a document that raises them?