Re: [Gendispatch] I-D Action: draft-halpern-gendispatch-antitrust-00.txt

Hi Christian et al,

> On 4 Sep 2021, at 10:12 am, Christian Huitema <huitema@huitema.net> wrote:
> 
> One of the biggest attacks against the Internet is the generalized spying on Internet users. Preventing that spying is a fairly important topic when debating drafts and standards. Yet it is very hard to debate that without also discussing  surveillance as a business model. So let's be careful with the potential backlash of good intentions!

The language used is 'business relationships between specific vendors and customers', noting the use of 'specific'. That does not preclude talking about general market dynamics such as surveillance capitalism.

If you want to use the relationship between a specific vendor and their customers or suppliers as an example of such a market dynamic, I can see how you'd be concerned about this language. I don't see a good way to carve that out without refusing right of reply to those actually involved in that relationship. 

I think the root of the issue here might be that 'should not discuss' is intended to give guidance about what's safe to do, but it can (and probably will) be used to squash some future discussions -- i.e., the text can be weaponised. 

A potential fix would be to separate out that line as well as 'details of particular supply chains' and 'specific market opportunities' into a separate section where 'should not discuss the following topics' is replaced with something like 'should be aware of the antitrust implications of discussing the following topics.'

Alternatively, we could try to qualify these statements to make it clear what words like 'specific' mean.

Similarly, this text caused me concern:

'IETF participants, particularly those in an IETF leadership position, should not engage in [...] behavior that may be considered abuse of a dominant position.'

My understanding[1] is that there is very little case law or regulatory guidance addressing how standards setting organisations should address abuse of dominance beyond the relatively well-understood minefield of patent law. However, it has become a very 'hot' area, thanks to things like the CMA's (and now Europe's) actions against Google regarding the Privacy Sandbox, and various complaints and proposals in the US.

Such a broad and pre-emptive statement on behalf of the IETF could take us to undesirable places; I can easily imagine this statement being weaponised along the lines of something like 'it's IETF policy that you don't behave in this fashion, some competition regulator has expressed concerns about this area, therefore you should stop it' -- even though a court hasn't yet ruled on the matter.

AFAIK it's also far from clear that as a standards development venue the IETF would be held responsible for someone behaving in such a way. All of the current case law and regulatory guidance is regarding cartel behaviour, not abuse of dominance, and I don't think we should pre-empt it.

I think the fix here is to either remove the statement, or change 'should not engage in the following' to 'should be aware of the antitrust implications of the following'.

Cheers,

1. With the proviso that I'm more familiar with European competition law than US anti-trust.

> 
> -- Christian Huitema 
> 
>> On Sep 3, 2021, at 4:36 PM, Rob Sayre <sayrer@gmail.com> wrote:
>> 
>> 
>> On Fri, Sep 3, 2021 at 4:29 PM Brian E Carpenter <brian.e.carpenter@gmail.com> wrote:
>> 
>> By stating that a draft is describing a proprietary technology and isn't suitable for IETF discussion unless the proprietor gives change control to the IETF. That's happened many times.
>> 
>> I think this description is too reductive. It could be "open source" and "royalty free", but still be in the interests of just a few entities. The "end run" provisions in other RFCs are also a concern here.
>>  
>> I don't see anything in the draft 
>> that forbids this. It just says: the underlying business arrangements should not be discussed in the IETF. To me, that's nothing new.
>> 
>> If it's "nothing new", I am not sure why it needs to be in the draft.
>> 
>> thanks,
>> Rob
>> 
>>  
>> -- 
>> Gendispatch mailing list
>> Gendispatch@ietf.org
>> https://www.ietf.org/mailman/listinfo/gendispatch
> -- 
> Gendispatch mailing list
> Gendispatch@ietf.org
> https://www.ietf.org/mailman/listinfo/gendispatch

--
Mark Nottingham   https://www.mnot.net/