Re: [hrpc] draft-irtf-hrpc-association-02

Hi Avri,

Thanks for the comments, response inline:

On 3/28/19 12:02 PM, Avri wrote:
> Hi,
> 
> Again apologies for the last minute comments.
> 
> Abstract: saying that something establishes  ‘the causal link’ is a very strong claim.  Are we sure we want to make that strong a claim?  Can we really prove a causal link and can we prove that it is ‘the’ causal link. I personally do not see this draft as establishing any causal links, though it does point toward where they might be found.

Could you elaborate why? There is a research question, proposition, a clear understanding of the concepts, and the relation is shown consistently across various cases. So why do you think causality cannot be shown here?

> 
> Introduction: “by investigating the exact impact of Internet protocols on specific human rights,” again a very strong claim.  Can we ever really know the exact impact in some possible causal chain?  An even if we want to position a possible impact, we need more definitive illustrations and argumentation.

I think we describe quite a few cases. What are you missing?

> 
> There is an ambiguity between  allowing a right to prosper meaning a right is enabled, and what looks like a claim to a ‘right to prosper’. This is just a language nit, and I have avoided commenting on those for the most part, but this jumped out at me.
> 

Good point. Changed into:

One one hand, the right to freedom of assembly and association protects collective expression. Likewise, systems and protocols that enable communal interactions between people and servers enable this right given that the Internet itself was originally designed as "a medium of communication for machines that share resources with each other as equals" {{NelsonHedlun}}.

> 4. Methodology
> 
> “been further validated through confirmatory research in the form of Human Rights Protocol Reviews.” I think this is the primary reason for the HRPC to do reviews, but I do not beleive we have established a methodology for doing those yet. 

The reviews gathered data that we use for confirmation, for that they do not need to be standardized themselves, it is about the data they produce with furthered the relation.

> At this point some people are doing reviews and delivering them to responses that vary from happy acceptance to irritation to shock and awe. As far as I can tell we have not yet developed a systematic methodology for measuring the impact of the considerations in test reviews.

But that is also not the topic of draft-association, nor needed for the reviews to increase the validity of the claim. 

> Nor are we yet studying the reviews to see how they apply the considerations.  We have not yet established a rigorous method for testing our ethnographically established hypothesis. 

Idem.

> 
> “Even though the present work does not seek to create new guidelines, the conclusions could inform the development of new guidelines such as is done in draft-irtf-hrpc-guidelines.” I think we need a tighter binding between the discussion in this draft and anywhere in the guidelines where a consideration is said to be relevant to freedom of Expression/Assembly.

We did improve the interrelation, as you suggested during last session, do you have an idea how this could be done better?

> 
> Perhaps this is a personal academic prejudice, but while I think ethnography is extremely useful in hypothesis formation, I do not see how it can also serve to test and verify relevance and usefulness. Seems circular process to me. I think we need to find other methods for testing and measuring.
> 

You seem to say here that you reject ethnography as a scientific method to establish relations? Ethnography is the systematic study of groups of people and exactly aimed at understanding relations, so I hope you can provide more background to this, quite strong, claim.

> 6.  Cases and examples 
> 
> Seems like a loose collection of internet features that may or may not have a positive or negative affect on expression & assembly.  

Are you criticizing the method of case selection? Because what you describe seems to be indeed a case study format. 

I do not see the persuasive argument that shows the possibility of a causal chain.  Perhaps I am just not seeing it and others do, I accept.  Even in the two discussion sections, 7 & 8, the draft seems to jump from assertion to assertion without the rigor of argument that makes the conclusions inescapable. 

Could you comment in text? Because I think the discussions are quite tight.

 I do not see the arguments that show why consideration in general or why specific considerations would be significant in enabling or disabling the freedom of association and assembly, though the guidelines does make those sorts of claims.

I don't think this documents makes a statement about considerations at all, neither in the cases, nor in the conclusion. It just shows that there are interrelated concerns in RFC8280 and this document. During last session you asked us to foreground these interrelations, as you did above, and so we did.

But the argument is not about considerations, but about the relation between protocols and architecture and the right to association and assembly,

> 
> Re considerations:  While I understand that the authors do not want to introduce new considerations related to assembly and association - there may not be any new considerations or changes to existing considerations, I still think the draft needs to show a mapping between internet protocols and features and existing considerations.

These are mentioned after every case.

> 
> Note: I have not called them out, but the draft needs a spelling and grammatical scrubbing.

Agreed, will do. But also hope to work on this with the RFC editor. 

> 
> Thanks for continued efforts on this draft.
> 

Thanks for the review and happy to discuss.

Best,

Niels

> Avri
> 
> _______________________________________________
> hrpc mailing list
> hrpc@irtf.org
> https://www.irtf.org/mailman/listinfo/hrpc
> 

-- 
Niels ten Oever
Researcher and PhD Candidate
DATACTIVE Research Group
University of Amsterdam

PGP fingerprint	   2458 0B70 5C4A FD8A 9488  
                   643A 0ED8 3F3A 468A C8B3