[gnso-rds-pdp-wg] The survey raw data issue

[Tapani Tarvainen]

Hi Maxim,

After some reflection I think I agree limited disclosure is
fine, even though I'd no problem with full disclosure either:
details such as IP addresses are not important.

What is important and should be public is who voted how on each
question, preferably in an easy-to-analyze form (spreadsheet).

It should not be difficult to generate such reports from
the polls.

I don't see any need for anonymity or privacy in this kind of
working group, any more than in parliamentary votes. If these
polls are used, as they are, in making decisions, people should
know who voted what, and it should not be made any harder than
necessary.

I do appreciate the problem of changing rules after the fact
though, so releasing such data from past polls would need a
stronger support from the group than for future ones.

Sincerely,

Tapani

On Jan 19 12:08, Maxim Alzoba (m.alzoba at gmail.com) wrote:

> Hello Sam, 
> 
> I think we might stick to (3)
> in format of choice between 3
> 
> 1. my name and affiliation is Ok to show
> 2. please show only my affiliation (could be group and not a company name_)
> 3 .please do not show any info. 
> 
> P.s: as an analyst I can say that the persons might be identified by the writing, so I see almost no value in hiding , and since we have 
> public records of meetings and we express the same ideas via voice and chat ... it is almost not possible to 
> push a particular idea without being identified.
> 
> Sincerely Yours,
> 
> Maxim Alzoba
> Special projects manager,
> International Relations Department,
> FAITID
> 
> m. +7 916 6761580
> skype oldfrogger
> 
> Current UTC offset: +3.00 (Moscow)
> 
> > On Jan 19, 2017, at 06:35, Sam Lanfranco <sam at lanfranco.net> wrote:
> > 
> > WG Colleagues, 
> > 
> > Here are my thoughts on the survey raw data issue under discussion in the RDS PDP WG. We face four options. They include: (1) no survey raw data disclosure (but still mean/std. dev. disclosure); (2) full survey raw data disclosure, (3) limited survey raw data disclosure, and (4) abandoning use of the survey. 
> > 
> > No disclosure (1) is the status quo. Full disclosure (2) maximizes transparency, at the risk of reduced survey participation and with little benefit over simple WG dialogue.  Limited raw data disclosure (3) is the RDS PDP WG Thick/Thin data challenge, only now with regard to our survey data fields. The design of a limited disclosure protocol is beyond the time and resources available to us, and details beyond mean/std. dev. probably mean a loss of confidentiality. Small participant size in these surveys means that disclosure beyond mean/std. dev. makes it harder for responses to remain confidential. Comments are already less than anonymous since we know each other’s proclivities and propensities. One does have a choice to not comment. A permission box [Show my name] is also problematic, given respondent numbers, since it makes it easier to identify “no name” respondents. 
> > 
> > Where do I stand on this? I am for either option (1) the status quo (no disclosure), or option 4 (no surveys at all). The survey is a quick aid to the WG dialogue and need not be seen as a binding measure of consensus. Survey results are not a vote. They are inputs for the WG dialogue grist mill, inputs that can facilitate the process of WG consensus. Confidentially poses no problem since the consensus process is still within the WG dialogue. The Chair of the WG, and the ICANN staff member, act as survey “scrutineers” and we should trust them to flag survey participation irregularities. 
> >   
> > If (1) the status quo (no disclosure) is not acceptable, I am in favor of (4) no surveys. Limited disclosure (3) is logistically problematic, and full disclosure (2) offers few benefits over simply conducting the dialogue within the  RDS PDP WG. To recap, I prefer either the status quo or no surveys at all. I look forward to other views on this matter. 
> > 
> > Sam Lanfranco, npoc/csih
> >