[UA-Measurement] Independent evaluation work in LAC starting

[Mark W. Datysgeld]

Thank you for the interest, everyone.

Matogoro, let's make sure to stay in contact as this project unfolds.

Nico, we'll have a preliminary meeting next week and I will see what the situation for UY is. I can put you in contact with them right after.

Roberto, this is being pushed forward by Sylvia Leite - sylvia at internautabrasil.org. She is actually very interested in this sort of information, so please reach out to her.

Michael, I intend to move forward with the creation of different schemas to standardize our work, and in this sense I find it essential that in the upcoming meetings we define the scope of what are our testing subjects.

As for MongoDB, I retrieved the minutes from the meeting in which the team decided on using it, and we felt that a nosql approach was interesting due to how scalable and approchable it is, in such a way that this can be worked on by different developers over time with no friction, something that seems ideal for UA. It is also pretty good at managing a cluster with N copies in a way that keeps data highly available.

In relation to some of your concerns, I believe there were issues with earlier implementations of this technology, but now it feels quite mature considering the projects we've been following. Nosql is being deployed by IBM, UDacity, Sony and others at a large scale.

Even so, we have been considering taking the database to MongoDB Atlas, which is a great solution for consistency and minimizing issues. This is something we definitely could discuss as this moves forward.

Best,
-- 
Mark W. Datysgeld from Governance Primer [www.markwd.website]
Representing businesses in IG together with AR-TARC and ABES

On July 11, 2019 10:59:11 PM GMT-03:00, Michael Casadevall <michael at casadevall.pro> wrote:
>I like the concept of having a centralized database with the state of
>tested sites and their current "pass/fail/other" status.
>
>It may be worth extending this concept to also include incorporation of
>a schema for testing of tools, standardized libraries and off-the-shelf
>products for EAI/IDN status and where possible, provide an integrated
>API where tests can be submitted as these type of evaluations can
>generated be automated through a test harness.
>
>As a further note, I have some (IMHO) minor concerns with MongoDB being
>used as a back-end as it can make searching the database somewhat more
>difficult than a standard relational database plus historic issues
>relating to Mongo in regards to replication causing data corruption.
>
>On 7/10/19 7:17 PM, Mark W. Datysgeld wrote:
>> Hello everyone,
>> 
>> an independent IDNs study group was assembled within ALAC-LACRALO.
>They
>> showed excitement for multilingualism when I first presented it at
>LAC
>> DNS Forum last year, and are now moving into working on the subject.
>> 
>> Their initial aim is to perform regional evaluations for e-mail
>> acceptance in top local websites of different LAC countries, like the
>> one my team did in Brazil last year (attached).
>> 
>> My key concern is that, as we galvanize UA, interest in these smaller
>> studies will grow, and there is a need to keep all that information
>> useful and accessible.
>> 
>> There are two priorities I am currently working on:
>> 
>> 1) Creating a 1-page document describing which websites are supposed
>to
>> be tested and how people should do it. We have followed some
>guidelines
>> excluding malware domains but not discriminating websites due to
>their
>> content, for example.
>> 
>> 2) In anticipation of the global e-mail tests report that I hope will
>be
>> released this week, I think any measurement report moving forward
>should
>> follow an appropriate metadata schema that allows us to aggregate
>data
>> over the years and across all regions. I am including a copy of the
>one
>> we developed attached to this e-mail (it is an excerpt from the full
>> report).
>> 
>> I would appreciate any information on how this collides with any
>> preexisting UA documents, as well as people or teams that are
>> potentially working in parallel in similar matters, and so on. I
>think
>> it's vital to be proactive at this early stage.
>> 
>> Regards,
>> 
>> 
>> _______________________________________________
>> By submitting your personal data, you consent to the processing of
>your personal data for purposes of subscribing to this mailing list
>accordance with the ICANN Privacy Policy
>(https://www.icann.org/privacy/policy) and the website Terms of Service
>(https://www.icann.org/privacy/tos). You can visit the Mailman link
>above to change your membership status or configuration, including
>unsubscribing, setting digest-style delivery or disabling delivery
>altogether (e.g., for a vacation), and so on.
>> 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mm.icann.org/mailman/private/ua-measurement/attachments/20190714/c0ba8237/attachment.html>