Really appreciate the detailed review. Thank you for taking the time!

--Rich

Richard Merdinger

VP, Domains - GoDaddy

From: ua-discuss-bounces@icann.org [mailto:ua-discuss-bounces@icann.org] On Behalf Of Tex Texin
Sent: Thursday, September 7, 2017 6:24 AM
To: 'Don Hollander' <don.hollander@icann.org>; 'Universal Acceptance' <ua-discuss@icann.org>
Subject: Re: [UA-discuss] Draft Media Release and Report Publication- email addresses

Regarding the email address report

A couple suggestions.

There are some styling issues. For example the font size changes in the last 2 paragraphs of the intro. Also, the table in Evaluation should be consistent about top or bottom cell alignment.
The document should be spell and grammar checked. For example: “Three attempt to explicitlys match”
I am not so sure about this “nobody attempts IPv6 literals”
This seems an odd aside “though one messes it up by not allowing for the enclosing [].”
The section on global vs local sites doesn’t seem to make a useful point. It mentions 4 companies, the Times of India being the only “local” or regional one. The authors expectations about the 4 companies aren’t met despite where the applications were developed, or the author’s view of their being modern or “staid”. This doesn’t seem to fit the heading global vs local sites. Singling out these companies does not seem wise nor there is a value to the section. With such a large number of sites failing, why mention these? It might be better to mention that the 749 sites included sites developed in non-English speaking markets and/or with target audiences in those markets, so the need for education and attention to UA is global (and the education itself must be multilingual).
Having castigated developers for accepting recommendations of regular expressions that did not work well for UA, isn’t it ironic that we recommend isemail for its popularity while admitting we don’t know how well it works and haven’t evaluated it?
I don’t think we should offer the mitigation sections without more clarity on the correct solutions. We certainly should not have them under a section called key findings as they are not findings.
It would be good to offer a reference or some other support for the claim “Regular expressions cannot fully validate an email address.” (I am not disagreeing. However, the claim should be justified.)

Hth

tex