[UA-discuss] Review of UASG018

Jim DeLaHunt list+uasg at jdlh.com
Sat Jan 5 01:36:15 UTC 2019


Don:

Thank you for sending this document out for review.

It looks like this is only part of the complete content. Either a 
separately-authored piece of content needs to be merged in, or a 
companion document needs to be distributed with this one and 
cross-referenced. In either case, I don't think this .docx file by 
itself fulfills the goal of UASG018.

Details below.

On 2018-12-31 13:30, Don Hollander wrote:
>
> Please find attached a link to a proposed revision of UASG018.
>
> http://viagenie.ca/ua/UASG_Programming_Language_Framework_Review_v1.0%20-%20VG.docx
>
> Comments welcome, of course.
>
> Don
>
> Don Hollander
>
> Secretary General – UASG
>
> Skype: Don_Hollander
>
My background: I gave a presentation on the assessments encouraged by 
this document in Oct 2017 to the 41st Internationalization and Unicode 
Conference (/Universal Acceptance of non-Latin email addresses and 
domain names: how does your framework rate? (IUC41 presentation)/, 
<http://blog.jdlh.com/en/2017/10/31/universal-acceptance-eval-iuc41/>).

Top level comments from a quick read-through of this version:

/About this Document/, p. 5: "Technical details required by those 
performing library evaluations are presented in a separate document. 
This separation of documents is purely due to technical restrictions in 
the document platform. [Footnote] The tables in the technical 
presentation are wide, and best presented in landscape form. Google Docs 
cannot at present mix portrait and landscape pages in a single 
document." What is this "separate document"? I don't see a 
cross-reference to it. I don't see it circulated with this document.

The purpose of UASG018 is, as I understand it, to describe an evaluation 
methodology in sufficient detail that some competent engineer could use 
it as a guide to performing an evaluation of a framework, and that 
evaluations done by different people following UASG018 for different 
frameworks would provide comparable insight about the frameworks. 
Without the "technical details required by those performing library 
evaluations", this document is incomplete for that task.

And, why accept this limitation of the Google Docs tool as a reason for 
making UASG018 a half-document?  There are many alternatives. 1. Find an 
authoring tool more up to the requirements of the content. 2. Author the 
landscape-format tables as a separate Google Doc making a separate PDF 
file, and merge the PDF files into one UASG018 in PDF form. 3. Author 
the tables in landscape orientation, on portrait-orientation pages, and 
let their text become tiny. Those reading online can zoom in to make the 
text readable. 4. Distribute UASG018 as two PDF or .docx files.

But I don't think this single .docx file as distributed is acceptable to 
do the work of UASG018.

/Page count/: was 24 pages in Version 0.96 (March 10th, 2017), the 
previous copy I had locally. It is 17 pages in Version 1.1 (July 13th, 
2018). The list of changes don't explain to me how 17 pages was cut. 
Some is editorial: a blank page 2 was dropped. But some must have been 
substantive. The revision history should give a clue if scope was changed.

/File name/: the document circulated has "1.0" and 2018-12-31 in its 
file name, /UASG_Programming_Language_Framework_Review_v1.0 - VG 
2018-12-31.docx/ . But the content of the doc says it is "Version 1.1 
(July 13th, 2018)". I suggest that both version number and date in the 
filename match the document content, or else it will be hard to find the 
correct version of the document.

/Font of code samples/: e.g. Appendix A, code sample 1. On my computer, 
most of the code is in the Roboto Mono font, but the line "const char 
*name = u8"普遍接受-测试.世界";" appears in SimSun font. This is probably to 
represent the Chinese text in the example well. However, SimSun looks 
very different from Roboto Mono. It has thinner lines, different 
character widths, the glyphs have serifs, and the quote marks appear 
slanted instead of as upright C-language string delimiters. The code 
samples should be formatted in a consistent font.

/Appendix B - References/, page 16: The references to external documents 
are given as links, with link text being the title of the document, and 
the link reference being the URL of the document. This is workabout but 
not terribly clear. It would be better to give the title, the date of 
the revision accessed, then URL of the document as both link text and 
link reference. This way, someone who copies the reference and pastes 
into a text-only document won't lose the link.  It is also closer to 
traditional citation style.

"/a set of comprehensive test data/",//Footnote 3, page 7: This footnote 
text links to 
<ftp://ftp.unicode.org/Public/idna/latest/IdnaMappingTable.txt>. If that 
resource is useful enough to link to, it should be in Appendix B as well.

/Test case strings in reuseable form/: I had suggested before that UASG 
provide the strings listed as test cases in this document (and in its 
"technical details required by those performing library evaluations" 
sister document) be also provided as reusable data files. I recall you 
encouraging me to do this, and I have not yet followed through. I still 
think it's a good idea. This review cycle might be an opportunity to get 
it done. Once done, this document should link to that resource also.

I hope these comments are helpful in getting discussion going. I look 
forward to hearing from others.

Best regards,
       —Jim DeLaHunt, Vancouver, Canada

-- 
     --Jim DeLaHunt, jdlh at jdlh.com     http://blog.jdlh.com/ (http://jdlh.com/)
       multilingual websites consultant

       355-1027 Davie St, Vancouver BC V6E 4L2, Canada
          Canada mobile +1-604-376-8953

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mm.icann.org/pipermail/ua-discuss/attachments/20190104/581e4692/attachment.html>


More information about the UA-discuss mailing list