[Neobrahmigp] IP review of Tamil LGR proposal v3.0

[Sarmad Hussain]

Dear Shanmugam, all,



Please find below the review of the proposal by IP.  Please visit here 
<https://www.icann.org/resources/pages/lgr-proposals-2015-12-01-en>  to see 
the Arabic proposal, in case you would like to review the no-mix rule.



Kindly let us know if you have any queries.



We look forward to the final version, incorporating the feedback.



Regards,
Sarmad





To: NeoBrahmi Generation Panel
From: Integration Panel
Subject: Review of Tamil LGR dated Dec 6, 2018

We have reviewed the updated proposal dated 2018-12-06 including XML and TXT 
files.

Here are our findings:

TECHNICAL / DOCUMENTATION ISSUE

Tamil is one of the few LGRs that assign an "allocatable" variant, yet the LGR 
did not defined a "no-mix" rule to prevent needless multiplication of variant 
labels where the same label alternates the variant used.



The LGR makes the two encodings of SRI/SHRI allocatable but it is not clear 
why it would be essential, or even desirable to support all 4 permutations in 
a case where SHRI/SRI occurs twice in the same label.



--- SRI ---- SRI ---

--- SRI ---- SHRI ---

--- SHRI ---- SRI --- 

--- SHRI ---- SHRI ---



Here "---" stands for some code points common to both labels.



It seems only necessary to allow one or the other spelling consistently for a 
single label.



If, unexpectedly, there is some requirement to cater to all possible 
permutations, then this needs to be documented very clearly and the cost for 
supporting it (potentially many more than 2 allocatable labels) must be 
spelled out.



Alternatively, the LGR should be amended to contain a "no-mix" rule, patterned 
after the Arabic LGR's no-mix rules, or other suitable restriction.



DOCx:

The main document shows no changes other than what looks like minor copy 
editing/editorial. No additional issues identified.

In section 5.2.1, only the variant sequences for 6.1.3 have been listed, but 
not the ones for 6.1.1 (U+0B92 U+0BB3) and 6.1.2 (U+0BC6 U+0BB3). This is not 
consistent and is at odd with the XML/HTML where obviously sequences appear 
along single code points.

The remedy is add these 2 sequences in section in 5.2.1 with references to 
6.1.1 and 6.1.2 as it has already been done for the other pair.



While the two sets are slightly different because the sets for 6.1.1 and 6.1.2 
are 2 to 1 variants unlike the one in 6.1.3, the difference appears immaterial 
for this purpose and having a full list of sequences in section 5 makes sense.



XML:

No changes found other than Date. One common editorial issue:



(1) as with all LGRs the placeholders in the <description> need to be updated 
with final date and URL:

       Neo-Brahmi Generation Panel, "Proposal for a Tamil Script Root Zone 
Label Generation Rule-Set (LGR)", [URL and Date TBD]



The XML passes the tool



TXT:

Test label file verified; existing TLDs show as valid.





  _____





-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mm.icann.org/pipermail/neobrahmigp/attachments/20190206/53583f63/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/pkcs7-signature
Size: 5026 bytes
Desc: not available
URL: <http://mm.icann.org/pipermail/neobrahmigp/attachments/20190206/53583f63/smime.p7s>