[Comments-devanagari-gurmukhi-gujarati-scripts-lgr-27jul18] A quick review of the Gurmukhi proposal

[梁海 Liang Hai]

- §3, “… but it has now been established, on the basis of its name, that the Indians did have a system of writing which must have been borrowed freely from local script.”: How’s this (and the following two paragraphs, and the whole §3.1) even relevant to the LGR proposal? Authors shall look for a proper place to publish their history research.

- §3.3, “… ligatures are formed only with following /h, r and v/ consonants.”: Has the well-known post-base form of ya already fell out of use in common text? Probably should mention this.

- §3.3.2, “Unlike Devanagari, Gurmukhi consonants are also used to represent consonant sounds where / ə / is not included in them.”: Both Hindi and Punjabi–Gurmukhi orthographies allow implicit dead consonants. It’s just Punjabi–Gurmukhi allows more. This level of spelling and reading rules are not really relevant to the proposal. An encoded pure killer (virama/halant) is only used when the mark or its conjunct-forming effect visually exists.

- §3.3.2, ‘In Gurmukhi, virama “੍” (U+0A4D) is used in place of halant "੍" (U+094D)’: This sentence only brings confusion. U+094D as a Devanagari-specific character has nothing to do with Gurmukhi. Are the authors going to clarify such relationship between other cognate graphemes too?

- §3.3.2, “In Gurmukhi, virama is not used with any consonant that represents only the consonant sound instead of consonant plus vowel sound”: Rewrite to “The grapheme of virama is not used in Punjabi text to strip a consonant letter’s implicit vowel.”

- §3.3.4: “Suprasegmental” is not an appropriate term here, since at least gemination is segmental. Also, according to §3.3.4.1 and §3.3.4.2, the nasality is not pure nasalization of vowels but is segmental nasal consonants also.

- §3.3.4.2, rule 1: The detailed phonetic spelling logic (eg, “… the forms of u, uu vowels after any other vowel …”) is not really relevant to text encoding.

- §3.3.4.3, “In these letters, NGA (ਙ) and NYA (ਞ) are nasal consonants so these are stressed or doubled by the nasal sign tippi.”: Suspicious explanation. What about na and ma then?

- §3.3.4.4, “But in Gurmukhi, these letters can also be written as a single unit …”: There’s a difference between writing and encoding.

- §3.3.5, “Some of the character combinations … are encoded using ZWJ and ZWNJ.”: How are multiple-vowel-sign clusters encoded using ZWJ/ZWNJ?

- §4.1.3: Visarga is used for marking abbreviations according to §3.3.4.5. Need to clarify this either in this section or in §4.1.3.

- §4.1.6, “These characters can occur as single character words, but in TLD, single character labels are not allowed, so these letters will not be added.”: Should introduce and better discuss the usage of them in “single character words”, as those words can presumably appear in multi-word labels too.

- §4.1.6: Also, since a/aira is also a vowel carrier, the section needs to be worded more accurately.

- §5.3, “It is very easy for a native language speaker to count the number of syllables in a sequence”: Don’t exaggerate. The split of phonetic syllables and orthographic syllales in Indic scripts makes it often confusing for native users to count a certain type of syllables.

- §5.3, “The definition is a combination of 2 rules”: Similar streamlined rules/patterns should be included in other scripts’ corresponding sections in their LGR proposals. Also, the “{CH}” part in the pattern is worth considering by authors of the other proposals.

- §5.3, 3rd table, row 2, “Zero or one Consonant + Virama/Addak sequence followed by consonant is a syllable”: `CA` is a preceding orthographic syllable and is not relevant to this rule. The rule above the table is not even consistent with the original introduction.

- §5.3, “Examples of combination of the rules”, “2. ਪਿਰੰਦਾ (parindā)”: The authors keep mixing up phonetic strucutres and written structures. There’s no V (already defined as independent vowel letters) in this word. It’s CCMDCM. Same problem in “3. ਅੰਦਰ (andar)”: it is VDCC, what are “Vm” and “CvC”?!

- §7: A comprehensible pattern for other reviewers to consider: `[ C[N]{HC}[M] | V ] [A|B|D]`

- §7.6: Probably too restrictive as this is about spelling conventions (note ਐ and ੈ are already special cases, and there can be more). It’s not future-proof to limit the usage when there’re no confusability issues.

Best,
梁海 Liang Hai
https://lianghai.github.io

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mm.icann.org/pipermail/comments-devanagari-gurmukhi-gujarati-scripts-lgr-27jul18/attachments/20181008/eae6f515/attachment-0001.html>