[hebrew/block] please add note referencing UTN #27 for U-ZARQA & U-ZINOR

[bdenckla]

[source] https://r12a.github.io/scripts/hebrew/block

Unicode Technical Note #27: Known Anomalies in Unicode Character Names

I actually think the summary of the issue in the Technical Note is still mildly backwards:

• It is U-ZARQA's name that is dead wrong, and thus it is this code point that absolutely should have been named U-TSINNORIT (or similar; I think Unicode's transliteration scheme doesn't preserve gemination, i.e. doesn't double consonants, so maybe U-TSINORIT).
• U-ZINOR's name is reasonable, although perhaps it should have been U-ZARQA, and if not, it almost certainly should have been spelled U-TSINOR, to correspond to the transliteration used in other code point names with initial tsadi, notably U-TSERE & U-TSADI itself.

For the record, I'll paste in the summary in the Technical note, even though it slightly further confuses this already-confusing issue by being mildly backwards. I'll add my editorializations in square brackets.

U+0598 HEBREW ACCENT ZARQA Perhaps should have been called Hebrew accent tsinnorit. [Mildly wrong: it absolutely should have been called Hebrew accent tsinnorit, not perhaps.] May also be used for zarqa when shown on accented non-final letter. See Appendix A.

U+05AE HEBREW ACCENT ZINOR [Perhaps s]hould have been called Hebrew accent zarqa (= tsinor). See Appendix A.

It is a much smaller (and debatable) issue, but I also take issue with the suggestion that U-ZARQA "[m]ay also be used for zarqa when shown on accented non-final letter." I plan to submit a UTN myself addressing this issue. Thank goodness, Unicode makes no analogous suggestion that U-QADMA may be used for pashta when pashta is shown on a non-final letter. I have seen texts that do this with U-QADMA and I think it is an abomination. Only slightly kidding. I plan to recommend that Unicode go further than just not mentioning this possible use of U-QADMA but actually discouraging it.

The overall principle I intend to promote is that all accents use the same code point in their primary role and their stress helper role:

• U-PASHTA should be used as the stress helper for itself.
• U-ZINOR should be used as the stress helper for itself.
• U-DEḤI should be used as the stress helper for itself (if anyone ever produced a text with deḥi stress helpers; AFAIK, no such text exists)

This "self-help" already has to be the case for the following code points, which have no overly-clever "impositive (central) cousin" option:

• U-SEGOLTA
• U-TELISHA QETANA
• U-TELISHA GEDOLA

(E.g. U-PASHTA has the (bad) option of using U-QADMA as an overly-clever "impositive (central) cousin" to represent its stress helper.)