I changed my mind: I'm willing to support the 2013 proposal #standardization


Nathan Galt
 

Assuming there's no way to get the Consortium to accept anything more than the bare minimum of letters, I'd like to say that I support the 2013 proposal (a copy's at https://www.quikscript.net/proposals/2013.pdf). It adds 9 letters.

If it's possible to get the entire chunk of sixteen instead of the minimal nine, I'd ask for:

  1. h
  2. l
  3. m
  4. n
  5. ʍ
  6. ɬ
  7. x
  8. ks
  9. gz
  10. ɔɪ
  11. b
  12. θ
  13. ð

My rationale? The fewer dissimilar-to-Quikscript-readers-but-not-dissimilar-enough-for-the-Consortium letters there are, the better things'll be for Quikscript readers and writers.

My response to question 10a, "Can any of the proposed character(s) be considered to be similar (in appearance or function) to an existing character?" would still be "no", of course, but Michael Everson has conditioned me to expect the Consortium to disagree.

Regarding section 6 — I take it there's a unified sort order (the kind that sort(1) would produce) and a Quikscript-specific sort order? I don't think I have anything to add here.

Notwithstanding the above, at this point I'm ready to support the 2013 proposal as-is.


Brad Neil
 

As I explained in my reply to Mr Everson last month, I'm not sure that any more than an extension of 5 letters would be required, if unification with Shavian were to occur:

  1. hw/ʍ
  2. x
  3. ɬ
  4. ks
  5. gz

The reason being, every existing Shavian codepoint directly corresponds with a QS letter (or sequence of two letters). If we are to go with unification, why add new encodings for letters that are direct, 1:1 (or many:1) equivalents? (even if they switch around shapes or replace them entirely.)

By all means, see my reply to his post "QuikScript and Unicode" for the full argument.

And just to play devil's advocate regarding your suggestion of 16 characters: /eɪ/ and /aɪ/ may have different shapes, but only barely. The top bar of Shavian 𐑱/𐑲 was switched out for a little loop at the top. I've never seen this an impediment to reading. Further, the only difference with /b/ is that it was mirrored horizontally in QS, with a loop added so that it ends on the baseline instead of the descender line.Those are pretty minimal changes in my opinion.

What do you think?


Nathan Galt
 



On Jan 28, 2020, at 10:34 PM, Brad Neil via Groups.Io <friedorange79@...> wrote:

As I explained in my reply to Mr Everson last month, I'm not sure that any more than an extension of 5 letters would be required, if unification with Shavian were to occur:

  1. hw/ʍ
  2. x
  3. ɬ
  4. ks
  5. gz

The reason being, every existing Shavian codepoint directly corresponds with a QS letter (or sequence of two letters). If we are to go with unification, why add new encodings for letters that are direct, 1:1 (or many:1) equivalents? (even if they switch around shapes or replace them entirely.)

If we’re using 𐑥 and 𐑯 (S·Mime and S·Nun) for QS·Ah and QS·Awe, we’re going to need new code points for m and n sounds. S·Ah and S·Awe don’t look like QS·May or QS·No, no matter how many beers you’ve had. Based on what Everson’s 2013 proposal, I do not believe the Consortium expects us to twist U+10477 (𐑷) into something that looks like a loop-de-loop (·No) or a clockwise spiral (·May).

If Michael Everson thinks he has a colorable case for adding those two letters to a Quikscript section in the SMP, I’m not about to contradict him. Doing so would merely make my job as a reader or one-man font foundry harder by arguing for maximally smashed-together character sets.

By all means, see my reply to his post "QuikScript and Unicode" for the full argument.

And just to play devil's advocate regarding your suggestion of 16 characters: /eɪ/ and /aɪ/ may have different shapes, but only barely. The top bar of Shavian 𐑱/𐑲 was switched out for a little loop at the top. I've never seen this an impediment to reading. Further, the only difference with /b/ is that it was mirrored horizontally in QS, with a loop added so that it ends on the baseline instead of the descender line.Those are pretty minimal changes in my opinion.

What do you think?

I think you’re better at this than I am ;)

More to the point, I’m tripped up/slowed down by more trivial things. While mnemonics like “the top bar of Shavian 𐑱/𐑲 was switched out for a little loop at the top” help, I’m more thrown by deviances than, I suspect, many people here.


John Cowan
 



On Wed, Jan 29, 2020 at 5:28 PM Nathan Galt <mailinglists@...> wrote:
 
More to the point, I’m tripped up/slowed down by more trivial things. While mnemonics like “the top bar of Shavian 𐑱/𐑲 was switched out for a little loop at the top” help, I’m more thrown by deviances than, I suspect, many people here.

That was an issue during Chinese/Japanese integration:  Japanese readers are thrown by differences that look insignificant to Chinese readers.  (In this case, "Japanese" and "Chinese" are cultural choices rather than languages: Chinese text in a Japanese context such as quotations from the Chinese classics uses Japanese shapes.)  The ISO committee, which represents the governments of all relevant countries (Japan, China, Taiwan, Hong Kong, Singapore, North Korea, South Korea) resolved the matter in favor of the "Chinese" viewpoint.[*]  

In practice, Japanese readers manage fine with Japanese-specific fonts that show the shapes they expect (a very rare character might look wrong to the minority of Japanese who know it).  QS and Shavian users would do the same.

[*] Except in the cases where a pre-Unicode Japanese encoding already distinguished close variants, which is not applicable to Shavian/QS.



John Cowan          http://vrici.lojban.org/~cowan        cowan@...
You cannot enter here.  Go back to the abyss prepared for you!  Go back!
Fall into the nothingness that awaits you and your Master.  Go! --Gandalf


Michael Everson
 

On 29 Jan 2020, at 22:27, Nathan Galt <mailinglists@ngalt.com> wrote:
On Jan 28, 2020, at 10:34 PM, Brad Neil via Groups.Io <friedorange79=yahoo.com.au@groups.io> wrote:

As I explained in my reply to Mr Everson
Michael.

last month, I'm not sure that any more than an extension of 5 letters would be required, if unification with Shavian were to occur:

• hw/ʍ
• x
• ɬ
• ks
• gz
The reason being, every existing Shavian codepoint directly corresponds with a QS letter (or sequence of two letters).
I understood that shape was important. There are shapes and sounds that are identical between the two scripts:

k s ʃ tʃ ŋ d z dʒ w ɪ ɛ æ ə ɔ ʊ r oʊ uː

There are shapes that are identical between two scripts but with different sounds:

θ/f m/ɑː n/ɔː (possibly ð/v)

There are five shapes and sounds that are new (not in Shavian):

ʍ ɬ x ks ɡz

There are shapes I judge to be different in enough in shape to disunify:

p t θ j b ɡ v ð ʒ h l m aʊ ɑː n iː eɪ aɪ oɪ ɔː

If ð/v are unified then maybe ʒ/ʒ should be too. But in a unified font based on Shavian you won’t get the loop in the QS ʒ. And you won’t get the upcurve on the QS v. This one is probably the unification/disunification to worry about.

If we are to go with unification, why add new encodings for letters that are direct, 1:1 (or many:1) equivalents? (even if they switch around shapes or replace them entirely.)
If we’re using 𐑥 and 𐑯 (S·Mime and S·Nun) for QS·Ah and QS·Awe, we’re going to need new code points for m and n sounds. S·Ah and S·Awe don’t look like QS·May or QS·No, no matter how many beers you’ve had. Based on what Everson’s 2013 proposal, I do not believe the Consortium expects us to twist U+10477 (𐑷) into something that looks like a loop-de-loop (·No) or a clockwise spiral (·May).
I don’t think the Consortium “expects” much. The proposal could suggest that the scripts be disunified. Osage was proposed as a separate script, then as Latin extensions, then as a separate script again.

If Michael Everson thinks he has a colorable
tolerable?

case for adding those two letters to a Quikscript section in the SMP, I’m not about to contradict him.
I think the maximum disunification that would permit fonts that each orthography could use happily would be the ones I gave in the Androcles-font image I sent out, possibly omitting v and ʒ (but in for a penny, in for a pound perhaps).

Doing so would merely make my job as a reader or one-man font foundry harder by arguing for maximally smashed-together character sets.
By all means, see my reply to his post "QuikScript and Unicode" for the full argument.

And just to play devil's advocate regarding your suggestion of 16 characters: /eɪ/ and /aɪ/ may have different shapes, but only barely. The top bar of Shavian 𐑱/𐑲 was switched out for a little loop at the top. I've never seen this an impediment to reading. Further, the only difference with /b/ is that it was mirrored horizontally in QS, with a loop added so that it ends on the baseline instead of the descender line.Those are pretty minimal changes in my opinion.

What do you think?
I think you’re better at this than I am ;)
I kind of disagree. Brad was wrong about 𐑱/𐑲 being glyph variants of the looped curves. And the implication that 𐑭/𐑷 could be treated as glyphs for 𐑥/𐑯, well that just either disunifies the two scripts entirely or leaves it for a simple font difference to be used, but I think that would make QS pretty illegible to people if a properly Shavian font style were used.

And then you have the problem of heavy ligation in QS, which can be handled by the font, since QS doesn’t treat the rhoticized vowels as atoic characters but as ligations of the vowel + r.

More to the point, I’m tripped up/slowed down by more trivial things. While mnemonics like “the top bar of Shavian 𐑱/𐑲 was switched out for a little loop at the top” help, I’m more thrown by deviances than, I suspect, many people here.
Perhaps I should take a paragraph of Alice in Shavian and try to show it in my De Vinne font with QS shapes. That would take some time though.

Michael Everson


John Cowan
 



(Private response)

> If Michael Everson thinks he has a colorable

tolerable?

_Colorable_ 'apparently true, potentially justifiable'.  It can also mean 'specious, feigned', but that's plainly not the meaning intended here.



John Cowan          http://vrici.lojban.org/~cowan        cowan@...
The internet is a web of tiny tyrannies giving an illusion of anarchy.
                --David Rush


Nathan Galt
 

Well, this is much more than I’d hoped for since 2013. Here’s a repost of the Androcles font with (some) Quikscript extensions on the third and fourth lines, just so everyone’s on the same page:



On Jan 29, 2020, at 3:56 PM, Michael Everson <everson@...> wrote:

On 29 Jan 2020, at 22:27, Nathan Galt <mailinglists@...> wrote:
On Jan 28, 2020, at 10:34 PM, Brad Neil via Groups.Io <friedorange79@...> wrote:

As I explained in my reply to Mr Everson

Michael.

last month, I'm not sure that any more than an extension of 5 letters would be required, if unification with Shavian were to occur:

• hw/ʍ
• x
• ɬ
• ks
• gz
The reason being, every existing Shavian codepoint directly corresponds with a QS letter (or sequence of two letters).

I understood that shape was important. There are shapes and sounds that are identical between the two scripts:

k s ʃ tʃ ŋ d z dʒ w ɪ ɛ æ ə ɔ ʊ r oʊ uː

There are shapes that are identical between two scripts but with different sounds:

θ/f m/ɑː n/ɔː (possibly ð/v)

There are five shapes and sounds that are new (not in Shavian):

ʍ ɬ x ks ɡz

There are shapes I judge to be different in enough in shape to disunify:

p t θ j b ɡ v ð ʒ h l m aʊ ɑː n iː eɪ aɪ oɪ ɔː

If ð/v are unified then maybe ʒ/ʒ should be too. But in a unified font based on Shavian you won’t get the loop in the QS ʒ. And you won’t get the upcurve on the QS v. This one is probably the unification/disunification to worry about.

If we’re fussing about the upcurve on the QS v then we ought to be fussing about the upcurve (or lack thereof) on 𐑟. Shavianists tend to treat the tail as…almost optional (looking in the back of my copy of Androcles and the Lion); in QS-land, quite often only the bottom half of QS·Zoo is what’s written if the previous letter ends at the baseline.

Also, if we’re fussing: I’d like Quikscript-specific rounded (as opposed to angled) ʊ and uː. 𐑵 looks an awful lot like QS·Eat.


If we are to go with unification, why add new encodings for letters that are direct, 1:1 (or many:1) equivalents? (even if they switch around shapes or replace them entirely.)

If we’re using 𐑥 and 𐑯 (S·Mime and S·Nun) for QS·Ah and QS·Awe, we’re going to need new code points for m and n sounds. S·Ah and S·Awe don’t look like QS·May or QS·No, no matter how many beers you’ve had. Based on what Everson’s 2013 proposal, I do not believe the Consortium expects us to twist U+10477 (𐑷) into something that looks like a loop-de-loop (·No) or a clockwise spiral (·May).

I don’t think the Consortium “expects” much. The proposal could suggest that the scripts be disunified. Osage was proposed as a separate script, then as Latin extensions, then as a separate script again.

If Michael Everson thinks he has a colorable

tolerable?

Colorable. 'apparently true, potentially justifiable’.


case for adding those two letters to a Quikscript section in the SMP, I’m not about to contradict him.

I think the maximum disunification that would permit fonts that each orthography could use happily would be the ones I gave in the Androcles-font image I sent out, possibly omitting v and ʒ (but in for a penny, in for a pound perhaps).

Maximum unification, surely? It’s trivially easy to make everyone happy when nothing is shared.

To repeat myself: The Androcles font as-is would be tolerable. Adding rounded Quikscript versions of ʊ and uː would make it great.


Doing so would merely make my job as a reader or one-man font foundry harder by arguing for maximally smashed-together character sets.
By all means, see my reply to his post "QuikScript and Unicode" for the full argument.

And just to play devil's advocate regarding your suggestion of 16 characters: /eɪ/ and /aɪ/ may have different shapes, but only barely. The top bar of Shavian 𐑱/𐑲 was switched out for a little loop at the top. I've never seen this an impediment to reading. Further, the only difference with /b/ is that it was mirrored horizontally in QS, with a loop added so that it ends on the baseline instead of the descender line.Those are pretty minimal changes in my opinion.

What do you think?

I think you’re better at this than I am ;)

I kind of disagree. Brad was wrong about 𐑱/𐑲 being glyph variants of the looped curves. And the implication that 𐑭/𐑷 could be treated as glyphs for 𐑥/𐑯, well that just either disunifies the two scripts entirely or leaves it for a simple font difference to be used, but I think that would make QS pretty illegible to people if a properly Shavian font style were used.


Agreed that it’d be pretty illegible to Quikscript people if things are being displayed Shavian-style.

And then you have the problem of heavy ligation in QS, which can be handled by the font, since QS doesn’t treat the rhoticized vowels as atoic characters but as ligations of the vowel + r.

More to the point, I’m tripped up/slowed down by more trivial things. While mnemonics like “the top bar of Shavian 𐑱/𐑲 was switched out for a little loop at the top” help, I’m more thrown by deviances than, I suspect, many people here.

Perhaps I should take a paragraph of Alice in Shavian and try to show it in my De Vinne font with QS shapes. That would take some time though.


If you’ve got the font and the Shavian-script source paragraph ready, I can do the transcription work, even if it means clicking on characters in a character picker. I’m running macOS here, and I’ve got Office 365, if either of that matters.