"Sagispeak"

[cmloegcmluin]

Is the "Plain text shorthand" here an update to the "Mixed short ASCII representation" that appears in the outdated Sagittal-2 Character map?

I don't understand the application for either. The long-ASCII form requires more characters but is so intuitive. Anecdotally, I have yet to see anyone communicating about Sagittal using either "short" style myself. And I also don't understand why the sets of characters used in them must be mutually exclusive, as I feel like I've seen someone say is the case.

I feel like this question may be a bit off-topic w/r/t to Sagispeak, but I just haven't found any mention of this plain text shorthand anywhere else.

[Dave Keenan]

Is the "Plain text shorthand" here an update to the "Mixed short ASCII representation" that appears in the outdated Sagittal-2 Character map?

Yes, it is. Note that we no longer use any "extended ASCII" characters, so there is no longer a short version for anything outside the Athenian single-shaft set.

At some stage, I decided "ASCII" was a dying term (given Unicode) and that, in any case, it was a computer-geek term that musicians shouldn't be expected to know, so I recommended using "plain text" as a synonym for "ASCII" as Apple Computer have done. But "short-ASCII" and "long-ASCII" roll off the tongue far more easily than "plain text shorthand" and "plain text longhand". And you're forced to use mixed-Sagittal when using the short-ASCII because it's too hard, and pointless, to encode any multi-shaft symbols.

I don't understand the application for either. The long-ASCII form requires more characters but is so intuitive. Anecdotally, I have yet to see anyone communicating about Sagittal using either "short" style myself.

Yes. We probably should have stopped with the Spartan set. I guess we filled out Athenian because we could.

I think I have only ever used \/ tf v^ as the down/up versions of the 5-comma, 7-comma and 11-diesis or as 1, 2 and 3 degrees of 72-edo. But I can see those 3 pairs, plus dq for the 13-dieisis, being useful for representing the prime-factor JI notation (formerly know as multi-sagittal or the one-symbol-per-prime notation) in email.

If you learn the sagispeak for the symbols you use, you've also learned the downward short-ASCII in many cases. And then it's usually not too hard to also remember the upward character.

It was also useful to have characters that could be used in filenames, when naming the files that contain bitmaps for the symbols, as used in Scala. So that's an argument for having short ASCII for all single-flag symbols, which can then be re-purposed to represent the flag itself.

And I also don't understand why the sets of characters used in them must be mutually exclusive, as I feel like I've seen someone say is the case.

They are not quite mutually exclusive, since "/" and "\" are used in both short and long ASCII forms. They essentially have the same meaning in both forms, so it's not a problem. But you don't want it to be possible to interpret some string of characters in two conflicting ways, depending on whether you interpret them as the short or the long form.

[cmloegcmluin]

Yes, it is. Note that we no longer use any "extended ASCII" characters, so there is no longer a short version for anything outside the Athenian single-shaft set.

Sorry, what are the "extended ASCII" characters? I assumed they'd be in the Character Map but I'm not finding them. Maybe I don't have to worry about them, though, if we no longer use them.

At some stage, I decided "ASCII" was a dying term (given Unicode) and that, in any case, it was a computer-geek term that musicians shouldn't be expected to know, so I recommended using "plain text" as a synonym for "ASCII" as Apple Computer have done.

ASCII works for me too, but I think you're right. I'll switch over to using "plain text" too.

It was also useful to have characters that could be used in filenames, when naming the files that contain bitmaps for the symbols, as used in Scala.

Okay, well that's a good use case.

But you don't want it to be possible to interpret some string of characters in two conflicting ways, depending on whether you interpret them as the short or the long form.

Got it, makes sense. I'll keep this constraint in mind when pondering the issue you posed to me elsewhere of the plain text for the new Olympian diacritics. Thanks.

[Dave Keenan]

Sorry, what are the "extended ASCII" characters? I assumed they'd be in the Character Map but I'm not finding them. Maybe I don't have to worry about them, though, if we no longer use them.

They are in the obsolete Character Map. You don't have to worry about them. See https://en.wikipedia.org/wiki/Extended_ASCII. The specific set we used was, I hope, the printable Unicode characters representable in 8-bits.

[cmloegcmluin]

Ah ha! So that's what the hexadecimal character code was about.

Well, I've gone over this thread and I have to say I'm a big fan of Sagispeak. The only thing I might like to change is the use of "k" as opposed to "c".

Generally I prefer "k" because its less ambiguous, always making the /k/ sound, whereas "c" makes /k/ ("fact"), /s/ ("face"), /ʃ/ ("special"), or /tʃ/ ("fettuccine"). However, since in Sagispeak the consonants always appear before an "a", and in these contexts "c" always makes /k/, it's not an issue.

And when its corresponding flag is doubled ( -> ) and we do the effect where the doubled letter is replaced with the letter and an "h" ("cc" -> "ch") the better choice is "c", because "ch" is a productive and distinct formation in more languages than "kh" is.

So in my notes I've retained "k" and "kh" as options, pronouncing "kh" as /x/ as in the German pronunciation of "Bach", but I prefer using "c" and "ch", pronouncing "ch" as /tʃ/. My plan is to soon prepare tooling and educational materials for Sagittal using these notes, so please let me know if you have opposing opinions.

Oh yeah, I did have another thing — I noticed that may be spelled either "sr" or "sl" however is only given one spelling option of "r". I assumed that was an oversight since there's no conflict in offering "l" as a spelling for too.

Oh wait, actually there was one more thing. On the New Olympian diacritics forum thread the Olympian diacritics are moved from the right side of the symbol to the left. I don't see this explicitly acknowledged there, but I assume that analogously they become prefixes now instead of suffixes? I see that is spelled "mi-ai" so it seems like this is indeed the case (by the way, are the hyphens necessary? I might prefer miao myself, and I don't see that introducing any ambiguity). So in that case would be "moovai" and pronounced with three syllables /mo o vai/ (perhaps we could deploy umlauts to disambiguate the syllable break there, i.e. "moövai"?).

I'm eager to hear opposing opinions, really. The only thing I was slightly disappointed about on this thread was that most of the talk was about the shorthand characters, and not the pronunciation (I didn't see anyone debating the "k" vs. "c" choice already). Maybe I'm a bit of a pronunciation geek and I was hoping for more entertainment!

[Dave Keenan]

Ah ha! So that's what the hexadecimal character code was about.

No. That tells what character the glyph was mapped to — what is called the "code point" in these days of Unicode. The extended ASCII characters I'm referring to, can be seen in the "Mixed short ASCII representation" column of http://sagittal.org/Sagittal2_character_map.pdf

Well, I've gone over this thread and I have to say I'm a big fan of Sagispeak. The only thing I might like to change is the use of "k" as opposed to "c".

Generally I prefer "k" because its less ambiguous, always making the /k/ sound, whereas "c" makes /k/ ("fact"), /s/ ("face"), /ʃ/ ("special"), or /tʃ/ ("fettuccine").

Yes. That's one reason why we didn't use "c".

However, since in Sagispeak the consonants always appear before an "a", and in these contexts "c" always makes /k/, it's not an issue.

An interesting point. I wasn't aware of that rule. I agree it checks out. Does it apply in languages other than English, that use the Latin alphabet?

However, another reason we didn't use "c" is that "c" doesn't look anything like while "k" does.

And when its corresponding flag is doubled ( -> ) and we do the effect where the doubled letter is replaced with the letter and an "h" ("cc" -> "ch") the better choice is "c", because "ch" is a productive and distinct formation in more languages than "kh" is.

Which is why we allow "ch" as an alternative spelling, with a corresponding alternative pronunciation as /tʃ/. And there is some resemblance between "c" and (if you ignore the shaft).

So in my notes I've retained "k" and "kh" as options, pronouncing "kh" as /x/ as in the German pronunciation of "Bach", but I prefer using "c" and "ch", pronouncing "ch" as /tʃ/. My plan is to soon prepare tooling and educational materials for Sagittal using these notes, so please let me know if you have opposing opinions.

Those two spelling options and pronunciation options are exactly what we suggest for . Note that both kh and ch can be pronounced /x/, but only ch can be pronounced /tʃ/. It should not be spelled "ch" in French as this would be pronounced /ʃ/ which would conflict with sh for . There was a suggestion to deal with that, in a previous post in this topic, namely spelling it as "tch" in French. I've now added that to the table.

But I don't want "c" as an option for , for the reasons given above.

Oh yeah, I did have another thing — I noticed that may be spelled either "sr" or "sl" however is only given one spelling option of "r". I assumed that was an oversight since there's no conflict in offering "l" as a spelling for too.

The "sl" spelling was allowed for because there is no consonant blend "sr". At least not in English. https://kiddymath.com/worksheets/sr-blend.

For we have the problem of "l" not looking like the downward symbol (you have to imagine the "r" rotated 180°), and we want a definite character for its short-ASCII, which is "r". However I have now added the text: "(may be pronounced ℓ)". Thanks for that.

Oh wait, actually there was one more thing. On the New Olympian diacritics forum thread the Olympian diacritics are moved from the right side of the symbol to the left. I don't see this explicitly acknowledged there, but I assume that analogously they become prefixes now instead of suffixes? I see that is spelled "mi-ai" so it seems like this is indeed the case (by the way, are the hyphens necessary? I might prefer miao myself, and I don't see that introducing any ambiguity).

I don't see the hyphen as necessary. This move from suffix to prefix was not discussed with George. It creates a possible problem when a mina diacritic is followed by a schisma diacritic, which gives mii mio, moi, moo, which may lead to the schisma diacritic not being clearly articulated. Any suggestions? Magrathean diacritics might also be considered.

I'm eager to hear opposing opinions, really. The only thing I was slightly disappointed about on this thread was that most of the talk was about the shorthand characters, and not the pronunciation (I didn't see anyone debating the "k" vs. "c" choice already). Maybe I'm a bit of a pronunciation geek and I was hoping for more entertainment!

Ha! Sorry. It's a shame you weren't around when we were designing it. We might have got there a lot quicker.

[cmloegcmluin]

However, since in Sagispeak the consonants always appear before an "a", and in these contexts "c" always makes /k/, it's not an issue.

An interesting point. I wasn't aware of that rule. I agree it checks out. Does it apply in languages other than English, that use the Latin alphabet?

According to this: https://en.wikipedia.org/wiki/Hard_and_soft_C#Other_languages it does.

There was a suggestion to deal with that, in a previous post in this topic, namely spelling it as "tch" in French. I've now added that to the table.

Shall we add the German spelling "sch" for "sh" as well?

However, another reason we didn't use "c" is that "c" doesn't look anything like while "k" does.

Fair enough. I suppose that consideration trumps the others.

The "sl" spelling was allowed for because there is no consonant blend "sr". At least not in English. https://kiddymath.com/worksheets/sr-blend.

For we have the problem of "l" not looking like the downward symbol (you have to imagine the "r" rotated 180°), and we want a definite character for its short-ASCII, which is "r". However I have now added the text: "(may be pronounced ℓ)". Thanks for that.

Except of course in the famous Sriracha sauce! I'm not sure if its as ubiquitous in Australia, but it's everywhere here (at least in California).

Again, the consideration for the resemblance of the shape to the letter is more important, so I'll delete my option of l for . I appreciate the consistency here, that every symbol with a single flag has just one consonant and just one spelling; only the symbols with multiple flags/consonants have multiple spellings. And then /l/ is an alternate pronunciation for "r" wherever it shows up.

This move from suffix to prefix was not discussed with George. It creates a possible problem when a mina diacritic is followed by a schisma diacritic, which gives mii mio, moi, moo, which may lead to the schisma diacritic not being clearly articulated. Any suggestions?

Bummer... I went back and edited my post just a couple minutes after first posting it to include an example of the "moo" problem. I suggested umlauts, like the New Yorker uses, although apparently I meant diaeresis: https://www.newyorker.com/culture/culture-desk/the-curse-of-the-diaeresis But I don't think it's the best solution. I solved the problem of breaking up vowel sequences in my own G.A Phonemic Transcription with apostrophes, so we could consider "mo'ovai" for

Magrathean diacritics might also be considered.

Welp, I came up with a proposal, spent about a half an hour on it, and it got lost when I tried to submit and I got booted to the login screen. I guess it wasn't meant to be. Perhaps I'll remember it later and chime in on the New Olympians thread where you shared Magratheans.

[cmloegcmluin]

Okay, it turns out I've got a little chunk of time before I have to get back to my day job, so while it's fresh in my head I'll try to dump this back out:

It's a shame that ambiguous "j" sucks up so many consonant sounds. What if we start using vowel sounds? "ei" and "eo" for half-tina and "ui" and "uo" for full tinas. They'd each be two syllables, the "e" pronounced /e/ and the "u" pronounced /u/. And we could combine "mi" and "bi" to get 3 minas (= 9 tinas). So here's the results:

tinas    up          down
0.5      ei          eo
1.0      ui          uo
1.5      eiui        eouo
2.0      uiui        uouo
2.5      eiuiui      eououo

3.0      mi          mo

3.5      eimi        eomo
4.0      uimi        uomo
4.5      eiuimi      eouomo
5.0      uiuimi      uouomo
5.5      eiuiuimi    eououomo

6.0      bi          bo

6.5      eibi        eobo
7.0      uibi        uobo
7.5      eiuibi      eouobo
8.0      uiuibi      uouobo
8.5      eiuiuibi    eououobo

9.0      mibi        mobo

9.5      eimibi      eombo

Or perhaps the last ones could be shortened to "eimbi" and "eombo", respectively... I itch to leverage the productiveness of the terminal consonant cluster "mb"!

Suitably insane, no?

[Dave Keenan]

Why do I find myself singing Old MacDonald had a farm?

[cmloegcmluin]

Okay, but Magratheans are out of scope for now...

What did you think of my more serious proposal to use apostrophes to break double vowels, e.g. "mo'ovai" for

And of adding “sch” to the table for German “sh” sound, as long as we’re adding “tsh” for French “ch” sound.