SCA2 questions

Discussions worth keeping around later.
User avatar
Qwynegold
Smeric
Smeric
Posts: 1606
Joined: Thu May 24, 2007 11:34 pm
Location: Stockholm

Re: SCA2 question

Post by Qwynegold »

Oh, I thought this thread had already been pruned, so I PMed Zompist. Maybe someone here could also help?
Qwynegold wrote:Hi! I'm working with SCA², and have come across a problem that I'm not sure if it's because of a bug, or if there's something that I'm just not seeing. Would you take a look at this please?

Categories

Code: Select all

C=mnpbtdkɡqʷpfbvtsdzfvszxɣrlʋ̥ʋθðqβ̊βʍwtʃdʒhʲʝɫŋjɮ
Rewrite rules

Code: Select all

<V_unstressed>|Ѵ
<V_stressed>|Ѷ
ɑ́|ắ
<P_voiced>|Ṕ
<P_unvoiced>|Ṗ
<A_unvoiced>|Ȧ
<A_voiced>|Ã
<F_voiced1>|Φ
<F_unvoiced1>|Ḟ
<bilabial>|B
pf|ȹ
bv|ȸ
ts|ʦ
dz|ʣ
<velar>|K
qʷ|ʠ
<alveolar1>|X
<guttural1>|Ⱪ
<V_unuvular>|Ṿ
<V_uvular>|Ṽ
ɒ̝|â
ɒ̝́|ấ
<uvular>|Q
<V_unrounded1>|Ɯ
<V_rounded1>|W
œ́|ő
ɞ́|ḗ
ɒ́|ǻ
<unlenited1>|DZ
<lenited1>|Θ
ʋ̥|ƒ
<labiodental1>|F
<bilabial1>|Ƀ
<labiovelar1>|Ẅ
<V_front1>|E
<V_back1>|O
β̊|ɸ
tʃ|ʧ
dʒ|ʤ
<O_voiced>|Õ
<O_unvoiced>|Ȯ
<C_unvoiced>|Ċ
ɛ́|ȅ
<V_front2>|Ε
<V_back2>|Ο
<V_unrounded2>|Ι
<V_rounded2>|Υ
ə́|ớ
Sound changes

Code: Select all

ou/øy/_(C(C(C(C))))j
əu/øy/_(C(C(C(C))))j
ɑu/æy/_(C(C(C(C))))j
ɑo/æø/_(C(C(C(C))))j
u/y/_(C(C(C(C))))j
o/ø/_(C(C(C(C))))j
[ɑɒ]/æ/_(C(C(C(C))))j
óú/ǿý/_(C(C(C(C))))j
ə́ú/ǿý/_(C(C(C(C))))j
ɑ́ú/ǽý/_(C(C(C(C))))j
ɑ́ó/ǽǿ/_(C(C(C(C))))j
ú/ý/_(ː)(C(ː)(C(C(C))))j
ó/ǿ/_(ː)(C(ː)(C(C(C))))j
[ɑ́ɒ́]/ǽ/_(ː)(C(ː)(C(C(C))))j
ou/øy/_C(C(C(C(C))))[ií]
əu/øy/_C(C(C(C(C))))[ií]
ɑu/æy/_C(C(C(C(C))))[ií]
ɑo/æø/_C(C(C(C(C))))[ií]
u/y/_C(C(C(C(C))))[ií]
o/ø/_C(C(C(C(C))))[ií]
[ɑɒ]/æ/_C(C(C(C(C))))[ií]
óú/ǿý/_C(C(C(C(C))))i
ə́ú/ǿý/_C(C(C(C(C))))i
ɑ́ú/ǽý/_C(C(C(C(C))))i
ɑ́ó/ǽǿ/_C(C(C(C(C))))i
ú/ý/_(ː)C(ː)(C(C(C(C))))i
ó/ǿ/_(ː)C(ː)(C(C(C(C))))i
[ɑ́ɒ́]/ǽ/_(ː)C(ː)(C(C(C(C))))i
Input lexicon

Code: Select all

ourtskjɑ́
urtskjɑ́
wɒjé
øvóːskjɑ
øvórːskjɑ
əurkstríː
ɑlmsprínː
ekúːstri
ekúnːspri
ekúlmstrik
I'm trying to do some umlauting, but none of the [j] causes it with this code, while [i í] does as expected. The output I get is:

Code: Select all

ourtskjɑ́ → ourtskjɑ́
urtskjɑ́ → urtskjɑ́
wɒjé → wɒjé
øvóːskjɑ → øvóːskjɑ
øvórːskjɑ → øvórːskjɑ
əurkstríː → øyrkstríː
ɑlmsprínː → ælmsprínː
ekúːstri → ekýːstri
ekúnːspri → ekýnːspri
ekúlmstrik → ekýlmstrik
So I tried some things, and it turns out that if I remove one optional C from the conditions dealing with j, then the code works for most part. I.e. with _(C(C(C)))j instead of _(C(C(C(C))))j. But not only that, with ou/øy/_(C(C(C)))j for example, [ourtskjɑ́] turns into [øyrtskjɑ́] even though [ou] and [j] have four consonants between them, and therefore ought not to be affected.

Another strange thing is that if the vowel is directly followed by [j] such as in the word [wɒjé], then the vowel will not change, even if only one optional C is specified in the code.
Image
My most recent quiz:
Eurovision Song Contest 2018

User avatar
R.Rusanov
Avisaru
Avisaru
Posts: 393
Joined: Sat Jan 05, 2013 1:59 pm
Location: Novo-je Orĭlovo

Re: SCA2 question

Post by R.Rusanov »

ts is counted as one character, due to your specifications
Slava, čĭstŭ, hrabrostĭ!

User avatar
Qwynegold
Smeric
Smeric
Posts: 1606
Joined: Thu May 24, 2007 11:34 pm
Location: Stockholm

Re: SCA2 question

Post by Qwynegold »

R.Rusanov wrote:ts is counted as one character, due to your specifications
Ohh, duh! Thanks for pointing it out. Now, if I change that [t] to another consonant, for the first two words in this list, it works like it should.

ourlskjɑ́ → øyrlskjɑ́
urkskjɑ́ → yrkskjɑ́
wɒjé → wɒjé
øvóːskjɑ → øvóːskjɑ
øvórːskjɑ → øvórːskjɑ

əurkstríː → øyrkstríː
ɑlmsprínː → ælmsprínː
ekúːstri → ekýːstri
ekúnːspri → ekýnːspri
ekúlmstrik → ekýlmstrik

But there are still unresolved questions left. Since [ts] is just one consonant, why don't words with that change? Or the words that I've marked with red here. It's as if the program thinks that (C(C(C(C)))) means that there must be exactly four consonants and no less.
Image
My most recent quiz:
Eurovision Song Contest 2018

zompist
Boardlord
Boardlord
Posts: 3368
Joined: Thu Sep 12, 2002 8:26 pm
Location: In the den
Contact:

Re: SCA2 question

Post by zompist »

Try rules like

ou/øy/_(C)(C)(C)(C)j

A quick test worked as expected for me-- the change happens with zero to four consonants before the j.

I'm not sure what you wanted with the nested parentheses, but I can tell you from the code that they certainly do nothing to help. :)

User avatar
Radius Solis
Smeric
Smeric
Posts: 1248
Joined: Tue Mar 30, 2004 5:40 pm
Location: Si'ahl
Contact:

Re: SCA2 question

Post by Radius Solis »

It might be worthwhile to rename this thread with a plural -s and move it to NOTA, because a) trials and tribulations people have run into in the past will still be on record to help anyone else who runs into the same ones a year later, and b) half the time it's something we can answer for each other anyway.

If nobody disagrees I'll do it tomorrow.

User avatar
Drydic
Smeric
Smeric
Posts: 1652
Joined: Tue Oct 08, 2002 12:23 pm
Location: I am a prisoner in my own mind.
Contact:

Re: SCA2 question

Post by Drydic »

I VIOLENTLY DISAGREE

You should do it now.
Image Image
Common Zein Scratchpad & other Stuffs! OMG AN ACTUAL CONPOST WTFBBQ

Formerly known as Drydic.

User avatar
KathTheDragon
Smeric
Smeric
Posts: 2139
Joined: Thu Apr 25, 2013 4:48 am
Location: Brittania

Re: SCA2 question

Post by KathTheDragon »

In which case, I also have a question. I have the rules ə/a/x_ and ə/a/_x, but is there any way to combine them? Also, why doesn't (say) S/R/_[VS] work? I've had to split it into two rules (one ending in V, the other in S), but there are numerous other pairs (or more) of rules just like that.

User avatar
Salmoneus
Sanno
Sanno
Posts: 3197
Joined: Thu Jan 15, 2004 5:00 pm
Location: One of the dark places of the world

Re: SCA2 question

Post by Salmoneus »

KathAveara wrote:In which case, I also have a question. I have the rules ə/a/x_ and ə/a/_x, but is there any way to combine them? Also, why doesn't (say) S/R/_[VS] work? I've had to split it into two rules (one ending in V, the other in S), but there are numerous other pairs (or more) of rules just like that.
To my knowledge, no, there's no way to do 'before and after' changes in one rule. And it appears that you can't put categories in square brackets. Of course, if V and S commonly pattern together, you can just define another category that includes both.
Blog: [url]http://vacuouswastrel.wordpress.com/[/url]

But the river tripped on her by and by, lapping
as though her heart was brook: Why, why, why! Weh, O weh
I'se so silly to be flowing but I no canna stay!

User avatar
KathTheDragon
Smeric
Smeric
Posts: 2139
Joined: Thu Apr 25, 2013 4:48 am
Location: Brittania

Re: SCA2 question

Post by KathTheDragon »

Thanks!

User avatar
Qwynegold
Smeric
Smeric
Posts: 1606
Joined: Thu May 24, 2007 11:34 pm
Location: Stockholm

Re: SCA2 question

Post by Qwynegold »

zompist wrote:Try rules like

ou/øy/_(C)(C)(C)(C)j

A quick test worked as expected for me-- the change happens with zero to four consonants before the j.

I'm not sure what you wanted with the nested parentheses, but I can tell you from the code that they certainly do nothing to help. :)
Ah, that was another thing I was gonna ask about, if (C(C(C(C)))) or (C)(C)(C)(C) was quicker for the computer to compute, or if it made no difference. But okay, one should avoid nested parenthesis.

Anyhow, I had tried things like ou/øy/_(C)(C)(C)(C)j before, and now I tried it again just to make sure. It works if there are exactly four consonants, but not if there are 0-3 consonants. This is what I get:

Code: Select all

ourlskjɑ́ → øyrlskjɑ́
ourskjɑ́ → ourskjɑ́
ouskjɑ́ → ouskjɑ́
oukjɑ́ → oukjɑ́
oujɑ́ → oujɑ́
But it works when you try it? :?
Image
My most recent quiz:
Eurovision Song Contest 2018

User avatar
KathTheDragon
Smeric
Smeric
Posts: 2139
Joined: Thu Apr 25, 2013 4:48 am
Location: Brittania

Re: SCA2 question

Post by KathTheDragon »

Tried it just now and it works for me in all those cases.

User avatar
Qwynegold
Smeric
Smeric
Posts: 1606
Joined: Thu May 24, 2007 11:34 pm
Location: Stockholm

Re: SCA2 question

Post by Qwynegold »

KathAveara wrote:Tried it just now and it works for me in all those cases.
That's strange. I tested it with both Chrome and IE just now, it does not work for me. I also asked a friend to try it, and it didn't work for him either. He has Firefox. What's different with you and Zompist? I tried bith with and without the rewrite rules, but it didn't make any difference.
Image
My most recent quiz:
Eurovision Song Contest 2018

zompist
Boardlord
Boardlord
Posts: 3368
Joined: Thu Sep 12, 2002 8:26 pm
Location: In the den
Contact:

Re: SCA2 question

Post by zompist »

As an experiment, start with the default categories, replace the rules with

ou/øy/_(C)(C)(C)(C)j

and the lexicon with

oucja
ouccja
oucccja
ouccccja
oucccccja

This works fine for me (PC, IE 10, but I tested it on Mac Firefox earlier). Does it work for you?

Also FWIW make sure you have the latest version (shift-Refresh), in case you have a cached older version.

User avatar
KathTheDragon
Smeric
Smeric
Posts: 2139
Joined: Thu Apr 25, 2013 4:48 am
Location: Brittania

Re: SCA2 question

Post by KathTheDragon »

I'm using the latest Firefox on Win7.

Cedh
Sanno
Sanno
Posts: 938
Joined: Tue Nov 14, 2006 10:30 am
Location: Tübingen, Germany
Contact:

Re: SCA2 question

Post by Cedh »

It works for me too (FF, Mac). Qwynegolds version didn't work, but I figured out why: the default categories don't include <k> because they're geared towards Latin orthography.

User avatar
Qwynegold
Smeric
Smeric
Posts: 1606
Joined: Thu May 24, 2007 11:34 pm
Location: Stockholm

Re: SCA2 questions

Post by Qwynegold »

OMG I finally figured out what the problem is. Thanks everyone whose been helping out. Now I realized that if I remove j from category C, then it works. It must be that when it says (C)(C)(C)(C)j, then in a word like "ourskja", j is counted as the fourth C rather than the obligatory j in the rule. >.<
Image
My most recent quiz:
Eurovision Song Contest 2018

zompist
Boardlord
Boardlord
Posts: 3368
Joined: Thu Sep 12, 2002 8:26 pm
Location: In the den
Contact:

Re: SCA2 questions

Post by zompist »

Hmm, yeah, the program doesn't know how to take multiple paths. In this case I'd just put j in a different category than C.

User avatar
Qwynegold
Smeric
Smeric
Posts: 1606
Joined: Thu May 24, 2007 11:34 pm
Location: Stockholm

Re: SCA2 questions

Post by Qwynegold »

Yeah thanks, I just did. :P
Image
My most recent quiz:
Eurovision Song Contest 2018

User avatar
Artaxes
Sanci
Sanci
Posts: 27
Joined: Tue Dec 21, 2010 3:31 pm
Location: Anshan Imparatorlugu
Contact:

Re: SCA2 questions

Post by Artaxes »

Hi!

How can make the assimilation of one category of a preceding consonants to the next immediately adjacent, with use of one-verse rules?
It's something like this:

agla → alla
abda → adda
erse → esse

User avatar
Rhetorica
Avisaru
Avisaru
Posts: 271
Joined: Sun Nov 03, 2013 3:33 pm

Re: SCA2 questions

Post by Rhetorica »

Artaxes: NS/SS/_
Where N is the assimilated category and S is the assimilating category.

User avatar
Artaxes
Sanci
Sanci
Posts: 27
Joined: Tue Dec 21, 2010 3:31 pm
Location: Anshan Imparatorlugu
Contact:

Re: SCA2 questions

Post by Artaxes »

Rhetorica wrote:Artaxes: NS/SS/_
Where N is the assimilated category and S is the assimilating category.
Thank you! But if two consonants are within the same category (ie. only voiceless plosives) must be assimilated by way "first to second" ?

P=ptʈckqʔ

atka → akka
atca → acca
etc.

User avatar
Rhetorica
Avisaru
Avisaru
Posts: 271
Joined: Sun Nov 03, 2013 3:33 pm

Re: SCA2 questions

Post by Rhetorica »

CC/CC/_ works by keeping the first value it encounters (atka → atta), so you'll need to write rules for every final voiceless plosive, e.g. Ct/tt/_

This is why you can't do metathesis with DC/CD/_ (where C = D). The special metathesis rule, DC/\\/_ can't be combined with other characters (e.g. DC\a\\/_ causes the output "a\\")

User avatar
dhok
Avisaru
Avisaru
Posts: 859
Joined: Wed Oct 24, 2007 7:39 pm
Location: The Eastern Establishment

Re: SCA2 questions

Post by dhok »

Not to bother this, but...can we make it possible to put pre-existing categories in nonce categories themselves? For example, if

P=pb
T=td
K=kg
and we have a rule
[KT]w/P/_
it won't do anything. But
Kw/P/_
Tw/P/_

works as expected. Is this a bug?

User avatar
KathTheDragon
Smeric
Smeric
Posts: 2139
Joined: Thu Apr 25, 2013 4:48 am
Location: Brittania

Re: SCA2 questions

Post by KathTheDragon »

Apparently that's intentional.

CaesarVincens
Lebom
Lebom
Posts: 204
Joined: Thu Feb 25, 2010 7:26 pm

Re: SCA2 questions

Post by CaesarVincens »

Apparently, optional categories don't play well coming before non-optional categories in the environment part.

Use the following code with the sample to see what happens.

Code: Select all

T=ptcbdg

V/V²/_(T)Tu
VT/\\/_(T)u
u//TV_
If the code is used as is, you will get fco, jco, and scendo; if you switch (T) and T, so T(T), then it runs fine.

Post Reply