Hebrew Grammar Distinguish at VOLT Level

Typograph's picture

Gohebrew: this is the place for you.
first will discus what glyphs are needed for such a project, and the best way of designin them.

Seconed, We will start at ground basics decisions we must make brfore we start.

then will talk about the sheva na\nach rules on by one in depth

So Gohebrew Fill free to start the discotion

gohebrew's picture

>> If the word is in the context of Tanakh then it's shva nach, but if it's in modern Hebrew it's shva na. How does the font know where it comes from?

This seems difficult to believe? Why is modern Hebrew and Biblical Hebrew have different grammar rules?

gohebrew's picture

>> fourth, fifth, sixth, seventh?

Are you saying that there are different systems, more than the Minchat Shai, R' Eliyahu Bachur, and R' Zalman Henna?

If my neighbor goes to a daf-yomi class, learns a little Hebrew grammar to sound like a grammarian, invents his own shita, gets hired by KP, then this is not a real shita. Do you mean this?

Minchat Shai, R' Eliyahu Bachur, R' Zalman Henna, and Koren Publishing?

Typograph's picture

Hey Hey, GoHebrew, don't put words into my mouth.
I said that i got the project to a very high level of precision.
I do not clain that it can be done to 100%;

With Teamim, I get to 98.8%
Meaning, that from every 1000 cases the font will fail at 2 cases.
without teaamim with meteg, precision level gows down and with text with aout teaamim and meteg precision gows down a little more.

This does not include the fact that i tend to ignore any thing that is in debate.

therefore, the font uses 2 stylistic sets and the ability at the keyboard to correct these cases.

now, here is a basic fundimental logic that i decided from the start.

A Normal qamats in a place that should be a qamats qatan is less of a mistake than a qamats qatan in a place where there should be a qamats gadol.

A Normal Sheva in a place that should be a Sheva na is less of a mistake than a sheva na in a place where there should be a normal sheva (Nach).

let me explain a little more.
Qamats + Hataf Qamats (Like TzoHorayim)
The font uses qamats gadol.

The reason.
According to the Ashkenazim its a qamats qatan(O), According to the Sefaradim it's a qamats gadol (A).

So the font does nothing;

Volt you can never know what is SHEM TOAR or SHEM ETSEM and what is the SHORESH.
In volt you cant build a data base to lookup from.

+ alot of decisions were simply made bt searching all the tanakh and determening wrather a certain check will result with the least mistakes.

then if the mistakes were'nt to many, i add them into a dictionary.

a lot of assumtions are made.
like he+qamats followed by Alef,He,Het,Ayin.
i would assume that this qamats should realy be Patach, But because of TASHLUM DAGESH it became Qamats. So this qamats is gadol and not qatan.

so if by checking this i resolve nost cases, and have a couple cases that fails in EYOV, So i do that, and either i ignore the mistakes or look them up in a dictionary.

Then you have also matters to do with MASORET.

So again, my project is 99.8% with idle text, but never 100%.

insetion of meteg by volt is hardly possible.

the absolute conclusion is:
sheva na-nach, qamats gadol-qatan is 90% contextual. but not fully contextual.
no matter what Pro.Dotan or Rabbi Shmuel Rabin tell you (if they toled you so).

gohebrew's picture

>> The reason.
According to the Ashkenazim its a qamats qatan(O), According to the Sefaradim it's a qamats gadol (A).

I diagree.

Why?

The function of a kamatz gadol and kamatz ketan is a grammar issue.
Saying (O) or (A) is cultural pronunciation.

For example,
the first word of kiddush on riday night is the word in Hebrew for day, 'yom'.
Litvish Jews (whose families came from Lithuania) pronounce this word as 'yaym', as if the cholam was not there, and instead a tzayrei appears under the aleph.
But nothing is really different. And the spelling and nikkud remain the same.

In the recital of kaddesh, many Jews (except Lubavitchers and Sephardim, who say, 'Amen') repeat the phrase, 'may He be blessed', 'breech hu'.

Hungarian Jews, though, say 'breech hee', as if the vav (shuruk) was not there, and instead a chirik appears under the hei.
But nothing is really different. And the spelling and nikkud remain the same.

Rule:

Pronunciation is cultural, and does not affect the rules of Hebrew grammar.
Hebrew grammar is not affected by Ashkenazic or Sephardic customs or stlyes.

gohebrew's picture

>> Volt you can never know what is SHEM TOAR or SHEM ETSEM and what is the SHORESH.

I agree. No one can create an accurate Volt fonts that corrects spelling or places nikkud. This is not the realm of Volt.

gohebrew's picture

>> insetion of meteg by volt is hardly possible.

You have not shown this logically. Faith is not reason.

If meteg follows rules, then Volt could do it.

Typograph's picture

Oy Veys, GoHebrew, GoHebrew...

The word צהרים, the qamats under the Tsadi Is Qatan Or Gadol ?????
the word שרשים, the qamats under the Shin Is Qatan Or Gadol ?????
The word באניה, the qamats under the Bet is Qatan Or Gadol ?????

gohebrew's picture

>> the absolute conclusion is:
sheva na-nach, qamats gadol-qatan is 90% contextual. but not fully contextual.
no matter what Pro.Dotan or Rabbi Shmuel Rabin tell you (if they toled you so).

Who should I believe?
The world-class experts, or
the whipper-snapper?

Again, it's like do I listen P. Dotan say jump off the boat and swim to save your life, because the boat will turn over, and you'll drown',
or
do I listen to Raphael, who says 'stay on the boat and relax'.

P. Dotan knows physics, water currents etc. and much more.
Raphael is a great typesetter.

Who will live, and who will die?

gohebrew's picture

>> The word צהרים, the qamats under the Tsadi Is Qatan Or Gadol ?????
the word שרשים, the qamats under the Shin Is Qatan Or Gadol ?????
The word באניה, the qamats under the Bet is Qatan Or Gadol ?????

Are citing these example because these questionable letters begin the word?
How many kamatz katan begin a word? 25. 50. Make an exception list.
Volt needs a few more nanoseconds.

Typograph's picture

Its not a matter if its in the begining of the word, and no need for exception list.
the check is if i have a qamats and then hataf qamats(or qamats and qamats + an ending of yod mem) if the first qamats is Qatan Or Gadol.
This is the debate at the grammer level, not the pronunciation of it.

gohebrew's picture

You are saying that a word which begins with a kamatz, for it have no ltter and nikkud before it, to clarify the two possibilities, is either
a kamatz gadol, or
a kamatz katan,
depending upon two schools of thought.

I would count the valid views both ways, and follow the majority view,
unless the minority view had an outstanding rationale.

Typograph's picture

Gohebrew: my logic of this is much simpler.
what ever is in dabate, i leave to the user to decide, and even if he did not make a any choice, still a qamats gadol in the place of qatan is less a mistake than a qamats qatan in a place where there should be a qamats gadol.

so these cases do not indicate that this particular qamats is a gadol, but is left to its default.

gohebrew's picture

>> what ever is in dabate, i leave to the user to decide

The whole point of an intelligent OpenType font properly in Volt allows thee user never to decide anything.

The fonts does all the work.

So too here.

gohebrew's picture

You are following logic. I am following rules.

Typograph's picture

you know what Gohebrew, I'll give you another example of debate.

words like
המלמד, המכסה, המעורר
ect'

The He (The first He) is HE HAYEDIA With a patah
the second letter is any letter who can get a dagesh but has no dagesh.
There is a grammer rule that says that the latter after the He Hayedia should be with a dagesh, and a letter with a dagesh and a sheva the sheva is NA.

Another rule says, that a sheva after a tenua ketana is a Sheva Nach.
But a word with a dagesh even if it is after a tenua ketana the sheva is NA.

So here is a debate.
In theses cases there is no dagesh in the second letter and comes after tenua ketana.
is this sheva NA??? because of the dagesh that was supposed to be there or we go after the tenua ketana and then the sheva is NACH.

Well, My Project does not make any decision, and leaves the sheva as is, and allows the user to decide.

And again you have things to do with MASORET

So you see, you cant have the font making all the decisions without allowing the user ever ro decide.

you absolutly have to allow the user to the posibility to do what he thinks is right according to his SHITA.

>The whole point of an intelligent OpenType font properly in Volt allows thee user >never to decide anything.

Gohebrew, What amaizes me with you is that as one who has absolutly no knoledge of hebrew grammer will be the one to decide what should be and without allowing the user to have his say in the matter.

and who says that "The whole point of an intelligent OpenType font properly in Volt allows the user never to decide anything."???? You????
where did you get that conclusion from????

>You are following logic. I am following rules.

What rules???? Whos rules???? do you even know the rules?????
I woulde believe that your only knolegde in the matter, is the knoledge you got from this threads and other threads here on typophile.

you argue with me as an equal, but your not. and you are starting to get on my nerves.
i fill like i am talking to a wall.
I try to explain to you different matters in detail, but you don't want to listen and continue to persist with this idea witch is simply not true.

>>Who should I believe?
>>The world-class experts, or
>>the whipper-snapper?

The correct ansawr is B) the whipper-snapper.
I don't believe that any of these experts actualy toled you that it is 100% posible, or its 100% contextual... ask them again in detail.

but when i say "the absolute conclusion is:... ..." i believe that 100%

Now, if you stop arguing and start listening and asking you might just learn somthing after all.

Typograph's picture

maybe i should take a rest, and allow David Hamuel take on from here???
what do you say David??? :)

david h's picture

Israel,

I think that I have a solution to the problem with North Korea & Iran (nuclear weapon); we record this thread and broadcast it 24/7; trust me, a week... and they are done; and Ahmadinejad is becoming a type designer.

> maybe i should take a rest, and allow David Hamuel take on from here???
> what do you say David??? :)

Eli,

are you tired? already?

Typograph's picture

Tired to discuss???, no.
Tired to explain a simple fact of life again and again??? yes.

>>I think that I have a solution to the problem with North Korea & Iran (nuclear >>weapon); we record this thread and broadcast it 24/7; trust me, a week... and they >>are done; and Ahmadinejad is becoming a type designer.

I think you actualy have a real solution.
thay will be so convinced that jews are creazy, they will let them of the loop.

Gohebrew:
>> I would count the valid views both ways, and follow the majority view,
>> unless the minority view had an outstanding rationale.

and this is how years of debate will finally com to an end...

david h's picture

> The whole point of an intelligent OpenType font properly in Volt allows
> thee user never to decide anything.

> The fonts does all the work.

Israel,

I told you before -- you can't do that. And Now -- you can't! Not this year and not ten years from now. And again -- "The fonts does all the work": you can't.

gohebrew's picture

>> I think that I have a solution to the problem with North Korea & Iran (nuclear weapons)

I repeat what the former Chief Rabbi of Israel, the Rishon L'Tzion, Rabbi Mordechai Eliyahu zatzal, said about the PA and its descendants of Amalek,
"Nuke 'em, they're all chayav meetah."

That'll wake up Eli.

gohebrew's picture

Eli,

Gohebrew:
>> I would count the valid views both ways, and follow the majority view,
>> unless the minority view had an outstanding rationale.

and this is how years of debate will finally com to an end...

===

Actually, it's based on the Tzemach Tzedek's reasoning.

gohebrew's picture

David,

>> I told you before -- you can't do that. And Now -- you can't! Not this year and not ten years from now. And again -- "The fonts does all the work": you can't.

You know that I respect your opinion very much.

But Prof. Dotan heard carefully my goal. He even asked me to repeat it. He thought carefully about it, for awhile. He concluded, "Yes."

David versus Aron. Who would you bet on?

david h's picture

> David versus Aron. Who would you bet on?

Israel,

Nice try; this isn't going to work, there is no "versus" :)

gohebrew's picture

David,

This was the question that Prof. Dotan thought about, after I emailed him various samples:

I asked: "There is a new font technology for computers. It is very advanced, and can be programmed before the users uses it.

"It can take a string of text, with nikkud etc., and replace it with a different string of text, with nikkud etc. I seek to program the font to include replacements with a shvah-nah. Is this contextual replacement possible."

He answered: 'Yes'.

Then, he added: "You must understand very well the relationship in the nikkud. It's all in nikkud".

I understand his words to refer to the different grammatical rules of the different types of nikkud, including the different types of dagesh.

Each thing builds upon another.

The shvah-na and shvah-nach is after the kamatz-katan and kamatz-gadol. The latter is in turn after the other types of nikkud and dagesh.

I have been taught the rules the rules of the various kind of meteg. So, I can not place it in the scheme of things.

I no longer see that the needed Glyph Groups contain all the different sequences. I now see that the needed Glyph Groups can merely contain a few dozen very small variables.

To get this info, I must pull the last tooth from the one-tooth monster, called David. And I am not a dentist.

William Berkson's picture

Israel what you quote does not say that the contextual replacement can be done 100% automatic, which is the issue you have raised. I think that you have misunderstood him.

gohebrew's picture

I did not ask him, Bill, if it could be 100%, or as Eli claims, 99.8%.

When Prof. Dotan said, 'yes', I understood that it was absolute. 101%

gohebrew's picture

Bill,

Basically, how is it 100% automatic?

My Volt font compiles whatever vocalized Hebrew text in that font by replacing each nikkud and dagesh with either of two types of nikkud and dagesh. Then, it inserts the shva-na and kamatz-katan as required. Then, it replaces each changed nikkud and dagesh (except for the shva-na and kamatz-katan).

Very simple.

I thank Eli and David, for from their words I derived the above formula.

Bill, your questions and Raphael's remarks was like the teaching in Avot about how the teacher learns more from the students' questions than from anything else. Why? Questions and statements cause the teacher to think again, and to rethink a lot. Hence, their questions cause his greatest learning.

Typograph's picture

GoHebrew
letes summarize:
A) you think that it is absolutly posible to to the sheva Na/Nach Qamats Gadol/Qatan Checks.

B) You Agree that it is not posible to automaticly insert metegs.

Is this correct????

gohebrew's picture

No,

1- it is possible to do A) .

2- I never said it is not possible to insert metegs automatical.

I think if we first sub-divide meteg into a few kinds, and replace all metegs accordingly,
then, we (after we determine what causes a meteg) insert accordingly.

Afterwards, we return all metegs to look like metegs of different forms of meteg.

david h's picture

> I have been taught the rules the rules of the various kind of meteg.

Israel,

Which rules? I didn't see here anything -- e.g. (Bible) light meteg/ga'ya? meteg/ga'ya on a closed syllable? meteg/ga'ya on an open syllable? heavy meteg/ga'ya? etc etc

are you talking about the siddur? -- we need to say stress mark (sm) beacuse I don't understand what is going on here; there is a big mish-mash :)

Typograph's picture

Ok GoHebrew;

So tell me the word
Hogla = He Qamats Gimel Sheva Lamed Qamats he, The first qamats is qatan
Natna = Nun Qamats Tet Sheva tav Qamats he, The first qamats is gadol

So if its all in the nikud, so what is the difference????
You might Claim that in the word Natna There should be a meteg under the nun after the Qamats, And Hogla has no meteg. Fine. But...
A) If you are inserting the metegs, how do you know???
B) With text without Metegs, How will you know??? (Un-Metteged text is most common)

gohebrew's picture

You are asking the blind to lead you.

First, in the Aleppo Codex, eyes notice a long mark, like a meteg gadol :)

Is it a siluk?

Rabbi Stein, a Reform/Orthodox grammarian, says: Don't forget the siluk in the font.

You say a ga'ya.

A stress mark. A meteg hamafsik. Sof pasuk.

Let's sort them all out.

I can take a single Unicode code for meteg, and grown a whole bunch of alternative glyph for each kind.

Wanna help? We'll call the font "GH Hanuel Grammar".

Typograph's picture

> You are asking the blind to lead you.

And still you say its all in the nikud????

gohebrew's picture

I didn't say. P' Dotan said.

You wore diapers when he taught at TA U.

Nikkud include dagesh and meteg.

Typograph's picture

> Nikkud include dagesh and meteg.

Really????

And if so, you still think you can automaticly insert Metegs, why not dageshs???.

> You wore diapers when he taught at TA U.

That is not a valid argument

gohebrew's picture

Eli,

>> > Nikkud include dagesh and meteg.

Really????

---

I said this to better understand P' Dotan's words.

>> And if so, you still think you can automaticly insert Metegs, why not dageshs???.

I think that I can, nikkud, dagesh, and meteg, if I know the rules.

---

>> You wore diapers when he taught at TA U.

That is not a valid argument.

Israel: Of course not. I said it to tell you that his wisdom exceeds yours, even though most Israelis think that they know more than their elders.

Typograph's picture

>Israel: Of course not. I said it to tell you that his wisdom exceeds yours, even
> though most Israelis think that they know more than their elders.

Pro.Dotan Never tried to make a program and Kol Sheken a volt project that would accomplish this, so there for there a some issues that maybe he is overlooking.

a lot of stuff in nikud is driven from knowing the source of the word.
What is the root of the letter.
what is SHEM ETSEM And What is SHEM TOAR Ect'
is it a hebrew letter or foreign.

Is there a MASORET on a specific letter or not, and so on and so on.

To simply say "its all in the nikud" is incorrect.

But 95% of it is in the nikud.

gohebrew's picture

Eli,

>>> Pro.Dotan Never tried to make a program and Kol Sheken a volt project that would accomplish this, so there for there a some issues that maybe he is overlooking.

I disagree. He doesn't write code, but he understood my intention immediately, and what Volt could do.

I want to write a contextual replacement program in a true-type font that makes the rules of Hebrew grammar defined as strings and not rules, as was done since Radak.

He and R' Rabin grasped this in a few seconds, that has taken me months to explain here.

P' Dotan is a giant. Maybe, when you reach his age, people will Fried is the next Dotan.

Typograph's picture

>I want to write a contextual replacement program in a true-type font that makes the
>rules of Hebrew grammar defined as strings and not rules

What do you mean by strings and not rules???

gohebrew's picture

strings are a series of glyphs
rules do not relate to a string or series of glyphs in a hebrew word

when a dagesh follows a kamatz katan, the dagesh is kal - this is a rule
when a shvah follows a kamatz gadol, the shvah is na - this is a rule

to make the rules relate to a string of glyphs in a hebrew word, we must first make the generic kamatzs either katan or gadol, we must first make the generic dageshes either kal or chazak, then we can inset...

after we insert, we globally change all the dagesh kal or chazak back to generic.

raphaelfreeman's picture

Oy, GoHebrew, you are missing such fundamental points of this argument that you are driving everybody nuts. Let me try and clarify.

Yes, you can probably create a font that puts in kamatz katan and shva na according to Prof Dotan's view.

But you keep missing the following point: NOT ALL TEXT FOLLOWS PROF DOTAN'S VIEW OF NIKUD.

You ask how is modern Hebrew grammar different from the grammar of Tanakh? BINGO!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

YES IT'S DIFFERENT.

And the grammar of the Amoraim is different from the grammar of Modern Hebrew (by modern I mean the past few hundred years).

And not only are they different, but in each period there are different opinions.

You got confused with צהריים, thinking we are talking about pronunciation. NO WE ARE NOT. The pronunciation of the kamatz katan is irrelevant to this conversation. We are talking about GRAMMAR. According to the Sepharadi view, their Masorah is that the first kamatz under the tzadi is a kamatz gadol and according to the Ashkenazi view it's a kamatz katan.

Now Prof Dotan doesn't care about the Sepharadi masora because he knows that the masora is based on mistake that the Scholars of that time made (around 300 years ago) due to the information that they had. In his opinion, this is irrelevant and therefore not a problem. For him, it's a kamatz katan. Prof Dotan also doesn't care about a masorah that we have about pronouncing Hebrew in the siddur according to the time period of the Amoraim and Tanakh. He is a Prof of Modern Hebrew grammar.

Prof Dotan isn't wrong, it's just that you didn't ask him the right questions because your knowledge of the subject is too limited to even know to ask the questions.

Does that clarify things a little?

gohebrew's picture

Yes, Raphael, thank you for your opinion of Prof. Dotan's views.

Since you say there are differences in Hebrew grammar in the current modern period, in Hebrew grammar in an earlier Amoraic period, perhaps also in Hebrew grammar in an even earlier Mishnaic period, and even a very early Biblical period.

Please show me some examples.

When the Radak defined Hebrew grammar, which period was he referring to?

If Prof. Dotan was referring to Hebrew grammar in the current modern period, less the Sephardic masorah, why did Rabbi Rabin concur with his conclusion, as he is a Lubavitcher baal-koreh, reader of the Torah scroll on the Sabbath for the congregation, when speaking to another Lubavitcher (and we know little about modern Hebrew grammar)?

So, I don't fall into a Christian-like trap of accepting things with blind faith, please show me some examples of the differences between the seemingly different Hebrew grammars.

david h's picture

> Prof Dotan isn't wrong, it's just that you didn't ask him the right questions...

Probably. Israel didn't ask him the right question, but.... "He is a Prof of Modern Hebrew grammar." ?

what was his last book/article/lesson about Modern Hebrew grammar? or any of his classes -- Tel Aviv University & Bar Ilan University -- about Modern Hebrew?

Typograph's picture

David, We should not underestimate Pro.Dotan and others, also i would just assume that he is well respected at the hebrew acadamy and i believe that he knows his stuff from his point of view.

But.. the Haredim don't follow the academys conclusions. they hav a masoret, and each one has is own masoret the sefaradim and ashkenazim.

I as a type designer, developed the project to give service to them with respect for each ones masoret from the one hand, and maximum precision on the other hand.

I do not have the pretension of deciding in these matters and especially not coming out and saying that the rules that i put in are to be studied as Hloche Lemoyshe Misinay with no ability of correction because it is carved in stone.

Now letes say that Gohebrew will succseed of doing this 101% according to Pro.Dotan, to others it will not be even 90% correct.

I worked hard to figur out a balance for an efficient help-tool and accuracy
I might be arrogant when i say that in a dabte with Pro.Dotan on this particular erea(if there is one), I do acualy know what i am talking about.

But it seems, that the only way to learn this, is by experiencing for him self.

Typograph's picture

another great reason to allow override of the font.

lets say i hav a name like קַנְיֶבְסְקִי
according to the rules the sheva under the samekh is NA.
but i want it to be nach

Names don't have to follow the grammer rules

gohebrew's picture

Perhaps, this clarifies matters; perhaps not.

My intention is very simple. Arrogant, maybe; simple, sure.

My font is intended to address verses and statements of chazal. Eliezer Ben Yehuda can be jealous. So can Shmuel Agnon too.

Override, you ask?? Na is nach, you desire??

Just highlight the letter and its nikkud...

... and turn OpenType off in Adobe InDesign CSx ME ...

behold the symbol for shvah-na disappears!

gohebrew's picture

The truth is that with very little modification, a font for each approach is possible:

Naming scheme:

GH E or B R or M

GH = GoHebrew
E = Enhanced = Dagesh, Nikkud, and Meteg only
B = Biblical = Taam too
R = Rabbinical
M = Modern

Example:

GH E Agnon - Bold M
GH E Agnon - Bold

Poetry - before & after

Before > GH E Agnon - Bold
After > GH E Agnon - Bold M

Typograph's picture

>Just highlight the letter and its nikkud...
>behold the symbol for shvah-na disappears!

Gohebrew, how well do you know Indesign???
Turning From OpenType To Normal effcts only MARK Feature not Any SUB Feature, if you are talking about the diacritic postioning.

Now let say that you allow it in a stylistic set.
If the user is using Character Styles the unchecking the stylistic set will turn off the grammer.

But if the user is working with Paragraph Styles then highlighting and unchecking theat stylistic set is no good, because in paragraph style it will turn the feature off for the intier paragraph and not the selected text.

Plus, You want to give the option of Forcing a sheva na\nach Qamats Gadol\qatan ect'

Thats in my case, in your case you are going to have 3 sets of glyphs for Normal sheva, a Sheva NA and Sheva NACH, turning off the stylistic set will result with 3 sets of shevas in one text
The Sheva NA
The Sheva Nach
And then the normal Sheva

gohebrew's picture

Let me consult my experts.

I think if there is an identical non-Grammar font as above, you can override on a character-by-character basis without turning off OpenType diacritical processing.

raphaelfreeman's picture

gohebrew, I gave you 2 examples several times. the example of אתה and the example of יברכך.

Syndicate content Syndicate content