Word list / spelling issue

Question

Our language is written with phonemes separated by a space: ie Sin = vood dam dang pid. In light of this feature, is there a way to use spell checker/word list for actual words? Right now it only checks phonemes…which is helpful but only slightly. I imagine how the software could do this without creating 1000 other problems, but I wanted to check to see if there might be a workaround or other thoughts.
Thanks!
D

舊文章 - 以原文顯示

Answer 1

I would think that you could use ~ character which in Paratext signifies that a no break space will be used for output and put it in the language settings menu as a word medial character.

Blessings,

Shegnada James

舊文章 - 以原文顯示

Comments

Thanks Shegnada,
That sounds like a good solution. I guess I would have to do that input manually throughout the whole NT?? Any other thoughts on how to speed up that aspect of this task?
Thanks!
wdavidhj

舊文章 - 以原文顯示

---

The ~ (no break space) will not allow any line break within the word that the ~ appears in. Is that what you want? I think the spelling status in Wordlist and Display Spelling status would be really helpful here for you. It seems that you have to do this manually, because you will only want the ~ character in certain places and not others. But once you get it right for a word, the Correct spelling

舊文章 - 以原文顯示

---

Ben+Pehrson:

The ~ (no break space) will not allow any line break within the word that the ~ appears in. Is that what you want?

While working on the project, it wouldn’t matter too much that long words would not break at the end of a line. At publication time, you could change all the ~ to spaces again.

Shegnada:

I would think that you could use ~ character which in Paratext signifies that a no break space will be used for output and put it in the language settings menu as a word medial character.

I wonder whether it would confuse PT, or create some sort of conflict that the ~ character, which represents a special type of space, is also listed as either word-medial punctuation or as a word-forming character/alphabetical character. That’s why I said “I wonder” in my previous post …

However, if PT can handle this, then presumably you could set up hyphenation (via the Word List) just as you would for any other language. While the non-breaking spaces would not allow the word to break, the optional hyphens after the non-breaking space would.

And, then again, maybe you don’t need words to break across lines at all. A lot of children’s books, except in languages with very long words, do not use hyphenation at all.

Ben+Pehrson:

I think the spelling status in Wordlist and Display Spelling status would be really helpful here for you. It seems that you have to do this manually, because you will only want the ~ character in certain places and not others.

Perhaps it would be less work to change every space to a ~, and then change back the spaces between words to ordinary spaces – since there are presumably less of these than there are spaces within words. Whatever you do is going to involve a lot of work because it seems that you currently have no distinction between a word-medial space and a word-final space: that distinction has to be made, and that can only be done with some manual human intervention.

Now, if the spellchecking in PT could recognise phrases as wrong spellings, you could use the spellcheck to correct each word wherever it occurs. Here’s an English example:

Say you consider the phrase “time frame” to always – or almost always – be wrong because the correct form is “time-frame”: then, if PT allowed you to register phrases for correction, you could register the correction:

• time frame ➢ time-frame

Similarly, for your language:

• bac def ➢ bac~def

舊文章 - 以原文顯示

---

Ben+Pehrson:

But once you get it right for a word, the Correct spelling

@Ben+Pehrson, did something get lost from the end of your post?

舊文章 - 以原文顯示

Answer 2

I wonder whether it’s possible to include space in the alphabet as a word-forming character, or as word-medial punctuation (select project, then Language Settings ➢ Alphabetic Characters / Other Characters), just as, for English, you can include and in the alphabet so that the second word in this sentence is treated as a single word:

Tim Berners-Lee’s invention: the World Wide Web

But what does your language do when a word ends? Does it also use a space to separate words? Spaces within words, and spaced to demarcate one word from the next: that potentially makes reading more difficult, I guess, though no more difficult than NT Greek with no spaces at all – perhaps a little easier than NT Greek, in fact.

[Sorry, I wrote this ages ago, but didn’t quite finish it until now.]

舊文章 - 以原文顯示

Answer 3

I think the no-break ~ space would probably be the best bet. I think searching and replacing might be the quickest route for inputting them, though, it will be quite tedious. I’d say 1/3 of our terms are multiple phoneme words. The rest are single phoneme terms.

舊文章 - 以原文顯示

Comments

I’d also add that you could use any other character that your language does’t already use, e.g. hyphen or apostrophe. These could be changed to spaces at publication time.

Thsi would be a good choice if you want to use no-break space for its original purpose: to prevent a line break between two words. If the following are words:

• bac def
• gih nop

… you might not want bac def gih nop to break across a line, which would necessitate typing:

• bac def~gih nop

But, while working on the project, this could be entered as:

• bac-def~gih-nop

… with the hyphen set as a word-forming character.

舊文章 - 以原文顯示

Answer 4

I think that what you are dealing with here are separate morphemes, not phonemes. A phoneme is one unit of sound in a language. A morpheme is the smallest unit in a language that has meaning. So in your example, /d/ in “dam” would be a phoneme. But “dam” cannot be a single phoneme, since it is composed of three phonemes - /d/, /a/ and /m/. It may be a small thing but I just want to help us not get confused.

Anyway, I think I understand your problem; the languages I work with often use several words for one thing. Believe, for example, is “taa kem”. I find that for phrases like that, it is better to look to the biblical terms list. You can enter a whole phrase as your rendering for the Greek or Hebrew words for “believe” or “sin” or whatever, and it will mark any places that don’t match the rendering you entered, including any spelling errors. For example, if they put “ta kem” in one verse instead of “taa kem”, it will mark it as incorrect. Would that maybe do what you are wanting it to do?

舊文章 - 以原文顯示