I've got the reverse interlinear sort of working, based on strong+morphology in the Greek. I'll need to test with the Hebrew, but to do that, I'll somehow have to work out how to create and install a JSword module.
However, I don't read greek, or hebrew, and as a result I'm not sure if the words are lined up correctly or not! There's some obvious UI improvements to make too like for e.g. the division "Acts 2:1-10" to be placed above and not left to float to the left.", but a lot of those things will be style.
Cheers
Chris
On 20 February 2010 19:06, Tyndale STEP Project <tyndalestep@googlemail.com> wrote:
Chris, we haven't done much posting in the past about interlinears.
There are a lot of issues with versification differences in the OT between English, Hebrew and LXX, though very little in the NT. THe Sword modules have the rules built-in and there are standard lists of differences.
OSIS does use a module with KJV and Strongs numbers which is Public Domain. They also use another, which is not PD, which tags NASB with Strongs.
You are correct that the OT tagging does not include full morphology, though it does have limited morphology for verbs. This is sufficient for most purposes. The full morphology used by commercial products is a copyright source and the owners are keen to keep it out of the PD.
Using the tagged translation together with a tagged Hebrew text, we can do the following:
1) make popups which give quick definitions and links to a search based on the Strongs numbers.
2) make highligting parallel versions like the one here which
a) highlight the whole verse on both sides (so you can see where you are)
b) highlight every word in the chapter on both sides which use the same root (ie the same Strong's number)
(highlighting every word in the chapter is really useful for finding the structure or verbal links in a chapter which is often obscured by translation).
3) make a very rough interlinear which has the Hebrew/Greek with English under it, as defined by the tagging.
(ie ignore any words which aren't tagged - so that it gives a help to the reader rather than attempting to provide a translation)
THe data we have is:
<H7225">In the beginning
<H430">God
<H1254!a">created
<H8064">the heavens
<H776">and the earth
ie this gives us an interlinear:
Gen.1.1
hâ)âretz we)êth hashâmayim )êth )elôhÎm bârâ) berê)shÎth
earth the and heavens the God created beginning the In
instead of
hâ)âretz we)êth hashâmayim )êth )elôhÎm bârâ) berê)shÎth
earth the and heavens the God created beginning the In
(The lines above are written in a monospaced font, so if things don't line up, try changing it to monospace)
The only difference is that "the" etc is not under the proper words.
Making a 'real' interlinear is more difficult, so I suggest we wait.
The guys at neXt Bible have got this sussed, and we may get permission to use their work.
But actually, I'm wondering if the parallel version with highlights isn't better,
esp if we have an extra line of limited interlinear on both sides which you can turn on or off.
David IB
At 13:43 20/02/2010, Chris Burrell wrote:
OK, so I'm now looking at interlinears and raking my brain. Do refer me to previous threads if we've already discussed this at length!
First off, there is not JSword module in Hebrew annotated with the strong numbers. Someone has kindly given us a link to the WLC Hebrew annotated in Strong numbers, but unfortunately has not yet seen the day as a proper module. So we could investigate that. The morphology is not in those files though.
So, secondly, for the classic interlinear, here's probably what the steps are and correct me if i am wrong:
- Lookup the hebrew text from the selected reference
- Lookup the english text for the selected reference
- Match the references based purely on the strong number and not the morphology.
The problem I see is that if words are repeated several times within a verse (assuming a verse is a correct marker for delimiting portions of text), then by mapping solely on the strong number, we may map the wrong part of the verse, hence making the English a bit weird/difficult/confusing/wrong to read.
For the Greek, we have got versions with the morphology, and so I'm guessing if we match on both criteria it would be easier and a better interlinear, although perhaps not full proof?
Similarly, we'll have the same Hebrew issues in the Greek
Thirdly, is that actually full proof? Can we trust versification to have been done the same in the original text and in the english? For example, in that one very long sentence at the beginning of Ephesians, are the same parts of the verse in the same order, meaning the same thing in the Greek and the English?
Just a few thoughts,
Chris
On 19 February 2010 09:43, Tyndale STEP Project < tyndalestep@googlemail.com> wrote:
- Good start on interlinear!
- The ultimate plan is to get interlinear working with NASB and hopefully, ESV + NET Bible
- but that's just a matter of using different data.
- If we start with the AV, we can add different datasets later.
- We also have a better Strongs with real Hebrew and abridged BDB
- (I supplied the one which is used by Crosswire)
- There are a few different ways to mix Hebrew & English usefully:
- * Follow the Hebrew order and put English under it (the classic way)
- * Follow the English order and put Hebrew under it (so-called Reverse interlinear)
- (see both of these here)
- * Put them side by side with interactive highlighting
- * Pop-ups or hover - useful, but by itself it doesn't help people to read the original
- (see side-by-side highlighting with pop-ups together here )
- Ideally we'd like to offer all of them!
- David IB
- --
- Posted By Tyndale STEP Project to Tyndale STEP - Programming on 2/19/2010 01:43:00 AM
--
Posted By Tyndale STEP Project to Tyndale STEP - Programming on 2/20/2010 11:03:00 AM
No comments:
Post a Comment