Asia Bible Society GNT Syntax Trees @ biblicalhumanities.org

David Lim
Posts: 901
Joined: June 6th, 2011, 6:55 am

Re: Asia Bible Society GNT Syntax Trees @ biblicalhumanities

Post by David Lim »

Jonathan Robie wrote:Suppose I could create a character-based vertical display like this:

Code: Select all

S ─ CL ┌ S  ─ np ┌ np  ─ noun ───────────────────────────────────────────────────────── Παῦλος
       │         └ np  ┌ np  ┌ np  ─ noun ───────────────────────────────────────────── ἀπόστολος
       │               │     └ np  ┌ np  ─ noun ─────────────────────────────────────── Χριστοῦ
       │               │           └ np  ─ noun ─────────────────────────────────────── Ἱησοῦ
       │               └ pp  ┌ prep ─────────────────────────────────────────────────── διά
       │                     └ np  ┌ np  ────────────────────────────────────────────── θελήματος
       │                           └ np  ────────────────────────────────────────────── θεοῦ
       └ IO ─ np ┌ det ──────────────────────────────────────────────────────────────── τοῖς
                 └ np  ┌ np  ┌  np ── adjp ── adj ───────────────────────────────────── ἁγίοις
                       │     └  np  ┌ det ───────────────────────────────────────────── τοῖς
                       │            └ CL ┌  VC ── vp ── verb ─────────────────────────  οὖσιν
                       │                 └  P   ── pp ┌ prep ─────────────────────────  ἐν
                       │                              └ np ── noun ───────────────────  Ἐφέσῳ
                       ├ conj ────────────────────────────────────────────────────────  καὶ
                       └ IO  ┌  np ── adjp ─ adj  ───────────────────────────────────── πιστοῖς
                             └  pp ┌ prep  ──────────────────────────────────────────── ἐν
                                   └ np  ┌ np  ─ noun ───────────────────────────────── Χριστῷ
                                         └ np  ─ noun ───────────────────────────────── Ἱησοῦ
Is that user-friendly and useful?
I think you should consider using a HTML table with rowspan and colspan to align the cells. As in your suggestion, the surface language elements should be in the same column on the right. This way we can copy it directly to any spreadsheet and can both view and process all the data easily.
δαυιδ λιμ
Jonathan Robie
Posts: 4158
Joined: May 5th, 2011, 5:34 pm
Location: Durham, NC
Contact:

Re: Asia Bible Society GNT Syntax Trees @ biblicalhumanities

Post by Jonathan Robie »

David Lim wrote:I think you should consider using a HTML table with rowspan and colspan to align the cells. As in your suggestion, the surface language elements should be in the same column on the right. This way we can copy it directly to any spreadsheet and can both view and process all the data easily.
Have you tried copying / pasting the text and editing it as text? You have to use a monospaced font.

I have indeed considered using an HTML table, and it's still a possibility, but the tables get very complex, and I haven't quite figured out how to get all the lines to come out right that way for a complex diagram. And I really like to be able to cut and paste into a text editor or email without needing a spreadsheet.
ἐξίσταντο δὲ πάντες καὶ διηποροῦντο, ἄλλος πρὸς ἄλλον λέγοντες, τί θέλει τοῦτο εἶναι;
http://jonathanrobie.biblicalhumanities.org/
David Lim
Posts: 901
Joined: June 6th, 2011, 6:55 am

Re: Asia Bible Society GNT Syntax Trees @ biblicalhumanities

Post by David Lim »

Jonathan Robie wrote:
David Lim wrote:I think you should consider using a HTML table with rowspan and colspan to align the cells. As in your suggestion, the surface language elements should be in the same column on the right. This way we can copy it directly to any spreadsheet and can both view and process all the data easily.
Have you tried copying / pasting the text and editing it as text? You have to use a monospaced font.
Yes I know that your current suggestion works well with a monospaced font. However it is hard to do anything except view it.
I have indeed considered using an HTML table, and it's still a possibility, but the tables get very complex, and I haven't quite figured out how to get all the lines to come out right that way for a complex diagram. And I really like to be able to cut and paste into a text editor or email without needing a spreadsheet.
Ah I see. Most email now accepts tables, but I guess your suggestion is one of the best ways for it to be compatible with a text editor like here in this forum. But if you want to do a version with a HTML table, I suggest using a light grey border and rowspan for each abstract node, instead of lines connecting the abstract node, and colspan to fill up the gap from the leaf abstract node to the surface element. It will still be as clear, because the entire subtree is spanned by its root node and the cell borders will make it easy to see. That will also make manipulating the structure very easy in a spreadsheet such as if I wanted to rearrange some parts of the sentence by moving subtrees. The only thing is that it is not so trivial to compute the rowspans and colspans correctly. If you want, you can send me a sample input file (I don't want to download the whole thing haha..) and I could write a Javascript function for this. However I can't promise when I can complete it because I'm very busy recently.

Also, for the purpose of text editors, there is another suggestion that I have, which is to string the surface elements all on one line and use brackets to denote the structure and place the node names on the brackets, like this:

Code: Select all

S( CL( S( NP( NP( NOUN( παυλος ) ) NP( NP( NP( NOUN( αποστολος ) ) NP( NOUN(  χριστου ) ) NP( NOUN( ιησου ) ) ) PP( PREP( δια ) NP( NP( θεληματος ) NP(  θεου ) ) ) ) ) ) ... ) )
And then when clicking on any of the abstract nodes, for example the 3rd NP, only that subtree will be displayed in a panel below with each child subtree on one line:

Code: Select all

NP(
  NP( NP( NOUN( αποστολος ) ) NP( NOUN(  χριστου ) ) NP( NOUN( ιησου ) ) )
  PP( PREP( δια ) NP( NP( θεληματος ) NP(  θεου ) ) )
)
This would be much easier to do than the HTML table, and I think it would be complementary to the fully expanded tree. :)
δαυιδ λιμ
Laisvunas
Posts: 5
Joined: August 22nd, 2013, 1:36 pm

Re: Asia Bible Society GNT Syntax Trees @ biblicalhumanities

Post by Laisvunas »

Maybe Scalable Vector Graphics (SVG) is the way to go? Now SVG is being supported by all major browsers.
Laisvūnas Šopauskas
Jonathan Robie
Posts: 4158
Joined: May 5th, 2011, 5:34 pm
Location: Durham, NC
Contact:

Re: Asia Bible Society GNT Syntax Trees @ biblicalhumanities

Post by Jonathan Robie »

Laisvunas wrote:Maybe Scalable Vector Graphics (SVG) is the way to go? Now SVG is being supported by all major browsers.
Certainly a possibility - and the first I considered. I'll probably try several and pick one or a couple.
ἐξίσταντο δὲ πάντες καὶ διηποροῦντο, ἄλλος πρὸς ἄλλον λέγοντες, τί θέλει τοῦτο εἶναι;
http://jonathanrobie.biblicalhumanities.org/
Jonathan Robie
Posts: 4158
Joined: May 5th, 2011, 5:34 pm
Location: Durham, NC
Contact:

Re: Asia Bible Society GNT Syntax Trees @ biblicalhumanities

Post by Jonathan Robie »

David Lim wrote:But if you want to do a version with a HTML table, I suggest using a light grey border and rowspan for each abstract node, instead of lines connecting the abstract node, and colspan to fill up the gap from the leaf abstract node to the surface element. It will still be as clear, because the entire subtree is spanned by its root node and the cell borders will make it easy to see. That will also make manipulating the structure very easy in a spreadsheet such as if I wanted to rearrange some parts of the sentence by moving subtrees. The only thing is that it is not so trivial to compute the rowspans and colspans correctly. If you want, you can send me a sample input file (I don't want to download the whole thing haha..) and I could write a Javascript function for this. However I can't promise when I can complete it because I'm very busy recently.
Here's a shot at a version that uses HTML tables - but I need to tweak this if we want people to be able to cut and paste from it, the CSS styles don't transfer with cut and paste.

http://biblicalhumanities.org/sandbox/g ... tax-trees/

I rendered the entire GNT. I chose a format that highlights the text rather than the analysis, and tried to set it up in a way that lets you read and be aware of the groupings without paying too much attention to the analysis.

If you hover over the blue Greek text, you can see the morphology (but only if you wait for the page to fully load).
ἐξίσταντο δὲ πάντες καὶ διηποροῦντο, ἄλλος πρὸς ἄλλον λέγοντες, τί θέλει τοῦτο εἶναι;
http://jonathanrobie.biblicalhumanities.org/
Jonathan Robie
Posts: 4158
Joined: May 5th, 2011, 5:34 pm
Location: Durham, NC
Contact:

Re: Asia Bible Society GNT Syntax Trees @ biblicalhumanities

Post by Jonathan Robie »

David Lim wrote:Also, for the purpose of text editors, there is another suggestion that I have, which is to string the surface elements all on one line and use brackets to denote the structure and place the node names on the brackets, like this:

Code: Select all

S( CL( S( NP( NP( NOUN( παυλος ) ) NP( NP( NP( NOUN( αποστολος ) ) NP( NOUN(  χριστου ) ) NP( NOUN( ιησου ) ) ) PP( PREP( δια ) NP( NP( θεληματος ) NP(  θεου ) ) ) ) ) ) ... ) )
That's essentially the same as what I did here:

http://www.ibiblio.org/bgreek/forum/vie ... 987#p11464

The brackets are shaped differently, and my version spans multiple lines. I chose this syntax because it is already used in several syntax diagramming packages.
ἐξίσταντο δὲ πάντες καὶ διηποροῦντο, ἄλλος πρὸς ἄλλον λέγοντες, τί θέλει τοῦτο εἶναι;
http://jonathanrobie.biblicalhumanities.org/
David Lim
Posts: 901
Joined: June 6th, 2011, 6:55 am

Re: Asia Bible Society GNT Syntax Trees @ biblicalhumanities

Post by David Lim »

Jonathan Robie wrote:
David Lim wrote:But if you want to do a version with a HTML table, I suggest using a light grey border and rowspan for each abstract node, instead of lines connecting the abstract node, and colspan to fill up the gap from the leaf abstract node to the surface element. It will still be as clear, because the entire subtree is spanned by its root node and the cell borders will make it easy to see. That will also make manipulating the structure very easy in a spreadsheet such as if I wanted to rearrange some parts of the sentence by moving subtrees. The only thing is that it is not so trivial to compute the rowspans and colspans correctly. If you want, you can send me a sample input file (I don't want to download the whole thing haha..) and I could write a Javascript function for this. However I can't promise when I can complete it because I'm very busy recently.
Here's a shot at a version that uses HTML tables - but I need to tweak this if we want people to be able to cut and paste from it, the CSS styles don't transfer with cut and paste.

http://biblicalhumanities.org/sandbox/g ... tax-trees/

I rendered the entire GNT. I chose a format that highlights the text rather than the analysis, and tried to set it up in a way that lets you read and be aware of the groupings without paying too much attention to the analysis.

If you hover over the blue Greek text, you can see the morphology (but only if you wait for the page to fully load).
It's nice! I just tried various things, but I don't understand how the clipboard works.. Somehow the following works for no apparent reason:
(1) remove the "<td></td>"
(2) put the word itself in one more nested table

Code: Select all

<table><tbody><tr>
<td><table><tbody><tr>
<td>
<table><tbody><tr>
<td>
<table><tbody><tr>
<td><table><tbody><tr>
<td>
<table><tbody><tr>

<td title="ἐν">
<table><tr><td><p class="text">Ἐν</p></td></tr></table></td><td>
<p class="analysis">prep</p>
</td>
</tr></tbody></table>
<table><tbody><tr>
<td><table><tbody><tr>

<td title="ἀρχή Dative Singular Feminine">
<table><tr><td><p class="text">ἀρχῇ</p></td></tr></table></td><td>
<p class="analysis">noun</p>
</td>
</tr></tbody></table></td>
<td><p class="analysis">np</p></td>
</tr></tbody></table>
</td>
<td><p class="analysis">pp</p></td>
</tr></tbody></table></td>
<td><p class="analysis">P</p></td>
</tr></tbody></table>
<table><tbody><tr>
<td><table><tbody><tr>
<td><table><tbody><tr>

<td title="εἰμί Imperfect Active Indicative Singular">
<table><tr><td><p class="text">ἦν</p></td></tr></table></td><td>
<p class="analysis">verb</p>
</td>
</tr></tbody></table></td>
<td><p class="analysis">vp</p></td>
</tr></tbody></table></td>
<td><p class="analysis">VC</p></td>
</tr></tbody></table>
<table><tbody><tr>
<td><table><tbody><tr>
<td>
<table><tbody><tr>

<td title="ὁ Nominative Singular Masculine">
<table><tr><td><p class="text">ὁ</p></td></tr></table></td><td>
<p class="analysis">det</p>
</td>
</tr></tbody></table>
<table><tbody><tr>
<td><table><tbody><tr>

<td title="λόγος Nominative Singular Masculine">
<table><tr><td><p class="text">λόγος</p></td></tr></table></td><td>
<p class="analysis">noun</p>
</td>
</tr></tbody></table></td>
<td><p class="analysis">np</p></td>
</tr></tbody></table>
</td>
<td><p class="analysis">np</p></td>
</tr></tbody></table></td>
<td><p class="analysis">S</p></td>
</tr></tbody></table>
</td>
<td><p class="analysis">CL</p></td>
</tr></tbody></table>
<table><tbody><tr>

<td title="καί">
<table><tr><td><p class="text">καὶ</p></td></tr></table></td><td>
<p class="analysis">conj</p>
</td>
</tr></tbody></table>
<table><tbody><tr>
<td>
<table><tbody><tr>
<td><table><tbody><tr>
<td>
<table><tbody><tr>

<td title="ὁ Nominative Singular Masculine">
<table><tr><td><p class="text">ὁ</p></td></tr></table></td><td>
<p class="analysis">det</p>
</td>
</tr></tbody></table>
<table><tbody><tr>
<td><table><tbody><tr>

<td title="λόγος Nominative Singular Masculine">
<table><tr><td><p class="text">λόγος</p></td></tr></table></td><td>
<p class="analysis">noun</p>
</td>
</tr></tbody></table></td>
<td><p class="analysis">np</p></td>
</tr></tbody></table>
</td>
<td><p class="analysis">np</p></td>
</tr></tbody></table></td>
<td><p class="analysis">S</p></td>
</tr></tbody></table>
<table><tbody><tr>
<td><table><tbody><tr>
<td><table><tbody><tr>

<td title="εἰμί Imperfect Active Indicative Singular">
<table><tr><td><p class="text">ἦν</p></td></tr></table></td><td>
<p class="analysis">verb</p>
</td>
</tr></tbody></table></td>
<td><p class="analysis">vp</p></td>
</tr></tbody></table></td>
<td><p class="analysis">VC</p></td>
</tr></tbody></table>
<table><tbody><tr>
<td><table><tbody><tr>
<td>
<table><tbody><tr>

<td title="πρός">
<table><tr><td><p class="text">πρὸς</p></td></tr></table></td><td>
<p class="analysis">prep</p>
</td>
</tr></tbody></table>
<table><tbody><tr>
<td>
<table><tbody><tr>

<td title="ὁ Accusative Singular Masculine">
<table><tr><td><p class="text">τὸν</p></td></tr></table></td><td>
<p class="analysis">det</p>
</td>
</tr></tbody></table>
<table><tbody><tr>
<td><table><tbody><tr>

<td title="θεός Accusative Singular Masculine">
<table><tr><td><p class="text">θεόν</p></td></tr></table></td><td>
<p class="analysis">noun</p>
</td>
</tr></tbody></table></td>
<td><p class="analysis">np</p></td>
</tr></tbody></table>
</td>
<td><p class="analysis">np</p></td>
</tr></tbody></table>
</td>
<td><p class="analysis">pp</p></td>
</tr></tbody></table></td>
<td><p class="analysis">P</p></td>
</tr></tbody></table>
</td>
<td><p class="analysis">CL</p></td>
</tr></tbody></table>
<table><tbody><tr>

<td title="καί">
<table><tr><td><p class="text">καὶ</p></td></tr></table></td><td>
<p class="analysis">conj</p>
</td>
</tr></tbody></table>
<table><tbody><tr>
<td>
<table><tbody><tr>
<td><table><tbody><tr>
<td><table><tbody><tr>

<td title="θεός Nominative Singular Masculine">
<table><tr><td><p class="text">θεὸς</p></td></tr></table></td><td>
<p class="analysis">noun</p>
</td>
</tr></tbody></table></td>
<td><p class="analysis">np</p></td>
</tr></tbody></table></td>
<td><p class="analysis">P</p></td>
</tr></tbody></table>
<table><tbody><tr>
<td><table><tbody><tr>
<td><table><tbody><tr>

<td title="εἰμί Imperfect Active Indicative Singular">
<table><tr><td><p class="text">ἦν</p></td></tr></table></td><td>
<p class="analysis">verb</p>
</td>
</tr></tbody></table></td>
<td><p class="analysis">vp</p></td>
</tr></tbody></table></td>
<td><p class="analysis">VC</p></td>
</tr></tbody></table>
<table><tbody><tr>
<td><table><tbody><tr>
<td>
<table><tbody><tr>

<td title="ὁ Nominative Singular Masculine">
<table><tr><td><p class="text">ὁ</p></td></tr></table></td><td>
<p class="analysis">det</p>
</td>
</tr></tbody></table>
<table><tbody><tr>
<td><table><tbody><tr>

<td title="λόγος Nominative Singular Masculine">
<table><tr><td><p class="text">λόγος</p></td></tr></table></td><td>
<p class="analysis">noun</p>
</td>
</tr></tbody></table></td>
<td><p class="analysis">np</p></td>
</tr></tbody></table>
</td>
<td><p class="analysis">np</p></td>
</tr></tbody></table></td>
<td><p class="analysis">S</p></td>
</tr></tbody></table>
</td>
<td><p class="analysis">CL</p></td>
</tr></tbody></table>
</td>
<td><p class="analysis">CL</p></td>
</tr></tbody></table></td>
<td><p class="analysis">S</p></td>
</tr></tbody></table>
Then copying and pasting into Excel (I don't have other spreadsheet software) not only preserves the structure, it somehow puts all the surface elements in a single column on the left! Other things I tried all created odd extra blank cells for reasons I can't figure out...
δαυιδ λιμ
David Lim
Posts: 901
Joined: June 6th, 2011, 6:55 am

Re: Asia Bible Society GNT Syntax Trees @ biblicalhumanities

Post by David Lim »

Jonathan Robie wrote:That's essentially the same as what I did here:

http://www.ibiblio.org/bgreek/forum/vie ... 987#p11464

The brackets are shaped differently, and my version spans multiple lines. I chose this syntax because it is already used in several syntax diagramming packages.
Yes I saw what you did earlier, but I was wondering whether it might be helpful to be able to show a particular abstract node with its child nodes on single lines.
δαυιδ λιμ
Laisvunas
Posts: 5
Joined: August 22nd, 2013, 1:36 pm

Re: Asia Bible Society GNT Syntax Trees @ biblicalhumanities

Post by Laisvunas »

Hi Jonathan,

The boxed display of the syntax of NT setences is almost unusable for more complicated sentences.

Tree display is much better. It would be nice if tree display would be accompanied with some highlighting functionality. That is, mousing over a word might not only display its morphological parsing, but also highlight the relevant higher nodes. Also mousing over some node might highlight relevant lower nodes. Also displaying different types of nodes in different colors might be helpful. Some functionlity of such kind has been implemented in Logos and Accordance Bible softeare packages.
Laisvūnas Šopauskas
Post Reply

Return to “Other”