Bug: Bible Browser results incomplete

Page 1 of 1 (9 items)
This post has 8 Replies | 0 Followers

Posts 1376
Forum MVP
Reuben Helmuth | Forum Activity | Posted: Thu, Dec 27 2018 2:15 AM

In testing before responding to this thread, I found that the Bible Browser results are incomplete (in addition to being limited to the NT). The screenshots demonstrate the problem.

According to BB there's only 1 hit in the entire book of 1 Cor.,

yet there should be at least 6 hits in the first two verses alone!

Is this possibly affected by spotty/slow internet? Is my dataset up to date?:

DB:SD-NT-SYNTACTIC-FORCE
2017-04-04T23:28:24Z
NT-SYNTACTIC-FORCE.lbssd

Posts 23844
Forum MVP
Dave Hooton | Forum Activity | Replied: Thu, Dec 27 2018 5:12 PM

Reuben Helmuth:
According to BB there's only 1 hit in the entire book of 1 Cor.,

And that is also the result from the Search  {Section <SGNTSyntacticForce = prop. name>} (which explains why it is limited to the NT!) i.e. 2 hits in 1 verse.

Dave
===

Windows & Android

Posts 7782
LogosEmployee

This looks like an error in the Syntactic Force dataset. If I right-click on Sosthenes in 1 Cor 1:1, it doesn't show "proper name" on the context menu:

A search for {Section <SGNTSyntacticForce = prop. name>} only shows results in one verse in 1 Corinthians (actually in all of Rom–2 Thess): 

Posts 1376
Forum MVP
Reuben Helmuth | Forum Activity | Replied: Fri, Dec 28 2018 2:04 AM

Bradley Grainger (Faithlife):
This looks like an error in the Syntactic Force dataset.

Thanks for the response Bradley. It looks like you're getting the same results as I am so I assume my syntactic force dataset is up to date...

Posts 7782
LogosEmployee

Yes; I've reported it to the data team.

Posts 7782
LogosEmployee

The Lexham SGNT Syntactic Force dataset (in Bible Browser) was automatically extracted from the analysis performed by Dr Lukaszewski in The Lexham Syntactic Greek New Testament, published in 2010.

It represents that particular scholar's analysis of the text, so Faithlife will not be correcting any problems or omissions. (If we ever work with him again to produce a second edition (which is probably unlikely), it might be corrected then.)

Posts 7782
LogosEmployee

I think I figured out what might have happened. It looks like the different editors of the analysis were not consistent in how they applied "proper name" tagging. If you continue using this dataset, it may be prudent to assume that proper name tagging only applies to the highlighted books:

And there may be other annotation differences between the three editors.

Posts 1376
Forum MVP
Reuben Helmuth | Forum Activity | Replied: Tue, Jan 1 2019 1:23 AM

Bradley Grainger (Faithlife):
If you continue using this dataset, it may be prudent to assume that proper name tagging only applies to the highlighted books:

In this case, I'd say it would be better to pull that dataset entirely. Perhaps this would be a good time to reopen the crowd-sourced tagging can of worms! Please let us do it even on a volunteer basis (repayment was a major drawback according to Eli a couple years ago when I submitted a proposal)!

Posts 23844
Forum MVP
Dave Hooton | Forum Activity | Replied: Thu, Jan 3 2019 6:44 AM

Reuben Helmuth:
In this case, I'd say it would be better to pull that dataset entirely.

It would be prudent to do so. Results are clearly unpredictable, and will only generate further comments.

Dave
===

Windows & Android

Page 1 of 1 (9 items) | RSS