Bring text comparison to the next level
Hello,
When I use the text comparison tool I am seldom interested in comparisons that are the same. So, for example, when comparing minor revisions (RSV→ESV) I am not interested in verses with no changes ("0% difference"). Instead of needing to visually identify and scroll past such verses it would be nice to hide them altogether.
Additionally, verses may convey close to the same meaning in different words. Or the meaning may be taken substantially differently, such as gen 3:16 "for" or "contrary to" in RSV vs ESV. Using modern inference engines, commonly called "AI", it would be possible to determine whether two verses "likely do" or "likely don't" convey the same meaning. The threshold or sensitivity would be something the user can adjust. Being able to hide verses based on this score would help the feature be useful even for text comparisons where the versions are quite different. The display of the text comparison, when the verse is not filtered out, could be the same as the current display.
While we are on the subject of AI, this technology can also be used to help align text for versions where the numbering schemes are quite different or the underlying manuscripts are quite different. This way the text comparison tool would know which verses or portions of verses to compare even if they do not have the same numbering. Perhaps Logos already has this mapping problem solved through some internal consistent numbering used under the hood, just thought I'd mention it. In addition, for sake of text comparison, sometimes groups of verses must be compared together because the looseness of the translation reordered phrases or made small adjustments to verse numbering vis-a-vis meaning. "AI" can help untangle all of this without tedious manual review, setting appropriate multi-verse boundaries for a comparison unit where necessary.
I am a programmer and could discuss these things in more detail with your tech staff if that is of interest.