Tool to distill the essential meaning?

I like to read theological books, but they are often wordy.

Just wondering, could it be possible to develop a tool that finds the essential, and abridges the text to a manageable size?

I tried to check the Internet, but could not find anything yet. And if something is currently available, I doubt it works properly.

This would be an excellent feature for those of us who try to "save" with deals and end up with a large library Indifferent

Find more posts tagged with

Comments

David Thomas

develop a tool that finds the essential, and abridges the text to a manageable size?

This used to be a tool available in Logos, but it was abandoned. We used to be able to set a rate of abridgement and Logos would summarize the resource to that percentage. I wish it still existed! A great feature that could come back in L9?

Dave Hooton

Just wondering, could it be possible to develop a tool that finds the essential, and abridges the text to a manageable size?

David's memory is more positive than mine!

Do you want to abridge each individual article or abridge/choose the articles that are essential? Either way, it is very subjective and you would be better served with a collection of theological books that suit your purpose. You can use that in the Theology Guide (Systematic Theologies section) and general Searches.

Veli Voipio

https://community.logos.com/discussion/comment/1086564#Comment_1086564

Thanks David and Dave!

Currently I cannot remember which book I was reading when that came into my mind, but my general process, if I have time is like this:

First I try to read fast and I try to catch and mark some essential points. After that I read around those marked points, and then on the third round to skim to get the whole picture.

What might be possible: the abridging tool could utilize discourse analysis and that way catch what is important, to mark the text. That would shorten the first part of the process.

Real abridging is very subjective and perhaps impossible for a a machine. And my experience with artificial intelligence and learning machines has not been encouraging.

David Thomas

https://community.logos.com/discussion/comment/1086566#Comment_1086566

Real abridging is very subjective and perhaps impossible for a a machine. And my experience with artificial intelligence and learning machines has not been encouraging.

I believe this is why the tool was abandoned. I found it helpful when I was doing seminars in graduate school. We would often have 4-6 textbooks to read in the weeks leading up to the in-class portion of the seminar. I would read for comprehension in the pre-class period then 1-2 days before the class, I would go back and review the book at 10% abridgment.

It was helpful for reviewing the big ideas of each text, but not for a meaningful first read.

Perhaps @Denise could verify my recollection. I believe she is still running Libronix in some manner.

MJ. Smith

Pick a book on critical thinking/argumentation etc. such as:

Cengage Advantage Books: Understanding Arguments: An Introduction to Informal Logic by Walter Sinnott-Armstrong and Robert J. Fogelin

You are looking for one that does a good job of listing argument markers in a particular language. For books that are focused on persuasion in the sense theological books often are, a search on the argument markers (all of them in an OR relationship) does a reasonably good job of allowing you to follow the essential argument and ignore the filler and examples.

DMB

https://community.logos.com/discussion/comment/1086567#Comment_1086567

Perhaps @Denise could verify my recollection. I believe she is still running Libronix in some manner.

That was 'BL' (before Libby/2005)? Or I didn't get the add-on. It's an interesting idea.

MJ's point on logic code words I think is good. Except for the high-level of wiggle-waggle wording in religious arguments. "Although most scholars have not completely rejected the contra-argument, this absence is definite proof the belief might not have existed."

MJ. Smith

https://community.logos.com/discussion/comment/1086596#Comment_1086596

[8-|]

J. Remington Bowling

This might not be what you're looking for, but here is a hacky solution that might build on MJ's suggestion.

Argdown works through Visual Studio Code to automatically create argument maps from Argdown syntax.

You could copy and paste the text you want to map into VSC then, as you read along, enter in the Argdown syntax. Once you're finished reading and marking you might have generated some argument maps or at least have the text marked up where it's easy to identify key-terms, premises, and conclusions.

I don't think there currently exists a program that can identify an argument with all the premises. After all, humans still have a hard time doing this because we actually aren't that great at constructing or parsing arguments. Sometimes premises or conclusions are implicit. And, most confusing for a computer, sometimes key-terms for identifying an argument have other non-argument usage and knowing whether a term is meant to designate a premise or a conclusion requires contextual knowledge.

In theory, it should be possible to create a program that does this. But I suspect that, once we do have such a program, what we'll find is that a lot of authors aren't very good at stating their arguments!

J. Remington Bowling

https://community.logos.com/discussion/comment/1088346#Comment_1088346

Just thought of another way to make this easier, though it requires hardware you might not have.

If you have a Stream Deck you can assign a syntax marker to each of the keys using the Text function under the systems menu. This makes it extremely easy to add the Argdown syntax to the text that you've pasted. (You might want to paste it into Word first and find-replace all periods with periods and line-breaks so your paste doesn't appear on a single line in VSC.

DMB

https://community.logos.com/discussion/comment/1086596#Comment_1086596

MJ's point on logic code words I think is good.

Subsequent to this thread-start, I've watched several authors. Agreed, a formal argument is a ways off. But code words are pretty predictable. English has them, and the greek, more so. Without the code words, the listener/reader looses the trail, and has to backtrack.

But I still think a user-drawn scheme is easily implemented, especially since Logos already handles highlighting, even across updates. Mine knows to draw lines, and with a page-turn, to compute the assumed line-angle, much like the Footpath app. You can also put in slash-cuts, where the author's logic jumped the tracks. "Since, we've shown Diatesseron as the source of the Old Syriac ....".