What's the best way of summarizing an article?

Many of you may have seen we're experimenting with summarizing articles as part of our Experimental Search engine.

To help us with that, which summarization style do you prefer:

Style 1, which tries to capture the writer's own style

I have come to some conclusions regarding Christian assurance. First, we should appreciate the mysterious nature that underlies this assurance and not try to reduce it to simple explanations. Second, by closely observing the various biblical statements in their contexts, we can protect our theology from reductionism and pastoral mistakes. It is important to consider how the Scriptures function in cases where believers seem to deny their faith. Third, it is misguided to make absolute Christian assurance the foundation of theological systems and then reinterpret Scripture to fit this preconception. This approach is as flawed as trying to develop a Christology solely based on Jesus' humanity. Fourth, while there may be a case for beginning with the doctrine of assurance in theology, it is strange to make it the touchstone for the entire structure of Christian theology. We could instead start with God, Christ, redemption, or revelation. Fifth, it is essential to emphasize that perseverance is not the basis of assurance, but rather failure to persevere can undermine assurance. The basis of assurance is in Christ and his work. Finally, the biblical writers provide believers with ample assurance grounded in the character of God, the new covenant, election, and more. However, they also challenge us to not be spiritually indifferent and to obey God's commands.

Style 2, which uses the third-person

In this text, the author discusses several conclusions related to Christian assurance. They emphasize the importance of appreciating the mystery behind Christian assurance and allowing biblical statements to stand in their full power without reducing them. The author also highlights the significance of understanding the functions of biblical statements in their contexts to prevent dangerous reductionism and pastoral malpractice. The text criticizes the approach that prioritizes absolute Christian assurance as the foundation of theological systems and engages in re-reading Scripture to justify this view. They argue that this approach is ill-conceived and can lead to distortions in theology. Additionally, the author discusses the relationship between perseverance and assurance, stating that while perseverance does not serve as the basis for assurance, failure to persevere can undermine it. They emphasize that the basis of assurance is Christ and his work. Overall, the text emphasizes the importance of biblical balance, the character of God, and the love of God in providing believers with assurance, while also challenging them to live out their faith through obedience.

Find more posts tagged with

Comments

MJ. Smith

Style 2 - the first seems to me to be a lie

scooter

https://community.logos.com/discussion/comment/1270438#Comment_1270438

Style 2 is best. I like how it reads, similar to a book review, or how an author sums up a second author's opinion.

Style 1's first person style seems contrived. If I were writing a paper, I would write, "S/he says that...'

Thanks for asking, Mark.

Carmen Gauvin-O'Donnell

https://community.logos.com/discussion/comment/1270443#Comment_1270443

That's an interesting question! I spent part of my childhood in Europe, in French Lycees in London and and Brussels. In grade 10 French literature class we learned how to do something called the Summary and Commentary. We were given some literary text which we had to summarize to 25% of the original (and you'd lose points by having too many words in your summary! Yes, they were that picky! We'd literally have to count the words in the original text and divide by 4!), then you'd comment and give your thoughts on the author's ideas.

But the point behind the summary was a "literal" one - we were to use his or her words exactly, which would really convey whether we had understood the meaning or not.

So for me, anything would do, but hey, the challenge (to Logos) of creating a more literal summary might help better convey the style of the author?

Here's a challenge to my fellow Logos users: you want to know if you've understood a Bible chapter you've been studying? SUMMARIZE iT to 25% original size. I did this for my personal study of Revelation. It was HARD work, but then I'd read my summaries to my small group and they would enourage me to keep doing it. According to them, I had "caught the flavour" every time! (Insofar as one can catch the flavour of Revelation, LOL!)

[:D]

Gregory Lawhorn

On balance, I would prefer #1 by a narrow margin. I understand the reasoning of the other posters, but the second option feels like a commentary, not a summary. I would assume that a summary is an abbreviated version of the original, but the second option feels more like a critique.

Tim Hensler

https://community.logos.com/discussion/comment/1270447#Comment_1270447

Way back in the Libronix days, there was a function that either (I can't remember which) highlighted part of the text of a resource, or reduced the visible text to a certain percentage to give the user a "summary" of the text. If I remember right, the user had the option to determine the level of summarization. Since it was working only with the original text, the result was more like Style 1 and in line with Carmen's point above. It was a neat and useful function, but with today's AI capabilities, wow, what a tool it could be.

I wouldn't want to use this as the sole basis for a Bible lesson or sermon, but as an introduction or evaluation tool, it could provide a handy sample (style 1) or summary (style 2) of a pericope, chapter or whole book (of a commentary, monograph or even (with caution) Bible).

► I would like to see, by a slim margin, Style 2 (summary) developed first, then Style 1 (sample).

It would be interesting to compare the results of this tool using the same chapter from different Bible translations.

JT (alabama24)

#2 because #1 is misleading.

MJ. Smith

https://community.logos.com/discussion/comment/1270447#Comment_1270447

We were given some literary text which we had to summarize to 25% of the original (and you'd lose points by having too many words in your summary!

This is one technique I was taught in order to prove I understood the text; I was taught it along-side paraphrasing, outlining and short précis. However, we were not required to retain the original wording - that was called a condensation.

Here's a challenge to my fellow Logos users: you want to know if you've understood a Bible chapter you've been studying?

This step is one I frequently include as a step in Bible study. However, I absolutely hate using chapter/verse boundaries. I am firmly wed to apply this at the pericope level. I am glad you brought the topic up.

MJ. Smith

https://community.logos.com/discussion/comment/1270448#Comment_1270448

I would assume that a summary is an abbreviated version of the original, but the second option feels more like a critique.

Interesting how we use language differently. An abbreviated version of the original is to me a "condensation" as in the Readers' Digest series of condensed literature. "Summary" to me implies a summary of meaning in a neutral voice -- think of summary of an initiative in votes guide where it converts from legalese to common English.

Gregory Lawhorn

https://community.logos.com/discussion/comment/1270486#Comment_1270486

Interesting how we use language differently. An abbreviated version of the original is to me a "condensation" as in the Readers' Digest series of condensed literature. "Summary" to me implies a summary of meaning in a neutral voice -- think of summary of an initiative in votes guide where it converts from legalese to common English.

True. I would rather read an author's abstract than one person's opinion on the author's meaning, but that's me. FWIW, I believe there is no such thing as a "neutral voice." Even in a reduced form, only an author's own words can express the author's meaning.

Dave Hooton

Providing it does reflect the writer's style, I prefer #1 because it is lost in the third person summary.

scooter

https://community.logos.com/discussion/comment/1270534#Comment_1270534

Providing it does reflect the writer's style, I prefer #1 because it is lost in the third person summary.

I find some authors' style beyond me: ever-flowing sentences with much substructure that leads to incoherence, or the possibility of more than one meaning, as the sentence coils and loops.

A summary should read like a John Frame book. There is a simple elegance to his style.

Then, people whose first language is not English must be considered. John Frame English would be excellent for them.

We have friends born in Asia. The lady was taking a community college course in business writing and I was her editor. The course was in English, her fourth language. Her sentence structure was exactly like John Frame's. Both people use a simple, logical sentence construction that results in eminently clear communication. [i.e.: just like this paragraph]

If the concept is difficult and the language is difficult [being the author's hard-to-understand verbiage], then a summary in that style does not work. A John Frame style summary will fit the most people's bill.

Mary-Ellen

I would prefer the old Libronix approach, which highlighted actual passages in the text that the AI deemed most significant. I just finished reading a lengthy book, with highlights of segments I want to incorporate into my research project. I remembered the old Libronix function, and thought it would be great to compare that "summary" or "condensation" or "digest" against my collected highlights to see if I missed anything important for my purposes.

Many, many resources already contain abstracts, introductions, conclusion sections and the like where we can find the author's own "summary" of the work. I'm not interested in AI doing my work for me. That keeps me from actually learning and as others have suggested, can actually be misleading.

xnman

For me, Style 2 by all means.

1Cor10 31

https://community.logos.com/discussion/comment/1270448#Comment_1270448

I agree with Greg's logic. Hence, prefer #1.

DMB

https://community.logos.com/discussion/comment/1271534#Comment_1271534

I would prefer the old Libronix approach, which highlighted actual passages in the text that the AI deemed most significant.

That'd definitely be my preference (granted, not what Mark offered). But it fits into the same UI-format as inline searches. Even currently, I use a Visual Filter, to track the logic, dependencies, and conclusions.

https://community.logos.com/discussion/comment/1271572#Comment_1271572

When I'm in a time crunch, I sometimes use the Community Highlights for "summary" purposes.

Mary-Ellen

https://community.logos.com/discussion/comment/1271588#Comment_1271588

When I'm in a time crunch, I sometimes use the Community Highlights for "summary" purposes.

What a great idea! Thank you! I'll try it on the book I just finished.