A small problem with searching The Complete Shakespeare

A Logos search for "a rose by any other name would smell as sweet" will not register a hit in The Complete Shakespeare because the quote spans two lines of verse, lines which in the Logos text are serparated by intervening spaces and a line number.

In other words, the Logos implementation of this text does not appear to recognise phrases which are broken by a line break as contiguous for the purpose of searching.

This makes it a bit difficult to pinpoint a favourite quote!

Find more posts tagged with

Comments

Jack Caviness

A Logos search for "a rose by any other name would smell as sweet" will not register a hit in The Complete Shakespeare because the quote spans two lines of verse,

Perhaps you should try a shorter phrase: "A rose by any other name" should achieve the desired result.

Tonya J Ross

A Logos search for "a rose by any other name would smell as sweet" will not register a hit in The Complete Shakespeare because the quote spans two lines of verse, lines which in the Logos text are serparated by intervening spaces and a line number.

In other words, the Logos implementation of this text does not appear to recognise phrases which are broken by a line break as contiguous for the purpose of searching.

This makes it a bit difficult to pinpoint a favourite quote!

The program should return search results for a phrase that includes a line break. I will submit a report to the correct department for investigation.

Thank you for bringing this to our attention.

Matthew C Jones

https://community.logos.com/discussion/comment/256676#Comment_256676

The program should return search results for a phrase that includes a line break. I will submit a report to the correct department for investigation.

Apparently this is not a one-point issue but a systemic problem. All line breaks appear to break the search string.

Giovanni Baggio

https://community.logos.com/discussion/comment/256712#Comment_256712

Apparently this is not a one-point issue but a systemic problem. All line breaks appear to break the search string.

The problem is the square root of search-o-matic result-ability of the organic frequency of the lines module...oh nevermind, I'm not a computer geek so I don't know what the heck I'm talking about...LOL...[:P] I'll delegate the task to the correct department.

Sam Henderson

https://community.logos.com/discussion/comment/256712#Comment_256712

You're right. It looks like the problem is the insertion of line numbers.

I had the same problem searching for quotes that span line breaks in The Iliad too (another text with numbered lines).

No such problem searching for multi-line quotes from Milton's Paradise lost (which does not have numbered lines/verses).

Bob Pritchett

It's the line number that causes this problem, I'm pretty certain, and that's a hard thing to code around without a lot of very detailed tagging (and complexity). We can look into it, though.

The best solution, I believe, is to simply search without the quotes. Our system returns ranked results now, and performs better the more words there are. So just search for

a rose by any other name would smell as sweet

and not only will you find the quote, you'll find articles, anecdotes, and other interesting references to it -- as well as possible slight misquotations.

Martin Folley

https://community.logos.com/discussion/comment/257161#Comment_257161

The best solution, I believe, is to simply search without the quotes. Our system returns ranked results now, and performs better the more words there are. So just search for

a rose by any other name would smell as sweet

A good idea but the ranking does not seem to work, at least as i would expect it to work, with the results presented as they are.

In the shot below, you can see that The Shakespeare quote is visually referenced by other resources a couple of times, but the quote itself is not actually found! (or at least ranked high enough to make it onto the first page of results, or the second, or the third.... ). In reality, it is actually found but you would not know it!

Hit 6 does point to the correct Romeo and Juliet resource, a few verses before the actual quote, but you would not know that it was a direct hit by looking at the search results. Which of these produce a direct hit? Which just use some of the words in the search? Is there a way of determining the relevance of a hit without loading each resource?

Hit 4 uses the quote directly ... but you could not se in what context, or even that it does use the quote!

Hit 5 and 7 both clearly use the quote, and I can see the context!

Hit 10 (a Spurgeon sermon) does not use the quote, but uses some of the words quite a lot! If I could see the places where the words are together, then I could discern the difference between direct reference and coincidental use.

Would it be possible for the previews to be longer, and centred around the text where there is the greatest proximity?

Greg

https://community.logos.com/discussion/comment/257161#Comment_257161

It's the line number that causes this problem, I'm pretty certain, and that's a hard thing to code around without a lot of very detailed tagging (and complexity). We can look into it, though.

Why can't we just do away with the numbers altogether? Or perhaps have them hidden in the actual text (like you do for page numbers). When reading the collection in Biblia, its incredibly annoying to have all those numbers after every line. Quoting lengthy excerpts would be mildly tedious too because they would have to be deleted.

One other thing. Since this is a digital resource, why are the character's names abbreviated? I can understand the need to conserve space in a print book, but not so much in a digital one. I've found it makes following the dialogue more difficult. Is it possible to replace the abbreviated name with the full one?

NB.Mick

https://community.logos.com/discussion/comment/257182#Comment_257182

Since this is a digital resource, why are the character's names abbreviated? ... I've found it makes following the dialogue more difficult. Is it possible to replace the abbreviated name with the full one?

This should be possible by using visual filters (there was a thread somewhere where people used this to replace "servant" with "slave" in their bibles). There should be some examples of doing such stuff on the wiki.

Sam Henderson

https://community.logos.com/discussion/comment/257182#Comment_257182

One other thing. Since this is a digital resource, why are the character's names abbreviated? I can understand the need to conserve space in a print book, but not so much in a digital one. I've found it makes following the dialogue more difficult. Is it possible to replace the abbreviated name with the full one?

I suppose the abbreviated character names (like the line numbers - which some people find useful for referencing) are an artifact of the source text from the original scan. In the Logos version it would have been nice to have the abbreviated character names come up in their full form as popups on mouseover as do many abbreviations in other Logos texts.