NEAR (proximity) search rendered useless?! Any suggestions?

When one performs a proximity searching using “NEAR”, or “NEAR/10”, command + G (which should find the next instance) does nothing. In order to find the next instance of any one of the terms I’m searching for I must do a FIND for TERM1, or TERM2, and cycle through all of the content (sometimes over 1,000 pages) looking at every instance of the terms to find the instance where both terms are used together. This of course renders the proximity search entirely useless. For the whole purpose of doing a proximity search is to find only the use of TERM1 where it is used near TERM2. I’m reduced to having to do a single keyword search and use command + G to cycle through the document looking for the place where TERM1, and TERM2, are used together. This is all quite depressing because this is, to be honest, one of the most useful features DN/DT, and it turns out it doesn’t even work.

Unless I’m missing something, but I’ve been working on this for hours and can’t figure out how to go directly to the instances where both terms are used within the given proximity.

I would greatly appreciate any help. Thank you.

1 Like

So, to be specific:

Is there away to quickly go to the results which exactly match a proximity search query?

If I use “rock NEAR/10 soil”, can I quickly go to the place in the document where those two words are used in proximity? As it stands I currently have to use a find, for the search terms, within the document(s) that met the search criterion, and cycle through every instances of the search term(s) attempting to catch the location where they are mentioned in proximity. I have documents containing 500,000-1,000,000 words this can get really tedious.

:cry:

Seriously,

No replies?

This topic has been discussed many times previously in several threads.

If you have set one or more proximity operators in a query, the results returned by the search will be documents that indeed meet the criteria.

There are, however, non-trivial logical issues with highlighting only those occurrences of the query terms used in the proximity criterion that meet the criterion. Users have noted that a couple of Windows search apps claim to be able to do that. But using real world documents in my databases, I found that a number of them were improperly or confusingly highlighted in the Windows searches, so that their approach was untrustworthy.

A possible workaround when n is small (4 or less) is to open a document that meets the proximity criterion under Preview, and search for one of the proximity criterion terms. Then inspect the text strings in the search list.

If the desired result can be defined as an exact string, use that string enclosed in quotation marks, instead of a proximity search for terms contained within the string.

Bill, thanks for the reply.

I do know of a piece of windows software which does this beautifully, and with perfect results.

Here’s an example:

Query: “life love;5”

This will find the words life and love where they are within 3 words of each other. Then there’s a results display window which shows excerpts that meet the criterion. Like this:

“I love my life more than I love yours…”
“If your life is full of love then you will…”

And so on.

The software in question is Bibleworks (bibleworks.com/). It has brilliantly written search functionality that I’ve never seen in any other program.

With regard to your suggestion, “Then inspect the text strings in the search list,” I think I mentioned that as my current “work around.” (Unless I misunderstood you.) However, it’s not really work around since it takes an incredible amount of time.

It would be great if in a new version you could add a results window which shows a list of excerpts containing the criterion. It would be much, much, easier to find what I’m looking for.

As it stands DT/DN are good for finding documents that meet criterion, rather than specific passages that meet criterion. Just something to think about.

I assume that this is very low priority for DEVONTech, and I’m very sorry to hear that since it is a very important issue to me.

Frankly, I could go without all of the tagging, categorizing, etc., because recall via powerful search is far, far, more valuable. I should be able to find exactly what I’m looking for within seconds, or at least have a list of passages that fit the criterion (rather than simply a list of documents that I must command+G, or Find, my way through.)

I happen to know that creating a results window of this sort would be rather easy. I hope to see it in the future.

See the Bibleworks software for a great example. The their search functionality here: bibleworks.com/content/videos/0003SearchWin.htm

After many years, proximity search still remains useless to this day.

This is unfortunate as Devonthink is supposed to be a super-searching machine.

Thankfully, I have Foxtrot that does the proximity search beautifully.