DEVONthink: stuck at the bottom of the hill

I tend to keep them all in one big folder and make sure the name on the document describes it sufficiently. I am not a tagger so I don’t use them in DT at all.

I also tend to lump until the folder gets too big (which is a relative thing and not a consistent size) then I’ll split it. If I only need the item once in a great while and it won’t take that long to search for it why not keep it all together in one big folder?

I am now fully in DTP from EN as well. That is a big deal for me. I started this thread!

The power of DTPO for me is the search. I have larger “buckets” and rely on tags rather than nested folders. I don’t think that matters as much as making sure all your Files are searchable. I took the time to OCR all PDFs to get PDF + Text.
In 6 months or six years you can just start typing words and it finds your file.


@Isaac LOL. I had to scroll back to the beginning of this thread. I was one of the people who was trying to tell you there is no reason to switch from Evernote. Several months later, I switched to DTPO, I don’t recommend Evernote, and I come back to this thread several times as a reference tool!


DTPO has a smart group for Duplicates. It found about 30 of them. I went to go compare them…obviously, if the note content is the same, I delete the copy.

Sometimes, I do have the same note in 2 places (for whatever the reason) I believe according to the DTPO, I should look into deleting 1 copy and using a replicant?

The issue I can’t figure out…is when I have 2 notes that are similar in content and title but are not duplicates. i.e. Example, I have several notes that are like this a little bit.

Note 1 - 800 word count
Note 2 - 1200 word count

Both notes have similar titles (I tried making it more unique but no luck). Out of the shared notes, there is a good potion that are shared content. (if the note is based on the same reading passage, but with a different date).

Any one have ideas on how to fix this? I would still like to keep the Smart Group for Duplicates, it’s helpful sometimes.

I don’t think there’s a “rule” for this. Personally, I will duplicate a document if I am going to change the content of one of the copies, or some attribute of the copy like name, tags, etc. But it’s up to you. Duplicates cause no problems in databases since every duplicate has its own unique identifier under the covers and they don’t interfere with one-another.

There’s no way to tell DEVONthink “these two files are not duplicates – stop telling me they are”.

I’ve never seen a complete explanation of why DEVONthink assumes two things are duplicates when you can see that they aren’t. It’s a regular frustration for users. (The question recurs over and over in their forum.) Obviously “content” is the major factor. The algorithm is probably overly liberal in deciding A and B are duplicates and leaving the rest up to you.

It would probably be more accurate if DEVONthink called files “similar” rather than “duplicate”.


Another user has a question about managing files in DevonThink 3. The thread is here DevonThink - avoiding lock-in while taking advantage of organizational ease and I thought you all could help him.

Thanks for the bump - it got me to reread this thread - striking to see @MitchWagner so enthusiastic back then and now reporting that he’s no longer using DT (in DevonThink - avoiding lock-in while taking advantage of organizational ease )

1 Like

Drives my wife and friends crazy.

ME: “I have discovered a wonderful new app! It has changed my life!”

[months later]

WIFE/FRIEND: “I started using that app. Thank you so much for recommending it!”

ME: “That app? It’s rubbish. I haven’t used it for ages.”


This is me…

Over in the Devonthink forums nobody had a good answer for me so I ask you… I really don’t get the concordance feature in DT. I mean, I understand what it does (basically showing me the word frequency in a text) but I just see no real use case and I would love to hear of an actual real world use case… The Devonthink folks position it as a valuable feature but I just don’t know what to do with it.


1 Like

I can think of two off the top of my head:

  1. See word use frequency on an article or paper I’m writing. Helps to identify overused words.

  2. If I have a folder on a topic like salvation, and I’m looking for one that quotes Paul a lot, I could look for the name Paul in a concordance as a good place to start combing through.


Anybody in the field of Philology would find it invaluable. There was an interesting discussion a few years ago involving ‘codes’ apparently embedded in Holy Books. I think these techniques were used amongst other: both to find the supposed codes and also to find, for debunking purposes, alternative ‘codes’: ones meaning the opposite of what the original claims were and that kind of thing. I forget the details.
I am not sure if something of the basic idea was involved in that awful book and film The Da Vince Codes? I saw it but didn’t take much notice of the details I am sad to say. Two hours of my life I will never get back…

Finding the frequence of certain words can be invaluable in forensics as a matter of fact. You can also use it to identify topics and relationships between items. I am assuming that DEVONthink’s excellent AI, if we call it that, uses it to ‘group’ and find relationships of degree. I always assumed that they had just made a kind of UI version of that capacity. You can use it to date a certain document. I don’t, I will say, use it much. I have done though but as I said, I assumed that it was just a UI peek into the AI.

This is an old post so I guess you have already made some decision. I would climb the hill. I had this app for nearly two years before it really ‘kicked in’ for me.

Thank you for elaborating! This makes sense indeed. :blush:

1 Like

In fact the Bible code example was a bit orthoganal and I could have found better examples. You will see the line of thought though and the analysis really is at the level of letters. Though for the statistics you would use word frequencies too. It turns out really that it would be very odd if you couldn’t find patterns arising at random as it were.

1 Like

The Take Control book that’s been mentioned here repeatedly has been updated for DEVONthink 3. It is, as promised, free.


The new version, as expected, is excellent. Author Joe Kissell, who now runs Take Control Books, is a pro, and he sweats the details. A bit of a surprise that not only is the update free, the ebook itself is free to anyone.


the book is sponsored by DevonThink :smiley:

side note: how on earth you make an account on the take control website? I only see option to login… :thinking:

edit: just got an email telling me they created an automatic account after the “purchase”… weird method :smile:

1 Like

Interesting about being sponsored. I think I paid for my earlier version of the book.

Now that you’re registered I check out some of their other ebooks. Their eBooks are not copy-protected, you can re-download any time, and you can get them in multiple formats (I usually download my purchases as pdf, and also as ePub, which looks better in the Apple Books app.) They’re usually high quality manuals/cheat-sheet ebooks, and the pages for all the (non-free) books have downloadable pdf samples to check out the table of contents and usually a chapter from the book. Highly recommended, and I’ve been buying from Take Control for many years.

Once or twice a year they’ll have 30%-40% discounts, but if you’re a member you can get 30%-off Take Control orders year-round.


It is good for examining words that you are trying to rank in SEO’s for