Hi. I’ve been using DTPO for sometime now. My workflow relies heavily on tags. However creating tags occurs in one of three areas : a)finder, B) DTPO inbox, C)individual inboxes. Keeping the text consistent is crucial so that my searches are complete. What I would like to know is if there is away, manual or automatic, to be able to merge or make all three areas contain the same tags?
Thanks so much,
Jay
The short answer is: kinda, and it won’t be perfect.
Here’s the deal. DEVONthink and Finder understand each others’ tags, and tags are a property of a given file.
What exactly this means depends a bit on some aspects of your workflow.
Finder->DEVONthink
If DocumentX is in Finder and you apply a Finder tag TagA, then TagA is a property of DocumentX. In Finder, if you click to see all the files with TagA, you’ll see DocumentX.
Now, if you IMPORT that file into DEVONthink, that file and its tag are now part of whichever database you imported it into. So in that database, if you click on the Tags group, you’ll now see TagA has been created (if it didn’t already exist), and that DocumentX is listed as having been tagged this TagA.
DEVONthink->Finder
Exactly the same as above except reversed. DocumentY is in DEVONthink, you tag it with TagB, when you export that file to a location in Finder (or drag it out of DEVONthink into Finder) then the document remains tagged, and the tag is created in Finder (if it didn’t already exist).
DEVONthink->DEVONthink
If you move a document between databases, then the same principles apply. The tag remains a property of the document and is created in the receiving database if it doens’t already exist.
Indexed Files
Things get a little finicky if you are INDEXING into DEVONthink files that lie in some location in Finder. In this case, all of the above principles apply, but it can be a bit flakey. Indexing is pretty solid but does require a little bit of care and is not bulletproof.
With indexed files, lets say a text file, if you edit the document in DEVONthink, the change is visible in Finder, of course. The same applies in the other direction. If you use Finder to open the text file in BBEdit and make a change, that change is visible in DEVONthink.
Similar to editing the contents of an indexed text file, since tags are a property of the document, then changes made in DEVONthink should be pushed to Finder and vice versa. In practice I find it’s a bit less seamless with tags.
If you have a file that is already indexed in DEVONthink which you subsequently tag in DEVONthink, or tag in Finder, you need to force DEVONthink to “Update indexed items” to get the tags for a given document to reconcile between DEVONthink and Finder.
In all of the above cases, deleting a tag (NOT removing the tag from a file, but deleting the actual tag) is not synced – that is, if you delete a tag in DEVONthink that had been applied to 50 indexed files, the tag will be deleted and removed from those files. Once indexing updates everything, the tags will also be removed from those files in Finder however the tag will still exist in Finder (it just won’t be applied to those 50 files from which it was just removed in DEVONthink). This means that deleting an actual tag in one location will not delete it from all locations.
This kinda covers the basic possible scenarios. What exactly is your scenario? Perhaps if you tel us whether you are indexing or importing, and other details, we can offer you a more specific an detailed set of suggestions.
HI Scott,
Thank you for the detailed response but that is not quite what I am looking for. Let me try explaining with a specific type examples:
tags between dtpo databases:
- In DTP files can go into a database lets call it db01. So for example db01 could be the global inbox or any other db
- Now lets say I select a tag or create a tag in db01 we will call it HappyTag1.
- Now lets say that db02 does not contain HappyTag1, It contains
- Now lets say I move the file from db01 to db02
- If db02 does not have this tag then it will be added to the tags available in db02
- when I go to add tags now I have a different tag list.
- in this example if I don’t remember the tag I want while the file is in the inbox I either have to move to the db1, select that file and fix the tag to match the ones available in db1. But now, after I fixed the tag in db1 if you look at the tab list you will see the tag yo
tags between dtpo and finder:
- In DTP tags are case sensitive. So lets say you have a tag called mytag. mytag and MyTag are different. No problem.
- Now lets say I have 100 tags in dtpo. I would like to sync (preferably) or “mirror” the 100 tags from dtpo to “finder”.
- I can enter them manually in finder but that is cumbersome and difficult with both the volume and the cap sensitive.
- If not automatic (i.e. sync) then to manual hit a command that sends all the tags in dtpo to finder so they are “mirror”. This also means I do not want extra tags in finder should I have typed them wrong.
I hope this is a little clearer. Thanks again for your quick responses.
Jay
Hi Scott,
Im wondering if you are able to answer the question above? Thanks
Jay
I have to admit, I’m not sure what more I can add beyond what I’ve already covered, and beyond your own observations in your second post.
You cannot “sync” tags, and you are subject to the limits and quirks that I, and you yourself, have outlined above.
tags will follow Documents between databases and between DTPO and Finder. If you created a tag in one Database, it MUST be re-created in another databse either by moving a file with that tag into that databse, or applying the tag to the new file.
Indeed, because they are case sensitive, this means that you may inadvertently create near-duplicate tags (“mytag” and “myTag”). This is where your taxonomic strategy will have to adapt. I, for example, only ever use lowercase tags.
You can also just periodically audit your tags. If you look at the list of tags in a given database, sorted alphabetically, spotting near-duplicates is pretty easy and you can simply merge them.
I’m sorry I don’t have a better answer for you but, again, I’m not 100% sure I know precisely what your question is at this point! That said, I do think that your answer lies in your own post: you’ve got to take care in your taxonomy to overcome them imitations of tagging in Finder and DTPO.
Hear, hear! This is sage advice for those prone to tagging. 
As a suggestion, the Tags view (CMD-6) is even better than sorting the list of tags in another view. The Tags view sorts alphabetically and puts everything in a flat list, which is helpful for those who have nested tags. It also sorts group tags in the same flat list, for those who have group tagging enabled.
Thanks to all of you. You did answer my question. I too have changed to all lowercase for everything! It is crazy how one program ignores capitalization it and another doesn’t. Also, with respect to Tags, it seems to me that both apple and microsoft are heading in that direction [finally].
In DTPO search with a tag criteria has several steps and is cumbersome for those who use it as a primary field to sort. A while ago I inquired about being able to place a modifying word like “tag:” similar to email: or phone: in html. and learned that this is not possible yet. Is there any discussion on placing tags on the front window of the search? It would save a lot of time. 
Thanks again and a Happy New Year to you all.
Jay
Yes we are discussing some syntax changes for future releases of DEVONthink. Cheers!
I YEARN for a nice, syntax- or token-based search system, contains: tag: name: (after all DTTG has this), instead of or in addition to the current, clunky, system of clicking around.
Glad to hear you folks are rethinking some aspects of search! 