Deleting tags doesn't seem to work?

DEVONthink Tips
1

I’m new to DT and spending a lot of time trying to wrap my head around it, so please bear with me. I’ve tried to read up on the documentation, but there’s very little of use available about the new version and since tags are the newest part it’s even more sparse.

Scenario:
I scanned in a document via exactscan and since I decided the best way to start getting a handle on tags was to just use them, I added some tags. Unfortunately I used tag1,tag2,tag3 as the entry string as that’s what I’ve used in other systems for specifying multiple tags. This caused the obvious problem of having a single tag created “tag1.tag2,tag3” instead of three separate tags.

'K

So I go into the appropriate DB in 3 pane view, go to that file and turn the tag bar on so I can see the tags associated with the file. It’s got the one initial concatenated tag and it’s oddly picked up other tags that have nothing to do with this file.

Ok

So I just need to select one of these light blue tags (which makes it a darker blue as it’s selected) and press delete. So that looks like it worked (I.E. the ones I deleted disappear from the tag bar above the bottom right pane in 3 pane view ), and I delete all the non-relevant tags. BUT when I go to another file and back to this one all the tags I deleted come BACK!

So I go to the TAG (which I haven’t really worked with before) view instead of 3 pane view, and on the far right I have a pane with what looks like THOUSANDS of useless tags (dozens of duplications etc) all stacked up! Looks like they were auto-created when I imported some archives from my old machine?

Is there a way to actually get RID of tags like that?

I REALLY want to use tags to classify and further refine the relationships of files/documents across groups etc, but honestly I’ve found the tags implementation in DT to be totally counter-intuitive and flat out confusing/frustrating so far. :frowning:

Can anyone provide some guidance of just HOW tags work and how to actually use them in an effective manner?

Is there a way to start over, wiping the tags clean avoiding auto-created tags based on file name/folder structure below the top level groups?

Thanks in advance for any assistance, we’re really trying to get started the right way with DT so I can convince my partner to use it, but right now I’m having a tough time selling it as tags seem useless and HIGHLY frustrating.

James

2

I’m only a little bit less confused then you are, so I’m probably not the most helpful person to answer your question about why tags come back, I’ve never had that happen to me once I’ve removed them.

I do very much want to reiterate a request I made a while ago, in the hopes that the DevonThink programmers will hear this.

viewtopic.php?f=2&t=9923

Please, PLEASE, PLEASE add an option to the Database Properties (i’m guessing that’s the best place to put it? That way the 2 databases I have where automatic tagging is useful, can keep doing that, while everything else can be switched OFF separately) that let’s us turn off the “automatic” tagging DevonThink does, so that I don’t have to go through groups by hand, turning tagging off. Doing it all by hand is incredibly tedious and annoying.

I would very much like to use tags however I want to, in order to gain another entry point into my information, sorted how I want. I would really love an option to only see the blue tags (mine) in Tags view and have the grey, automatic tags turned off. So far the grey tags have done nothing at all which is even vaguely useful to me except automatically generate hundreds of useless tags many of which contain the same name, many of which contain names like x983-J6k4fyD and seem to attach themselves to web pages from sites I’ve downloaded to Downloads and only, at best, 5-10% of what DevonThink automatically generates is in any way useful.

I’d like to just TURN OFF AUTOMATIC TAGGING without turning off MY tags, on a per-database basis.

Thanks for hopefully reading this and maybe acting on it.

3

That seems to be 1/2 of what I’m fighting with as well. It looks like it’s auto-assigning tags based on content that it read when it OCR’d the files?
If that’s the case that’s REALLY obnoxious and counter the the fundamental purpose of tagging, which is to allow the Human element to specify individual tags based on what HUMAN judgement views as the important aspects of the file in question.

Auto tagging seems to be generating soooooooooo much “noise” tags that it renders the entire tagging system nigh useless.

How does one go about turning off the auto tagging? And will that remove the thousands of auto-generated tags that are currently making my tagging system unusable?

James

4

Forgot to mention that the auto-tagging of files that are multiple nested folders makes for an ungodly mess.

I’ve got a LOT of archives of different versions of software systems for things like vBulletin versions and plugins, Joomla! versions and extensions etc, and as they are often in computer generated folders with names like “0” or “axufte1236” it causes an incredible amount of useless noise in the tagging system since there are multiple instances of the same value which each generate an identical tag but each is showing up separately. Total bedlam.

I’d suggest at a bare minimum turning OFF auto-tagging by default and letting people define WHEN it should be used. One option would be to be able to specify how many levels deep to enable auto-tagging.

I can see it being a LOT more useful if I could tell it to only auto-generate tags for the first 2 levels of the file system hierarchy, or maybe 3. That would vastly reduce the noise while still maintaining the usefulness of dropping individual files into a filing system and have them auto tagged.

James

5

One reason you are seeing so many tags in tag view is that in DT tags and groups are essentially alike, i.e., groups (folder names) automatically become tags. For example, you have a group called finance. Inside the finance group is a subgroup called receipts. If you import a receipt from Best Buy and tag it audio equipment, then place it in the receipts group, the file will carry three tags: finance, receipts, and audio equipment. You can turn off the automatic tagging of groups by clicking on a group, clicking on the info button, and checking the box ”exclude from…tagging“ in that pane.

6

Understood.

Problem is, I’m dealing with an archive that has tremendous numbers of deeply nested files in it. Things with a path like:
Joomla/extensions/wordpress_in_joomla/working/archive/2_3_8/default/images/header-img.php

And that’s a fairly short one. Most of the vBulletin plugin paths are longer as they need to be fully extracted before uploading via FTP.

So the problem is I’ve got an archive with groups that contain tens of thousands of files (the Joomla group for instance has 49,602) all deeply nested as it’s multiple versions of many extensions along with the changes I’ve made to the original files for various reasons.

Problem: I can’t turn off tagging for the entire group, as turning off tagging on a parent group does NOT affect the tagging on child groups. So I’d have to spend literally days sorting though all the hundreds and hundreds of folders (groups) and sub folders (sub groups) and turning off tagging for each and every one?!

That just doesn’t seem practical, and it’s absolutely sure that neither I nor my partner have the patience to waste days/weeks of time just turning off tagging on groups :wink:.

I’m sor of understanding the concept behind DT’s tagging, but wanted to provide some constructive feedback to point out that for large archives with deeply nested structures of files (esp where you have duplicate group names because you’re looking at different versions of the same software package) it’s essentially useless.

We use tagging systems for images, for threads/posts, blog entries. etc etc. In my understanding/usage Tagging as a concept is generally used to allow a human agency to draw linkages between elements in different logical containers to group disparate elements into a single group because they are related in some fashion. DT’s tagging system seems to totally miss the whole point of tagging? IMO tagging should be a secondary classification system that allows one to enhance the hierarchical filing system and create virtual groupings of files to reflect their associations. I tend to think of them a bit like a “View” of a SQL database. A way for me to pull together multiple things from different tables and present them as if they were a totally separate table.

Tagging should be an organic, human created structure in order for it to be used effectively in my experience. You certainly want to provide a mechanism to help the user use existing tags rather than creating slight variations due to spelling changes etc, but it shouldn’t be fully automated or it loses the point, and the USEFUL, human designated tags are lost is the vast quantity of “noise” created by the auto-generated tags.

Honestly, I’ve got 10 tags that were auto-generated in this particular database that are all named: “1”. Ten separate tags with just the name “1”, how is that useful? With hundreds and hundreds of duplicates like that, it’s impossible to FIND any of the tags I created by hand.

Seems to me that the ability to specify on a database or group level how many levels deep to create tags based on the file hierarchy would go a long way to making it more useful.

The other big problem was files picking up tags that had nothing to do with them? I scanned in some tax documents, put them in a “Financial/Taxes” group and they automatically had tags of “Joomla”,“wysiwygpro”,“working”, and “project”. The file in question was NOT in any of those groups, and I was unable to delete the extraneous tags, they just came back when you looked at another file then came back. It seemed to happen after I ran the PDF through the OCR, so it seemed like it was assigning tags based on text content?

James