Jump to content

Feature request: Autotag notes with tag hierarchy of nested tags


Recommended Posts

I've begun nesting tags, sometimes 3-5 levels deep, as a way to keep them organized. Something that would be really helpful is if a note was auto-tagged with the tag hierarchy when the user adds the tag to the note.

 

For example, if I have: IT > Vendors > IBM and add the IBM tag to a note, the note would be autotagged with IT and Vendors.

 

Thanks for the consideration!

Link to comment
  • Level 5*

I've begun nesting tags, sometimes 3-5 levels deep, as a way to keep them organized. Something that would be really helpful is if a note was auto-tagged with the tag hierarchy when the user adds the tag to the note.

 

For example, if I have: IT > Vendors > IBM and add the IBM tag to a note, the note would be autotagged with IT and Vendors.

 

Thanks for the consideration!

 

I would approve of this suggestion only if it was available as an option next to how it is today, as your suggestion unfortunatly would ruin the way I and some others use nested tags.

 

I have set up my 2000+ tag system in the way that I for example have a parent tag named US, next level has Florida and 49 other tags for each state, and the final level has Miami and other Florida cities etc. Now if all the notes that are tagged with a tag under US were also tagged with US, there would be so many notes showing when I press US that I would have a very hard time finding notes that are about the US as a country, which is what the tag is called and should show.

 

As both alternatives calls for semi-good solutions to their respective weaknesses, my suggestion is to try and think of an alternate way to tackle your problem and what it is you are after. If you for example use some tags with child tags more often then others, like say Vendors, my suggestion is to create a saved search containing all tags under and including Vendor, so that when you press that saved search, it will show all notes tagged with either one of these tags. 

Link to comment
  • Level 5*

My take on this has always been the other way 'round: rather than automagically adding tags (though user setup, rules-based auto-tagging or auto-notebooking, a la GMail or Outlook would be welcome), I would prefer that the search grammar be modified to be able to take into account tag hierarchies.For example, a specialization of the tag: search term -- I've suggested +tag: -- would allow a search of the named tag and all notes containing its subtags; in the example cited above, this would be +tag:Vendors. This could also be supported by actions in the mouse-click actions tag tree; say, Alt+Click on a tag (since Ctrl+Click and SHift+Click are already taken) could cause the search of the selected tag and all of its subtags.

Link to comment
  • 5 months later...

Hello-  Long time EN user here.   It's time for EN to step up and provide some actual functionality for the Nested Tags.  As we know, today it is just a visual hierarchy with no underlying semantic relationships of the items in the "tree".

 

I definitely see a need for both bconrey's and gustavigi's approaches.  I have one modification to add.  The ability to select how many levels 'up' or 'down' the tree you want to add to the new note.  This would simultaneously address both requirements.  I am not a big fan of jefito's approach for several reasons:

  1. a primary objective of both bconrey's and gustavigi's proposals (and mines) is to actually add the tags that you want to a note.  Jefito's approach would not add the tags.
  2. the approach would return all notes that had the 'subtags' in the tree in question whereas I may only want portions of the tree (this is very similar to the concern that gustavigi expressed) 

 

I'm neither unrealistic or greedy and would be very happy with a phased approach from EN.  Maybe start with gustatvigi's suggestion of adding bconrey's suggested functionality but also keep the current functionality.  Then in a subsequent version add the feature I am proposing of being able to specify how many levels 'up' or 'down' you want to include.

 

Either way- its time for EN step up it's game regarding nested tags.  Thousands of us need greater functionality from tags.  The Taggify solution for Mac is a great proof point for this.  

 

Here's the real deal for EN (or value proposition if you want to keep if more formal)  Those of us that are advanced users of EN know that the next big thing about to explode on the scene are "one-stop" solutions for content management and task management (email, social (Facebook, Twitter, etc.) , blogs, rss, etc. Contenders like IQTELL, and Unified Inbox are coming on fast.   EN is still the overall best 'one-stop' but must continue to innovate to stay ahead.  It is my opinion that adding more functionality for tags would be one of those "Kano" 'nice-to-have' features that would put EN solidly in the lead for a long time to come.  As we all know 'nice-to-have' features eventually become 'must-haves' so EN could really solidify its lead by being first out of the gate with advanced Tagging.  

 

Eventually one of the "one-stop" contenders will search deeply for EN weaknesses and discover this one about its mediocre tagging solution and exploit that weakness to gain market share (that's the way market disruptions work folks).  So IMO, beefing up Tagging is a solid, no-brainer, win-win proposition for EN.

Link to comment
  • Level 5*

I am not a big fan of jefito's approach for several reasons:

  • a primary objective of both bconrey's and gustavigi's proposals (and mines) is to actually add the tags that you want to a note.  Jefito's approach would not add the tags.
As I said in my post, I am in favor of adding automated rules which would do auto-tagging, among other things.

  • the approach would return all notes that had the 'subtags' in the tree in question whereas I may only want portions of the tree (this is very similar to the concern that gustavigi expressed) 
I have no idea how you would specify parts of a tree without referring to sub-trees. Example? And since my suggestion would be part of the search grammar, you'd be able to filter even more, as you can now.

 

Eventually one of the "one-stop" contenders will search deeply for EN weaknesses and discover this one about its mediocre tagging solution and exploit that weakness to gain market share (that's the way market disruptions work folks).  So IMO, beefing up Tagging is a solid, no-brainer, win-win proposition for EN.

You may consider it 'mediocre', but it's almost exactly how GMail's labels work, so at least it's in good company. That being said, I'd be in favor of being able to take advantage of the nested tag tree, and have been for nearly as long as I've been using Evernote and posting here. Auto-tagging? Check. Enhanced search syntax? Double-check.
Link to comment
  • 3 months later...

I would like both (and believe it is possible).  Nested tags (call it a tree or whatever, I use hierarchy) with auto-aggregation.  I use the term Smart Tag or Tag Groups (could be dynamic tag or virtual tag or filter tag) such that is primarily a visual UI concept of nested where clicking the tag shows all notes of the tags below it.  Nothing is added to the note, it only in the user-defined tag groups. It would allow a cleaner UI (save screen space by no duplication of tag terms). Dependant only on a list of nested tags and would include those in any tally counts.  This would be especially useful for temporary or test a new naming system (esp. if manual ordering is allowed) and prevent orphan/unused tags.  Depending on the implementation, it could work offline on the client and as a access control security.  I envision Saved Search to allow more than just tags, has not implicit structure, and likely requires reading the search help file.

 

Tag Groups/Smart Tags are very different from auto-tags. Auto-tagging by itself has no structure and can be hard to work with groups without a good filtering UI(even just remembering the tag name), but combined with Smart Tags could be very empowering.

 

see more at

Feature Suggestion: Hierarchical Tags (Ordered Lists) and Auto-Correction for Tags and more
Link to comment
  • 1 month later...

 

Jefito said "I have no idea how you would specify parts of a tree without referring to sub-trees...."

The way you would do this is as follows:  Lets say you had a tag tree that had 10 levels L1 - L10.  Let's say you want to tag a new piece of content with only the last 3 levels.  The feature would allow you to drag your note to the tag tree and then provide a pop-up box (or part of the search syntax) that would allow you to specify up or down and a number for the number of levels.  So for example - to execute the first example that I mentioned - I would drag the new note to L8 tag in the tree and then type "down3" (or perhaps -3) in the pop-up box.  The feature would then populate my new note with the tags for L8 - L10.  If all I wanted to do was see all notes with the L8 - L10 tags - I would just type 0 in the pop-up box and the feature would then only display notes with tags L8 - L10.  

 

 

Jefito said "You may consider it mediocre but it's exactly the way Gmail labels work so at least it's in good company."

Not necessarily.  I have always considered Gmail labels to be somewhat underpowered.  Yes, they are definitely a step forward for email but they're almost like a tease - in the sense that they tend to leave you wanting more functionality.  For example, wouldn't it be great to have nested labels in Gmail???  And wouldn't it be great if the nested labels worked the way we are describing them in this thread where new mails could 'inherent' specified parts of the "label tree"???  So bottom-line, mediocre is still mediocre no matter who has done it.  Just because Google has thus far only taken labels to a mediocre level doesn't all of a sudden turn a mediocre implementation into excellent.  Remember- the point of my post was to encourage EN to take the LEAD.

Link to comment

There is an over-arching philosophical point in this thread that could be getting lost.  ONE SIZE DOES NOT FIT ALL.

 

I like many of the approaches and themes laid out in this thread.  The point is - we are all different and even if were all the same - we all have the need for a different 'tool' for different jobs in our busy lives at varying points in time.

 

There may be cases where we only want to look at the notes with tags in a tag tree.  There may be times we want a simple way to add the tags in a tag tree to a note.  There may be times we only want to add specified portions of the tag tree to a note.  The point is - we need this variety of capability available to us.

 

It's really very simple.  Think about a spreadsheet.  Spread sheets used to just do simple math.  But as we all know we now can not only do simple math but statistics, trigonometry, advanced financial calculations and so forth.  We sometimes tend to limit ourselves by thinking that one person is advocating ONLY for a particular solution, when in reality we need a broad selection of capabilities available to us.  Nested tags (with all of the functions described above) is one such very important tool-suite that we need in our tool-kit.

Link to comment
  • Level 5*

Jefito said "I have no idea how you would specify parts of a tree without referring to sub-trees...."

The way you would do this is as follows:  Lets say you had a tag tree that had 10 levels L1 - L10.  Let's say you want to tag a new piece of content with only the last 3 levels.  The feature would allow you to drag your note to the tag tree and then provide a pop-up box (or part of the search syntax) that would allow you to specify up or down and a number for the number of levels.  So for example - to execute the first example that I mentioned - I would drag the new note to L8 tag in the tree and then type "down3" (or perhaps -3) in the pop-up box.  The feature would then populate my new note with the tags for L8 - L10.  If all I wanted to do was see all notes with the L8 - L10 tags - I would just type 0 in the pop-up box and the feature would then only display notes with tags L8 - L10.
Sorry, but yeesh, that's a scary UI design. In the first place, dragging a note (or set of notes) to the tag tree doesn't do a search, it applies the tag to the notes, so that's breaking prior user experience. Second (supposing that we overlook objection #1), does this mean that dragging a note to a tag that lives in a tree will require that the user interact with the sub/super tree UI? It's one thing to make difficult operations feasible, but you should try not to do it at the expense of keeping easy/common operations easy to do.

 

Jefito said "You may consider it mediocre but it's exactly the way Gmail labels work so at least it's in good company."

Not necessarily.  I have always considered Gmail labels to be somewhat underpowered.
How does this contradict in any way what you quoted from me? to "be in good company" just means to "be in the same situation as someone important or respected", if you can believe Google.

Yes, they are definitely a step forward for email but they're almost like a tease - in the sense that they tend to leave you wanting more functionality.  For example, wouldn't it be great to have nested labels in Gmail???  And wouldn't it be great if the nested labels worked the way we are describing them in this thread where new mails could 'inherent' specified parts of the "label tree"???  So bottom-line, mediocre is still mediocre no matter who has done it.  Just because Google has thus far only taken labels to a mediocre level doesn't all of a sudden turn a mediocre implementation into excellent.  Remember- the point of my post was to encourage EN to take the LEAD.

I certainly didn't say that GMail label handling was excellent. All I'm saying is that an established, well-known, and widely-used company has seen fit to keep their labelling system fairly simple. Could it (and by extension, Evernote's similar system) be made more powerful? Of course. I've said that many times. Does it need to be? I'm not so sure about that.
Link to comment

 

Jefito said "Sorry, but yeesh, that's a scary UI design."

Yes, correct, my description could have been clearer.  In my proposal you would only drag notes to the tag tree if in fact you wanted to add tags to the note.  If you don't want to add tags then simply clicking anywhere in the tag tree would pop-up the same dialogue box where you could enter things like +3, -3, etc. (e.g. +3 would mean all tags up to three levels above the current tag and -3 would mean all tags up to three levels below the current tags) and EN would only display notes with those tags.

 

Actually, this is a great UI design- because it frees the user from needing to do queries when they have something quick and simple that they want to do.  Remember my philosophical point - we should give folks plenty of ways to do stuff.  This would certainly not complicate the UI but in fact make it simpler and more usable.  Also keep in mind that you have also agreed that you are for expanding the functionality of EN - my proposal would certainly expand the functionality of the sub/super tree UI but in a very simple and intuitive way.  There is nothing in my proposal that complicates the UI.  Let's see:

1) Folks are already dragging notes to the sub/super tree UI so no changes there.

2) Folks are already very familiar and adept with friendly easy dialogue boxes so no complications there.

 

Hmmm, I guess I fail to see the complications that you are referring to - on the contrary I only see enhanced functionality provided in a very consumable, easy and user friendly fashion.

 

Regarding your other points about labels in Gmail - I think we're basically on the same page and essentially saying the same thing.  We both agree that there could be more functionality to both Gmail labels and EN tags.  Where we disagree is that you are still not certain that expanded functionality (for tags) is necessary whereas I definitely feel that it is both necessary (more ways to skin the cat) and that it would be a huge market win for EN.

Link to comment

One additional point.  There is another reason why the nested tag functionality that is being proposed in this thread is so important.  It allows folks to develop their own semantic taxonomy standardization in a very quick and simple way.  Without going too deep let me briefly explain:  Lets say that I'm a project manager at a large company. Lets say I have developed my own personal way of categorizing projects.  However incorporated into the 'semantics' of my categorization are my own personal 'standards' for how to handle things like project lifecycles and the time dimension.   Here would be an example tag tree: (wild example - not much thought)

 

Project A

   Governance Phase

        Steering committee

                  New Fiscal Investment proposal

                           Prep

                                 Review

                                      Final

 

This hierarchy would mean certain things to me.  So for example if I want to look at all Steering Committee actions across all Projects that have new Fiscal Investment proposals that are at least in the Prep phase I would click on Steering Committee - the dialogue box would pop up and I would type in -2.  EN would then return all notes with the Steering Committee, New Fiscal Investment proposal and Prep tags.  Or, let's say that I wanted to give a new project only the following tags; Governance Phase, Steering Committee, New Fiscal Investment - then I would drag the new note to Governance Phase and type -2 in the dialogue box.

 

These are very simple, straight-forward actions but it allows me (and others) to manage the semantics of their own personal taxonomies.  I personally believe that this is a very very important part of content management.  Because as our lives become more and more digital - the only thing that will separate organized folks from unorganized ones is the level of detail with which they develop their personal semantic taxonomies and the rigor with by which they apply them.

 

Another way to think about it is the following;  your personal semantics and taxonomy becomes equivalent to the Cloud but instead of a Cloud for content it is a Cloud for your idea-organization-system.  The ability to create these then frees you from being locked into any one tool because once you develop the taxonomy you can implement it in almost any tool.  However, the tool where you create, manipulate and store the taxonomy will end up becoming core to your life - THIS IS EXACTLY WHERE EN WANTS TO BE - IT WANTS TO BE CORE TO OUR LIVES.  And this is why expanded tag functionality has a lot more value underneath the covers than might be initially apparent.  BLAZE THE PATH EN - BECOME OUR PERSONAL TAXONOMY ENGINES.

Link to comment
  • 1 year later...

Completely understand and agree with Karetep, with two points to add:

 

1) When it comes to shared notebook with different departments in the organisation, it's a TEAM TAXONOMY ENGINES. It's like boasting the associative brain power of the team!

 

2) If there can be something a Dynamic Tag-Cloud where, upon selecting on certain tag(s) and seeing the list of all the notes that has that tag(s), the tag-cloud would then automatically changes in such a way that it displays all the other tags that are associated with all the resulting notes.  It's like the one in Leap (by ironicsoft)

 

I would use such tag-cloud for 3 functions:

1. to narrow down my search (by filtering out notes that do not have all of the tags that I select)

 

2. to bring out the existing but hidden associations between notes that one may not even realise otherwise. It's kind of like the new "context" function, only this is determined by the network of tags used by individual, or by a team or family (in case of the using of a shared notebook)

 

3. to create saved searches, something like the smart folders in OS X.

 

see: https://discussion.evernote.com/topic/64741-tags-cloud/

Link to comment
  • 5 months later...

Completely understand and agree with Karetep, with two points to add:

 

1) When it comes to shared notebook with different departments in the organisation, it's a TEAM TAXONOMY ENGINES. It's like boasting the associative brain power of the team!

Hello nc40-

Yes, love your contribution here.  The TEAM TAXONOMY ENGINE  is way-cool and definitely is another strong usage scenario where EN could blaze the path!!

Your other two points are also rock-solid, I'm Loving It!!

Link to comment
  • 11 months later...
  • Level 5*
6 hours ago, krisdestruction said:

This feature can be so easily achieved. Just create hook into IFTTT and allow the user to configure tag rules based on detected input triggers.

Its difficult to figure out what feature you're talking about and how using IFTTT can achieve it.
Can you provide more details?

Link to comment

Any type of auto tagging feature would be achievable if provided with the right triggers on IFTTT.  Detecting which notebook or area the note is added to. Even getting the entire hierarchy as a variable is possible in IFTTT. Auto adding tags as the output is possible if the devs added support for it.

 

I guess I didn't read the OP as in depth, but basically I was looking for auto tagging based on note title matching. 

Link to comment

Archived

This topic is now archived and is closed to further replies.

×
×
  • Create New...