Jump to content

Simple, powerful, obvious,... and missing from Evernote


Recommended Posts

I'm a frequent Evernote user, on multiple platforms.  I depend on it, and I desperately want to like it.  But it frustrates me every single time I interact with it.  I write this in the hopes that I'm just missing something.

 

To me, the most transformative benefit that came with storing things digitally was the availability of an arbitrarily deep hierarchy.  No longer did I have to limit myself to a sheaf of notes stapled together inside a manila folder inside a file drawer.  I could create a lush hierarchy that let me quickly navigate to my current area of focus, and brought complete freedom to subdivide that focus, or not, as deep or shallow as I chose.*

 

Here's the key:  When I add a note, I may not remember how I've organized things in this "region" of my brain precisely.  I want to see that organization explicitly, so I can decide where the new information belongs.  If there's already a note on this topic I may want to add to it rather than force myself to look in two places for it.  If there are related ones that are diverging into multiple clusters I may want to split them.  Throwing my new note into an unsorted pile and hoping search can find it later feels like throwing it away to me.

 

Evernote is a big step backward from files stored in filesystem folders in this respect.  I acnowledge all the powerful data entry tools that come with EN -- that's why I use it.  But in terms of organization we're all the way back to manila folders (Stacks) containing sheaves of paper stapled together.

 

Why?  What's so hard about putting notes in a folder hierarchy?  Is there some downside to this that I can't see?

 

-J

 

* I fully recognize that tags are in one sense more general than a folder hierarchy in that they allow a single item to be associated with multiple organizational nodes.  But that strength is also their weakness -- they don't allow you to rediscover the hierarchy as you navigate it.

Link to comment
  • Level 5*

But in terms of organization we're all the way back to manila folders (Stacks) containing sheaves of paper stapled together.

Hardly. Tags make organization *much* better than manila folders (better than hierarchical folders, in my opinion, though tags could be better, too).

Why?  What's so hard about putting notes in a folder hierarchy?  Is there some downside to this that I can't see?

Hard? Downside? Dunno; it's more a choice than a difficulty, I believe. Oddly, Google GMail evidently agrees with this type of architecture. But who knows, they may change their minds someday. I wouldn't bet my workflow on that, though.
Link to comment

* I fully recognize that tags are in one sense more general than a folder hierarchy in that they allow a single item to be associated with multiple organizational nodes.  But that strength is also their weakness -- they don't allow you to rediscover the hierarchy as you navigate it.

The problem with hierarchies is you often have to "rediscover" it, if/when you don't remember how you organized it.  IE, did you file your cell phone bill under "phone", "telephone", "cell phone" or "Verizon"?  So you have to go to the bills folder & look for it one of those options.  OTOH, with tags (and Evernote's amazing search engine), you have the luxury of not having to "rediscover" how you filed it.

 

http://discussion.evernote.com/topic/48008-search-by-tag-or-just-search/?p=244648

 

IMO, you're overtagging. IMO, some folks "overtag". Often, tags are not needed if you use descriptive titles, some tags, some notebooks/stacks & keywords. I have over 62,000 notes & have never reached the 250 notebook limit. I'd guess less than half my notes have any tags at all. Probably only about 1/3 of my notes have tags. Those that do have tags normally have only 1-3 tags. IE, all my bills are in Evernote. Although they are in a "bills" notebook, I can quickly & easily find my Cox cable bill from June 2009 by searching ALL notes because the title of that note includes the vendor (Cox) and the date of the bill in YYYYMMDD format. So a simple search of

intitle:cox intitle:200906*

will quickly find the ONE note which is the bill I'm looking for from over 62,000+ notes. And I didn't even use a tag.

You can also net the tags for organizational purposes only. This allows you to collapse the top tier so the nested tags do not show up. Or even the "hide unassigned" tags may be helpful.

Link to comment

Thanks for the responses

> Jefito: 

I get that some people like a completely flat organization accessible only via search, and using tags as search aids.  I don't.  But the point is they aren't mutually exclusive -  there is no reason I can't tag and search things stored in a hierarchy.  But actively preventing me from using a hierarchical organization is, to me, unnecessarily restrictive.

 

Google is an interesting case in point.  It's a company constructed on the fundamental belief that search solves all problems, and gmail under the hood has tags but not folders.  But interestingly, they allow a hierarchical organization of mail via tags that does a good job of emulating folders, yet retains the key benefit of tags, which is the ability for a message to be in multiple places in the hierarchy at once.  This is a fine solution, and something Evernote could do too.

 

>Burgers...

No, you're never forced to rediscover a hierarchy, because you aren't forced to use it.  You're welcome to throw everything in a big pile and search it if you like.  Or, with a Gmail-esque implementation, if you can't remember which tags you've used to refer to something in the past, you can use the hierarchy to remind you.

Link to comment
  • Level 5*

Whether you prefer hierarchies or not, the fact is that Evernote doesn't support anything beyond stack/notebook/note and it seems given the history and comments from  Evernote that this isn't going to change in the short to medium term (at the very least).

 

So, you have a choice, work with Evernote as it exists now and probably for the foreseeable future or find another tool that suits you better.

Link to comment
When I add a note, I may not remember how I've organized things in this "region" of my brain precisely.  I want to see that organization explicitly, so I can decide where the new information belongs.

 

 

You almost sound like you're looking for a mind mapping tool.  There are a few integrations that work to expand upon our flatter architecture--Mohiomap springs to mind: http://discussion.evernote.com/topic/37800-mohiomap-public-beta-launched/

 

And I'd check out the App Center as well: http://appcenter.evernote.com/

Link to comment
  • Level 5*

...It's also possible to set up a virtual folder hierarchy by using inter-note links (although that works only on the machine that Evernote is installed on).  Create your preferred structure of parent - child notes and link them together,  then whenever you add a new note just link to it from the relevant child node.  You'll be able to 'see' your structure then.

 

- But it's much better to rely on tags and search.

Link to comment
  • Level 5*

> Jefito: 

I get that some people like a completely flat organization accessible only via search, and using tags as search aids.  I don't.  But the point is they aren't mutually exclusive -  there is no reason I can't tag and search things stored in a hierarchy.  But actively preventing me from using a hierarchical organization is, to me, unnecessarily restrictive.

Oh, I understand that they don't need to be mutually exclusive. There's also no reason that a hammer can't have a screwdriver attachment either (I'm sure they exist); even so, companies still make pure hammers, and people buy them and use them as such.

 

 

Google is an interesting case in point.  It's a company constructed on the fundamental belief that search solves all problems, and gmail under the hood has tags but not folders.  But interestingly, they allow a hierarchical organization of mail via tags that does a good job of emulating folders, yet retains the key benefit of tags, which is the ability for a message to be in multiple places in the hierarchy at once.  This is a fine solution, and something Evernote could do too.

The way that I understand GMail labels is that they are stored in a hierarchical fashion, but unless I've missed something, search doesn't respect the label hierarchy, just as Evernote doesn't (though I've long thought that it would be great if Evernote did). And I don't see them emulating folders in any way that I can recognize either; at least it doesn't seem to work that way for me.

Link to comment

Wow.  Clearly I've inadvertently stumbled onto a religious issue hereabouts.  i wonder why a discussion about the digital equivalent of how best to sort one's sock drawer sparks such passionate denouncement of an alternative approach.

 

>Jefito:  "The way that I understand GMail labels is that they are stored in a hierarchical fashion"

I don't think this is the case.  I believe Google's method is, that in lieu of a folder hierarchy like

Folder1

      Folder1.1

      Folder1.2

             Folder1.2.1

 

They instead create a "label" (i.e. tag") for each explicit folder, where the tag is a concatenation of the folder label, e.g.

tag1 = "Folder1"

tag2 = "Folder1/Folder1"

tag3 = "Folder1/Folder2"

tag4 = "Folder1/Folder2/Folder1"

 

While the contents are still stored (I believe) flat, with this method, sorting and displaying them in a hierarchy is easy, since the location in the hierarchy is implicit in the construction of the tag.  Search can be constrained to a single folder, but not to a folder tree.

 

Metrodon:

So thanks for the "my way or the highway" sentiment, but could you explain (if you know) why Evernote is so averse to hierarchy?  Does it violate some other fundamental premise of their value proposition?

Link to comment
  • Level 5*

What always amazes me about these discussions (and there have been many,  many more over the years) is

  • Someone wants a specific feature added
  • Evernote read the posts here, so the suggestion will be noted; but they don't typically comment on if,  when or whether said feature will (ever) be added
  • Everyone chimes in to point out work-arounds - though sometimes it can sound a bit overly defensive
  • Original poster strongly objects to not getting own way immediately,  wants to know why* feature can't be implemented

*Answer: because they own the product,  don't have to defend or comment on every decision they take,  and (sensibly) do things that fit in with their long term plans,  without getting distracted by requests for 'easy to implement' changes.

 

I've got my own little list of things that I'd like to see added,  plus a much longer list of things that need to be improved.  Since I do have a day job though I'm going to continue using my workarounds - and quite a lot of other software - while Evernote do their development work.  I'm on the Beta track for both desktop and PC because that's where the newest features are - but some of my workarounds are safety belts in case things go sadly wrong.  

 

(Don't try this at home kids unless you're a Data Professional like what I am.)

Link to comment

What always amazes me about these discussions (and there have been many,  many more over the years) is

  • Someone wants a specific feature added
  • Evernote read the posts here, so the suggestion will be noted; but they don't typically comment on if,  when or whether said feature will (ever) be added
  • Everyone chimes in to point out work-arounds - though sometimes it can sound a bit overly defensive
  • Original poster strongly objects to not getting own way immediately,  wants to know why* feature can't be implemented
*Answer: because they own the product,  don't have to defend or comment on every decision they take,  and (sensibly) do things that fit in with their long term plans,  without getting distracted by requests for 'easy to implement' changes.

 

I've got my own little list of things that I'd like to see added,  plus a much longer list of things that need to be improved.  Since I do have a day job though I'm going to continue using my workarounds - and quite a lot of other software - while Evernote do their development work.  I'm on the Beta track for both desktop and PC because that's where the newest features are - but some of my workarounds are safety belts in case things go sadly wrong.  

 

(Don't try this at home kids unless you're a Data Professional like what I am.)

 

 

Perfect! I'm warning you already that I've Evernoted this post & the next time this issue comes up (b/c we know it will), I'm going to link to this post.

Link to comment
  • Level 5*

 

 

Metrodon:

So thanks for the "my way or the highway" sentiment, but could you explain (if you know) why Evernote is so averse to hierarchy?  Does it violate some other fundamental premise of their value proposition?

 

No inside info, but I've been around for a while and reading between the lines it's a decision based on the architecture of their storage/processing and their own design choice.

 

Given that they are a private company and one of many who offer note/data capture it really is a case of "their way or the highway". There is no user right here to have Evernote work as they'd like it to, this isn't a service where all users are equal. By Evernote's own admission, this is an app that the company writes the way that it believes to be best and that works the best for it's employees first.

 

I quite like it, so I use it a lot, but I'm also realistic enough to know that Evernote is far from perfect, doesn't fit everything I need to store, most likely isn't anywhere near being really a 100 year company and is always going to have bugs, problems and things that annoy me.

 

My advice, use it as it is if you find it useful, use it for the things it is useful for now and don't spend lots of time agonising about a $50 a year software solution.

Link to comment

@jhd

 

What stops you from using the gmail's aproach as you described:

tag1 = "Folder1"

tag2 = "Folder1/Folder1"

tag3 = "Folder1/Folder2"

tag4 = "Folder1/Folder2/Folder1"

 

You can define these tagnames in evernote as well and you can organize them in a tag tree in the tag view...

Eric

Link to comment
  • Level 5*

Wow.  Clearly I've inadvertently stumbled onto a religious issue hereabouts.  i wonder why a discussion about the digital equivalent of how best to sort one's sock drawer sparks such passionate denouncement of an alternative approach.

Actually, clearly you've misunderstood what's being said here. For me, this is not a "religious issue", it's a matter of practicality. Evernote has chosen an architecture for their product -- that's their business, not mine (a user of Evernote), except that I do need to understand the facilities that they offer in order to 1) use Evernote effectively, and 2) help others to use it in ways that fits for them. There is no "passionate denouncement" of hierarchies (though they are problematic in some ways). Again, for me, they're just not necessary, and Evernote are perfectly allowed to make those choices in the design for their product.

So whether they have or don't have a purely hierarchical organizational system really matters very little to me. I don't have a stake in Evernote's choice, except where it comes to being able to find my stuff in their system. I've certainly used a great many hierarchical systems over the 30 years been a software developer, and I can tell you that I'm pretty comfortable and experienced with hierarchies, and I recognize that they can be useful; on the other hand, they're just not necessary for lots of situations. This is one, for for my usage anyways (GMail is another, apparently, and what you said about them actually reinforces the notion that it's not a lot different from Evernote in that respect). In Evernote and GMail, I don't really care how things are stored, just that I can locate things effectively, wherever they are located. So, it's OK, but not ideal, but that has nothing to do with lack of a full-fledged hierarchical organizational scheme. I do, however, have issues with the search capabilities, particularly with respect to a search language that is relatively inexpressive in a number of areas (exploiting the hierarchical organization of tags, for example), yet is arcane in other areas (the attachment search stuff). I have much stronger opinions about search than I do about the storage scheme.

So my defense of Evernote's current system is not that hierarchies can't be useful (they are undeniably so), it's really that they're not necessary. They are problematic in a number of ways: for example, they can lead to over-complicated workflows ("great, another depth-first search through someone's idea of a "lush hierarchy" that doesn't help me to find what I'm looking for") and they force you to put items in exactly one place in the tree where they might reasonably go in more than one place. I find nothing transformative about them, nor do I revel in the joys or rediscovering the hierarchy as I trawl through it -- that's just a bookkeeping detail that I'm going to forget ten minutes after I've found what I'm looking for, unless I spend a few moments thinking "why did I/they organize things *that* way". So yeah, they're useful, but sometimes dealing with them is just... so... old. Since they're not necessary, then I have no problem with the fact that Evernote doesn't use them. I don't miss them in the least. And when the latest bright spark comes along and says "why doesn't Evernote have hierarchical storage? That'd be brilliant. It's so obvious. It's such a no-brainer. Etc., etc.", my thought is that, hey, Evernote are smart folks and if it was so obvious, simple to implement, useful and necessary, don't you think they would have done it already?

Anyways, given that context, Evernote works for me in a lot of ways. I don't need hierarchies (though stacks are occasionally useful, thanks to a quirk of the search grammar where you can search only the contexts of all notes, a single notebook or a single stack), indeed, I keep my notebook count to a minimum: my rule of thumb: use notebooks where you need to (e.g. offline notebooks, local notebooks, sharing), and use tags for everything else. It's not all perfect (search language), but it works for me. I understand that it may not work for all, but ultimately, it's up to Evernote to make the determination of how much the effort to add this sort of functionality is worth to their business.

Link to comment

@jhd

 

What stops you from using the gmail's aproach as you described:

tag1 = "Folder1"

tag2 = "Folder1/Folder1"

tag3 = "Folder1/Folder2"

tag4 = "Folder1/Folder2/Folder1"

 

You can define these tagnames in evernote as well and you can organize them in a tag tree in the tag view...

Eric

Thanks eric99 - I think that's just what I will do.  I haven't used EN enough to know what a tag tree is, but I'm off to find out.

 

Those whose Ire I've (unintentionally) raised seem to have missed the point that hierarchy and tags are orthogonal - one does not preclude the other, and you can have either, or both.  But failing that, one way to view (and emulate) a nested hierarchical organization is that it just implicitly puts one tag on every new item by virtue of where in the hierarchy you put it. After that it's just down to whether the U.I. lets you see the hierarchy.

 

Best,

-J

Link to comment

J

You might want to check out Mohiomap which gives you a really cool graphical representation of your Evernote data. I understand what you mean about it being a "religious issue". When I went from using a regular email program to gmail it took quite a while to relax into relying on search and not bothering with hierarchy. Prior to that I spent hours filing stuff away into a labyrinth of folders and about 20 levels of sub-folders! Having said that I've now reached the point with Evernote where I often find myself unable to find something despite having everything tagged and knowing all of the search tricks. Maybe it's time to review my strategy for notebooks and tags.

--

Best rgds

Mark

www.todohack.com

Link to comment
  • Level 5*

Those whose Ire I've (unintentionally) raised seem to have missed the point that hierarchy and tags are orthogonal - one does not preclude the other, and you can have either, or both.  But failing that, one way to view (and emulate) a nested hierarchical organization is that it just implicitly puts one tag on every new item by virtue of where in the hierarchy you put it. After that it's just down to whether the U.I. lets you see the hierarchy.

No fear -- you haven't raised any ire, at least from my end. But some frustration, sure, in that you seem to have missed the two main points that I made in my prior post. First, that the independence of the hierarchical structure and the tag structure schemes is not exactly earth-shattering news. And second, defending Evernote's choice (because it's 1) a reasonable architecture, though not perfect for everyone, and 2) it's their product to define, and they live or die based on their choices) is not the same thing as saying that a hierarchical structure is a bad thing and shouldn't be implemented.

At this point, even though I don't like to say "it's never gonna happen", I think that Metrodon's basically correct: you won't be seeing this any time soon, if ever. I wouldn't bet my workflow on it appearing in Evernote if that workflow depended on hierarchical notebook organization, that's for sure.

BTW, the suggested method of encoding a hierarchy in your tag names is also pretty well known in the forums. It's got some benefits with respect to searching (used with wildcards, this gives you the ability to do some mixed AND/OR searching, something that is otherwise impossible to do with the Evernote search language), but some big awkwardnesses in terms of reorganizing your hierarchy (potentially lots of tag renames).

Link to comment

Jhd, this is a bit different than what your proposing but an area Evernote could go a long way on to remedy your needs and that is that tags be integrated to the Notebooks and Notes they're associated with. I've again proposed this today actually (http://discussion.evernote.com/topic/48952-a-notebooks-own-tag-view/) 

A good number of Android apps  tightly integrate Notebooks with viewable, accessible tags from those Notebooks....as in this Springpad example. Evernote's have no connection. 

It's a logical organizational structure and one in which  they wouldn't have to create another organizational layer. 

Sigh... 

Link to comment

Metrodon:

 

So thanks for the "my way or the highway" sentiment, but could you explain (if you know) why Evernote is so averse to hierarchy?  Does it violate some other fundamental premise of their value proposition?

 

I have heard that's it's quite a coding challenge /nightmare 

Link to comment
  • Level 5*

Again, no great inside knowledge here, but it appears to me....

 

Early on they made a design decision that they would build an organisational model built around tags. A very simple notebook structure was put in place that was later supplemented with Stacks (which are in the main a UI rather than structural change I think - with a little added search grammar).

 

I would guess that allowing a more complex hierarchy at this point would be a challenge from a processing/storage/app design and UI perspective and that given the success of the app in it's current form there is not a significant enough drive to make what would be most likely a very expensive change.

 

Most likely of all though is that this is a core design decision that they feel gives them the flexibility to grow and adapt without the cost and complexity of having to support tags plus hierarchies.

 

I feel some of your pain by the way. I use a hierarchy to store data on my hard drive, I have more folders in my email than I probably need and probably more notebooks in Evernote than I need too. I like a hierarchy.

 

But, I also like Evernote, I understand it's limitations (as far as my usage and requirements are concerned) and I adapt my usage accordingly. I like that it's pretty generalised, simple, that it has few user config options. These are things that I know drive other people crazy, they want to be able to change the colour of notebooks, add specific workflows, support GTD methods, have a night-time reading mode, write books, draw pictures, store photos etc etc etc. And you know what? Evernote does an OK job at all of those things, it's not perfect (a long way off in some ways) but because of it's generalised nature and because it's available in so many places it's really useful. You can adapt the way you work to fit in with it and do lots and lots of these things with it and for me that makes is very useful.

Link to comment
  • 5 months later...

I'm a frequent Evernote user, on multiple platforms.  I depend on it, and I desperately want to like it.  But it frustrates me every single time I interact with it.  I write this in the hopes that I'm just missing something.

 

To me, the most transformative benefit that came with storing things digitally was the availability of an arbitrarily deep hierarchy.  No longer did I have to limit myself to a sheaf of notes stapled together inside a manila folder inside a file drawer.  I could create a lush hierarchy that let me quickly navigate to my current area of focus, and brought complete freedom to subdivide that focus, or not, as deep or shallow as I chose.*

 

Here's the key:  When I add a note, I may not remember how I've organized things in this "region" of my brain precisely.  I want to see that organization explicitly, so I can decide where the new information belongs.  If there's already a note on this topic I may want to add to it rather than force myself to look in two places for it.  If there are related ones that are diverging into multiple clusters I may want to split them.  Throwing my new note into an unsorted pile and hoping search can find it later feels like throwing it away to me.

 

Evernote is a big step backward from files stored in filesystem folders in this respect.  I acnowledge all the powerful data entry tools that come with EN -- that's why I use it.  But in terms of organization we're all the way back to manila folders (Stacks) containing sheaves of paper stapled together.

 

Why?  What's so hard about putting notes in a folder hierarchy?  Is there some downside to this that I can't see?

 

-J

 

* I fully recognize that tags are in one sense more general than a folder hierarchy in that they allow a single item to be associated with multiple organizational nodes.  But that strength is also their weakness -- they don't allow you to rediscover the hierarchy as you navigate it.

 

I totally agree with this. For those who say "tags are better", you should really say "tags are better FOR ME", which is fine. Tags with for some people, folders work for others. By only supporting tags, EN basically supports only one type of user; how is this good???

 

Tags have two basic problems:

-- They are not hierarchical. This makes tag organization for anything non-tricial very difficult.

-- They are global.

 

The global one is significant. I have a Food+Wine notebook for recipes. I also have a consulting notebook for software development. When i search by tag in the consulting notebook, why does EN prompt me for "Breakfast" as a tag? That's just idiotic, but is caused by tags being global.

 

I'm really not impressed by people who say "EN doesn't need folders". I don't use tags in EN, but I don't post her saying "EN doesn't need tags". Different people have different needs/wants.

 

So why don't I just abandon EN and use a different tool (e.g. DEVONthink) that does have folders? Because there is so much else to like about EN. But lack of folders is crippling it for my workflows.

Link to comment

 

I'm a frequent Evernote user, on multiple platforms.  I depend on it, and I desperately want to like it.  But it frustrates me every single time I interact with it.  I write this in the hopes that I'm just missing something.

 

To me, the most transformative benefit that came with storing things digitally was the availability of an arbitrarily deep hierarchy.  No longer did I have to limit myself to a sheaf of notes stapled together inside a manila folder inside a file drawer.  I could create a lush hierarchy that let me quickly navigate to my current area of focus, and brought complete freedom to subdivide that focus, or not, as deep or shallow as I chose.*

 

Here's the key:  When I add a note, I may not remember how I've organized things in this "region" of my brain precisely.  I want to see that organization explicitly, so I can decide where the new information belongs.  If there's already a note on this topic I may want to add to it rather than force myself to look in two places for it.  If there are related ones that are diverging into multiple clusters I may want to split them.  Throwing my new note into an unsorted pile and hoping search can find it later feels like throwing it away to me.

 

Evernote is a big step backward from files stored in filesystem folders in this respect.  I acnowledge all the powerful data entry tools that come with EN -- that's why I use it.  But in terms of organization we're all the way back to manila folders (Stacks) containing sheaves of paper stapled together.

 

Why?  What's so hard about putting notes in a folder hierarchy?  Is there some downside to this that I can't see?

 

-J

 

* I fully recognize that tags are in one sense more general than a folder hierarchy in that they allow a single item to be associated with multiple organizational nodes.  But that strength is also their weakness -- they don't allow you to rediscover the hierarchy as you navigate it.

 

I totally agree with this. For those who say "tags are better", you should really say "tags are better FOR ME", which is fine. Tags with for some people, folders work for others. By only supporting tags, EN basically supports only one type of user; how is this good???

 

Tags have two basic problems:

-- They are not hierarchical. This makes tag organization for anything non-tricial very difficult.

-- They are global.

 

The global one is significant. I have a Food+Wine notebook for recipes. I also have a consulting notebook for software development. When i search by tag in the consulting notebook, why does EN prompt me for "Breakfast" as a tag? That's just idiotic, but is caused by tags being global.

 

I'm really not impressed by people who say "EN doesn't need folders". I don't use tags in EN, but I don't post her saying "EN doesn't need tags". Different people have different needs/wants.

 

So why don't I just abandon EN and use a different tool (e.g. DEVONthink) that does have folders? Because there is so much else to like about EN. But lack of folders is crippling it for my workflows.

 

 

AFAIK, "EN doesn't need folders", you're misquoting/misinterpreting.  First, EN doesn't have "folders".  It has stacks & notebooks. 

 

If you meant to say "I'm really not impressed by people who say "EN doesn't need folders subnotebooks".", I'm pretty sure no one has ever said that so please quote where this has been posted.

 

What has been said (countless times) is EN does not have sub-notebooks & there are two options.  Either adapt to EN as it is now or, if sub-notebooks are a deal breaker, then you will need to find an app that better suits your needs.  The fact that you haven't found a viable alternative speaks volumes. Don't you suppose if it were so easy to make an EN that lived on all the platforms & did all the things people are griping about what EN does not do (IE sub notebooks, better editor, etc) and given that EN now has over 100 million users that someone would have done it by now...???

 

I have yet to find an example (and many have been posted) where tags/notebooks/descriptive titles & keywords won't easily accomplish what people are trying to do with sub-notebooks.  But the bottom line is that EN has chosen not to implement them & may never choose to do so.  It may be simply a decision or (more likely) there's a valid tech reason.  IOW, since EN lives on so many platforms, they may be trying to follow the KISS method. 

 

It's really easy to do Monday morning quarterbacking.

Link to comment

Only downside I've heard is that it's a real coding/software chore.

For me,in terms of organization my brain could wrap itself around better, it'd go a long way if (on Android) either 1. Notebook notes could be sorted by tag (heLLO Evernote that's what tags do visually in a file cabinet) or 2. within Notebooks somewhere there were a list of clickable tags used in that notebook.

Unfortunately Evernote has its own timetable and it has little if anything to do with user feedback.....as I've sent this to them at least twice, as well as posted it here.

Link to comment

"Either adapt to EN as it is now or, if sub-notebooks are a deal breaker, then you will need to find an app that better suits your needs."

No those aren't the only 2 options.

This is a f o r u m , other options are *use Evernote or not use Evernote AND express the fact that you'd need/would like subfolders (regardless of how many times it's been brought up) and hope to generate your own discussion on the matter.

If your tired of the topic as is evident in your post:skip the thread. Subject is new to the OP.

Link to comment

I totally agree with this. For those who say "tags are better", you should really say "tags are better FOR ME", which is fine. Tags with for some people, folders work for others. By only supporting tags, EN basically supports only one type of user; how is this good???

 

Tags have two basic problems:

-- They are not hierarchical. This makes tag organization for anything non-tricial very difficult.

-- They are global.

 

The global one is significant. I have a Food+Wine notebook for recipes. I also have a consulting notebook for software development. When i search by tag in the consulting notebook, why does EN prompt me for "Breakfast" as a tag? That's just idiotic, but is caused by tags being global.

 

I'm really not impressed by people who say "EN doesn't need folders". I don't use tags in EN, but I don't post her saying "EN doesn't need tags". Different people have different needs/wants.

 

So why don't I just abandon EN and use a different tool (e.g. DEVONthink) that does have folders? Because there is so much else to like about EN. But lack of folders is crippling it for my workflows.

I understand but it could also prompt you a "must read" or a "reference site" tag...

If I want dedicated tags I set prefix to them. Here prefixes could be something like fw and sd.

Link to comment
Posted · Hidden by jefito, May 25, 2014 - Not even remotely on topic.
Hidden by jefito, May 25, 2014 - Not even remotely on topic.

drtimhill:

 

As you will have realized by now, this is actually not a forum where Evernote users support each other, nor is it a place where you can suggest features or desired improvements.  After my interchange above, a couple of friendly users messaged me privately to express regret for that, and to let me know that I was wasting my time.  Perhaps someone has reached out to you by now also.  If not, to save others from hitting their head against the same wall, I'll repost the (suitably redacted) text of one of those messages below.  Pretty much says it all.

 

======================================================================
Your comment: "Wow.  Clearly I've inadvertently stumbled onto a religious issue hereabouts.  i wonder why a discussion about the digital equivalent of how best to sort one's sock drawer sparks such passionate denouncement of an alternative approach."

Great comment! But get ready for the onslaught of responses from the "Evangelists". They will whine about how they monitor the forum as unpaid support staff. They will tell you how much time they spend creating responses. I believe a couple of them have a few screws missing "upstairs" when it comes to any honest discussion about Evernote.

"Evangelist" xxxxxxxxxx often resorts to snarky and troll-like comments to try to start a fight. I've reported her several times via the Report option but Evernote has never responded.  She insists on getting in the last word. She seems to enjoy belittling new users. And if she realizes she is losing the discussion, she will add her final comment and then lock down the thread.

Whether Evernote realizes it or not, a corporate online community in today's world is the face of the company. Deputizing folks that are not skilled in customer service or people skills reflects very poorly on the business.

Strange way to run a company.

Regards
xxxxxx
======================================================================

Link to comment

drtimhill:

 

As you will have realized by now, this is actually not a forum where Evernote users support each other, nor is it a place where you can suggest features or desired improvements.  After my interchange above, a couple of friendly users messaged me privately to express regret for that, and to let me know that I was wasting my time.  Perhaps someone has reached out to you by now also.  If not, to save others from hitting their head against the same wall, I'll repost the (suitably redacted) text of one of those messages below.  Pretty much says it all.

 

Edit: PM removed  

FYI, posting a PM publicly is bad form. A PM is just that and carries no weight when mentioned like this. And anyone who feels the need to PM other users like this is simply a whiner themselves & should grow up.

Link to comment

As one of those "Evil moderator Evangelists", I feel I need to step in a little here.  (Gotta own the title, right? :P)
I don't think it is right to post the content of a personal message.It is by definition, personal. I think this is doubly the case with a message such as this, which contains personal attacks and is bound to start an argument.
If the sender wants to send them, then that is their prerogative (though I don't see how scaring away new members like this is in anyway better than what they describe, but that is neither here or there) and it should stay private.

 

 

As always, if you have an issue with a post, feel free to hit the "Report" button. The moderators AND ADMINS (Read Evernote Staff) look at all reports and act on them accordingly. Thank you.

 

 


 

 

Trying to bring the topic back around...
If I remember correctly, Jefito had a pretty good explanation of why the current system could not use multiple levels of notebooks. Down to the way that they are coded, based on his interpretation of the API, I will have to try and find it.

Could it be changed? Possibly. How easy would that be? I have no idea...

Link to comment
  • Level 5*

I know I'm obviously irretrievably prejudiced,  but rereading the initial posts in this thread,  it seems like a reasonably free and frank exchange of views with several people trying to be helpful.  Anyone is free to make comments here (despite the views of your mystery supporter),  but if you do,  you might get responses which agree or disagree with you.  That's how discussion works.  We despised Evangelists aren't employed by Evernote at all,  so tend to hide behind that "freedom of speech" nonsense to voice our own opinions - sometimes pure speculation on what Evernote might or might not choose to do.  

 

At the moment,  because of the nature of the request and the likely adverse implications of trying to fulfil it,  we really don't think Evernote is likely to consider this option at all.  If there's a sudden landslide of support here for a folder hierarchy Evernote might take note and decide to do something about it,  so if you do want a folder hierarchy please do post here.  It'll show the level of demand. (I promise not to bite.)  I'll watch this thread with interest to see who weighs in with support.

Link to comment
  • Level 5*

Folks, post the contents of someone else's PM is inappropriate, for any number of reasons. More importantly, this one isn't even remotely on topic. I've therefore hidden the post (it's already been reported) and removed a quoted copy of the PM from another post. This will be dealt with by an Evernote staffer in due course, probably gbarry (it's a long holiday weekend in the US). That person may decide to reverse what I've done here; fair enough.

 

This topic has veered quite a ways off topic. The usual recourse is to lock the topic. Please, let's not go there. The original topic is worth discussion, and most of the posts here have been topical. For those of you who wish to discuss off topic material, please feel free to use the subforum that's expressly designed for that, elsewhere on the site. And for anyone who feels that any post violates the forums code of conduct (link is at the bottom of this page), go ahead and use the Report button on the post. Evernote does indeed take reported material seriously.

Link to comment

Evernote does indeed take reported material seriously.

Just he more thing.. to clarify, simply because a particular user thinks something is offensive does not mean Evernote agrees. IOW, just because you report a post doesn't mean it will be deleted or the poster raked over the coals & hung up by their thumbs. There will always be those who have an opinion that differs from yours. Generally, all posts are allowed unless they violate the forum code of conduct.

https://discussion.evernote.com/index.php?app=core&module=help&do=01&HID=13

Now, back on topic...

Link to comment
  • Level 5*

jhd,

 

You might want to check out our TuskTools Treeliner offering - it's currently in a closed beta but downloads are being actively distributed now, so if you fill out the request form, you should get a download email within a few days.  It allows you to outline any or all of your notes in a hierarchical manner in a tree; it also lets you classify notes into "Item Types" to help make the structure even richer - so you can easily visualize and filter folders, projects, etc. within a notebook.

Link to comment

Phil, does the Treeliner work on Android? If not, do you have any plans to make it available on Android?

jhd,

You might want to check out our TuskTools Treeliner offering - it's currently in a closed beta but downloads are being actively distributed now, so if you fill out the request form, you should get a download email within a few days. It allows you to outline any or all of your notes in a hierarchical manner in a tree; it also lets you classify notes into "Item Types" to help make the structure even richer - so you can easily visualize and filter folders, projects, etc. within a notebook.

Link to comment
  • Level 5*

This current beta is for Windows.  Yes, versions for iOS and Android are definitely planned.  At present, these are envisioned more as supplements to the Windows version than as full standalone solutions, due to the need for a good amount of screen real estate in order to effectively view and manipulate tree-based outlining.  (Although, this plan could change as more thought is put into the mobile versions...)

Link to comment

 

* I fully recognize that tags are in one sense more general than a folder hierarchy in that they allow a single item to be associated with multiple organizational nodes.  But that strength is also their weakness -- they don't allow you to rediscover the hierarchy as you navigate it.

The problem with hierarchies is you often have to "rediscover" it, if/when you don't remember how you organized it.  IE, did you file your cell phone bill under "phone", "telephone", "cell phone" or "Verizon"?  So you have to go to the bills folder & look for it one of those options.  OTOH, with tags (and Evernote's amazing search engine), you have the luxury of not having to "rediscover" how you filed it.

 

http://discussion.evernote.com/topic/48008-search-by-tag-or-just-search/?p=244648

 

IMO, you're overtagging. IMO, some folks "overtag". Often, tags are not needed if you use descriptive titles, some tags, some notebooks/stacks & keywords. I have over 62,000 notes & have never reached the 250 notebook limit. I'd guess less than half my notes have any tags at all. Probably only about 1/3 of my notes have tags. Those that do have tags normally have only 1-3 tags. IE, all my bills are in Evernote. Although they are in a "bills" notebook, I can quickly & easily find my Cox cable bill from June 2009 by searching ALL notes because the title of that note includes the vendor (Cox) and the date of the bill in YYYYMMDD format. So a simple search of

intitle:cox intitle:200906*

will quickly find the ONE note which is the bill I'm looking for from over 62,000+ notes. And I didn't even use a tag.

You can also net the tags for organizational purposes only. This allows you to collapse the top tier so the nested tags do not show up. Or even the "hide unassigned" tags may be helpful.

 

Except that you have to remember what to search for versus a visual reminder, which would be the case with a Hierarchy.

Link to comment

jhd,

 

You might want to check out our TuskTools Treeliner offering - it's currently in a closed beta but downloads are being actively distributed now, so if you fill out the request form, you should get a download email within a few days.  It allows you to outline any or all of your notes in a hierarchical manner in a tree; it also lets you classify notes into "Item Types" to help make the structure even richer - so you can easily visualize and filter folders, projects, etc. within a notebook.

My request has been submitted, looks great. 

Link to comment

Folder hierarchies are basically an archaic left-over from the Windows Explorer days (or Mac-equivalent). Gmail was the 1st to make that clear with its labels system, which now seems to have been adopted by many services, so obviously users like it.

 

I agree that labels, or in the EN system tags, is the way forward & allows much better & efficient searches than folders. Apart from being able to assign tags to multiple notes, you can create hierarchies with tags too, as pointed out above. You can "nest" tags, for example, by creating "parent tags" with several "child tags" nested below it.

 

The parent tags can act as a cluster name (e.g. "COMMUNICATION") and then have child tags "phone", "email", etc. below it. Notice I assigned capital letters to the parent (which is really just a regular tag but with no notes assigned to it) & the child tags are assigned lower case letters for clarity.

 

Sometimes a tag could theoretically be assigned to more than 1 cluster but in practice that is not possible, so as a rule of thumb assign it to the cluster that best defines the tag's primary objective.

 

The are notes that cannot be assigned to any tag (remember you do not have to assign a tag), and all of those can be grouped in a cluster named "UNASSIGNED" or whatever.

 

Using this system you can can quickly decide which cluster(s) a new note belongs to & therefore which of the tag(s) should be assigned to it. Conversely, when looking for a note you can look at your tag hierarchy & quickly determine which cluster, and therefore which tag(s) it is most likely to belong to.

 

I used it successfully, and then went a step further & did away with conventional tags altogether & am now only using what I call short code title tags & text tags. Actually for the latter I have another name but some might find it offensive.

 

This system also makes notebooks & stacks redundant, you really do not need them.

1 last remark: jhd you are (relatively) new to EN he will have to "experiment" with which system you are most comfortable with. It is unlikely you find the right system when you 1st set out using EN.

Good luck.

Link to comment

With this system you really only need 5 notebooks at most:

* personal synchronised

* shared synchronised (if needed)

* personal local (not synchronised) for confidential data

* default synchronised notebook to receive clippings etc. (if you don't want to use thee personal synchronised for that)

* trash notebook.

Link to comment
  • Level 5*

With this system you really only need 5 notebooks at most:

* personal synchronised

* shared synchronised (if needed)

* personal local (not synchronised) for confidential data

* default synchronised notebook to receive clippings etc. (if you don't want to use thee personal synchronised for that)

* trash notebook.

For those of us who share notebooks, you may need more notebooks. Also for those of us who have a desire to maintain a subset of their notes on a mobile device (using offline notebook), you may need more notebooks. I do both of these, but I still don't need a lot of notebooks.

Link to comment

Jefito, you may be right. I pointed out what should be the bare minimum (which is what I have myself) but for others more may be necessary. Most of the discussion in this thread (excepting of course the personal aggression in some replies) was about organising one's Evernote, and particularly the use of tags. So the focus of my reply is tag organisation.

Link to comment
  • Level 5*

Jefito, you may be right. I pointed out what should be the bare minimum (which is what I have myself) but for others more may be necessary. Most of the discussion in this thread (excepting of course the personal aggression in some replies) was about organising one's Evernote, and particularly the use of tags. So the focus of my reply is tag organisation.

Sure, agreed. Those who have a leaning towards a minimalist approach should create notebooks only when they need to, and that leads to the particular cases where you can't do anything any other way -- basically because of Evernote's rules (or limitations, perhaps). Then do the rest by tags. That's pretty much my approach.

Link to comment
  • Level 5*

There's some good reasons behind a KISS approach - the fewer options available,  the more likely you/ users are to apply them consistently and use them correctly.  "Losing" data in a large database is often more a case of not 'filing' it correctly with the appropriate tags and headers and in the right location.  With fewer options that's less likely to happen.

Link to comment

I don't need subnotebooks but many many do...or think they do.

If I felt as passionate about the need as some do I'd bring it up too,regardless of whether of not it's been addressed before. That's perfect fair game on a forum...to express need and frustration.

If members can't answer those expressed needs without a condescending tone cause they've been through it soooooo manyntimes then let other people do it. I could do it repeatedly and easily...

Evernote has set the stage for this. Of course they don't have to explain anything....but it's just plain dumb not to....and ups the ante of the notion that their course of action has little to do with user feedback...which it doesn't.

You can keep saying they don't need to directly explain or answer user feature requests and the obvious response to that is...We'll, duh, of course they don't, that's obvious.

All it does is keep reinforcing Evernotes poor (nonexistant) customer responsiveness.

But I can buffer it in the meantime.

Send every need/demand for sub/nested folders to me . I'll shown you how it's done. I'll take care of it. It's a lot of what I do.

 

What always amazes me about these discussions (and there have been many,  many more over the years) is

  • Someone wants a specific feature added
  • Evernote read the posts here, so the suggestion will be noted; but they don't typically comment on if,  when or whether said feature will (ever) be added
  • Everyone chimes in to point out work-arounds - though sometimes it can sound a bit overly defensive
  • Original poster strongly objects to not getting own way immediately,  wants to know why* feature can't be implemented
*Answer: because they own the product,  don't have to defend or comment on every decision they take,  and (sensibly) do things that fit in with their long term plans,  without getting distracted by requests for 'easy to implement' changes.

 

I've got my own little list of things that I'd like to see added,  plus a much longer list of things that need to be improved.  Since I do have a day job though I'm going to continue using my workarounds - and quite a lot of other software - while Evernote do their development work.  I'm on the Beta track for both desktop and PC because that's where the newest features are - but some of my workarounds are safety belts in case things go sadly wrong.  

 

(Don't try this at home kids unless you're a Data Professional like what I am.)

Link to comment
  • Level 5*

Re the notion that Evernote hasn't discussed the topic of subnotebooks: please see this post. Sure, it's five years old, but it's from the Evernote CTO, and as best I can tell, the stance hasn't changed, and they generally don't like repeating themselves. Most of the discussion here has been along the lines of "why I prefer tags" or "why you might not need notebooks as much as you thought", and not that subnotebooks (or sub-whatevers) are inherently bad or that it's wrong to express a desire / need / requirement for them.

 

On more leftover from earlier in the topic:


If I remember correctly, Jefito had a pretty good explanation of why the current system could not use multiple levels of notebooks. Down to the way that they are coded, based on his interpretation of the API, I will have to try and find it.
Could it be changed? Possibly. How easy would that be? I have no idea...
I have thought about this a bit. I had looked at the Evernote architecture and discovered that a stack has no actual existence except as a name in a notebook; there is no Stack object in the same way that there is a Notebook object and a Note object (see http://dev.evernote.com/doc/reference/Types.html#Struct_Notebook). My first instinct was to think that that pretty much ruled out full hierarchy, since there is no place to put a stack's parent. But thinking about it some more, there's no intrinsic reason that the name in the current notebook's stack field couldn't just refer to another notebook name.
 
Sure, there's some bookkeeping involved to ensure no cycles occur, and probably more to do with sharing permissions, searching (you can search the notebooks in a stack all together; you'd probably want something similar in the notebooks contained by a parent notebook), you'd need to account for existing use of stacks (currently you can have a stack with the same name as an existing notebook; how to disambiguate between stack and notebooks names) and so on, but there doesn't seem to be any architectural reason why it couldn't be done. Note that this isn't an argument either for or against -- I just wanted to put Scott's statement in some context. And nothing is implied about the ease or difficulty with which this could be accomplished.
Link to comment
  • Level 5*

I used it successfully, and then went a step further & did away with conventional tags altogether & am now only using what I call short code title tags & text tags. Actually for the latter I have another name but some might find it offensive.

So curious... ;)

 

1 last remark: jhd you are (relatively) new to EN he will have to "experiment" with which system you are most comfortable with. It is unlikely you find the right system when you 1st set out using EN.

Good luck.

 

This is really the most important thing one must accept when setting up an in-depth, thorough, and effective data management system. It's gonna take trial and error. It's gonna take time. It's gonna frustrate at first. But eventually, if you keep trying, you will hone in on a method of organization that works for you, and it'll be an epiphany (perhaps a slight exaggeration, but it's close enough).

 

I'm a tagger. I've got an Inbox notebook, a Tickler notebook (for stuff I'm only keeping temporarily, like tracking numbers), and a Stack with my Shared notebooks. I tried the hierarchical tagging method, with tags like:

  • design
  • design.web
  • design.inspiration

But then I ran into issues where I would clip an article that I'd want to tag as blog.inspiration -- so I had two inspiration-related tags. Gross. At least, I feel it is :D So, I dumped the hierarchical tags and now just use basic keywords like inspiration, blog, design, money, receipt, etc. That way I can intermingle and sort the tags as I need. Of course, this is just my user case, but I post it in the hopes of showing those missing subnotebooks that there is another way. Gmail, Evernote, and public libraries -- three very large databases of information -- all use tags to sort items, just to name a few. Seems like maybe the think tanks realized that tags truly are a better way.

Link to comment
  • Level 5*

But then I ran into issues where I would clip an article that I'd want to tag as blog.inspiration -- so I had two inspiration-related tags. Gross. At least, I feel it is :D So, I dumped the hierarchical tags and now just use basic keywords like inspiration, blog, design, money, receipt, etc. That way I can intermingle and sort the tags as I need. Of course, this is just my user case, but I post it in the hopes of showing those missing subnotebooks that there is another way. Gmail, Evernote, and public libraries -- three very large databases of information -- all use tags to sort items, just to name a few. Seems like maybe the think tanks realized that tags truly are a better way.

Re "heirarchical tags": that's pretty much the way that I see it. The method definitely has one big benefit going for it, as I see it, namely that you can do some nice stuff with search using wildcards (e.g. "tag:a.b.*") which can get around Evernote's lack of general AND/OR filtering, but conceptually it doesn't sit right with me.
Link to comment

 

 

...

I totally agree with this. For those who say "tags are better", you should really say "tags are better FOR ME", which is fine. Tags with for some people, folders work for others. By only supporting tags, EN basically supports only one type of user; how is this good???

 

Tags have two basic problems:

-- They are not hierarchical. This makes tag organization for anything non-tricial very difficult.

-- They are global.

 

The global one is significant. I have a Food+Wine notebook for recipes. I also have a consulting notebook for software development. When i search by tag in the consulting notebook, why does EN prompt me for "Breakfast" as a tag? That's just idiotic, but is caused by tags being global.

 

I'm really not impressed by people who say "EN doesn't need folders". I don't use tags in EN, but I don't post her saying "EN doesn't need tags". Different people have different needs/wants.

 

So why don't I just abandon EN and use a different tool (e.g. DEVONthink) that does have folders? Because there is so much else to like about EN. But lack of folders is crippling it for my workflows.

 

 

AFAIK, "EN doesn't need folders", you're misquoting/misinterpreting.  First, EN doesn't have "folders".  It has stacks & notebooks. 

 

If you meant to say "I'm really not impressed by people who say "EN doesn't need folders subnotebooks".", I'm pretty sure no one has ever said that so please quote where this has been posted.

 

What has been said (countless times) is EN does not have sub-notebooks & there are two options.  Either adapt to EN as it is now or, if sub-notebooks are a deal breaker, then you will need to find an app that better suits your needs.  The fact that you haven't found a viable alternative speaks volumes. Don't you suppose if it were so easy to make an EN that lived on all the platforms & did all the things people are griping about what EN does not do (IE sub notebooks, better editor, etc) and given that EN now has over 100 million users that someone would have done it by now...???

 

I have yet to find an example (and many have been posted) where tags/notebooks/descriptive titles & keywords won't easily accomplish what people are trying to do with sub-notebooks.  But the bottom line is that EN has chosen not to implement them & may never choose to do so.  It may be simply a decision or (more likely) there's a valid tech reason.  IOW, since EN lives on so many platforms, they may be trying to follow the KISS method. 

 

It's really easy to do Monday morning quarterbacking.

 

 

I'll ignore the comment about "Monday morning quarterbacking" as ad hominem. In my post I said two basic things: I have found the lack of (hierarchical) folders to be a significant shortcoming for the way I prefer to work, and that I continue to use EN despite this as I like many of its other features. Should I not have posted this? If not, why not?

 

As for your comments on notebooks/subnotebooks etc, I find this argument strange. Notebooks and folders are conceptually both containers, which are quite different from tags which are attributes of a note. Both have a role in any good structural hierarchy (and both can be abused). But when you only have one or the other, then you are forced into contortions to make up for the lack of the other; hence the discussions here about magic tag names to hide a hierarchy in the name, or conversely (in folder only systems) the use of unwieldy folder hierarchies to try to track metadata that really belongs in tags (e.g. an "Invoice" folder rather than an "Invoice" tag).

 

I would characterize EN as having rich tags and weak containers. This fits with some workflows, but not others, including mine, which was why I posted a "wish" that EN had a more flexible container model. And let's be real here, EN has added stuff to the container side of things; multiple notebooks were added, then stacks, so someone in EN must see a need for stuff beyond tags.

 

If you want a good example of a product that integrates folders and tags, look at DEVONthink, which supports both well, but has vastly inferior sync and device support compared to EN. My ideal product is one that has the organizational flexibility of DEVONthink and the sync/device support of EN.

 

--Tim

Link to comment
  • Level 5*

If you want a good example of a product that integrates folders and tags, look at DEVONthink, which supports both well, but has vastly inferior sync and device support compared to EN. My ideal product is one that has the organizational flexibility of DEVONthink and the sync/device support of EN.

 

--Tim

Devonthink has a terrible interface, an (apparently) terrible iOS app, terrible support for languages other than English, and no cross-platform capabilities. It is in need of some major work, in my opinion, and I find it to be (at the moment) a difficult app to fit into my workflow.

But, it gets so much right, as you mentioned. In terms of organization, you can do just about anything with it, and the ability to index files that aren't even in the app is brilliant. I also really like their approach to security, and the recent upgrades they made to improve syncing (through Dropbox with other Macs) was a welcome improvement. If they could at least get the iOS app on its feet it would be a huge help.

All of this is to say that some seemingly simple and obvious things (this is actually directed towards the title of the thread) are not so simple and obvious in practice. Devonthink has known for years now that its iOS app is awful and desperately needs work, but here we are in 2014 and it still isn't updated. Even the behemoths on the block, Microsoft and Google, can't even get their apps right (Google on iOS is anemic and Outlook is almost useless on the Mac or iOS without some support for PDFs). My guess is that there are all kinds of obstacles and tradeoffs that users don't see in the making of an app, so when we talk about adding notebook hierarchies to Evernote, I have to wonder what impact that would have on database performance, interfaces, etc. Nothing is ever as simple as it sounds.

Anyhow, a wish list is great! I have got all kinds of suggestions and hopes for Evernote's service. I don't necessarily expect they will be implemented, because I am not the one doing the heavy lifting, but I still think it is good to make the case for changes.

Link to comment

What always amazes me about these discussions (and there have been many,  many more over the years) is

  • Someone wants a specific feature added
  • Evernote read the posts here, so the suggestion will be noted; but they don't typically comment on if,  when or whether said feature will (ever) be added
  • Everyone chimes in to point out work-arounds - though sometimes it can sound a bit overly defensive
  • Original poster strongly objects to not getting own way immediately,  wants to know why* feature can't be implemented

*Answer: because they own the product,  don't have to defend or comment on every decision they take,  and (sensibly) do things that fit in with their long term plans,  without getting distracted by requests for 'easy to implement' changes.

 

I've got my own little list of things that I'd like to see added,  plus a much longer list of things that need to be improved.  Since I do have a day job though I'm going to continue using my workarounds - and quite a lot of other software - while Evernote do their development work.  I'm on the Beta track for both desktop and PC because that's where the newest features are - but some of my workarounds are safety belts in case things go sadly wrong.  

 

(Don't try this at home kids unless you're a Data Professional like what I am.)

 

Yes, tags are great. I love tags. But it's not an either-or. People are not saying "tags are bad, let's have a deeper hierarchy instead". They are saying that the choice should be a fairly vital part of an efficient note and document organising application. Sometimes tags are absolutely the right tool to use --- and sometimes a tree structure is much neater and assists maintenance. Whenever it's pointed out here, yes, we get helpful workarounds (thank you everyone) -- usually forcing tags to replicate something akin to a tree structure -- but it's nearly always a second best. We don't want to constantly have to search for things. Frankly, I don't always know how best to search for something and will find it quicker if I can visually trace it in a hierarchical grouping. We do this all the time in a standard file hierarchy, when a search isn't always simple and neat. Heck, sometimes it's just useful to be able to survey what we have in our systems by doing a general visual overview. We can spot gaps and anomalies in a split second that way.

 

And anyway, the fact is that Evernote already has the concept of notebook stacks, which is itself a hierarchical grouping structure -- albeit a shallow one. So the complaint is not that they should introduce a conceptually new feature, but should complete something they have started.

 

By pointing out that the discussion has come up many times before illustrates how much pain it causes people. It really does reduce Evernote from being pretty much the perfect tool to being a really good tool with one very annoying flaw. i don't want to be told that if I don't like it I should go and use something else. That's a little dismissive. EN is so nearly everything I need, especially with its brilliant paperless / scanning features. I won't throw the baby out with the bathwater.

 
As for: "*Answer: because they own the product,  don't have to defend or comment on every decision they take,  and (sensibly) do things that fit in with their long term plans,  without getting distracted by requests for 'easy to implement' changes",
 
...that strikes me as arrogant and rude of EN. There is never an excuse to ignore your customers, especially when the same question is constantly raised. I use/have used many software forums and it's very common for the company to respond to a request that is constantly made. Even if it's to explain why it's not present (e.g. too difficult to implement or contrary to some underlying guiding principle), or that it's on some roadmap, would be respectful to their customers and be sensible customer-service. Perhaps best of all, it might stop some of these interminable correspondences.
Link to comment
  • Level 5*

...that strikes me as arrogant and rude of EN. There is never an excuse to ignore your customers, especially when the same question is constantly raised. I use/have used many software forums and it's very common for the company to respond to a request that is constantly made. Even if it's to explain why it's not present (e.g. too difficult to implement or contrary to some underlying guiding principle), or that it's on some roadmap, would be respectful to their customers and be sensible customer-service. Perhaps best of all, it might stop some of these interminable correspondences.

Hi. I don't think they have been ignoring their customers, though users have complained about a lack of response from Evernote employees since its early days. A search on the forums will uncover this.

https://discussion.evernote.com/topic/28871-feature-request-nested-stacks-multiple-notebook-levels/?p=17168

That was back when they only had a few hundred thousand people. Now they have over 100 million. I don't think it is realistic to expect regular responses every time the topic appears on the user forums. Fortunately, a search of the forums will unearth lots of discussions about this including employee responses.

https://discussion.evernote.com/topic/28871-feature-request-nested-stacks-multiple-notebook-levels/page-23

In fact, I am guessing that even Evernote employees would like to see sub-notebooks, as evidenced by this comment from a user who later became an employee. I think you will find other comments by employees thanking users for their input, even if the request ultimately isn't implemented -- I think this is quite polite and to be expected with 100 million + cooks in the kitchen.

https://discussion.evernote.com/topic/28871-feature-request-nested-stacks-multiple-notebook-levels/?p=179140

Even though I pretty much live in one notebook, I'd like to see deeper hierarchies as well to increase the popularity of the service, but I am guessing that there are technical reasons why they went with stacks instead of sub-notebooks, and there are technical reasons why they resist expanding the hierarchical levels.

https://discussion.evernote.com/topic/28871-feature-request-nested-stacks-multiple-notebook-levels/?p=200376https://discussion.evernote.com/topic/28871-feature-request-nested-stacks-multiple-notebook-levels/?p=200376

Anyhow, hopefully the discussion in the thread I linked will be of assistance. I doubt it will stop the "interminable correspondence," and I don't think the discussion should stop. Bring it on! It's good for Evernote to hear customer voices.

Link to comment

Nested tags allow visual hierarchy. I use some tags this way and others as we all consider as tags are.

My thought is it's overall semantic concerns.

 

Thanks -- yes, I'm aware of nested tags, and they are better than nothing, but not ideal.

Link to comment

 

 
Anyhow, hopefully the discussion in the thread I linked will be of assistance. I doubt it will stop the "interminable correspondence," and I don't think the discussion should stop. Bring it on! It's good for Evernote to hear customer voices.

 

 

Thank you, there are some useful looking links there. I'll check out some earlier responses.

 

As for a current solution, I've realised that my best best is to look at 3rd party tools (a couple mentioned in this thread look promising).

Link to comment

After mulling this for a bit I'm pretty sure that EN built their infrastructure on an object-store rather than a more conventional filesystem.  The hierarchical organization many of us are used to is an artifact of the tree structure of many filesystems, and the tagged, flat structure of gmail and evernote are similarly a pretty direct result of their object-store roots.

 

The implementation advantages of objects for a scale-out storage infrastructure are many - a flat namespace means you can scale to very large size, partitioning across server infrastructure is trivial (via hashing), and changes to one object tend not to create ripple effects or conflicts with others.  Object stores deal with the problem of simultaneous editing of objects by two users by simply not allowing edit-in-place at all.

 

So, when we ask for "hierarchy", what EverNote's engineers may hear instead is "distributed filesystem" -  something which is indeed a very difficult thing to build.  And aside from building it, converting EN's user data into such a structure would be prohibitively disruptive.  But the association of hierarchy and filesystem in this special case seems to me to be a false one:  You could build a hierarcical set of relationships between objects using tag concatenation or something similar, without dragging along all the other cruft that filesystems usually imply.  The fact that some (relatively small) 3rd parties have done that seems to show that it's feasible, at least at small scale.

 

One way to think about hierarchy is that by placing a new object in the tree, you've implicitly add a (potentially large) number of tags to it.  It occurs to me that since in EN's case the hierarchy is *not* tied to the underlying storage architecture, one could conceivably have the best of both worlds -- a given note could be aliased into multiple hierarchies, just as a given note can have multiple tags.

 

-J

Link to comment

 

If you want a good example of a product that integrates folders and tags, look at DEVONthink, which supports both well, but has vastly inferior sync and device support compared to EN. My ideal product is one that has the organizational flexibility of DEVONthink and the sync/device support of EN.

 

--Tim

Devonthink has a terrible interface, an (apparently) terrible iOS app, terrible support for languages other than English, and no cross-platform capabilities. It is in need of some major work, in my opinion, and I find it to be (at the moment) a difficult app to fit into my workflow.

But, it gets so much right, as you mentioned. In terms of organization, you can do just about anything with it, and the ability to index files that aren't even in the app is brilliant. I also really like their approach to security, and the recent upgrades they made to improve syncing (through Dropbox with other Macs) was a welcome improvement. If they could at least get the iOS app on its feet it would be a huge help.

All of this is to say that some seemingly simple and obvious things (this is actually directed towards the title of the thread) are not so simple and obvious in practice. Devonthink has known for years now that its iOS app is awful and desperately needs work, but here we are in 2014 and it still isn't updated. Even the behemoths on the block, Microsoft and Google, can't even get their apps right (Google on iOS is anemic and Outlook is almost useless on the Mac or iOS without some support for PDFs). My guess is that there are all kinds of obstacles and tradeoffs that users don't see in the making of an app, so when we talk about adding notebook hierarchies to Evernote, I have to wonder what impact that would have on database performance, interfaces, etc. Nothing is ever as simple as it sounds.

Anyhow, a wish list is great! I have got all kinds of suggestions and hopes for Evernote's service. I don't necessarily expect they will be implemented, because I am not the one doing the heavy lifting, but I still think it is good to make the case for changes.

 

 

I completely agree, and as a 30-year veteran software developer/architect I'm very aware that every new feature has significant hidden costs. And of course the Evernote team will have to weigh every new feature using a cost/benefit analysis. Part of that analysis, of course, comes down to "how people want this feature", which was really what prompted my post .. to raise awareness that there are some people out there who would like better containers.

 

And yes, DEVONthink has really serious shortcomings in it's overly-complex UI and abysmal iOS app. But it shines in other places. In fact it's interesting that DEVONthink and Evernote are almost complementary; where one is weak the other is strong and vice versa.

 

--Tim

Link to comment

After mulling this for a bit I'm pretty sure that EN built their infrastructure on an object-store rather than a more conventional filesystem.  The hierarchical organization many of us are used to is an artifact of the tree structure of many filesystems, and the tagged, flat structure of gmail and evernote are similarly a pretty direct result of their object-store roots.

 

The implementation advantages of objects for a scale-out storage infrastructure are many - a flat namespace means you can scale to very large size, partitioning across server infrastructure is trivial (via hashing), and changes to one object tend not to create ripple effects or conflicts with others.  Object stores deal with the problem of simultaneous editing of objects by two users by simply not allowing edit-in-place at all.

 

So, when we ask for "hierarchy", what EverNote's engineers may hear instead is "distributed filesystem" -  something which is indeed a very difficult thing to build.  And aside from building it, converting EN's user data into such a structure would be prohibitively disruptive.  But the association of hierarchy and filesystem in this special case seems to me to be a false one:  You could build a hierarcical set of relationships between objects using tag concatenation or something similar, without dragging along all the other cruft that filesystems usually imply.  The fact that some (relatively small) 3rd parties have done that seems to show that it's feasible, at least at small scale.

 

One way to think about hierarchy is that by placing a new object in the tree, you've implicitly add a (potentially large) number of tags to it.  It occurs to me that since in EN's case the hierarchy is *not* tied to the underlying storage architecture, one could conceivably have the best of both worlds -- a given note could be aliased into multiple hierarchies, just as a given note can have multiple tags.

 

-J

 

It's actually not quite like that, though you are correct that a hierarchy presents issues when it comes to synchronization. In fact, modern file systems are not as hierarchical as they might appear, and are much closer to databases in design. For example, on NTFS, HFS+ etc, files are really stored in a flat table. The folder hierarchy is distinct and simply contains references to the files within that table. (HFS+ uses this, for example, to maintain file links and aliases even when the file is moved around the hierarchy.)

 

Most of the issues around a hierarchy involve identity and tree management during sync. Case in point: [A] You have a folder "Foo" with a dozen notes in it. You rename it "Bar", then back to "Foo", so you get back where you started. When you sync, EN doesn't really have to do anything.  You start with the same folder "Foo", create a new folder "Bar" and move all the notes in "Foo" to "Bar". Then you delete the now empty "Foo" and rename "Bar" to "Foo". Now to a user, this looks just like [A], since you still have all your notes in the same folder "Foo". Trouble is, depending on how the application does its book-keeping, the new "Foo" folder may be identified as a different folder when you do a sync, and it's tricky to get this kind of corner case right.

 

I never said folders were simple; I just said they were useful :)

 

--TIm

Link to comment

 

After mulling this for a bit I'm pretty sure that EN built their infrastructure on an object-store rather than a more conventional filesystem.  The hierarchical organization many of us are used to is an artifact of the tree structure of many filesystems, and the tagged, flat structure of gmail and evernote are similarly a pretty direct result of their object-store roots.

 

The implementation advantages of objects for a scale-out storage infrastructure are many - a flat namespace means you can scale to very large size, partitioning across server infrastructure is trivial (via hashing), and changes to one object tend not to create ripple effects or conflicts with others.  Object stores deal with the problem of simultaneous editing of objects by two users by simply not allowing edit-in-place at all.

 

So, when we ask for "hierarchy", what EverNote's engineers may hear instead is "distributed filesystem" -  something which is indeed a very difficult thing to build.  And aside from building it, converting EN's user data into such a structure would be prohibitively disruptive.  But the association of hierarchy and filesystem in this special case seems to me to be a false one:  You could build a hierarcical set of relationships between objects using tag concatenation or something similar, without dragging along all the other cruft that filesystems usually imply.  The fact that some (relatively small) 3rd parties have done that seems to show that it's feasible, at least at small scale.

 

One way to think about hierarchy is that by placing a new object in the tree, you've implicitly add a (potentially large) number of tags to it.  It occurs to me that since in EN's case the hierarchy is *not* tied to the underlying storage architecture, one could conceivably have the best of both worlds -- a given note could be aliased into multiple hierarchies, just as a given note can have multiple tags.

 

-J

 

It's actually not quite like that, though you are correct that a hierarchy presents issues when it comes to synchronization. In fact, modern file systems are not as hierarchical as they might appear, and are much closer to databases in design. For example, on NTFS, HFS+ etc, files are really stored in a flat table. The folder hierarchy is distinct and simply contains references to the files within that table. (HFS+ uses this, for example, to maintain file links and aliases even when the file is moved around the hierarchy.)

 

Most of the issues around a hierarchy involve identity and tree management during sync. Case in point: [A] You have a folder "Foo" with a dozen notes in it. You rename it "Bar", then back to "Foo", so you get back where you started. When you sync, EN doesn't really have to do anything.  You start with the same folder "Foo", create a new folder "Bar" and move all the notes in "Foo" to "Bar". Then you delete the now empty "Foo" and rename "Bar" to "Foo". Now to a user, this looks just like [A], since you still have all your notes in the same folder "Foo". Trouble is, depending on how the application does its book-keeping, the new "Foo" folder may be identified as a different folder when you do a sync, and it's tricky to get this kind of corner case right.

 

I never said folders were simple; I just said they were useful :)

 

--TIm

 

Agreed.  EN has said in fact that they use a conventional MySQL database to contain user-level metadata, with pointers to the actual notes (objects) sharded on large cluster.  They claim the benefits include getting the ACID characteristics of a RDBMS for doing operations on notes or notebooks, and the scalability of noSQL at the same time, but I think that would not be the case if the hierarchy were a general one.

 

-J

Link to comment
  • Level 5*

For those interested, the basic Evernote model is described here: http://dev.evernote.com/doc/articles/data_structure.php. It's a little out of date, as it doesn't describe how stacks fit in, but you can figure that out via the API docs: http://dev.evernote.com/doc/reference/ (look for the Notebook struct). The main page for the developer doc is here: http://dev.evernote.com/doc/. They have a Tech Blog here: http://blog.evernote.com/tech/; some interesting articles on the nature of their server architecture and technologies used.

 

A note on containers: I understand the point, I know the difference between a container and an attribute (e.g. tag), and internally I cringe when I hear someone talk about a note "being in a tag". But this abstraction is a little leaky: there's not a heck of a lot of difference there, except for the rule that a note has exactly one notebook but can have an arbitrary number of tags. The system will enforce the former but doesn't really care about the latter. Moreover, the actual under-the-hood implementation need not mirror the outer abstraction. An Evernote Stack is one example: there is no Stack object in the Evernote architecture, even though externally, users interact with stacks as if there were. In any case, as far as users go, a notebook and a tag are both just named handles by which you pick up a group of notes. That feels a lot like a container to me, in either case.

 

It is possible to build hierarchies that "contain" notes, similar to commonplace folder hierarchies. You can build them as tightly (each note resides in exactly one tag hierarchy) or as loosely (notes can belong to multiple hierarchies) as you want (the thing that you can't do is use the same tag name in two different parts of the hierarchies; that may be a problem for some use cases). Now -- and I've been saying this for about as long as I've been coming to the forums -- it would be better if the Evernote clients better supported using tags as hierarchical objects. I've always wanted support for exploiting the tag hierarchies in the search language. It would obviate the need for the one of the tag naming schemes around, where the hierarchy is encoded in the name, so that you could take advantage of wildcards in the search language.

Link to comment

...

 

A note on containers: I understand the point, I know the difference between a container and an attribute (e.g. tag), and internally I cringe when I hear someone talk about a note "being in a tag". But this abstraction is a little leaky: there's not a heck of a lot of difference there, except for the rule that a note has exactly one notebook but can have an arbitrary number of tags. The system will enforce the former but doesn't really care about the latter. Moreover, the actual under-the-hood implementation need not mirror the outer abstraction. An Evernote Stack is one example: there is no Stack object in the Evernote architecture, even though externally, users interact with stacks as if there were. In any case, as far as users go, a notebook and a tag are both just named handles by which you pick up a group of notes. That feels a lot like a container to me, in either case.

 

It is possible to build hierarchies that "contain" notes, similar to commonplace folder hierarchies. You can build them as tightly (each note resides in exactly one tag hierarchy) or as loosely (notes can belong to multiple hierarchies) as you want (the thing that you can't do is use the same tag name in two different parts of the hierarchies; that may be a problem for some use cases). Now -- and I've been saying this for about as long as I've been coming to the forums -- it would be better if the Evernote clients better supported using tags as hierarchical objects. I've always wanted support for exploiting the tag hierarchies in the search language. It would obviate the need for the one of the tag naming schemes around, where the hierarchy is encoded in the name, so that you could take advantage of wildcards in the search language.

 

Agreed, and having designed and implemented many hierarchy systems over the years it can be shown that any tag system can be transformed into a hierarchy system and vice versa. I think for me it boils down not to internals but to the manner in which containers and tags are presented to the user as an organizational metaphor, and one of those presentations is the concept of focus (or context, if you like).

 

Most container models (the Finder or Explorer, Mail systems etc) that use folders use them as a metaphor for focus. When you choose a folder from a list, you see the contents of that folder. This focus is modal (persists until a new container is chosen), and provides an implicit scope for operations (Select All, for example, just selects all the items in that container). This model is useful as it allows users to focus only on the project/topic/area of interest and ignore other distractions. This is, after all, the best way to use Notebooks in EN and it is even recognized in the UI when the Search box auto-fills the current notebook as part of the search term. This is really the most useful aspect of folders or containers: clearly delineated persistent scope. Thus containers act as a "where" term in a search (metaphorically speaking).

 

Tags of course excel at searching and helping categorize data, in a way that containers cannot (or should not). Firstly, users understand the metaphor of having multiple tags on an item whereas having an item in multiple containers can be confusing (even if its fine for techie types like me). Tags are easily understood as a classification, or "what type" metaphor.

 

This is why I think tags and folders are complementary, rather than conflicting, features. When you you only have one, you can kinda-sorta bend the other to work as a poor-mans version, but it's never as good. As you note, you end up with funny looking tag names, or folders that should really be tags. Yuck.

 

One final follow-up on your point about tag hierarchies. I do find the flat model of tags limiting. First, I don't like that you can organize tags into a hierachry in the UI but this has no meaning when it comes to search terms (and is confusing to the user imho). Second, while some tags are clearly global in scope ("Note", "Meeting" etc), others are not (I have a "Breakfast" tag for me recipe notebook, why does it show up as a possible tag in my Software Development notebook?). I would love EN to have global tags and per-notebook tags to help with this.

 

As I've said, I really like many EN features, but I wish I had more ways to help manage the vast amount of data I'm accumulating there :)

 

--TIm

Link to comment

Archived

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

×
×
  • Create New...