I must mirror the comments others have made here: hierarchical tags would be very useful. Tag trees, groups, or categories would open up a number of exciting possibilities both for organization of research and accessibility (post-organization).
I also use collections and subcollections (with recursive option enabled) extensively. Collections can do a lot, but it's not a substitute for hierarchical tags. I recognize that this may represent a significant--or even herculean--effort, but I also see the potential for huge benefits to many Zotero users.
I too would very much benefit from a hierarchical system of tags. I sort a lot of references by biological nomenclature, which is hierarchical (i.e. order, family, genus, species). It would be nice not to have to enter all of these keywords each time I want to "file" an items. If I could simply add the species name that would make things a lot easier for me.
I use a proprietary program to manage my images called IMatch. They are in the beta stage for their new release IMatch 5, which will have hierarchical keywords as well as categories.
Imatch doesn't do a lot of what zotero does but I believe it is far better in terms of how documents can be categorized and then sorted and found. Perhaps zotero could pick up on some of the strengths of that program?
I'm currently exploring whether to expend the effort to tag articles in Zotero in a detailed manner. While doing this, I came across this old thread and I'm hoping to get some feedback.
I'm having trouble seeing why hierarchical tags are strongly preferred by some people compared to a flat tagging system, assuming that in the flat system you can combine multiple tags to see their intersections.
Possible hierarchical tags benefits: 1. It's easier to get an overview of top-level categories when you only see other tags a parallel level, rather than seeing hundreds or thousands of tabs all at once. 2. It's easier to enter tags because if you enter a child tag directly, all parent tags are automatically added (assuming that hierarchical tags are implemented in this way)
Are there any other benefits? Are these two benefits larger than I currently realize?
Auto-complete helps with 2, at least, and I can imagine a drawback being that I don't know all the parent tags that would be added when I a child tag.
For me, both are important. One key difference between tags and collections (which are already hierarchical) is that tags can be searched by typing in their names, unlike collections. It would be great to be able to search something hierarchical (be it tags or collections) by typing their names.
Can you elaborate on why it's important to be able to look at sub- group of parent tags? I can't see what research task that would actually help me with.
I agree that searching for tags by typing is useful, but why is the hierarchy actually useful to you?
Sure. Hierarchy helps when I'm adding the tags, because I can find the tags easily, starting with the top-level tag. Without hierarchy, I have to come up with the names of the tags among many.
For me, a good alternative to making the tags hierarchical would be to enable searching collections by typing their names. That might be even preferable - to have both hierarchical and flat organizations. If you make the tags hierarchical, you may need to add an option to view it in the flat format, for users who are accustomed to it.
It's now been 12 years and 5 months since the first person proposed this idea!!! Still, I would love, as many other, if that function could be released. That's why I hereby add my voice to their's.
I study in the field of public transportation, doing a masters' thesis. In the last months, I've been searching about student transit fares, and I know, many, many, MANY, tags emerged from that search. And I kind of lost control. It would be best if I could, just by attributing one tag to an item, make many others follow automatically. That would tidy a lot my library.
For example, if I were to write: "University fare program (U-Pass)", it would be best if it could trigger all the hierarchy of tags above it "Public transportation", "Transportation", "Transit pricing", "Transport Economics", "Transportation planning", "College students"...
Let's cross fingers it happens! Meanwhile, does one of you know a way around this problem? (Perhaps a slightly different solution exists now.)
I'm missing this feature, too. For a long time now.
Meanwhile I use structured tags instead of hierarchical tags, and use a special character ">" for structuring.
For example to >MED>therapy>lithotripsy or >MED>therapy>endoscopy. So you find all medical therapy sub-tags used in your library in one place under >MED>therapy and not distributed over all other tags.
It is not perfect, but it works for a workaround (for me). To display all items with >MED>therapy as parent tag you have to use the "full-text search". It is not able to do this in the tag-cloud (as it should be for real hierarchical tags). On other problem I have, that these structured tags could not fullly displayed in the tag-cloud, because they are simply to long.
Thanks for your answer! That's a clever workaround that you've got, even though it still has some technical issues with it. It is better than what I have.
Although, I think I might have a hard time to apply it to my system. It's because I am used to create exhaustive tags, by inserting as much synonyms as I knew into the same tag. It's to be sure I don't miss anything when I try to recover articles about a given topic. For example, I've created "Transport collectif (transport en commun, transport public, public transportation, public transport, transit, mass transit, public transit)" (French and English words) for identifying articles that are related to public transit. That way, if for some reason I forgot one the synonyms and I search for "mass transit", I will recover the same articles as if I choose to go with "public transit".
Unfortunately, if I combine my method to your method, I can end up with excessively long tags... Hence, I am wondering, how do you personally work around the fact that some words have multiple synonyms?
What you are requesting would be better than splendid but it is profoundly complex to implement -- especially with multilingual synonyms, hierarchical taxonomies, etc.
For further information and enlightenment Google these concepts
It would theoretically be possible for someone to write a Zotero plug-in that would allow you to incorporate an open-access thesaurus program's capabilities with an advanced search of the Zotero database. This would be a tremendous amount of effort -- constructing a personally useful thesaurus, linking the thesaurus to a query system, and linking the custom query system to Zotero.
There is an open-access, free multilingual thesaurus construction program, TemaTres, that is absolutely wonderful (https://www.vocabularyserver.com/) but that is only one very small step toward what you would like to have in your Zotero keyword/tag capabilities.
My safety literature database uses parts of TemaTres for its synonym and hierarchy capabilities but this requires a substantial continuous amount of work.
There may already be term taxonomies that already exist that may help you. The U.S. Transportation Research Board in partnership with European and Australasian transport research agencies have constructed the Transportation Research Thesaurus, (http://trt.trb.org/trt.asp?), that covers much more than public transportation but would be useful to give you ideas that have been demonstrated to help with classifying and indexing transportation-related literature.
one way to find things in your database with similar tags is surely to incoorporate a thesaurus into zotero.
But this is not the way for my "workaround" regarding hierarchical tags. You asked me, how I personally work with synonymes or similar tags.
I use tags to enhance every item with additonal information relevant for ME. I ask me, how do I like to find my stuff in my database. Then I add this information on each item (as tags) to be able to find my things.
Practically: Therefore I have designed a system of structured tags as a "hierarchical tree", which I use like folders. Before using them "correctly" in my own sense, I have to define (elsewhere, just for me) which similar topics should be found inside this "folder". (it would be nice, if zotero would have an extended tag handling, where this definition could be placed into). In my structered tags I have totally different "hierarchical sub-trees", one for "projects", or "companies", or "products" or different thematically sub-trees like "physics", "math" or "medicine"... By this definition of a structered tag I have not the problem to tag everything in every language with every synonyme which could be possible.
On the other hand I use "flat" tags (i.e. normal one word tags) to enhance the item tags with additional (mostly very special) information.
Just wanted to add my voice to those requesting a hierarchical option for tagging- I wonder what compromises might allow some limited introduction of this. I have been building personal databases with tags of academic articles in psychology since around 1998, fairly extensively, and my observation is that the act of tagging and knowing one's own taxonomy is a huge primer for knowledge construction and understanding. All I want is a separate extensions that would allow me to 'view' my tags and create a hierarchical structure for them so I could curate them. Otherwise I will simply manage the hierarchical structure offline and deeply learn it (which could turn a weakness intro a strength)- it's manageable for me as I use at most 60-70 tags and strongly resist over-differentiation (that's what two decades have taught me)- I am working with what level of generality for knowledge construction is appropriate for me and my subject. But yes, a simply taxonomic visual interface with the ability to select a certain family of tags with one click would be amazing.
Implementing hierarchical tags (in web library too) it's absolutely excellent, future-oriented idea. I need this function in my work. But user must have possibility to turn on/off hierarchical view.
I need only possibilities of creating my own tags tree and switching between flat and hierarchical view. That's all. What do you think? Is it difficult to implement?
Hello you can try the "nested tag" feature in Zotero-style add-on https://github.com/MuiseDestiny/zotero-style Still, It would be better to have hierarchical or nested tags built in zotero itself for easier management and being more future proof compared to lack of plug-ins updates in some cases.
Upgrade Storage
I also use collections and subcollections (with recursive option enabled) extensively. Collections can do a lot, but it's not a substitute for hierarchical tags. I recognize that this may represent a significant--or even herculean--effort, but I also see the potential for huge benefits to many Zotero users.
I use a proprietary program to manage my images called IMatch. They are in the beta stage for their new release IMatch 5, which will have hierarchical keywords as well as categories.
Imatch doesn't do a lot of what zotero does but I believe it is far better in terms of how documents can be categorized and then sorted and found. Perhaps zotero could pick up on some of the strengths of that program?
I'm having trouble seeing why hierarchical tags are strongly preferred by some people compared to a flat tagging system, assuming that in the flat system you can combine multiple tags to see their intersections.
Possible hierarchical tags benefits:
1. It's easier to get an overview of top-level categories when you only see other tags a parallel level, rather than seeing hundreds or thousands of tabs all at once.
2. It's easier to enter tags because if you enter a child tag directly, all parent tags are automatically added (assuming that hierarchical tags are implemented in this way)
Are there any other benefits? Are these two benefits larger than I currently realize?
Auto-complete helps with 2, at least, and I can imagine a drawback being that I don't know all the parent tags that would be added when I a child tag.
I agree that searching for tags by typing is useful, but why is the hierarchy actually useful to you?
For me, a good alternative to making the tags hierarchical would be to enable searching collections by typing their names. That might be even preferable - to have both hierarchical and flat organizations. If you make the tags hierarchical, you may need to add an option to view it in the flat format, for users who are accustomed to it.
I study in the field of public transportation, doing a masters' thesis. In the last months, I've been searching about student transit fares, and I know, many, many, MANY, tags emerged from that search. And I kind of lost control. It would be best if I could, just by attributing one tag to an item, make many others follow automatically. That would tidy a lot my library.
For example, if I were to write:
"University fare program (U-Pass)",
it would be best if it could trigger all the hierarchy of tags above it
"Public transportation", "Transportation", "Transit pricing", "Transport Economics", "Transportation planning", "College students"...
Let's cross fingers it happens! Meanwhile, does one of you know a way around this problem? (Perhaps a slightly different solution exists now.)
I'm missing this feature, too. For a long time now.
Meanwhile I use structured tags instead of hierarchical tags, and use a special character ">" for structuring.
For example to >MED>therapy>lithotripsy or >MED>therapy>endoscopy.
So you find all medical therapy sub-tags used in your library in one place under >MED>therapy and not distributed over all other tags.
It is not perfect, but it works for a workaround (for me).
To display all items with >MED>therapy as parent tag you have to use the "full-text search". It is not able to do this in the tag-cloud (as it should be for real hierarchical tags). On other problem I have, that these structured tags could not fullly displayed in the tag-cloud, because they are simply to long.
Thanks for your answer! That's a clever workaround that you've got, even though it still has some technical issues with it. It is better than what I have.
Although, I think I might have a hard time to apply it to my system. It's because I am used to create exhaustive tags, by inserting as much synonyms as I knew into the same tag. It's to be sure I don't miss anything when I try to recover articles about a given topic. For example, I've created "Transport collectif (transport en commun, transport public, public transportation, public transport, transit, mass transit, public transit)" (French and English words) for identifying articles that are related to public transit. That way, if for some reason I forgot one the synonyms and I search for "mass transit", I will recover the same articles as if I choose to go with "public transit".
Unfortunately, if I combine my method to your method, I can end up with excessively long tags... Hence, I am wondering, how do you personally work around the fact that some words have multiple synonyms?
For further information and enlightenment Google these concepts
Bibliographic database thesaurus
Bibliographic taxonomy
Synonym-ring
It would theoretically be possible for someone to write a Zotero plug-in that would allow you to incorporate an open-access thesaurus program's capabilities with an advanced search of the Zotero database. This would be a tremendous amount of effort -- constructing a personally useful thesaurus, linking the thesaurus to a query system, and linking the custom query system to Zotero.
There is an open-access, free multilingual thesaurus construction program, TemaTres, that is absolutely wonderful (https://www.vocabularyserver.com/) but that is only one very small step toward what you would like to have in your Zotero keyword/tag capabilities.
My safety literature database uses parts of TemaTres for its synonym and hierarchy capabilities but this requires a substantial continuous amount of work.
There may already be term taxonomies that already exist that may help you. The U.S. Transportation Research Board in partnership with European and Australasian transport research agencies have constructed the Transportation Research Thesaurus, (http://trt.trb.org/trt.asp?), that covers much more than public transportation but would be useful to give you ideas that have been demonstrated to help with classifying and indexing transportation-related literature.
one way to find things in your database with similar tags is surely to incoorporate a thesaurus into zotero.
But this is not the way for my "workaround" regarding hierarchical tags. You asked me, how I personally work with synonymes or similar tags.
I use tags to enhance every item with additonal information relevant for ME. I ask me, how do I like to find my stuff in my database. Then I add this information on each item (as tags) to be able to find my things.
Practically: Therefore I have designed a system of structured tags as a "hierarchical tree", which I use like folders. Before using them "correctly" in my own sense, I have to define (elsewhere, just for me) which similar topics should be found inside this "folder". (it would be nice, if zotero would have an extended tag handling, where this definition could be placed into). In my structered tags I have totally different "hierarchical sub-trees", one for "projects", or "companies", or "products" or different thematically sub-trees like "physics", "math" or "medicine"... By this definition of a structered tag I have not the problem to tag everything in every language with every synonyme which could be possible.
On the other hand I use "flat" tags (i.e. normal one word tags) to enhance the item tags with additional (mostly very special) information.
It is very enlightening to learn about what is a stake when talking about hierarchical tags.
I hope, one day, Zotero, or someone, will succeed to implement a plug-in that works.
you can try the "nested tag" feature in Zotero-style add-on
https://github.com/MuiseDestiny/zotero-style
Still, It would be better to have hierarchical or nested tags built in zotero itself for easier management and being more future proof compared to lack of plug-ins updates in some cases.