Currently, I use mostly tags. In L3, I used collections. But I haven't missed anything I did with collections in L3 that I cannot do with tags in L4. Hence, the question: why bother building collections? I'd value your input.
Hi Francis
While tags are clearly powerful, the main benefit of collections I see is that they are dynamic.
So, if I create a collection based on a set of rules any new resources which match those rules are automatically added to the collection. If I were using tags for this I would need to manually tag the resource.
Graham
Hi Francis While tags are clearly powerful, the main benefit of collections I see is that they are dynamic. So, if I create a collection based on a set of rules any new resources which match those rules are automatically added to the collection. If I were using tags for this I would need to manually tag the resource. Graham
As a "for instance" case if you set up a collection with the rule of title:theology then anytime you add a resource to your collection that contains theology in the title it is automatically included in that collection. Then when you go to search you can just search that collection and the Theology book you bought and forgot to tag is covered. You can also have a rule of title:theology AND mytag:theology. Then if there is a book in your library that doesn't contain the term theology in the title but you tag it as a theology it is included in your collection.
you can also combine collections. Suppose you created a collection with a rule of author:MacArthur and a collection with a rule of author:Piper. You can search all the resources in your library by each of those authors. Or you can make a collection that contains all the resources in both of those collections so that you can search your "MacArthur and Piper" collection.
It seems to me impractical to set up an equivalent collection for every tag I have. As I go through resources, I encounter topics that may not warrant a collection because either there are not or I do not own enough resources of the kind. Why do the work of tweaking the right syntax for a collection criteria in such a scenario? I would tend to think collection for larger collections (say, Bible commentaries, theology or Church Fathers) but perhaps not for "Smoking" (though I may actually tag a resource that speaks of it).
So far, I have given a tag or more to every resource I own (over 1,600). I add tags as needed (though I tend to use favourites as well when it comes to that). Perhaps it would be good if there could be some kind of intelligent under the hood conversation between these three: collections, tags and favourites (for instance, a pop-up asking me if I want to tag something I put in favourites)?
Can you folks elaborate on what your practices have been?
My approach has been different in that I have focused on creating collections for areas of interest - some I have one for "Church History" for example:
I find it useful then anything I purchase with "church history" (for example) in its title will get added automatically.
I actually haven't focused on using tags anywhere as much as I should have - for example it would probably have been more efficient for me to tag the resources which I have had to add manually and then just add a "tag" rule to the collection definition.
I am only using tags at the moment to note resources which I specifically want to read through.
In some cases, I have used tags in setting up collections. I divided my commentaries in several groups through tags (Critical, Exegetical, Notes, etc). I then open a commentary in each group in a layout. I can then use the Parallel Resources button to cycle through all the commentaries in a particular group. The downside to this is that PRAs are in serious needs of optimization.
Also, Logos has tagged resources by series, subject, type, language, etc. All these can be used in setting up collections which I then use through the Cited by tool.
As your library grows, tagging each resource can become a huge burden. With over 3500 resources, I don't want to tackle that task.
It seems to me impractical to set up an equivalent collection for every tag I have. As I go through resources, I encounter topics that may not warrant a collection because either there are not or I do not own enough resources of the kind. Why do the work of tweaking the right syntax for a collection criteria in such a scenario? I would tend to think collection for larger collections (say, Bible commentaries, theology or Church Fathers) but perhaps not for "Smoking" (though I may actually tag a resource that speaks of it). So far, I have given a tag or more to every resource I own (over 1,600). I add tags as needed (though I tend to use favourites as well when it comes to that). Perhaps it would be good if there could be some kind of intelligent under the hood conversation between these three: collections, tags and favourites (for instance, a pop-up asking me if I want to tag something I put in favourites)? Can you folks elaborate on what your practices have been?
It's actually quite easy to create a collection from a tag without having to "tweak the right syntax for a colection" -- just use mytag:XXX. But you don't need to set up an equivalent collection for every tag. As you've probably discovered, you can restrict your search to just the resources that have a particular tag by selecting mytag:XXX from the Search In dropdown. You can also use these sort of "auto-collections" in making custom Passage Guides, etc. I believe they are available anywhere you can select a collection name. It's just nicer to be able to have a more descriptive name for these collections. Many of my collections are simply the set of all books that have such-and-such a tag. That's how I do most collections that aren't easily definable using the metadata. There are some cases where tagging would be totally redundant. For example to get a list of all my Bibles I can use type:bible. No need to tag them as well with a tag "Bible." To get a collection of all my English Bibles, all I need to do for a rule is type:bible lang:eng.
Even though you could theoretically use tagging instead of collections in most situations, what you can't do with those tags is AND or OR them together in places where a collection dropdown menu is offered. Suppose you wanted to categorize all your commentaries into four different kinds, according to the schema suggested by Morris Proctor (Scholarly/Critical commentaries, Homiletic/Exegetical commentaries, Application/Devotional commentaries or Language commentaries). You could, of course, make up a tag for each of those categories (which I've done, in fact) and tag all your commentaries appropriately. Then you can have sections for each of these types of commentaries in your passage guide. But what if you simply want to search in all your commentaries? There'd be no way to do that unless you had also created another separate tag, say "AllComm," which you'd assigned to all your commentaries. That gets cumbersome when there are overlapping or nesting categories that you might want to restrict your searches to. With my four different commentary tags, I can create a single collection of all my commentaries by using the rule mytag:(CritComm, HomComm, AppComm, LangComm). That will include all resources that match any one of those tags. Actually, the way I do it is I have collections for each sub-category of commentary and my All Commentaries collection includes the other four as nested collections.
You might want to watch Mark Barnes' excellent videos on collections and nested collections to see why they can be beneficial.
Thanks for your feedback Rosie. I don't think that using mytag: in rules for collections is great unless it is used with OR not AND. My understanding is that new resources that I have not tagged would then not be added dynamically.
If one use the rule used for a collection as a criteria in Library, these resulting resources can be group tagged. I say this for those who already have gone through creating their collections. But it would be much handier if "collections" was a sortable field in the library as tags, subjects, types or titles are.
Right, but if you tagged a new resource with a tag that was used in the definition for multiple collections, then that resource would instantly be added to all of those collections (instead of updating each collection one at a time). So that is a kind of dynamic update.
In collections, mytag can be usefully used with AND as well as with OR. It really just depends on what types of tags you're using and whether they are disjoint in purpose. For example, I've got a tag called ScholMon which I apply to all the books in my library which I deem to be "scholarly monographs" which would include all the JSOT Supplements, most other T&T Clark / Sheffield resources, etc. I also have categories for Arts, Science, Jewish Background, Hermeneutics, etc. I might want to make a collection of only the scholarly resources on Jewish Background to help target my searches on hits that are likely to be more helpful for my academic work. So I'd use AND in that case: mytag:ScholMon AND mytag:JewishBkgnd, for example.
It is true that new resources you have not tagged would not be added dynamically. But if you're going to all the trouble to tag all the resources in your library already, you are probably doing that tagging work every time you download new resources. I know I do. I use the star rating system to let me know which resources I haven't tagged yet. If there are no stars, I need to tag them. As soon as I've assigned all the relevant tags I can think of, I assign a default of 2 stars. That means I don't know yet how good this resource is. Once I've determined that a reasource is not that useful, pretty good, very good, or essential, I mark it with 1, 3, 4, or 5 stars. If I later decide on a new category that I want to start keeping track of I can go through and find resources that I already own (using the subject or title metadata and filtering in the library) to assign that tag to. And I'll add that to my (mental) list of tags to look out for new resources coming into my library that should also have that tag.
My system is not error prone. I don't think anyone's is, even the tweaked dynamically generated collections. Some titles will slip through that don't belong, and some will be missed. But just as learning Logos is an iterative process (see my signature below), creating collections is an iterative process. When you find that you've missed something, you fix the tagging so that it will be included in the future. And so forth.
The other thing that collections is useful for that I don't think has been mentioned yet is the parallel resources. I was a Libronix user and loved that feature and was excited when Logos brought that back into L4. I personally use it with lexicons and commentries. Someday I do hope the serial resources will be brought back as well.
For those who don't know yet, it is easy to convert a collection to tags.
1) Create a collection, fine-tuning the rule syntax until you are satisfied (too many resources to put in the minus column is probable evidence that the rule should be bettered).
2) Select and copy the line in the rule box.
3) Open the Library and paste. The collection resources will appear. Do CTL-A to select all. Click on the information button (the i ). Under "Add these tags to all" type in the tag that corresponds to your collection.
4) Get a bowl of ice cream and watch your favourite movie.
Just to clarify for any who might be wondering, that's a celebratory optional last step, not something you need to do to wait while Logos finishes adding all the tags. [:)]
