Improve Replication

The ability to create replicants is arguably one of DT’s most innovative features and it’s the main reason why I use DT over all the similar products that don’t offer this capability. Since I use this feature constantly and I suspect many other users do as well, I’d like to propose a couple of improvements that would make it more streamlined:

  1. There should be a dedicated toolbar button to create a replicant. Clicking on the button would display the database hierarchy and allow you to select the folder you want to replicate the selected item into. I know you can do this with the Actions button, but it requires you click the button, then navigate down to the Replicate To menu item. For a feature that’s used this frequently, it should really have a its own dedicated toolbar button.

  2. In my thread on improving wiki linking, I mentioned the idea of integrating wiki linking with search, so rather than having to navigate through the entire database hierarchy, you could type a few characters and it would display a list of matching documents, allowing you to select one to link to. I’d like similar functionality for creating replicants. When you click the proposed Create Replicant toolbar button, in addition to display the database hierarchy, it would also display a search box at the top. You could type some characters into this search box and it would display a list of matching groups. You could then select a group from the list and it would create a replicant in that group.

I’m not sure how frequently they are used. I personally rarely use them, except in a Support capacity.

like the op, i use replicants a lot, and i also find them to be one of dt’s outstanding features. i find them particularly useful in cases where i’d normally rely on tags in other apps. honestly, i think dt could handle tags a lot better. there is currently no support for them in dttg, for example, and dt doesn’t classify using them, so you cannot rely much on them for organization if you want to take advantage of the ai. without replication, my workflow would be completely up-ended, so i consider it to be a critical, core feature in dt. in addition, the ability to make replicants clearly distinguishes dt from other apps.

as for the op’s suggestion, i don’t know what i think. i don’t find toolbar buttons in any app especially useful. i’m glad they are there, especially for infrequently used items (i won’t remember the shortcut keys for those), but in the case of replicants, right-clicking is fine with me. you’re going to have to pick the location anyhow, so right-clicking and choosing or choosing with the toolbar icon are essentially the same thing. there is also cmd + l, which is arguably faster than either method, and then you can use the excellent ai to more easily sort it. ideally, you won’t even have any typing or dragging at all to do, thanks to dt’s ai.

maybe the op’s suggestion is an improvement. i don’t know. but, i can’t imagine it working better in my use case.

Perhaps it is my lack of imagination for I do not use replicants much but I would be very interested in learning what uses people make of them and how it helps.

not a lack of imagination at all. their use is certainly not obvious, and they are not part of our regular use of files and folders outside of dt.

basically, in my case, they are good for files and folders that are not easily filed away into a single location. the most obvious use case within dt is the mobile sync – you replicate files and folders into it and sync those with your ios device. it’s basically, in this example, a way to tell dt that the files and folders belong in more than one place without actually moving them. other apps often accomplish the same thing with tags, though i haven’t found an app yet that does tags well (well enough for me, that is).

another use might be a receipt. let’s say i have a group for receipts i need to keep (expensive stuff). but, i also have a group for receipts i need to submit to my employer for reimbursement. and, maybe i have a group for a business trip. a single receipt could conceivably fall into all three groups. this is where replicants come in handy. if i annotate the receipt in one location, the annotation is reflected in all three. that’s really nice.

personally, i find replicants most useful with my research materials, but that is probably not a use case as familiar as receipts.

Thanks for your explanation. Most of my work is research so ideas on this would be appreciated.

I think that will be a long forum post coupled with a longer blog post (s). I’ll get on that as soon as I meet my deadline today!

I never understood Replicants initially. It took me a while reading through threads/posts by Bill and the like over here, to start understanding it – but it was my syncing to DTTG that really brought things home.

The ‘original’ files stay where they are in their group/folder structure, unannotated. Replicate to the DTTG sync folder. Get them over onto the iPad. Annotate them in Goodreader (or whatever), sync them back to DTPO. Because they are Replicants, all instances of those files are treated equally. So while the Replicated files are being annotated on the iPad, the ‘original’ files are sitting there, clean. As soon as the sync back from DTTG happens, and the annotated files land back in DTGO, the annotations are instantly (and automagically) transferred to those ‘originals’. They too are now annotated. In my case, I then simply delete the files in the sync folder, and the ‘originals’ again become the only instance of that file – but they are now annotated!

The beauty of the above is that I never needed to move any files, or interfere with my folder/group structure. No moving/duplicating and forgetting where what is where, when.

A second example – I might want to use files (articles; book chapters; caselaw etc) in a particular lecture. Again, they might not yet have been processed by me. If the lecture is going to see me “using” them in the same research context as my main research, I can Replicate them into a temporary (or permanent) Lecture-specific folder, where I store all the resources I will be using for that lecture series. I Replicate them, since any annotations I make, will still be aligned with the main research project, so I am happy to have the annotations transfer over to the original file.

If the lecture series was to use similar sources, but the emphasis of the series was slightly different to my central research, I would Duplicate those files instead, and the annotations would (obviously) NOT transfer over to the originals.

A third and final example, related closely to questions surrounding Tagging and Single/Multiple Databases – Article X might be a comparative analysis spanning several jurisdictions. I could either Duplicate that article into different groups/databases – and only annotate that section of the article that speaks to the jurisdiction of the enclosing group/folder I have placed it in. Each instance of that article would then remain separate and independent of the others, and would only have certain parts of it annotated (depending on which jurisdictional group that instance was in). Or I could Replicate it, and “see” what I annotated in all the various parts, and various jurisdictions, as I work my way through the different instances of that article, in its various groups. This obviously differs from article to article - if the comparative discussion is a “standalone” section - then Duplicating it works a charm, since I would only be interested in seeing annotations in that specific section. But if it’s constantly jumping in and out, interweaving the comparative jurisdictions, then it is more useful to “see” what I had annotated on a previous occasion, when working through the article as a Replicant, in a different group. I don’t think Replicants work across different Databases - it simply becomes a Duplicate then AFAIK, so the latter example really depends on how one’s DBs are setup - but this hopefully explains things somewhat.

Regardless, must agree with what was said above. The ability to Replicate, as “simple” as it is, is (imo) what makes DTPO so powerful and versatile – and is certainly one of the key distinguishers from other, seemingly similar information managers. I think Devonthink needs to make far more of it as a feature, than what it currently is. Along with the wikilinks!! :slight_smile:

Thank you to both Frobgoblin and especially Cassady whose description of the various ways he uses replicants is most useful. :slight_smile:

I also use replication all the time, indeed every pdf-file which is entering my database is normally replicated into several groups. I prefer this over simply tagging them (which I use alongside, but less frequently).