+1 for 3PV (more than urgend and requested by so many people) (please add that before all your customers are gone) (DT is a brilliant product but this is missing)
Thanks a lot in advance 
Is DT still following this topic ?
What I think is a great pity and what shakes me fundamentally is, that I think DT has not realized yet what is really behind all of this feedback. Dozens of individual things have been discussed, all of them are invalidated (solved) because the function is still there or because it is supposed to come soon (what are we doing meanwhile?)
I can see from myself and from the people in the forum that they are not espacially obsessed with the 3pane view but that they are just looking for handy way how to continue using the program with their individual setup. And they are not happy.
Of course there are new ui concepts and you have to get used to them but at this point I miss the feedback and the vision of how DT imagines this to be in the future. The user interface looks tidy now because everything has been packed into the side bars, but it has become impractical because of the lack of individual customization. This is neither modern nor flexible.
You can replace 3pane and even make it even better, but not in this way. The new interface looks good but for me it is a step backward in the handling of the program - here I miss both the idea of how to improve this in future with this ui concept as well as the feeling that real problem is recognized.
Also I cannot follow the statements that DT is flexible and you can use it the way you want to. This is no longer the case. The UI is more fixed than ever before. I currently have to do a lot of restructuring simply because it has become absolutly impracticable to use with my old folder structure. When trying new approaches I run into new problems again. I feel a little stuck in the middle and don’t know what to do best… and I don’t have the time for this.
So: I don’t need 3pane but I need a better alternative. As long as this is not foreseeable, 3pane should be retained. My feeling is that DT ist concentrating to much in cleanness and features but losing focus of some really basic questions. Here is enough feedback and they should take this serious.
Important to mention: I like DT and it’s support is outstanding and I would like to continue using it in future… but at this problem I need to point out clearly that something is going really wrong.
4 Likes
The Moderators and tech guys are on it all the time as far as I can see. I often get a reply from @BLUEFROG within minutes. I think maybe they are letting this thread just run. I would. I don’t miss 3 pane view and I think in some ways the point of DEVONthink 3, for me anyway, is to use the AI capacities and search rather than build up Folder structures that one can do in Finder? Am I missing something?
1 Like
I think that the real point of DEVONthink 3 is the AI. I use it a lot, if I didn’t I would totally agree with what you say. I think there are though some really nice new features on DEVONthink 3 that I find really useful. Metadata and Notetaker… many more.
Not sure what you’d be missing, but from the thread it’s clear that many people are missing what for them was an easy-to-use, low-effort way to work with libraries comprising lots of documents and folders. That’s what I miss.
Comparing the so-called “AI” (it’s not really) and the layout of 3-pane view is apples and rock salt. One thing has nothing to do with the other, IMO.
I think it is ‘really’ AI and quite a sophisticated one at that; what would you prefer one calls it? I would be quite happy with a new nomenclature, little hinges on it. There is no hard and fast definition of AI and a lot of very sophisticated stuff relies on similar principles that DEVONthink 3 does. Google translate for example is really a ‘brute force’ data crunch not ‘intelligence’ in the human related sense? To the extent we understand that anyway.
I think the term fits the capacities of DEVONthink 3 at present well enough. As I said my own approach stays light on folder systems for the reasons I gave already. I think it is a better use of DEVONthink 3 to avoid deep folder systems and so on. I am not going to get into an argument about it though and frankly think the 3 pane debate is now a bit passe.
It all depends on what you are using DT3 for- I have different databases, some light on folders and some heavy on folders.
If you are collecting articles on a topic such as for academic research in a given area, using AI to find documents may work best.
If you are doing professional or consulting work with client documents organized by case, then detailed organization by folders and/or metadata is very important and AI is not likely to be helpful.
2 Likes
For you perhaps, and I’m happy it all worked out for you.
But not passé (“out of fashion”) for anyone who depended on it. It’s best not to be dismissive of other user’s requirements.
Fair enough and I won’t opine on usages that I haven’t studied.
I will say, partly to @korm that Ffolders’ are not ‘really’ folders either. I won’t press the point. They are UI artefacts to some degree. I am tending more and more, as I keep saying, to exploit smart groups and other techniques. However it is clear that some users are comfortable with Skeumorphic folders. I will confess I am still afraid to leave them go myself in many cases. I accept that the human mind does function quite well iconically. I have no objections to 3 pane view being reinstated of course on that basis I suppose. It was no loss to me as I said and I think really the debate is passe. I did say I would not continue so I better practice what I preach.
Smart groups are wonderful in many situations. On the other hand, there are some situations where users must use folders or “non-smart” groups.
For example, a specialist in some area of knowledge might use DT3 to evaluate a case as an expert witness in a litigation matter. Such an expert might use smart groups for his personal collection of articles in order to identify articles on a given topic. But when working on a specific case, where the expert’s file may be subject to discovery and reviewable by other parties, a “non-smart” folder system would be much preferred so there can be no confusion or misunderstanding about exactly which documents are part of the case file. Moreover, a detailed folder/subfolder system may be desirable in order to allow the expert to quickly find a given document in a deposition or mediation setting.
2 Likes
I see that point, I do that kind of thing in fact sometimes. I will be honest with you I have forgotten all about 3Pane view and really am struggling to recall how I used it…
OK I think I understand you. I assume that if one were to use TWO Smart Groups for that task what would start to feel muddled and unsafe? Is that how you would feel? I do this kind of thing using Smart Groups: one file goes into two Smart folders under different criteria.
I admit I never use duplicates and replicants because I find them a bit confusing. I did have a couple of folders with eccentric contents: I had no problem finding a Smart Group (s) for them though. That might have been luck I accept. As I said the issue here is not so much 3 Pane view in my opinion as it is forming but something more fundamental?
I might try getting rid of all my skeuomorphic folders and just use Smart Groups. I will take a back up carefully first though… I think I could well end up in a muddle though in fairness.
I ran into some performance issues when I kept thousands of PDFs in a single group. You could avoid opening the group containing all of the files and only use Smart Groups to see into that repository, but I found I had to open it often enough that it causes trouble for me.
I now arbitrarily sort those PDFs into semester/year added, and use search and smart groups, etc., to browse them.
Just something to keep in mind when moving away from group hierarchies!
1 Like
Thanks Ryan. Really useful information. I really think developing strategies like yours, a smart one in my view and one I hadn’t thought of, is more useful than pining for 3Pane view? I should focus my own efforts on that constructive path I think. Thanks for pointing the way.
I appreciate there is a wide range of ways of using this app. I admit I hadn’t given any thought to folk with very large databases in responding here. Thanks for the corrective. That cuts both ways though regarding 3Pane view, for example there might be under the hood problems regarding performance, like the issue you mention, for other views and modes? I have no idea but suspect there are.
My own database is not very large in relative terms. It could be massive but I resist the temptation to keep every paper and quote I ever come across for some future reading date that never arrives.
As Korm observed upthread a little, DTech stopped contributing to this thread a while ago. Given the speed and helpfulness with which they respond elsewhere on the forum, it’s reasonable to assume that this is a topic that no longer interests them.
I agree that it would be nice to hear about DTech’s plans in terms of replacing the workflows that were taken away by the new UI. It’s not likely though. Sidebar improvements were mentioned several months ago, and there have been incremental improvements in the past 4 point releases. We can hope for more. None are likely to replicate or replace 3PV, that much seems clear at this point.
A folderless future may work for you, but there are many use cases where that would not be feasible. Some outlined above in some detail. It also completely glosses the point that users have constructed databases using a folder structure over many years. You’re rather cavalierly asking them to re-engineer and rethink how they interact with thousands upon thousands of carefully curated data, which is quite the opposite of a trivial (or should I say passé?) task.
Many years ago, when I started using DTPO2, I adopted a modified version of Rachel Leow’s approach to Devonthink and academic research. It’s a quite fantastic mix of folders and smart groups. I see that DTech are now promoting a new (and paid) guide to using DT3 for academic research, in this case – as in Rachel Leow’s – historical research. There’s a business case for putting together a guide for replacing 3PV and selling it, if I could only figure out how to actually do that 
2 Likes
I suspect (with no inside knowledge) that it’s more that they don’t have anything concrete to contribute beyond what they’ve already said, which is, in essence “We thought hard about how to make DT better, and this is the result”.
Many of us many disagree that this is better, but it doesn’t add anything to the thread for the DT folks to have to repay that in different ways.
And another factor will be that they’re working hard on stuff (DTTG stuff, I hope) and they don’t need all the splash back that would come from posting in the thread without offering a specific solution and date (mind you, they never offer dates anyway - and quite right too)
6 Likes
Thanks for those guides. I didn’t know about them and it is serendepitous since I don’t, like some say about the DT team, look at this thread any more: well I did recently of course. Those guides might be of great use to me. Whether my approach is cavalier is again a matter of perspective. I do appreciate the point though I will say to you and I hadn’t thought of it in that way in terms of continuity if you will. It would have been best if I had stayed away from ripostes with you initially 
I agree with your last point in essence: there must be some kind of work round or ‘solution’. I used 3PV all the time but I just don’t miss it. I actually delayed upgrading for months and then found I prefered DEVONthink 3 anyway. I will think about this some more then. I take @ThatGuy 's analysis too. I am sure he is right. I think they probably will come up with a work round or add on of some kind. We are working blind to some extent and don’t really know the technical issues. I think our comments ARE read by the way. Especially these constructive ones. Thanks @Kinsey for holding the line so to speak. In fairness this thread has taught me a lot about DEVONthink 3 and helped me think it through. I do have confidence in the team at DEVONthink 3 I will say. I suppose I can put myself in your shoes by imagining that they got rid at some point of the ‘mouse free’ note taker that is now there if you know the shortcuts. If they spoiled that I would be pretty aggrieved, it was the feature I actually wanted initially.
As a curious side note. The topic of Rachel Leow’s research and interesting book is very near to my heart and something I have been directly engaged in though not as a researcher.
Bear with me then. I am sure I am going over old ground. What is difference between 3PV and this that I can do on DEVONthink 3? It just happened automatically if you like when I upgraded. I am not being facetious or pointed in any way: sorry about doing that previously by the way. To me this feels just like the 3PV did. I accept that I might not be using it as others do. I screenshoted with the ‘view’ menu showing to show what was ticked. I can’t see any difference between this: even to look at you might say. I only have one database plus the Inbox of course and that might matter?
1 Like