I have a hard time trusting anything, honestly. I’ve been thinking for months and its a very difficult decision especially with this change. I use Notion HQ for clipping and doing basic research. with a lack lack of API and built in automation would take me decades integrating my projects let alone starting them. I sometimes embedd a mind manager web view file in a notion database and work from there. Its unique but you can’t make connections below the title of your mind map (in the sense of relating stuff from mind map to notion’s databases and vice versa) and that blows. Tinderbox is great but can’t alias/replicate more than 1 level (think scrivener collections but with nothing in the collection itself…just a link). Tinderbox cannot handle serious projects and so I use it more of a QDA tool, sometimes. Started looking at a couple of tools which blew me away, however, trust is a big problem for me. Workflow is naturally developed based o what a tool allows me to do. if they break functionality (as the 3PV saga here) you cannot simply export your crap and move to a different tool; you’re setup is dead (views, class. system, tagging system, automation…everything). Roam Research & Observable HQ. are some neat tools. Rom can break textual content at any point - be it group, paragraph, sentence or even a word - and will replicate it anywhere you want and with some sort of auto complete function. IT’s tweaky and freaky but do take a look. To fresh of the shelve and will need to wait see where it all goes. Hope that helps a bit.
As I’m sitting here working in DT3 trying to get the sidebar to work for me, and having glanced over the above comments, a thought for a tweak that perhaps has already been mentioned (maybe even by me - I’m losing track) -
Being able to right click on a database in the “Open Databases” list in the sidebar, and choose something like “Isolate Database.” Doing so would hide everything else in the sidebar. As I’m working now, I realize that part of the issue is the fluid nature of a database in the sidebar, because of its positional dependency on everything else in the sidebar. I’m finding that makes it harder to get my bearings, and is making navigation burdensome as I work on a case. It would be nice to lock everything into place.
Thanks for taking the time to out this together, much appreciated. Given I have so much time invested in crafting multiple DT databases, I’m nowhere near ready to adopt an alternative. I’m still trying to find a comfortable way of working with DT3. But I will be looking at some of the tools you’ve mentioned, to see if and how they can augment my workflow.
I have stayed out of this, but I too am suffering. Yes SUFFERING because of the removal of the 3 Pane view.
“Isolate Database” as an option certainly has my vote! 
I think it the usual term would be Focus or (more old-school) Hoist.
The touchpad / mouse-intensity required to use DEVONthink 3 is a drag. I’m using DEVONthink 3 a lot less – basically, not at all – for new projects because of all the clicking around needed. It got to be a repetitive-stress issue for me, in addition to just the slowness of navigating my files.
That is the problem. After using DEVONthink 2 for five years nearly I only recently started using 3 pane view would you believe? Ironic indeed, I was cautious about it for exactly this reason. Then in a short time I got really into it.
I will update to 3 at some point of course. Since I know 3 pane won’t be there I am getting used to doing without it again, ironic? I can only think there was a good ‘under the hood’ reason for dropping it? I do trust the developers of this app and that is one reason I stay with it. I agree with your analysis really. What I can’t work out, now after getting used to 3 pane view is how essential it is? Is 3 pane view essential to me: I really can’t decide. I did find it strange at first in fact and counter intuitive. My use of this powerful app is, by now, instinctive and I can’t often unpick how I use it. In the same way as one can’t tell people what finger one uses on a particular key. I touch type.
I am getting tired of the Apple upgrade cycle too. I don’t need a new system every year. I really don’t. I haven’t felt a single productivity improvement for a few years now. I think I might decide to skip the next one altogether and get my own cycle going. But I think it makes things more difficult for developers of complicated apps too? I don’t know that but I suspect.
Have you gone back to DTPO2, @korm?
This is very interesting – I catch myself micro-pausing multiple times a day while using DT3 as I try to remember how to navigate to my files. I’m trying to figure out if I merely need time to adjust to DT3’s new interface, and am conscious that any major update will require adaptation. And I think it’s worth stressing here that this isn’t a case of getting used to new features. I am learning to appreciate DT3’s updates more every time I use it. This is a question of navigating to my files.
I’m also conscious that sidebar improvements are coming, though of course they may take some time. The devs and support at DTech are likely overwhelmed with support queries and UI/UX tweaks, based on all the feedback here on 3PV and multiple other issues.
No. Not yet. I have been all-in on DEVONthink 3 for most of the year, and unwinding back to DEVONthink Pro Office v2 is not desirable since I would have to redo sync across 4 devices (after DTTG trashed my sync store). I keep v3 open all the time. Just don’t use it for work anymore.
I hope I’m wrong, but I just don’t see how massaging the way the sidebar works will give us back the elegant solution that we already had with the 3PV. And that’s after reading the stuff proposed above. The whole thing has become very clicky-scrolly. Which is for me, and obviously quite a few others, a productivity killer.
I said a little earlier that DT had gone from being my most-used app to my least-used, and I am interested that I am not the only one who has found the app less useful. And in your case, coming from a long-time stalwart, I find it particularly telling. The length of this thread also says something – how many others reach a tenth of the length, I wonder. Sic transit gloria mundi? Nothing lasts for ever, I guess.
What has replaced DT to do the things you used to use DT2 for?
Thank you. I’ve been hanging around these forums for seven or so years now and learned a lot from reading your answers to many queries on the forums. I hope to learn more in the future!
I haven’t really found a replacement yet, and I’m still experimenting. But I am beginning to think that the whole paradigm of using “static” folders is the wrong way to go. I am beginning to think that smart folders could be better, particularly combined with tagging.
I should perhaps say that I work in psychology and history, and that an important facet of my work is finding links or associations between things, be they documents, articles, ideas, or whatever. Following the paradigm I learned at school, I would “put together what goes together”. The fact that DT allowed replicants was an idea that struck a chord with me because one item is often related to more than one idea, so it could be made to appear in many places via replicants. But I have come to find this a very cumbersome way of going about things. If you use tags instead, it does not matter where on the file system an item is, it can still be “gathered” together with similarly tagged items through the magic of smart folders and saved searches.
In a sense, this is borrowing the idea of incremental formalisation, which is much talked about on the Tinderbox forum. For me, there is a lot of angst in deciding where to put things, because it is prejudging which of the ideas they contain is most important. But the approach of incremental formalisation allows you to put a couple of tags on an item, and then when you later realise that another aspect is also important, you can add a tag. You don’t have to move things around into different folders, or juggle with replicants, and so forth. The Finder, smart folders, and one or two other tools are quite good enough for this. Indeed, it might be argued that smart folders, which are similar in some ways to Tinderboxes agents, are superior. Smart folders are always up to date with the latest items without you having to do any moving around. I have got a bit tired of “See Also and Classify”. Tagging is a lot quicker than moving stuff around for me.
So what I did was to delete my databases and move stuff into the Finder. I created a few “large category” folders, and decided to use tagging for more “micro” classification. I used Leap for some of my tagging work, and HoudahSpot (which has a very nice interface in my opinion, and which I have used for years) for finding stuff. Hazel takes care of some of the automatic work of tagging files, renaming them, and moving them around (when needed).
This is a bit of a crude summary, and as I said, I am still experimenting. Indeed, I have been playing with DT again, partly because it will do things on iOS that the Files app cannot. And clearly there are some features of DT (like the concordance) that you are not going to get on the Finder filesystem (unless there is a utility that does that). It has sometimes been argued that it is better to use a range of utilities, each of which does one thing well, rather than one program that tries to do a range of things and falls between stools. There are pros and cons. But I certainly think that it is often possible to create one’s own tailor-made suite of tools for a particular kind of work, and that it is sometimes easier than trying to bend an existing application into a shape that it doesn’t like.
In summary, I’m not sure that the old paradigm of folders-in-which-you-put-things (which really comes from the old days of filing cabinets) is now the best way to do things. A methodology that automatically gathers relevant material into a smart folder (no matter where you have put it on the hard disk) has something to recommend it. I realise that there may be a problem of preserving confidentiality for some types of work, but I would have thought a solution could be found to that.
Sorry for the long post, I have been thinking out loud. I’m sure there are pitfalls I have not considered, but I am in search of a system, and experiment is the only way of finding what works for each of us (which will no doubt be very individual). It will be argued, of course, that the new version of DT can do what I have described above, via tagging, smart searches, and rules. But it is now doing it in a window that I don’t much like, and if I am not using other features of DT, why am I using it? Sadly, I have begun to find that DT is getting in my way rather than helping me – but that is just me, and I fully acknowledge that others will have a very different experience.
Very much how my thinking has gone. I mention it on fora quite often I find. I use and rely on search, smart folders, AI and that kind of capacity all the time now. I found myself recently actually undoing static folders as any attempt to ‘classify’ using my own system is less useful than ‘search’ and/or AI. It is actually quite disturbing in some ways, to make a deeper and incidental point. I think the issue for a lot of us is slightly deeper than we are expressing it to others and ourselves and the sense of control is a factor: habit even from the old filing cabinet days. As I keep saying the current AI solutions are a lot more like the natural ‘it is somewhere in THAT pile’ system than it is like the filing cabinet one.
If this works for you, or anyone, then sure. But I find piling to be more labor intensive than filing. For example, when I want to archive a document, I pretty much already know where it needs to go and I file it there. If I’m unsure, it goes into the Inbox for later categorization. As opposed to dumping that file in a directory that contains scores or even hundreds of other files, and in order to make sure I can find it again, I need to take the time - at the time of piling - to label it or give it tags, just so I can find it again, or have it display in the appropriate search results.
For me, “pile and search” is not an effective replacement for good filing in the first place. Even with as good of algorithms Google has, it still doesn’t always return the results I’m looking for and I have to go digging for it.
And I find the DT3 AI to be really good for stuff beyond simply searching for a file.
No doubt it depends to some extent on one’s field of study or work. When I worked in history it was a bit simpler because I could use a chronological system of organisation for material. Working in psychology on historical materials made things more complicated.
I am aware that I have drifted off topic for this thread, but with the indulgence of the moderators, I will enlarge a bit, with the excuse that the removal of the three-pane view has not merely been cosmetic, but has triggered a review of the way I work, and has made me question DT’s usefulness to me.
There was an interesting discussion on the Tinderbox forums some while ago, which can be seen here:
I hope the posters there will not mind my quoting an observation which I found telling: “DT works on a fairly naive principle – documents are related to each other because they have words in common – whereas [Tinderbox] notes are related because they have a meaningful relationship to me.” This struck a chord with me – I may have documents about cat food that I find relevant to discussions about Sigmund Freud, but the AI would not know that. And the connection may not become obvious to me until a later date. In my world, it is not simply about filing and finding – it is about discovering meaning. Tags help me do that, and it is incremental. Adding a tag does not destroy the original meaning that I gave to a document, it adds to it. On the other hand, putting a file in one folder gives it a certain meaning, and if I move it to another, it destroys that meaning and replaces it with another.
To drag things back to the original topic, DT at the moment is not helping me much with my “meaning-making” because I find the interface uncongenial. Last night I made much more progress with a combination of Leap, the Finder, and HoudahSpot than I have recently made with DT.
I haven’t quite given up on DT, but the pull is elsewhere at the moment.
Points well taken.
I can understand that. My research, collection, and retrieval isn’t nearly as complex or intertwined – word commonality is working for me (at least for now), but it makes sense where deeper connections need to be made. Personally, just coming up with the tagging for that would probably drive my over-analytical self into melt down, but I grok its significance.
For me, replication solves this, but I could see where this might not work in other disciplines of study. Although, I’m not so much as looking for an intertwined-and-yet-hereto-unseen connective meaning, as I am looking for related topical information.
I haven’t used Leap our HoudahSpot, but wouldn’t using tags within DT3 accomplish the same thing (minus the bad UI in what is the missing functions of the 3PV)? IOW, presuming DevTech addresses the issues of the missing 3PV functionality, would DT3 accomplish the same?
It would, though I find Leap’s interface a lot nicer to work with in general. A simple example is that in DT everything seems to be grey. It makes it a lot harder to find the right button or icon. Granted, Leap is doing much less, but I find it much easier to work with. Take a look and make the comparison: https://ironicsoftware.com/leap/ If I’m tagging a lot of files, give me Leap’s coloured buttons and tag cloud every time.
Flogging this dead horse, and the point has been made several times in various ways up this thread, but I’ve come to the conclusion that I and others feel that the sidebar is now being asked to do way too much in DT3, whereas DEVON thought it was doing too little in 2.
That middle pane really was the sweet spot for me. It was a visual and psychological separation of top level > mid level > detail. Mashing the top and mid levels together makes it very cluttered to my eye and mind, and fundamentally changes the way I comprehend what I’m looking at.