No bad at all. I was joking. Sorry if I appeared to be slapping you on the wrist—not my intent at all!
Don’t sue me!
I have to admit I couldn’t see anything arrogant or dismissive about that reply from you. I find this quite often though here. Sometimes one has to circumnavigate the truth so much that one loses course I find. By reassuring a person they have done nothing wrong as such… Most of my problems with tech, DEVONthink 3 especially have been user error frankly, same with apple over all actually. The biggest user error is not having proper back up. I consider that now a ‘user errror’ and have little patience with it. Same with ‘password1’ for a password.
2 Likes
Your use here is comparable to what I do with my PDF journal articles, Your struggles match somewhat those that I face now in trying to converge documents from three different citation management apps (Bookends, Papers, and Mendeley) into just one database and in finding a robust, reliable workflow to translate the annotations out to just one “brainstorming + reviewing” app (Obsidian).
The one reason that I am using DT is for my need to bring PDF, LaTeX, and other source documents from three decades into a single, trimmed-down collection where the documents are categorized (sorted) by type (PDF, Word, LaTeX, …), theme (assignment, exam, lecture notes, examples …), and topic (e.g. in materials science: composition, bonding, crystallography, phases …).
I appreciate the follow-up. It helps me to know that I am not alone in at least one of my expeditions.
–
JJW
That’s the problem with apps - difficult to find a perfect one. Not a surprise. I have some notes in Apple Notes (historical reason, and works best with pencil & iPad), Obisidian for journaling & research (so customizable & keeps on improving, and tempted but lazy to move everything out of DEVONthink to Obsidian), DEVONthink for storage (good dumping ground for everything but worst editor around for serious note taking especially on Mac; iCloud periodically has issue forcing me to use Bonjour only), Agenda (best for note taking & oh-so-beautiful UI), Draft Pro (best for markdown notes especially with so many actions available). Terrible to have so many apps but each have their own benefits. It’s a messy situation.
I have used DT since an early v2 release, not sure when that was but feels like 10 years ago. At the time was working for a big industrial in engineering management role which required me to support our patent strategies (just support, i am not a patent attorney).
Using DT, i built a database with our own as well as competitors patents grouping and subngrouping along relevant technologies. Tagging each patent with year and company name. With a small script i could extract various trends such as which technologies competitors appeared to focus on. As the database grew, new patents were quickly classified into its relevant group by DT’s classification function.
With ongoing engineering in a competitive space, staying on top of the patent landscape is rather important and that DT database was instrumental answering engineering questions about existing art but also what we might improve on. For any new ideas i asked engineers to describe these in their own words and then i let DT classify it so that i had a chance to review if the proposal was likely already covered by existing patents, or how the idea could be restructured and/or which patents came close and could possibly harvested for some new ideas. Also when working on a new product line, being able to oversee the existing patents can be a great way to create new technologies.
This was such a great tool and allowed for a great workflow throughout the departments that a few of our attorneys switched from windows to mac just to have access to DT themselves.
7 Likes
I am also an occasional DEVONthink (version 2) user, as I previously described here. Like @Bmosbacker and @Jeremy said, “I also use it as a tool, not as a full on repository”, which means I don’t store files in DEVONthink’s repository nor in a cloud service, but instead store files in the Mac file system and index them in DEVONthink. (See Douglas Barone’s classic 2009 essay “File System Infobase Manager” for more arguments in favor of the “good old file system”.)
My answers to @andy4222’s questions:
- “Is it worth the money?” I have found it to be well worth what I paid for it; I think it was $60 then.
- “Do I pay for the upgrade or is it free?” All of the upgrades within a major version were free. I am still using version 2, so I have never paid for an upgrade.
- “If I want to keep using the old version, how would that work…?” It has always been possible to download a disk image of DEVONthink from the DEVONtechnologies website and use it to reinstall on your Mac if necessary. This is how all apps were installed on Macs before the App Store, and it is still possible to install apps this way.
- “How does it compare to Evernote?” I don’t have much experience with Evernote, but an advantage of DEVONthink is that you can store all your files in the good old Mac file system as mentioned above.
- “How is the search?” It’s really great: it is the best way to quickly see the search results inside many kinds of files. Its “See also” suggestions of semantically similar files is also fantastic and a big reason to use it.
- “Is the app slow/bloated?” DEVONthink takes longer to launch whenever it periodically makes a backup of its database as I instructed it to do, but once launched it is a fast native Mac app. I love good native Mac apps, which is what DEVONthink is. Someone else described replacing DEVONthink with a different, non-native (Electron-based) app; I have tried that app, and I think it is the definition of bloated (over 300MB) compared to DEVONthink (60MB for version 2 that I use). An example of one of DEVONthink’s native Mac features is its excellent support for Mac file system tags, which (unusually) it can categorize in its database. Yes, you can categorize your categories in DEVONthink, which is more useful than it sounds.
- “FINALLY, how do you use it? What is the size of your database in it? What do you store in it and how do you backup?” I use it as an interface to my personal knowledge base. The database size (excluding backups, and also excluding the files that are indexed, which as mentioned are just stored in the Mac file system) is about 2GB. You can set how many automated backups of the database you want DEVONthink to make; I set it to make just one backup, and the database and its backup are themselves again backed up during my regular backups of my whole hard disk.
There has been a lot of discussion in this thread about data loss and DEVONthink, so I will note that I have never had data loss; the database did get corrupted once but I just replaced it with a backup, and it didn’t affect the files themselves since I just store the files in the Mac file system as mentioned.
1 Like
I’ll really have to take the time to read through this interesting topic. For now, I’ll just mention that DevonThink was one of (if not the) reason why I made the move from Windows to Mac less than a year ago. And, ironically, I haven’t really used Devonthink yet…
Here are some of the reasons that have contributed to that:
-
It seems to require that you develop at leat a rudimentary workflow/ system of what to store where and how and then stick to it. Fair enough, but for me, figuring out what that workflow should look like seems to require quite some effort of planning and changing long established ways of doing things. I still believe, that DT can make my life easier in the long term, but it requires quite some investment (in addition to the money).
-
I also still find myself using evernote a lot, for the simple reason that I can access it from all devices, i.e. also my android phone and any other computer that doesn’t have DT installed and my DB synced. Maybe the DT server edition (which costs even more) can replace Evernote? Not sure if it technically can and if it can, if it’s worth the money.
-
One specific use case where I had hoped DT would help me was downloading the monthly bills from my online bank (in my country, bills are rarely sent to you on paper but instead electronically to your bank where you check whether its correct and click OK to pay it. No need to fiddle with bank account numbers and what not. Downloading the pdf version of every bill every month is quite labourious because i need to find the right folder for each one. I thought saving it in DT instead of the file system would make this a lot easier but so far I have not managed to tell DT to download the pdf. It always downloads the whole webpage. Can probably be solved, but my point is: its not as easy as you’d like it to be.
-
When it come to personal knowledge management and note taking, I use Obsidian. My idea (based on what other obsidian users do) was that Obsidian is for my own thoughts and DT is for other people’s thoughts. But in my case, those other people’s thoughts are either somewhere on the web (in which case they go into Evernote - see above) or articles and books in pdf format. For various reasons I want to store my pdfs in the file system, not in the DT database, and I manage them with Zotero as my reference manager. No need for DT. But I did index my library folders in DT and I like the way DT shows you similar articles based on word counts. That does indeed yield some interesting and useful results sometimes. But since I rarely open DT, this experience is also rare.
-
When it comes to search, I use Alfred a lot (even managed to get my SMB shares into the index) and while I don’t think that Alfred search is that great (on Windows I used “Everything” by voidtools, which is fantastic), it does the job most of the time. (My biggest quibble with Alfred is that it doesn’t search the entire file path, just the file name.)
Honestly as regards your first point it doesn’t require that at all. DEVONthink 3 can be overhauled and totally reconstructed quite easily. I have done it a few times. I really suggest that one gets stuck in. I am also going to recommend MacSparky’s field guide. I am a long time user of DEVONthink 3 but I found, as I always do with David, that there was stuff I didn’t know. Also it is laid out in a very clear way. I use the Field Guide as a reference already, rather than a step by step introduction which I don’t really need. It is 49$ well spent even for a long time user.
Really though, believe me, you don’t need to fix a workflow at the start. The opposite in fact I would say. DEVONthink 3 will help you if you work with it.
I’d like second that i.e.; no need to fix your workflow from the start. I’ve learned to just start gathering info into DT and slowly bring structure to it as you become more familiar with the data you have (and learn DT). But be prepared to overhaul and refactor your approach once you discover a better way of working with the tool and data (you may need to refactor several times). DT can capture many different workflows, you may find yourself adapting to several for different reasons and for different information streams.
1 Like
My answers to @andy4222’s questions:
- “Is it worth the money?” There is NO question about that, Yes well worth it.
- “Do I pay for the upgrade or is it free?” Again yes, no-brainer for me.
- “If I want to keep using the old version, how would that work…?” Already answered by others
- “How does it compare to Evernote?” It is too long ago since I used Evernote so my experience is most likely no longer relevant. At the time I perceived Evernote as a cute, more light-weight information system which in the end did well at gathering info but not at structuring and retrieving. Last time I used it was 6 years ago so my observation may not be relevant.
- “How is the search?” Excellent search results even when you barely have a taxonomy set up.
- “Is the app slow/bloated?” I never experienced inproportional slowness. I.e., if you have a large DB and run a deep query you may expect it takes some time. My largest database is only 1.1G consisting of mostly PDFs and resolves most queries in seconds.
- “FINALLY, how do you use it? I described my most favorite use-case above but here a few more hopefully it serves to spur some imagination;
- As part of (software) engineering we had to take on large code bases from third parties (part of M&A take over) These could be wildly different; some built through fancy IDE some on shoe string budgets. Having DT index all the source and make-files to make sense of where is what is invaluable. Particularly with code infrastructures with many different file types (not only code but docs and pdf describing bits and pieces of implementations or requirements) and which lacked decent search, DT was instrumental.
- For hardware development, being able to retrieve and keep all used component specifications is a must (to be able to do part replacement many years later when the component is no longer available). Similarly with agency and approval specs the product needs to be tested against - must be kept for many years. DT’s ability to store documents with group as well as tag taxonomy is invaluable to search and retrieve relevant documentation quickly. I used DT’s web server access for the departments needing this info.
- Competitive info gathering is important but boring and laborious. Devon Agent & Devon Think combo is fantastic for following a multitude of competitor sites automatically. New found material (brochures, guides, specs you name it), was automatically dumped in a database which I could then skim for the good stuff and cull the unimportant.
Devon Agent is a bit of a mystery for most of us, sort of DevonTechnology’s step child. It allows multiple parallel searches on a variety of search engines with Boolean logic to specify the search query. Its interface is odd and creating search operations beyond mere boolean queries requires me to study it again every time I use it - definitely this is ripe for some improvements. However, it is very powerful as it is and although google remains my goto search tool, DA is not far behind when a first few google searches yield no results.
Finally, many of DT users are also DTTG users, it sorta is part of the whole experience; I have a love-disappointed relationship with DTTG. I don’t hate it, I just don’t always love it. Certainly the last major release improved it beyond the level of “can’t use” but I do think the developer can do so much more with it. I use DTTG very much as an Info gathering point where as DT is not only info gathering but also sorting, classifying, retrieving, massaging etc. A gathering point must be cheerful, quick and easy - DTTG is a bit too fiddly for me sometimes, it sits in my way, I am fixing to make it work. That destroys the work-flow
6 Likes
Useful remarks. I have to say DTTG is the weak link for me. I haven’t tried it a while though. Is it worth trying again? I own a copy still I think?
If you are not on DTTG version 3, i would certainly look into it if you have a mobile DT workflow/use case.
1 Like
Friendly reminder that DTTG 3 is a paid upgrade, so you need to upgrade to take advantage of the new features if your owned copy is ver 2.x.
Here’s the blog post on the new features and pricing policies, maybe it’d be useful to you:
1 Like
Let’s be honest… the moderator sometimes feels like a hostile lawyer.
7 Likes
Why DEVONthink? I think some good cases have been made.
I used DEVONthink until v3. The cost annoyed me so I attempted to use Foxtrot Pro, Hazel and Finder. this worked to a degree, but in the end had too much friction. I purchased DT3 and have not looked back.
I use DT3 for storing all my information. I tend not to use it for working in, but as a repository of all my data. The handy links allow me to use them liberally in my projects and work, and they even work if the files is synced to DTTG.
I’m working towards a paperless office and without DT3 this would not be possible for me. Scansnap and DT3 is a killer combination. I recently needed to archive a couple years of paperwork. I scanned, DT3 asked me where it should go and what it should be called, I am even able to alter the file’s creation date to the date the document was actually created. I can make notes about the scanned document in the annonation field, tag the document, create new groups or use existing one’s to house the files. I worked through thousands of pages fairly effortlessly. I find without doubt that for me DT3 is essential to going paperless. DTTG is essential to have those files on my mobile device when needed.
I also use DT3 to store all my email. I don’t delete email, I archive it. Too many times I’ve been caught out needing an email I had deleted. I have some 85K emails. I just archive the email boxes daily. I don’t process email in my email client apart from flagging to replying where quick replies are needed. Once flagged, I archive. Once all email is archivedin me email app I import into DT3. I have smart groups in my DT3 email database. One for each calendar year. Another for the last month, another for the last week, one for today and finally one for flagged email. This allows me to access all my email. Flagged emails become part of my work routine and are linked to project files and document, and synced to DTTG where needed. I then delete all email from my email client. At the end of the day there are zero emails in my email client. I manage about five different email addresses. Without DT3 this would be extremely difficult to set up. DT3 smart groups can help me list items for any criteria I choose. Finder is not even close to this functionality.
This is why I would heartily recommend DT3.
8 Likes
In a way, that’s what I have done and every now and then, when I have something that I don’t know where to put, I put it in DT. Maybe the reason I think I need a workflow is that I already have a workflow (big word, but I’ll stick to it for simplicity’s sake) for a lot of things. Like: I’m saving certain documents in certain places.
So if I want to have those documents available in DT, the first thing I’ll have to decide is whether I just index that folder and otherwise keep my workflow or save them all in the DT database and continue doing so in the future.
To make that decision, I need do understand the pros and cons. For the time being, I’m thinking that I will keep large files (audio, video, images) outside the database. The smaller files can go in, but not if I want to be able to access them from different apps via the file system. All my pdf articles and books fall into that category, but all other files, I’ll have to make the decision on the spot and a lot of the time, it’s just easier to stick to the file-system because it feels more flexible. Maybe that’s not a bad thing but it means that most of my stuff, while being indexed in DT, is readily available without using DT (see my previous post).
But let’s say I get to the point that I decide to store a certain class of documents in the DT database. I will now have to come up with a way of grouping those documents. Even if I don’t need a fully fledged system, I need a way of listing all documents of that class (e.g. all electricity bills). When stored in the file system, all of those are in a specific folder. So I need to make up my mind whether I should create a group or a tag or perhaps two (#bills and #electricty)?
Or to use a better example, where I actually expect some real benefits once I manage to use DT for this: everything that has to do with an association which I’m a member of. Currently, if I want to find the minutes for a specific meeting, I can search my emails, because they’re all in there somewhere. The problem is that in some cases I will get several different versions of those minutes and need to make sure that I use the final version. Or it may happen that “minutes” has been misspelled or the final version has been sent by a different person than usual, so that this is clearly not a good strategy. So I try my best at saving those documents when I received them - in a designated folder in the file system.
Now, if I want to move all of that into DT, I’ll have to figure out whether I would just archive all important emails (including their attachments) or just the attachments (e.g. minutes). Most importantly: there are many important emails without attachments. Do I save those to DT as pdf, or text, or as an email?
Of course you can say that I can just do it some way and change it later, if necessary. But that’s a bit like getting a new chest of drawers and throwing stuff in randomly to get ot out of the way, instead of giving some thought to roughly what kind of things should go in which drawer, wouldn’t you agree?
Just to be clear: what I’m whining about are, of course, not about insurmountable problems. I’m just explaining what is slowing down my adoption of DT. I’m happy to be proven wrong.
1 Like
I wonder if you’re overthinking this? DT does not need database like rigidity. Search is phenomenally powerful as are smart groups. For example a smart folder can be set up to find emails with text, “minutes” or “minuets” or however it may be misspelt. This is just the tip of the iceberg. You can then tell it you only want a certain file type from various senders, with or without attachments between such and such a date with this word “xx” near word “yy” and so on. This makes exact filing unnecessary and broader categories completely useable. Of course setting up groups here there and everywhere is just as easy as is adding tags, flags and labels and don’t forget per file annotations and finder comments.
The point I’m trying to make is that if you place it in DT there’s a tonne of ways to find it. It’s never failed me yet.
Let me give another use case scenario. Every year I need to file my taxes and have to find all my amazon purchases. As all my emails are in DT I create a smart group of all emails from amazon. I then add the first month of my tax year. Duplicate this smart group 12 times and change the month from April to March ( my tax year). Now when I go through my bank statement and find amazon transactions, I select the smart group of the month and easily find my transactions. Maybe, I’ll add Very or paypal too. It’s just that flexible.
This can be applied to meeting minutes, project material, people management, correspondence, etc. For me DT helps me manage all my data and slice and dice it any way I want and it’s on iOS too!
4 Likes
I really agree with @svsmailus and can only endorse what he says. I do find emails quite searchable as they are though. I actually use Houdah Spot with DEVONthink 3 for search. It will now search apple native emails if that app is open too. I have been able to find anything I had on my computer using that combo to date. I am not sure of how I set it all up either, but Houday Spot can be set to serach within DEVONthink 3. Some might say it is overkill but, you know, in real terms Houdah Spot is cheap and really, for what it is, DEVONthink 3 is too.
I put everything into DEVONthink 3 now apart from certain categories of emails. Very much like your situatino interestingly. My wife chaired a HOA board, well I must say, for four years using only her iPhone and native apps… Life is a mystery to me really since it is I who can’t find stuff or remember who, where and when…
1 Like
The release notes for DEVONthink 3.8 include this tidbit:
They don’t quite come out and say it, but I have to wonder if this feature would have alerted users to the zero length file error @OogieM and others encountered.
2 Likes
I believe it would have identified the problem soon enough to recover from my multiple backups.
It’s not enough for me to move my main note system back into DEVONThink. I’ve learned that I get far more value out of my notes when using and interacting with them in Obsidian. I enjoy using them so I’m more likely to do so compared to how I used DEVONThink.
I will consider whether to keep my archive of emails (over 81K of them) in it’s own separate database like I have been doing. I was struggling with how to migrate that out and into what tool. For now I’ll run the checksum routine at least weekly and leave them in DT.
3 Likes