DTP Office as a replacement for ALL my folders/files

I am considering setting up multiple DTP databases to ingest all my various documents and other bits of media. Since DTP leaves files intact for use in their native apps, I am thinking of moving everything into various databases, archiving off all the files in folders on my hard drive and just using DTP as my go to repository for my documents.

I realize many people do just this and that this is the power of DTP, but was wondering if anyone has completely gone over to DTPO and ditched their traditional file and document setups but saving them into archives for backup.

Is this a good idea? Any experience doing this? Is it better to just Index my folders into DTPO and keep them on my drive? Is there a good reason not to do this?

I thought it would be great to have everything in DTPO where I can find linkages, annotate, quickly search, etc. rather than dealing with everything through the Finder. Right now I have some things in DTPO, but the majority of my files live outside of the app right now.

Thanks for any advice or suggestions!

Try a search for finder replacement and you might find some discussions related to what you’re asking.

It’s your data. Are you willing to lose it all? :frowning:

I’ve been contemplating the same thing, actually.

For my workflow, I’ve achieved somewhat of a happy medium by having DTPO index (and subsequently synchronise) a few key ‘research’ folders that exist in Finder.

Be curious to hear what others think.

I would definitely have backups…more than one of the data. My goal is to just streamline things so that they all live in DTPO where I can see linkages and find things quickly. I can also just take a database on the go with me without having to worry about copying this folder or that file, etc.


Given the newly added feature ‘Smart tagging’ I’m considering the same. I have been tinkering with ‘tagging software’ for a long time now since ‘nested folders’ is a very impractical approach nowadays with these heaps of data coming in every day and it forces you to cling to a fix structure setup some time ago which is difficult to change altogether if you follow a certain logic with your ‘nesting’.
Besides sometimes you have to store certain files in more than one place/ folder if you want to create ‘logical links’ for certain docs. And finally - the search of Metadata in OSX is still very inconvenient (e.g by Spotlight) and doesn’t live up to my expectations a all.
Hence DTPO has none of these shortcomings but on the contrary a couple of asses in its sleeves making the storage and finding of data very convenient and lightning fast - I guess the question to migrate ones data to DTPO in general is very well justified.
I guess the main concern migrating ones data to DTPO seems to be that this would create a new System in the (OSX) System and having them stored in the native OSX environment seems more reliable compared to the quite new DTPO System. Well, here is what I thought why there no real risk involved moving ones Main repository into the (not really closed) DTPO world instead of keeping files in the native OSX environment and accessing them by the Finder/ Pathfinder:

  1. Set up several databases covering the main topics you are storing files for (e.g. work, companies, private, Banks, Invoices, IT Stuff, etc.)
  2. tick the box ‘exclude groups from Tagging’ in Database properties
  3. You can save files directly to DTPO Global Inbox from the save-menu, tag them there and move them to the final database where you want to store them.
  4. Access the files now neatly tagged by Launchbar (viewtopic.php?f=2&t=5610&p=45050&hilit=launchbar#p45050), in the DTPO window itself or by Spotlight (even without proper tagging-search capabilities).
  5. Since DTPO supports OpenMeta there is no risk even if one should decide to repopulate the Mac OSX Finder again (or its successors with Tagging capabilities ) one day in the future. The tagging wouldn’t be in vain (most likely)
  6. All the files stored in DTPO are stored in its original file-format in the DTPO file easily accessible by viewing the package-content of a DTPO database.

Well, lets summarize:

  1. No files will be converted into any weird proprietary file format but remain in the format you have chosen with the app you were using (eg iWorks, Photoshop etc.) to create it;
  2. Provided a proper backup routine is in place, all the files in their original format will be backed up regularly and can be accessed by ‘viewing the Package content’ of any database.
  3. If one is not satisfied and would like to go back to the ‘nested Folders’ System and Finder (or anything similar) - no prob. Just set up the Nested Folders structure and put the files in (o.k that would be some effort but who would do that anyway since tagging files is so much superior to nested folders.

Conclusion: There is no risk involved, since by storing the files in DTPO you literally just wrap them up with a DTPO pod keeping the original files; the only thing you’re losing is the nested folder structure (but even this can be achieved if somebody wants to stick with ‘nested folders’…: q.e.d. :wink:

P.S. I know, people working with version control software (like subversion of Flow) who won’t agree to the statement above since Flow wouldn’t work this way, but I guess there is no free lunch isn’t there?

Thanks for this detailed overview Wallenberg. Much appreciated. This does look like a feasible implementation. Like you, I have been messing about with tagging for awhile on my Mac. Now that DTPO supports tagging and OpenMeta tagging in particular, this becomes a great asset for considering DTPO as a finder replacement or as a storehouse for all one’s files.

As you mentioned, since DTPO maintains file integrity of stored items, it makes it less scary to put a lot of information in it, knowing you can get native files back out when and if necessary.

I am tempted to try this out.


As I said, there is no risk of loosing data involved as long as your backup routine stands. The only thing you are loosing is the nested folder structure but since that is no loss at all since tagging provides a so much faster access of the data searched for, it is actually a no-brainer. IMHO - the days of finder are counted anyway since this particular approach doesn’t coincide with today’s requirements anymore. Tagging rules!!!

I’m new here. With very much interest I studied the messages.
I’m testing DT.
I’ve put into DT about 17.000 files in a database.
Now I would like to find the “Programm” of the “Mitgliederversammlung” od “2008” of the “xxx” organisation.
I didn’t find any way to look for it!
Must I create all the folders (or “groups” - is it the same?) as xxx/2008/mitgliederversammlung for to put the programm in it?
But the question of a quick search rests.

Another question: Why does DT try to connect to google, when I open my database?

It seems that the document-format as odf and ods is not supported.


A couple of years on and especially in view of Mavericks and its tagging system, does anyone use DTPO as a finder replacement?

No. And I never would. Less than 10% of the documents in my working drive (500 GB SSD) are inside DEVONthink. Finder is deeply integrated with the OS, especially Spotlight and Mavericks Tags. DEVONthink is not. Everyone’s software is built to work with the filesystem mediated through Finder or its services – no one’s software except DEVONtech’s is built to work with DEVONthink. Though DEVONthink is open 100% of the time and I use it all day every day, it is a lot easier when working with programs to interact with the filesystem than with DEVONthink. Without special fiddling, there’s one gateway from save dialogs into DEVONthink - the global inbox. Without opening documents from DEVONthink there’s no way to use any software open dialogs to select documents inside databases. DEVONthink excels at what it does best: manage the information I put inside it. It could never replace Finder, though, IMO.

Like korm, I do not use DEVONthink as a Finder replacement.

There are many files on my Macs that would add no value to a DEVONthink database, for example zipped files, Microsoft Office user data, applications (which cannot be added) and the vast majority of my image, audio and video files.

I have a number of DEVONthink Pro Office databases that I created to meet specific interests and needs. I don’t put all the files on my computers into those databases.

My own preference is for self-contained databases. I almost never do Index captures, in order to minimize problems in moving databases among my computers, or sometimes running them off an external drive mounted to the computer I’m using at the moment, without worrying about ‘missing file’ message that would result if dangling external files had their Paths broken.

When I Import files from the Finder, I’ll delete them from the Finder, or archive them to an external drive. Literally, I no longer care about those files.

But I’m a stickler about data integrity. I keep the operating system clean and error-free. Every two or three weeks (or immediately if something flaky happens) I’ll run an OS X maintenance utility (I use C0cktail, but others such as OnyX can be used) to clear caches and do a number of other procedures that help keep OS X clean and efficient, and then restart the computer. Once in a while I’ll run the Verify Disk routine of Disk Utility, to check for any possible disk directory errors, and repair them if found.

I use a multilevel backup system for my DEVONthink database, which contain the most important files on my computers.

First level: DEVONthink’s internal Backup folders, which are set for daily backup, and which I sometimes manually invoke when I’ve done a major surgery on a database and verified it is error-free by running Tools > Verify & Repair. On occasion I’ve been known to do something dumb to a database. The Tools > Restore Backup procedure allows me to return to an earlier state of the database. Any new content I’ve added after the date of that internal backup will be available in an Orphan folder, after I’ve run Verify & Repair. But this first level wouldn’t protect me were my hard drive to crash.

Second level: A backup system that stores my data on an external medium. I use Apple’s free Time Machine to automatically backup all changes to an external drive. Now I’m protected against a hard drive failure. My preference is for backups that I can control, so I don’t like cloud backups of my databases. A related issue is that it would take a long time to recover them from cloud backup. Because Time Machine would faithfully backup my databases even if they were damaged, once in a while I run Verify & Repair to ensure they are in good shape. But this level of backup wouldn’t protect my data if a burglar stole all my computer equipment, or in case of a fire.

Third level: Periodically, I backup my most important databases as Database Archives (the smallest possible zipped versions of databases), which are updated to a portable drive that I store in a bank safety deposit box.

I’m in my 12th year of using DEVONthink. My databases have been stable and error-free, and they are better protected against loss than any other content on my computers.

What criteria do you use to decide what goes into DTP and what stays on the Finder? I think this is what bothers so many of us - having two places where documents live, and trying to remember which is where. I’d love to have just one system that does it all.

I’m an academic researcher, so I put a lot of my research and source material in DTP, so I can use Smart Groups, tagging and AI. I also use it as a general web clipper. But then I keep some stuff in the filesystem because I want to access it from other computers, or I want to access it quickly via Alfred or Spotlight. That’s a rough divider for me as to what goes into DTP and what stays in the Finder.

I’m curious how you divide up your stuff, since it sounds like DTP has a very specific purpose for you.

I have six working databases now, and several dozen that are retired because no longer needed. Three of them are for work products I develop under three major contracts. These databases are several years old – they grow as new tasks or work products are developed for this or that contract. I index anything I need to share with others through some cloud or another, or anything I want available on iOS. (I don’t use DTTG anymore.) Two of the other databases are for long-term research and writing projects. The sixth active database is the Global Inbox, which I use as a dumping ground for things that need a bit more organization than Finder.

I frequently make databases for special projects – all supporting data for tax returns, for example. Or research for travel and field work. When the project is done, the database is closed and forgotten. I always set up a Spotlight index for databases, which makes the content locatable even if the database is closed.

The rest of my data (about 3 TB) is distributed among (1) original media organized in libraries managed with Lightroom, (2) certain important archives carefully organized in hierarchical structures with Finder, (3) email archived in MailSteward databases, or (4) stuff thrown into piles indifferently organized because I don’t need regular access and when I do want to locate anything Spotlight is more than adequate.

It doesn’t bother me to have different storage methods. I have different categories and uses for data and merely use whatever methods seem right for the purpose at hand. The only times I lose or can’t find a file is when I never bothered to save the thing in the first place. Otherwise, things just take of themselves,

Like korm, I don’t capture all my files into DEVONthink databases. I have a number of DEVONthink Pro Office databases that meet particular needs and interests. I place files that relate to those needs and interests into databases because that environment increases their utility for me. DEVONthink provides tools for organizing, searching, linking and annotating them. Especially in databases that I use for research, I find the See Also and See Related Text assistants invaluable for exploring ideas and their variations in the literature.

I get a lot of mail. When I go though it I immediately discard junk mail. I’ll scan into a database items such as bills, receipts and others that will be useful in my financial database, correspondence related to projects and so forth that I need to keep. I may treasure and keep some items such as Christmas cards and family photos, but those probably won’t be scanned into a database.

I get a lot of email. I immediately delete junk email. Most of my email is at best of transitory interest; I’ll look at it and delete most such items. Some of it is important for addition to my financial database, and is captured there. I keep email correspondence from friends, and have a database for that purpose. I’ll put business related email into an appropriate database. But I don’t use databases to collect trivia.

I spend a good deal of time looking at information resources on the Web. I subscribe to online journals and often run DEVONagent Pro searches related to my continuing interests or to a current project. I’m always looking for additions to my research collections, but don’t dump everything I see into them. Years ago I tried RSS, but stopped when I realized that the amount of time spent in going though floods of data that contained only a small percentage of really useful content was inefficient.

There’s a lot of truth in that old Yiddish saying: 99% of everything is drek. I don’t see any purpose in adding drek to my databases, unless I add especially notable examples of junk science or bad academic writing to a database created for my amusement. :slight_smile:

The databases that I use for research have been evolving for years. I’m continually adding new content to them. But I’m continually pruning out obsolete or useless content, as well.

I have played with scenarios along these lines over the last year or so.

I have a DTPO database that holds my paperless household document management & hobby resources of some 8000 items

Included in this database:

Bills, receipts, warranties, manuals and other documents all OCRed PDFs.
Emails are achieved here from mail.app
Clipped images, web articles clipped to PDF are included
Documents (that can be viewed with Quicklook) are also held.
This database now also holds my RSS feed archives with over 15000 articles going back 5 years.

I originally had all of the above items imported into and stored internally in the database in a logically structured sets of groups. This all worked well for me especially the Classify functionality. I would say I had pretty much superseded the finder for my document management.

Recently I wanted to embrace the enhanced file tagging functionality of OSX Mavericks, use Hazel app and use spotlight tag searching more rather than opening DTPO.

So I moved all document types except emails out of the database into a single folder, with this folder indexed back into DTPO.

I found that I can make greater use of the hazel app to tag my documents on arrival on my computer. I can maintain tags both inside DTPO or on the indexed file. I have found that tagging gives me greater detail than the structures of groups in DTPO or a hierarchy of folders in OSX Mavericks.

I am not certain if I will continue with this setup or change to something else in the near future.

I suppose that is the power of DT - you have the flexibility to order your document in so many different ways :smiley: