Hey guys, thanks a lot for the input. I will give it a try. I start with the time machine backup and as you said the files are still in a folder structure in dt3 I will backup that folders zipped in onedrive and on my nas. I will get my brother ads-2800w tomorrow and will need to set it up first. An open question is still to use dt3 ocr or better abbyy that comes with the scanner. I will for sure generate the databases for my wife’s su jevts but I also plan to have one for the daily family documents. When I u der stood ndr correctly it makes sense to better have separate databases for separate tasks instead of having one big for all?
The OCR in DEVONthink 3 Pro is an ABBYY OCR engine, so it should be comparable.
Regarding database construction, there isn’t a 'one size fits all" answer. Check out Help > Documentation > Gettings Started > Build Your Database for some insight.
My posting indeed had reached the length at least of a novella. Point taken.
This time I try to keep it significantly shorter:
This topic encouraged me to play around a little with the *Daily Backup Archive" Apple Script DEVONthink comes with. And while backing up with it is not perfect it does work, both as a Smart Rule or as a Reminder. So backing up a database as a Zip file to a folder of your choice (could very well be a folder synced by a cloud service) in regular intervals can be automatized!
But I stick to my aforementioned opinion that backing up as a Zip file should be featured prominently and easy to use without any tinkering necessary.
1 Like
Then DT3 is for you!
Saludos desde tierras no tan lejanas 
@ndr42, currently I don’t use any program for backup at home, not even Time Machine as for example it was never has been able to complete an entire copy from my 2017 MBP… and had issues when restoring and/or reinstalling and recovering from Time Machine.
I periodically ZIP my Documents and Desktop folder, add the date and move into my NAS and one of my clouds, but not very frequently.
My current main security copy strategy now is use FreeFileSync, run it once a week and copy all wanted folders into a) OneDrive folder, b) Dropbox folder c) my Synology NAS and an external2TB Samsung T5 USB disk with different policies (for example, in NAS is add, in T5 is clone) from my master iMac. I manually run it Friday night and Saturday morning is done.
@rfog, that sounds reasonable, even if something happens to your hardware at home (e.g. theft) you have a backup. I never heard of FreeFileSync - I will have look.
I tend to forget things that are not automated so every early morning at 4 a.m. Carbon Copy Cloner looks for one of several possible hard drives connected to my iMac and makes a complete clone (takes about 30-40 minutes for checking and copying the changes from my hard disk with about 2,2 TB of data, the first clone takes longer (about 10h)). It really copies everything - I can boot my Mac from this backup disk and everything is the same. If my main hard drive fails I can continue the work from the backup disk.
If you want an entire copy of your Macbook you could try it out - if I remember correctly it has some time of free use (14 days?) and because you make a real clone you don’t need CCC anymore to recover your data, just copy it over from the clone. I have used it for a long time and it had not failed me. Another app I use is ChronoSync that can make entire clones too, but I use it just to sync single folders between volumes - it is a bit more complicated/sophisticated. I heard others recommend super duper for cloning.
I try to advocate for backups in my family for ages but nobody does it - for my kids and wife I automated the backups but as my parents get older I’ve given up - every now and then I collect their passwords in 1password (so that they can call me if they don’t remember them) and let them bring their MacBooks and make a clone…
1 Like
My employer uses CCC for all his non office stuff, and he is very happy. Perhaps I give a try.
(Of course, office/work stuff follows a very rigid backup policy -in Windows. 3 daily incremental copies of source code (10 million of lines more or less) in 3 different disks at different hour each, two times/week full copy of my main disk in two different disks and a weekly full offline copy I bring home each Friday. No clouds here. Plus local GitHub repository).
BTW, FreeFileSync has continuous sync but never tested that. Has command line support as well and could be automated.
Try using ocr on a document that hasn’t been created by your scanner.
Does it throw up an error?
In any case, it’s really convenient to start OCR from Devonthink, (some of this can be mimicked with custom scripting, but only if your OCR program is scriptable).
Before spending a lot of money on a scanner I recommend using your smartphone. I use my iPhone and it is precise, and quick. I can ‘scan’ multiple documents as a group, or scan individual pages. It did take a little practice but is actually quite easy. Also, I recommend you consider whether your wife will use an iPod as well as a computer to access documents (I suspect she may want to use an iPod while at school). If you do use a scanner instead if a smartphone, think about the format that is best for your needs. I find that saving as PDF + Text works best for me because it allows me to highlight when using the iPad (or computer). Not all formats seem to support this.
Never seen that error… And I use DT to OCR little scanned documents, from my old book scanner (running in a Windows XP Virtual machine) and from photos taken with my smartphone from books I’m reading.
For that I use Scanner Pro from Readdle. I’ve scanned entire old books with that, generated PDF then OCR with special tools (mostly PDFpen and Cisdem) and put in DT (I don’t use DT to OCR because it generates low quality OCRed PDF files and I want full resolution facsimiles).
A word on DT integrated OCR: for tiny documents (one to ten pages or so) is perfect. It balances image quality with file size, but with big files it does not perform very well in file size.
1 Like
Some licensing models are friendlier than others. Maybe the version distributed with Scansnap is uniquely restrictive-- and the version bundled with e.g. Epson scanners is more flexible.
Suppose that someone sends you twenty or thirty PDFs without embedded text… It’s relatively simple to batch process them with the built in DT engine. But a third party app might require you to manually perform the same task twenty or thirty times over.
It’s a convenience thing. How convenient it is depends on how often people send you graphics without embedded text.
For batch OCR when DT is not useful to me, I use “Cisdem PDF Converter OCR”, that is specific for batch stuff (it OCR and/or converts to word, excel, text, …). And it is not expensive for the things it does.
In general is the ocr also working with handwritten documents?
OCR of handwriting depends greatly on the quality of print and expected characters, so the answer is: possibly.
THat’s great advice, thanks! DT3 tells me that there is 1 file missing. How can I fix that? If I click on repair, the error remains. Thanks 
You’re welcome
And not to hijack this thread, but check out Help > Documentation > Troubleshooting > Repairing a defective database .
1 Like
Hi @Swoti. You mention a NAS. The easiest thing for you to do may be to set up your sync-stores on the NAS using WEBDAV as your connection. Then whatever backup solution you have on your NAS will take care of your DT data as well. That is what I do. I have a Synology DS918 I use for my data syncing (I’m almost completely free of the Google Universe) and I have the NAS back important stuff up to an external USB drive I keep on a long USB cable so it is less likely to be stolen if someone walks off with my NAS.
The main difference between DT and a cloud service like OneCloud or Evernote is that DevonTHINK allows you to set it up any way you like. You can use any file structure you like, you can put the databases where you like, you can back your data up any way you like… With “open” comes more options, but also more responsibility on your part to make sure you are using your tool in a way that best suits your needs. It’s not a one size fits all like the others try to be.
I have 8 databases open at any given time syncing and indexing somewhere around 45GB of data and DT handles it like a champ. I broke Evernote long ago, which is why I switched to DT and have never looked back.
you can put the databases where you like, you can back your data up any way you like
Except cloud-synced folders for the databases and a snapshot-style backup is best for backups.
With “open” comes more options, but also more responsibility on your part to make sure you are using your tool in a way that best suits your needs. It’s not a one size fits all like the others try to be.
Amen! Preach it! 
