DTPO, Bookends & ScanSnap

I would like to share with you my workflow using DTPO, Bookends (BE) and ScanSnap.
I have a database in Bookends of more than 12,000 references, 3,500 of which with attachments of different apps (most of all pdf, but also RTF and HTML are present). All BE attachments are in a huge folder without any structure of subfolder. Obviously each document is attached to a single reference, so the data retrieval is committed to BE. I like DTPO in particular for the better management of docs than BE and for the stunning Artificial Intelligence.
Whenever I add a new doc in my BE database I index it into DTPO. I am well aware of the intelligent caveat by Bill_DeVille against indexing docs, but I do not want so many duplicate docs on my HD (I have more or less 3,500 attachments in BE). I think that if the original docs are reposited all in a folder (always the same in the Bookends folder, inside the Documents folder), the issues due to synchronzation and lost links would be minimal (or so I hope).
The scans with ScanSnap is tricky with this structure of files. I prefer to scan docs with the OCR built-in DTPO, which is better than one offered by Adobe Acrobat. The new docs appear inside my DTPO database as a PDF+text file, so I must
• drag-and-drop the file from my DTPO database to my desktop
• delete the file from my DTPO database
• link the file on the desktop to the corresponding reference in BE (BE automatically deplace the file into the Attachment folder into the Bookends folder)
• Index the file into DTPO database
So I have the new doc correctly scanned in the BE attachment folder and the index file into my DTPO database, for in-deep analysis research activities.

Have you some suggestion to boost this workflow?
Do you find out any risk in this workflow?

Thank you very much

I too index rather than import files, both in order to avoid duplicates, but also to organize my PDFs by source and to make them available outside DT (easily found in Spotlight, for example). I use Acrobat rather that DT for OCR, because generally I’ve found that Acrobat produces smaller file sizes.

The recent upgrade in Bookends allows you not only to save a PDF to a folder, but to pick the subfolder in which to save it. This is perfect for me. BE also now allows me to specify how the file is named—I’ve created a special output style “Author, Title,” that I use to rename PDFs automatically (very handy with those downloads that just have a series of numbers for a filename). I now just create the BE reference first, attach the file, and BE does the rest.

One thing in my setup might be helpful for your workflow—I have DT index the PDF automatically, once BE puts it in its proper subfolder. I do this with a DT Applescript called “Action Index.scpt” which is attached the subfolder as a Folder Action. This script was included with DT once upon a time, though if I recall correctly it was orphaned at some point—perhaps the deleterious influence of Bill DeVille? :slight_smile:—and I had to retrieve it from an old backup. For anyone who is interested, here’s the script:

on adding folder items to this_folder after receiving added_items
tell application “DEVONthink Pro” to launch
repeat with theItem in every item of added_items
set thePath to theItem as text
if thePath does not end with “.download:” then
tell application “DEVONthink Pro” to indicate thePath
end if
end try
end repeat
end try
end adding folder items to

To use the script, copy the above into Script Editor, save it to ~/Library/Scripts/Folder Action Scripts. In the Finder, right-click on the folder you want to save something in, and choose Enable Folder Actions. Then you’ll have to right-click again to Attach a Folder Action, and choose the script. Then next time you put something in that folder, it will index it to DT automatically. Just open DT and you’ll find it in Home—you can then drag it to any folder you want. I use this all the time, and it saves a lot of hassle.