Indexing or Importing?

I have been roaming the User Forums for hours, without finding a clear answer to the following question: What are the pros and cons of indexing vs. importing pdf files from a bibliography manager’s database? (I use Endnote). I have read a number of posts explaining indexing of such files. But I haven’t seen a good account of the advantage of keeping the pdf files in Endnote (say). Why not just import them to Devonthink and delete them from Endnote? What’s the downside? (I assume that in the future I would always import new pdfs to Devonthink rather than endnote, but keep my bibliographical info in endnote, unsurpassed for production of final manuscripts). In short, what should I know about the pros and cons of either method? I would be very grateful for your answers!

Hi, Hedda.

Others may can correct me if I’m off but I see it this way: The main disadvatage to indexing is that if you move the file, DT will continue to look for it where it was before the move. If you import it then it becomes part of the DT database (which means it becomes larger) and you will generally need DT to continue to easily access it properly.

Tom S.

I haven’t used Endnote for a while, so I’m not up to speed with latest developments, but the essential thing to check is this: if you import your pdfs into DT, will Endnote still be able to find them?

I use Bookends for my bibliographic work, and I have Bookends set up to download pdfs, put them into an “Attachments” folder, and rename them with the author’s name, the year of publication, and the unique ID number provided by Bookends. If I wanted to, I could then import the pdfs into DT, but as far as I understand it, this would mean that I would no longer be able to read the pdfs in Bookends, or even see which articles had pdfs that I had downloaded, because Bookends would no longer know where they were on the hard disc. To avoid this, I leave the pdfs in the “Attachments” folder, and have DT index the “Attachments” folder, which means that the pdfs are viewable in both DT and Bookends. In my view, this gives me the best of both worlds. Bookends is already set up to search databases like Jstor, Pubmed, Google Scholar, etc., so it makes sense to search for articles and download them in Bookends. But indexing allows me to read them and move them around inside DT as well.

So I would advise you to look at how Endnote handles pdfs, and see if that limits your options.

I hope that helps,

Very helpful! It seems that indexing is the way to go.

Sorry for the remedial question, but when indexing a file instead of importing it, does DEVONthink incur the complete file size or just the size of a reference to the file? In other words, if I add an index to a 1 MB file, does DEVONthink use up 1 MB of storage in its database? It seems like it does — which is a bit confusing to me. See the screenshot at

The reason I ask is that I have lots of large files in Dropbox which I don’t want or need to import into DEVONthink. But I want to use DEVONthink as an “index” into those files. Should I (a) index the files into DEVONthink, or (b) copy the URLs of those files and add them to DEVONthink as bookmarks?

  1. It is reporting the sizes reported in the Finder (as it should).
  2. You can verify the size of the database file in the Finder. Here are the database properties and the Finder size for an indexed database…

  3. The size depends on the content. This DB has 7 text files and one 14.4MB video. If this was a bunch of PDFs (with text layers) or text-based documents, the size would be larger since the content is part of the metadata.