I have a scanner now in which I scan lots of articles and notes. It is useful to have them scanned straight to DTPO, but it raises one question. I used to keep my PDF’s in one big folder, alphabetised by author (in file name), and index them in DT. That way (a) DT’s size was kept in check, otherwise it becomes too slow, and (b) I have a safety copy which I can easily copy to my laptop with the same folder structure.
Now, however, I have to export them again out of DT, and remove them from DT, then locate them in my PDF folder to add them to the reference manager (Bibdesk in my case) - extra steps I’d like to avoid.
If there was an option to OCR and Index a file into DT, rather than wholesale importing, that would help. Even better would be some reference management support - I realise this is asking for much, but I am sure that there are many people in the same boat. DT does not replace reference management, so how does it facilitate an easy workflow?
I personally prefer to import all docs into DT. But I think that better collaboration between DT and reference managers (Sente, Bookends, BibDesk,…) is more and more important. I’d like to have a way to easily create a new reference in the reference manager from a document in DT: a script that created the reference and automatically linked the pdf/doc to it. The reference manager should know how to go back to the document in DT (a local url) and DT should link back to the reference. You can now do most of this by hand, with some quirks, but it’s too time consuming for me.
I think we’ve got some of pieces of software that do nice and useful things in this area (DT, Sente-Bookends-BibDesk, Papers, Yep-Leap,…) that may either evolve overlapping (reference managers start to manage pdfs and notes, pdf/paper managers start to build reference management, etc.) or by working together nicely. Unless someone is able to create the perfect all-in-one super-duper research and writing workbench machine (without becoming bloated), I think it be better to go the second way.
My Research database is already 9Gb, and noticeably slower than others which are between 1-4 Gb, which is the reason why I try to keep it slimmed - hence index rather than import. Otherwise it would be much easier to keep it all in DT.
I fully agree that a one-stop solution is unlikely, perhaps even undesirable, but some sort of collaboration should be possible. DT, which its incredible Applescript support, is what I pin my hopes on. I find it more easy to locate specific entries in Bibdesk than in DT, but DT finds further references, so both are helpful.
Update: I realise now that I can use Abbys Finereader to scan directly to the folder in which I keep the PDF’s, and have that folder indexed by DT. So that is one problem solved—especially since Abbys Finereader is, according to DT, of higher quality than IRIS.
I have been trying to get Sente and DTPO to work together and the best I have been able to come up with is Indexing the Sente document (with the Index script). A plug-in for either, to make DT “look” at the xml structure of Sente storage or vice versa, would be brilliant!
Because my version of Finereader does not work as well as IRIS (pages rotated in different ways, which is very rare in IRIS), I returned to scanning directly into DT.
