New PDF management software: Papers

Looks quite nice for scientific users out there…hopefully we humanities people will be able to get such goodness someday. I imagine many of its features would be deemed “feature bloat” for our PDF-manager of choice, DT Pro (Academic version, of course)…but here’s to dreaming.

For what it’s worth: Bookends allows you to attach a PDF file, and even renames files from PubMed in user-friendly names (e.g., transforms the PubMed ID to “Smith 2007”). Bookends stores PDF files in a folder of your choice, which makes it easy to index using DT. You can download from PubMed right in Bookends, and Bookends of course also provides bibliographic management (it’s core function). Although I would like to see a cocoa version of Bookends (sigh) I am not sure the benefit of using a program such as Papers when this functionality appears to already exist in Bookends.

You can of course use DT to index the pdfs referenced in Bookends, so you can use DT to search, and also use Bookends for bibliographic management.

And although it’s sort of a convoluted workflow, if I scan something using DT (say a photocopy of a book chapter), I get the benefit of OCR in DT, and then can export it in to Bookends as a PDF with searchable text saved in a folder of my choice, which as mentioned above can be indexed by DT.

For some things I just store them in DT, but if I expect to ever need to use a bibliographic citation, I scan the in DT, then export as a PDF and then create a record in Bookends, which is then indexed by DT. It works great.

Hi, talazem. I’ll bet that what caught your attention was the citation-like display – Author, Title, etc. I know that academics would like to see more metadata of that kind in the DEVONtechnologies applications. (I’ve been an ‘academic’ too, and have published bibliographies containing thousands of references.)

As Eric has noted several times, the developers are thinking about more metadata for the future. No promises on what or when.

Papers is an interesting project. It’s got a built-in search engine for Pub-Med, which is a major source for medical and biological science references. It depends on Spotlight for searching.

DT Pro is a more general-purpose document database. It’s search engine is typically faster than Spotlight in presenting search results, doesn’t distinguish results by file type, and presents the search results inside the database. And it has AI features that I find very useful.

DEVONagent also has a Pub-Med search plugin, along with more than a hundred others. The DA query features allow for very specific and powerful queries. So, for example, one can easily create searches that will pull reference collections by Author from Pub-Med, or by Journal source, allow one to transfer those results to a location in a DT Pro database, and then allow one to track and add future references by Author or Journal, for example, to the collection.

But, as DEVONagent is also much more general-purpose than Papers, it has many plugins and provisions for adding additional URLs so that it can cover the literature in other fields, as well.

Please don’t interpret these remarks as defensive, or dismissive of weaknesses in handling scholarly research needs. If I were doing scholarly literature research I would also be using bibliographic or citation software in addition to DT Pro. Nowadays I do a great deal of literature research, but I’m not constrained by the niceties of citation format. :slight_smile:

But I think the special vision of Christian and Eric has been to help users get at the information content of a document collection, with associated tools for note-taking and for gathering more information into the collection, especially from Web sources. In those respects, I consider DT Pro and DEVONagent to be much more mature and powerful than Papers.

But why not use both? I do a lot of research in Pub-Med.

A future release of DT Pro (which I promise not to talk about) will actually let DT Pro incorporate the same files that have been pulled down by Paper or similar types of software, as well as share other file types that are not currently stored in the Finder, such as text and HTML files. So one could use the citation-type advantages of a Paper application, and the ‘research assistant’ powers of DT Pro without conflict. As a practical reality, one need not wait for that ‘next version’ of DT Pro’s database to do that. If you have bibliographic software that pulls down references (usually PDF or other file types readable by DT Pro), just index those files also into DT Pro.

Hi Bill,

Thanks for the response, as usual. What strikes me about the Papers program is its integration of browsing, downloading, filing, etc., and how it is a step towards integration, between the PDF, the reference info, and notes. That’s what I (and others) would eventually like, either in one package, or through collaboration. But that’s been discussed repeatedly, so I’ll leave it at that.

Besides that, I already am an avid user of the DT + Bookends combo delineated by both you and roberthoodphd above. I’m in the humanities, and Papers is still in the works, so it’s not an option for me, but I just thought other readers here might find it interesting.

But I do look forward to the improgements you mention. As things stand now, there is still some overhead in using the indexing of the BE’s folder (making sure you don’t “double-index” files, having to sort by recent additions, etc. – is there a way of just indexing the whole folder, and having DT monitor it for changes, so that duplicates don’t happen?) Then there is the matter of problems if the file name is ever changeed. But, c’est la vie. It’s workable. Papers is just an inspiration (amongst many) of how nice things could be in the three areas you delineated above if united in one interface: indexing the PDFs, management of the PDF’s, and attaching notes taken on that PDF to it. But I suppose that when manipulable metadata columns and linkable sheets become a reality (sooner than later hopefully), a lot of these issues could be dealt with by the individual user.

Bill, first of all, thanks for this accurate vagueness. I really, really appreciate it.

And let me add that DTPro combined with Bookends (articles are attached to a record in Bookends and then indexed into DTPro), while improvements will of course be welcome, I’m grateful for the ingration I have now.

as a medic i use pudmed and hubmed ( a the moment i am using the DTPro as repository of all the pdf documents.i was wondering if it is possible to integrate the DTPro and Papers in some meaningful way,for example keeping the same pdf in 2 places with crosslinking and so on.

I found this thread searching for ideas how to use DTPro, EndNote and Sente together. To my eyes, Papers looks very similar to Sente (beside the price).


As I recall, Papers stores the PDFs into a designated Finder folder. If so, that folder could be Indexed into DT Pro.

This wouldn’t really ‘duplicate’ the PDFs as to required storage space, as DT Pro would only need enough memory/storage space to display the externally linked PDF.

But that’s the extent of integration. You would gain some advantage in citation information in Papers, and of course in DT Pro you would add the text content of the indexed PDFs to your other content for searching and analysis. DT Pro wouldn’t ‘know’ that the PDF is shared with Papers, nor would Papers ‘know’ it was sharing content with DT Pro. (Unless you add a comment to the file metadata in one or both databases.)

Since my last posting, I have used Papers to learn its features and like it even more.

Individual programs excel in separate features.

The easy way how Papers can download pdf files, integrate them in your collection and rename the file according to you preferences (Author_Year_PMID in my case) is very time saving. ot the quick way how a pdf file can be matched to a PubMed citation.

I think I will still use Sente for preprogrammed automatic searches on specific keywords and authors I want follow and then will export collections in to Papers for pdf archiving work.

EndNote (which btw works very well in its X version on my MacMini OSX10.4.9) steps somehow to the back, maybe just for the actual manuscript writing.

What pushes me in depression after I have spent last weeks more with learning these tools of trade than doing my actual scientific work, is that I can’t find a way how to integrate everything with DTP.

To give you an idea of my workflow which may be I believe similar to many of you:

  • searching the literature (Pubmed, journals) [Sente, Papers, Firefox]
  • collecting the finds [Papers]
  • reading the sources and coming up with ideas [Papers, sometimes Acrobat for better text cutting]

BUT what’s next ?

I admit I saw the videos, went dutifully through the DTP manual, so have rough idea what DTP can do. But how to use DTP for this step ?

I made a new database “Ideas”, indexed the Papers collection of pdfs and hoped that I will be able to do the following:

generate multiple notes, link in them to the specific pdf giving me the idea, review the content of the notes side by side, integrate the ideas into a new note.

But I can’t find a way how to use DTP for this step. it may be that just using Word with EndNote module would accomplish this goal.

Do you have a suggestion ?
Maybe I am rolling back from the steep learning curve of DTP…



Bill…As a medical student I also do a lot of research in PubMed…almost exclusively. I also use many online text resources. The FAQ section at Papers mentions they are not trying to replace Sente or Bookends, but work in conjunction with them. I must be missing the connection. I like the easy UI of Papers and Sente. Papers has some great features for predefined search terms. Do you see the need for both…or would Papers be adequate combined with DTPO? I also have DA, but to be honest haven’t spent the time to learn the UI. DA has a more powerful search engine and is much more mature…but I like the predefined criteria and simplicity of searching in Papers.

So I guess the question I have in all of this is do you think Sente, Bookends, and Papers are more complimentary or more competetive?