Urania

A blog named for the muse of Astronomy containing musings by an astronomer

Papers2: Not ready for prime time

Posted on March 08, 2011 by Juan

papers2 really still a betaOne of the programs I use almost daily now that I am on sabbatical is Papers by Mekentosj B.V. Papers really helps you organize your reading, allowing you basically an iTunes-like interface for organizing PDFs.  While it originally seemed to be aimed at medical researchers, it has supported ADS abstract searches for quite some time.  Its ability to store metadeta outside the PDFs in a searchable database made it a powerful tool for me.

For the last few weeks, Mekentosj B.V. has been promoting the release of Papers2 and it has some really interesting new features.  The ones that stood out for me were:

  • Ability to attach other sorts of files (such as ASCII data tables) to a paper in the database.
  • The ability to search multiple online search engines (ADS abstracts and Google Scholar, for example) simultaneously.
  • “Magic Manuscripts” which basically seemed like a citation manager to integrate Papers into almost any writing environment.

This morning, I downloaded and purchased the update to Papers2.  It is clear the developers had some idea of how important Papers is to people in that they made the update create a whole new database instead of wiping out your old one.  You don’t even have to shell out any money, as Papers2 has a fully functional 30-day trial mode.  This means if you are a Papers user, trying out Papers2 is a safe thing to do.  And while I paid for the update (since I trust the developer to improve the product over time), I would suggest you NOT do so until you have throughly tested it, especially if you are a Papers user already.

Papers2 is a complete re-write, so problems were bound to exist.  This is after all a .0 release.  However, in a few hours of experimentation with Papers, I have found it to be much buggier than Papers.  Papers2 is also missing some features I had come to rely on.  The developers are clearly aware of this as one of the web pages they link to within Help menu of the app is a link to their “Future Roadmap,” a lot of which seems geared toward bringing features back that were in Papers.

Among the problems I have experienced this morning:

  1. Activation problems: The license I purchased is supposed to allow registration and activation of Papers2 on two computers.  The activation failed on the second computer and I have not heard back from their (probably overwhelmed) tech support about this yet.  I am sure this will be resolved as they have very good tech support, but it is annoying. [Tech support didn’t respond to my email but a re-attempt a bit later did activate my second copy of Papers2.  I suspect an overloaded activation server. – March 10 3:45 pm CST]
  2. No synching available, yet: I use Papers on my iPad to carry my collection of papers with me on the road.  Apparently, syncing, even though advertised in the app as a feature.  When I reported it, the response was that synching is not available until the next update of the iOS app (which has already been submitted to the Apple for review).  I understand that Mekentosj B.V. can’t control Apple’s App Store acceptance timetable, but then make it clear the feature is not yet available or is pending the release of the new version of the iOS app.  It does say this in small print, but it is not clear enough in my mind.
  3. Missing AAS citation template: Its a small thing, but of the 1400 citation styles built-in in the Manuscripts Preferences, only one is astronomical and it is not one I use.  The American Astronomical Society format (used for the Astronomical Journal, Astrophysical Journal, and Publications of the Astronomical Society of the Pacific as well as many conference proceedings) is documented, I’d like to see it added.  I suspect I can add my own, but it would have been nice for it to have been built in.
  4. Refreshing Smart Collections: I have a smart collection set up to identify all papers not already in a collection.  When I file a paper in a collection, this smart collection doesn’t reflect that until I either restart Papers2 or control click on the smart collection and say “Edit Collection” to re-setup the collection.
  5. GUI bugs: There are various GUI errors I have run across.
    1. Occasional Lockups in “Match to Repository”: Attempts to “Match to Repository” an article usually brings up a one line interface for attempting to match an article’s metadata, but occasionally the program doesn’t seem to respond properly.
    2. Metadata column disappearance bug: I have a reproducible bug [as shown in the video below] where if I use the “View Mode” button more than 3x on a given paper, it locks up with the “View Mode” button grayed out and disabled, no longer allowing you to view the meta data.  Furthermore, it is clear that this bug is stored in a preference file, because relauching the app leaves you with the same disabled View Mode button.  I have reported the bug to the support forums and have at least one other person seeing the bug there.  [WORK AROUND DISCOVERED: I have confirmed that if you move your mouse to the right edge of the window after the right metadata column has disappeared, you have the ability to drag the panel back open as suggested by “Good and Bad”in their comments in the support forum. – March 10 4:00 pm CST]
    3. Failure to reimport article: An attempt to connect a arXiv preprint to the final peer-reviewed paper produced an interesting bug.  The new paper was found, and when I try to add the PDF, it reports it can add it as supplemental material, but all choices I make produce a “error” beep and the program doesn’t offer a way out of this.  Makes importing articles dangerous.
  6. Searching for metadata is crippled: Searching for articles on the ADS abstract server appears to only allow searches by paper author and maybe title (this problem was also noted by Bryan Gaensler).  It is no where near as robust as searching in Papers, which allowed your to indicate the title, author, year, and other pieces of metadata for your search.  That said, this crippled search is on their list of “Missing Features” in their “Future Roadmap” so it is clear they are working on fixing this.  

    [I will note that my critique below, while valid for how poor the search capability is compared to Papers, may also be a bit unfair, since Papers2 has a very different approach to metadata matching.  In the days since the release of Papers2, the folks at Mekentosj have been releasing videos to better explain the new search approach.  There are two videos of note regarding this: (1) a tutorial video here that explains Papers2’s approach to metadata and  (2) a video illustrating how to use specific tags in searches.  I don’t know yet if BIBCODEs can be used with ADS searches, but it is clear that it doesn’t default to this and so it is still not as easy as Papers was in its searching, at least for astronomers. – Added March 13 12:15 pm CDT]

    But this problem also affects the ability to match papers to their metadata.  When the “Match to Repository” dialog is used, its ability to search is so crippled [as shown in the video below] that I have not managed to get and article matched to its metadata yet (since I can’t pull up the correct article to match it with.  This reduced search capability is also giving some people pause on the support forums.  “Jonathan” on that forum page stated something I agree with:

    It appears as though the developers tried to simplify and automate the matching process, but in doing so, oversimplified and compromised a critical function. Although I have been looking forward to the new version, I have had no choice but to revert to Papers1 until this is addressed (hopefully a.s.a.p.).

    “Matias” from Mekentosj B.V. does respond that

    We’re looking into this — degrading the matching user experience certainly isn’t the plan and we will either re-introduce the matching view from Papers1 or improve the automated matching behaviour in an upcoming update (most likely both). Please also keep Papers1 still around and only migrate once you’re happy with the experience that Papers2 gives you.

Because of the GUI interface glitches and the compromised matching ability compared to Papers (version 1.9.7), for now I will be sticking with Papers for my work.  Am I concerned about the future of Papers2?  Not at all! Mekentosj B.V. has had a very good record, at least as I have seen with Papers, of addressing bugs and adding features as time goes on.  I am sure, in a few month, Papers2 will be part of my arsenal for organizing my research reading and writing… but it is not there yet.

Leave a Reply


  • Translate

  • Astro Pic o' the Day

  • Archives

  • Admin



↑ Top