Platforms, Interoperability, and Academic Workflow

One of the things I do to feel productive (read: procrastinate) during periods of writer’s block is try to improve my workflow.  (Another is blogging, clearly.)  This largely entails seeking out and tinkering with various software tools to increase the usefulness and accessibility of the research materials I use to write and teach. An e-library of hundreds (thousands?) of articles and books doesn’t just organize itself, you know.  Plus, highlighting and annotating on my tablet is far easier and more flexible than reading on a laptop. And of course, I want to make sure I avoid duplicating or losing my research work no matter what device I’m using or where I’m using it, so I need good cloud storage/backup and synchronization tools too. So on days when the words aren’t flowing smoothly from my mind to the page, I often find myself in a situation something like this:

Workflow Automation, per xkcd

A propos: lately I’ve been tinkering with combinations of software to achieve three distinct goals:

  1. Organizing my library so old sources and notes are easy to find and new sources are easy to add, wherever I happen to be working on a particular day, without depending on an always-on high-speed Internet connection.
  2. Linking my library to my tablet for easier reading and annotation of new sources.
  3. Synchronization and backup storage to propagate any work I do on one device across the entire system and insure it against loss from hardware failures.

And the most satisfactory software solutions I’ve found for these problems are, respectively:

  1. Zotero
  2. GoodReader for iPad
  3. Dropbox

Of course, there’s a problem: two of these solutions will work together tolerably well, but getting all three of them to work together is a tremendous pain.  And this appears to be by design.  Let me explain.

There is money to be made in software to improve academic workflow. I’m willing to pay for it, and my institution is willing to pay for it on my behalf (within limits). The problem is that many companies offering software that targets different aspects of academic workflow are trying to monetize based on the same, overlapping feature: cloud storage.

Zotero is one of several library- and citation-management solutions (some others, each of which I’ve tried, include Mendeley, Papers, and new entrant Readcube). Of these, Zotero is the most prominent one that is both open-source and based at an academic institution rather than a for-profit company (indeed, academic publishing behemoth Elsevier‘s recent acquisition of Mendeley, which was admittedly never a non-profit, drove me and others away from what had been our preferred research management tool).  It also works within Firefox, which is an especially nice feature, and it has an active community of developers and supporters (including research librarians) that ensure reliable user support, regular updates, and periodic new features.

But even an open-source, non-profit project that relies heavily on volunteers has costs, and they need to be recouped somehow.  Rather than sell its software, Zotero (like many but not all of its leading competitors) has adopted a freemium model: the software itself is free, but certain desirable features come at an additional–and preferably metered–cost. As any tech entrepreneur will tell you, freemium models can be among the easiest to scale rapidly in a way that allows price discrimination–which is key to extract revenue from wealthier or more intense users without driving away other users who contribute to the advantageous network effects of a large user base.  Obviously, the easiest feature of a library management system to meter is library size.

So for a software solution whose primary advantage is the ability to access and manipulate a library of source materials from different places and on different devices, the killer premium feature (from the developer’s point of view) is cloud storage.  And ample cloud storage really is essential to the usefulness of these software packages; without it, I can’t achieve the “work anywhere” freedom I’m looking for.

So I understand why Zotero (and others) look to cloud storage to monetize their services.  Really I do.  But here’s my problem: I already have a far, far superior, more flexible, more powerful, and cheaper cloud storage solution.  It’s called Dropbox.  It works in the background, it works on every device I use, it handles every type of data, and it gives me far more control over how I organize that data.  It even works tolerably well with the pdf annotation software on my tablet.  If I could just put my Zotero library into my Dropbox folder–that is, if the two software packages were interoperable–I’d be all set.  But then, of course, Zotero (and other library managers) would have a much harder time monetizing their software.  So–like Apple, which competes in the cloud storage space with its iCloud service–these library management software providers appear to have no interest in making it easy to use Dropbox or similar commercial services with their software.  Indeed, avoiding the undesirable outcome of paying twice for largely redundant cloud storage services turns out to be rather difficult.

Difficult, but not impossible.  There is a user-developed extension to Zotero, called ZotFile, that helps bridge the divide between Zotero and competing but popular cloud-storage services (hooray for open-source!).  It’s functional, if a bit kludgy.  It has taken me some time and some mining of user support forums to configure everything to my satisfaction.  I have run into several problems along the way. And I’m still not sure if it is going to work in the long run, so I’m maintaining a duplicate copy of my library for the time being anyway. But I am, as they say, cautiously optimistic that my workflow will actually improve as a result of this tenuous, jury-rigged solution.

In the meantime, the entire experience has been a useful illustration of the issues many of my friends and colleagues write about in their scholarship: the power of platforms and the benefits of interoperability; the accumulative momentum of networks and the dissipative entropy of competition; the economic and ideological battles between walled gardens and open-source.  The workaday academic–whose task, after all, is to build bridges between disparate sources of knowledge in order to create something new–seems particularly at the mercy of these forces. Having reached an uneasy truce with them for the time being, I suppose it’s time for me to get back to work.