Zotero + Ebsco: a match made in purgatory?

I love Zotero, and I use it for all kinds of projects:  collecting background research for patrons’ research questions, sharing citations with staff for ongoing projects in the library, collaborating with folks at other institutions.  Right now I’m using it to collect cites for an article I’m working on with a colleague at Colorado State University.

Sadly, Zotero is kind of flailing, because I’m trying to export records from an Ebsco database and…it’s not working.  Not by clicking the browser bar icon in the search results (list or individual item), not by going to the database’s “Export to bibliographic manager” options and choosing “Direct export” and hitting “Save,” not by saving as BibTeX or generic or XML or MARC21 or any other format, and saving as .txt or .html and then importing directly to Zotero.  Not anyway, not anyhow.  Every single time I try it, I get:

(I’ve already gone through all the suggested fixes for translator issues–it’s not any of them.)

Looking into the Zotero help forums, I see repeated questions from users about how to get Ebsco databases to talk to Zotero, going back two years.  The problems get resolved, then new problems crop up.  Some days it works, some days it doesn’t.  (That’s been my experience too–some days I can export directly and easily from an Ebsco database, some days I can’t.  No idea why.)

Ebsco is kind of a major database vendor, no?  And Zotero is pretty widely used, yes?  Shouldn’t this be a no-brainer?  Especially considering that the database I’m using is Library, Information Science & Technology Abstracts.  Isn’t it kind of ironic that I (an information professional) have spent forty minutes struggling to get this to work?

I’ll be posting to the forums, and continuing to try–if I get this figured out, I’ll post more here.

ETA:  After all that, an update to Zotero fixed the problem.  Sigh. Keeping this post up as a lesson to myself.

UC says no to Nature Publishing Group.

The question is, what will Nature Publishing Group say back?

The UC system has sent a strongly-worded letter to NPG, in response to NPG’s plans to increase journal subscription costs by 400% next year.

Four hundred percent.  Four.  Hundred.  Percent.

These are not cheap journal subscriptions to start with, let me remind you.  In 2007, average journal cost for math and computer science journals was $1313.  Average journal cost for physics was $2865.  [Source:  Library Journal.]

If we even use those outdated costs (journal subscription inflation tends to be in the neighborhood of 10%, so they are outdated) then that would put the average cost of physics journals at $11460.  Per year.  Per journal.

And, as a reminder:  much of the content published in these journals is produced by faculty and students at the universities that are then asked to pay exorbitant subscription costs.  Much of it is funded by federal, state, or other government grants.

I’m tagging this one “social justice.”  And will be interested to hear what happens with The Big No.  (My prediction is that this 400% is an opening gambit by NPG, designed to make their final finishing place look more “reasonable,” although it will probably still be an enormous hike over the current year’s already-exorbitant rates.)

Flickr Creative Commons photo by U-g-g-Boy.

Blogging Code4Lib PNW

Duncan Barth, University of Oregon on Authenticaion vs. Authorization

  • authentication:  is the user who s/he says s/he is?
  • authorization:  what does the user get to do on the site/resource?
  • passwords stink.  hard for end user to use unique passwords, hard for web manager to manage passwords and reset mechanisms
  • stay out of the authentication business if you can!
  • UO intranet:  MediaWiki, Drupal, custom applications.  use Apache standard authentication mechanism, pointing to LDAP server
  • using Shibboleth as authentication system (single sign-off) set up with EZProxy.  ultimately want to make system one log-in for all systems (including ILLiad, Xerxes, etc.)
  • for non-university systems (public libraries etc.) without user id databases, can use OpenID (Google provides OpenID.)  I.e., Basecamp login can be done with OpenID from Google account.
  • how to promote OpenID as a mechanism?
  • can use identity information from Facebook API to log in.
  • OpenID has authorization component called OAuth

Blogging Code4Lib PNW

Mark Dahl, Lewis & Clark:  Developing a Digital Initiatives Program at a Liberal Arts College

  • libraries/special collections transitioning from managing a local collection toward helping users navigate and contribute to global information network
  • less time buying, archiving, and providing gateway to resources (redundant across libraries) and more time in instruction, consultation, managing unique assets (unique activities at individual libraries)
  • thematic digital collections:  academic projects focused on a niche teaching/research area (NY Neighborhoods from Columbia University, accessCeramics image resource (Lewis & Clark), William Stafford archives, etc.)
  • different direction for libraries: moving from institutional collections, uniform practices, to non-standard practices & models, focused collections, consultative services
  • higher ed is seeing centers and institutes popping up around digital humanities projects:  UVa, Kenyon College, Columbia, Hamilton, Univ of Richmond
  • Lewis & Clark started with digitizing student theses, moved toward special collections & archives (ContentDM), visual resources, thematic projects closely related to research & curriculum on campus.  New challenges such as getting grants, hiring web developers, getting artists to contribute images to the database.
  • want to use the expertise from these projects to expand more widely to other projects on campus.  conversations with faculty revealed projects ranging from sciences (spiders) to history (oral histories) to many disciplines.
  • trends that emerged from these conversations:  interest in web mapping (showing historical points of interest on a map).  Traditional scholarly communication still very important, book publishing still more important than digital projects esp for pre-TP faculty.  Everyone wants web design help.  Scientists have have places to put data.  Scientists would like publicity for their work.
  • must choose how to prioritize projects; want to have high impact on student experience (tuition funds work) and faculty.
  • considering levels of service:  consultation only; assistance with discrete project; long-term collaborative effort for larger collection
  • i missed some points here…
  • potential pitfalls:  overloading staff, mission creep for library, unfair concentration of resources, sustainability of resources
  • upside is how wide the reach can be: visits to the site from all over the world; clear and direct benefits (as opposed to the “just in case” attitude of much library buying); increase library’s relevance; increase faculty support for library; directly advance academic mission; heighten institutional profile; highly gratifying; creative work = more cool projects


Question of Identity, Deborah Horrell, 1979
AccessCeramics project

Blogging Code4Lib PNW

Transitioning to Evergreen with Beth Longwell: Sage System Administrator & David Drexler:  Systems Librarian, EOU

  • Transitioning from current ILS Millennium using circ, cataloging, reserves, acquisitions, serials, etc. modules, to Evergreen ILS.
  • Other projects on the go at the same time: Evergreen development, marketing, catalog cleanup, etc.
  • Going live December 2010
  • Don’t want to lose features that patrons like in the current system
  • Will require new PIN for all users
  • Too many to-do tasks to list here.  Some include:  ensure that ll records have 008 & 007 entries where applicable, clean up patron & item records, set up bibliographic & holdings templates, import authority records and test interface with LTI, test how Marc editing works, modify main Evergreen search page for Sage libraries, modify relevancy algorithm to improve order of keyword search results, test EZProxy for authentication, train staff & do mock practice runs, etc.
  • Missing pieces:  remote access proxy (using EZProxy instead), cataloging interface, circulation has fixed due dates, need more sophisticated fund accounting, no serials module, reserves functionality very basic, reports are minimal, patron data is minimal (?), Summer interaction piece is unclear
  • Advantages:  can upgrade functionality without having to buy/budget for a new module purchase (i.e. bookings module); make best use of money, control over and access to consortial data, ability to choose how much support is contracted out, participation in open source project.

Blogging Code4Lib PNW

Al Cornish and Jon Scott (WSU) on LCSH Autocomplete in WorldCat.org:

  • Autocomplete = program predicts word/phrase that user wants.  Ubiquitous.
  • Subject autocompletion in WC based on LCSHeadings
  • When terms appear in the a-c drop-down they’re ordered not alphabetically but by “fitness” (approx. count of # of records returned in WC.)
  • Also available for MeSH and other subject headings systems
  • Test here.
  • Can type a % in that test window to see the macro subject areas listed in order of fitness.  “History” is #1.
  • Populates database using LCSH preferred headings from online list provided by LCSH.

I love this–amazing how much it helps focus and direct users’ searches toward correct LCSH, and helps clear up confusion about what a “subject” is (or, actually, removes the whole fiddly conversation about why subject headings matter) while still using LCSH to structure data, and getting people to the right item records.

Blogging Code4Lib PNW

Karen Estlund, from the University of Oregon, on preserving electronic records.

  • UO president retired last year, so the archivist’s office got a huge number of electronic (and physical) records that are key to retain.
  • Initially, concern was only preservation, not access.  Incoming president has made records management and access a higher priority, so need to respond to that.
  • Email archives:  how to show the chain of communication, who responded to what, etc.  Considering watermark or something similar to make this apparent.
  • Want to make it possible to find records independent of format: paper/electronic.
  • Versioning is a huge issue for electronic records.
  • Trend:  more product, less processing.  Not describing things on the item level b/c of sheer number of records.
  • No clear turnkey solution for these problems.
  • Staffing shortage = not enough support/resources for enterprise solution
  • Have a ContentDM instance for e-record data stores.  But highly labor-intensive to get records up and numbers are too high to use student labor as in the past.
  • Needed new process for handling the data.  Office of President, President’s personal files, and 2 executive assistants to President.  First vetted by executive assistant.  She tagged sensitive files, organized files in structure according to hot topics.
  • Used Excel, Droid, renaming utility (Renamer), Adobe to batch-convert files.
  • Email handling:  looked for extraction tools but didn’t find much that was useful.  Kept files in .pst (open standard) and used Outlook rather than changing files to read-only.  Will use Outlook for viewers to see President’s email.
  • Using Archivists’ Toolkit for versioning, provenance, etc.