Don’t talk to me about citations, what I love about Zotero is that I can write a translator that will extract useful structured data (and perhaps images or snapshots) from any old collection database and add it to my own research library. Match that EndNote! It’s Zotero’s capacity as a research manager that really excites me.
A few years ago I wrote a Zotero translator for the National Archives of Australia’s RecordSearch database. It’s been through several versions and can now do some pretty neat stuff. For example, using it and the Zotero add-on for Omeka, I was quickly able to create this mini-exhibition of some of my favourite letters in the Archives. With the arrival of the web API I can imagine even more exciting possibilities — NAA files have unique barcodes, so… barcodes, smart phones, metadata, digital images, Zotero, join the dots!
More generally, writing the translator really set me on a different path because it got me thinking about new ways of extracting, sharing and re-using collection data. With the web API and translators for archives and museums databases, for example, Zotero could become a platform for ‘routine’ crowdsourcing. Enriched metadata created and shared by researchers as part of their own projects could be harvested back into descriptive systems. Users of archives could create their own parallel finding aids alongside the institutional systems.
But there are some problems. The rigidity of the item types system is frustrating, and there really needs to be some way of creating semantic relations both between Zotero items and between an item and some external entity (it’s been talked about for a while).
I’d like a discussion about the future of Zotero that didn’t get too hung up on citations. A discussion that explores Zotero’s capacity to share, not just references, but research, that sketches some of the apps we might build and the collaborations we might create.
5 comments
Skip to comment form ↓
ajlyon
May 16, 2011 at 2:54 pm (UTC -5) Link to this comment
I’ve been thinking a lot about researchers working with archives to crowdsource digitization. A clean system that encouraged the production of usable metadata, and wasn’t too hard for the researchers, could change the way some archives work. And I think that Zotero could play a real role there– I tried to hack something together using tethered shooting, but I still haven’t gotten to where I’m happy with any workflow.
The API could change all that, since then we can use the bibliographic back-end of Zotero and write the front-end in whatever we please…
Anyway– I wish I were a bit closer to the site, and I look forward to helping make the session’s dreams a reality.
Frank Bennett
May 16, 2011 at 8:52 pm (UTC -5) Link to this comment
Comments in this line are gratifying. With citeproc-js and CSL 1.0 settling in chez Zotero 2.1, there are very few citation issues that can’t be handled gracefully by the system. We see this on the Zotero forums: many issues that once provoked long discussions can now be addressed by a simple link to the CSL specification. With a few modest extensions to infrastructure, we will be on track to build out solid, deep, thoroughly tested support for all major styles. There is work to do there, but not a lot to discuss.
I too am excited about the possibilities for harvesting, crowd-sourcing and recombination of metadata. Like Avram, I look forward to lending a hand, as I can, to help expand the possibilities in this area.
Trevor Owens
May 18, 2011 at 10:27 pm (UTC -5) Link to this comment
Sketching ideas for apps that could be built on top of the increasingly decoupled Zotero system is a cool idea. As I understand it, some of the next phases of work involve the write API and running translators as a web service. Those two things enable a ton of potential.
In the same space, I have increasingly been using Zotero to do data collection. Specifically, collecting 300 of what Flickr says are the most interesting Zombie images on a given day, archiving and annotating web pages as part of more ethnographic research on rpgmakervx.net, or gathering and coding yelp and tripadvisor reviews of the Albert Einstein Memorial. In each case, Zotero’s web scrap book functionality, full text search, tagging and annotating are really cool. Now, I think there is a lot that some of the upcoming work on Zotero could help take these nice data collection capabilities and augment them with very cool data display and visualization capabilities.
Craig Bellamy
May 20, 2011 at 2:00 am (UTC -5) Link to this comment
some good points made here Tim. I will have to use Zetoro some more
Sarah Werner
May 29, 2011 at 4:04 pm (UTC -5) Link to this comment
I’ve never used Zotero as much as I feel like I should, in part because using a citation manager has never been part of my work flow. But, as someone who relies heavily on library catalog records of rare materials, I would love to be able to use Zotero more readily to generate the sort of crowd-sourcing possibilities you mention–comparing the physical characteristics of many different copies of the first edition of Paradise Lost would be really helpful. I very much the idea of breaking away from a citation focus to see what else Zotero can do for us in the sort of ways that you describe.