Create an open access repository

Joe and I are overhauling the Technology & Social Change Group website. I took a step back this week to think about what’s most important for this first version, and how we’re going to transfer over our existing content. I’ve dubbed TASCHA website 1.0 the “does-not-suck version” in order to keep us focused on the basics, pull together all of our content, and push discussions about feature requests to the point in time where we have something up that works and something concrete to react to. Here are my conclusions:

  1. The backend — Clean WordPress install, giving our researchers the ability for our folks to post their own content. They should not have to fight with the system.
  2. The audience — Donors are our most important audience in this round. We have to be able to tell our story quickly — who we are, what we’re good at, and what we’ve done — to prove that we’re a good investment.
  3. Show off our stuff — We have a lot of great research assets: reports, evidence narratives, briefs, datasets. Make it easy to connect those to people and to project. And make it easy for folks to access them.

Originally number three was me thinking about the best place to store our pubs. Adding them as separate WordPress posts did not feel right. Not robust enough. All sitting in folders by year and month. No no no. Then I thought I’d put them on the iSchool server in a folder called resources. Nope. Too rudimentary. Then I remembered: You are not the only person to have had this problem. Get on google and poke around. Do your homework. What I found is good, really good.

I started by looking for something that would play nice with WordPress, which led me to Joss Winn’s Displaying a dynamic publications list from a repository on a staff profile page.

This in turn tipped me off to the whole world of cloud repositories. Of course. Duh! There are people out there and all they do is manage collections: repositories. Increasingly digital ones. So… from here I found EPrints. Created by the School of Electronics & Computer Science at the University of Southampton, it appears to be a wonderful open-source application to manage, specifically, open-access repositories. They  have a cloud version and a version you install on your own server. (Did I mention that open research is important to us? Yeah. Well it is. Remember that.) So they are cool three ways:

  1. They’re a little consulting shop in an academic department. I’ve worked for two of those. We can be friends.
  2. They’ve created an open-source app to manage information.
  3. They’re promoting open access.

Plus they were super super well reviewed in this overview by Neil Godfrey and this survey by the Repositories Support Project, an initiative of the UK’s Joint Information Systems Committee (JISC) to support the development and growth of a national repositories network. I called EPrints and they were super nice. (JISC, by way, also has a project called Dura, to embed institutional deposit into researchers’ academic workflow. Cooooool…)

MIT’s DSpace also looks pretty cool. IDRC uses it to manage their digital library. I’ll look into them before deciding. But my view with this sort of thing, increasingly, is this: Don’t spend months trying to find the perfect tool. Find a good-enough tool with an import/export feature and get to work!

In February 2010, JISC and Eduserv held Repositories and the Cloud, an event to discuss the policy and technical issues associated with cloud computing and the delivery of repository services in UK universitites. Adrian Stephenson took rocking-good videos. Watch a few to get an sense of what’s going on in this space. (Also note how they used TwapperKeeper to archive and summarize event-related tweets. More on social reporting.)

Okay, so what’s exciting about this and why should you care if you’re in development?

There are several massive repositories now, mostly stovepiped by donor (topic for another post). This has to stop. The model now should be aggregation. The open data model serves us well here: Standards are important. Build systems that can talk to each other and share with each other. Imagine if we could aggregate knowledge assets from a whole host of places.

So why should you care if you’re a development research group or consulting firm?

Or any organization that produces a lot of publications or digital assets? Ohhhh, here’s where it gets really exciting. Some ideas:

  • People profiles that automatically update when you post new content to the repository. This is good good news for all you academics out there with beautiful long publications list. You’ll never have to update it again!
  • Project descriptions that automatically display related research/project outputs, generated based on fields you’ve defined (see TASCHA’s draft of research project chunks).
  • Track downloads of each asset (example).
  • RSS feeds to display on various parts of your site.
  • RSS feed to share with anyone else, so they can display your latest resources.

See where I’m going with this? Will keep you posted.