Wednesday, November 25, 2009

IR+

Well, we did it. We released our new open source repository which was about 3 years in the making and I’ll have to say I’m very happy with the results – we’ve been running the system for about four months now without any major problems! We have had over 290,000 downloads since our August soft launch and that’s not including web crawler downloads. I’ll talk more about statistics in later posts.

We exported all of our 6,500 records from our old DSpace repository and imported them into our new IR+ repository which we call UR Research – you can see the new site here:

UR Research


We did some spot checking and made sure that we maintained all of our data. We now have features we always wanted as part of the system including: Authority Name Control, Researcher Pages, Multiple File Uploads, Statistics, Simpler Collection Model, Collaborative Authoring Workspace, Contributor Pages, Embargos and an easier to use interface.

We are testing the software on ourselves and hope to create an open source downloadable version in the next few days. It’s always good to run a system in production to make sure everything is running well.

The best news is that our existing users (Administrators, Faculty and Staff) have consistently told us it is much easier and more fun to use. For the first time we are getting asked by our faculty and graduate students about our repository and how it works.

This was a huge team effort, and it could not have been accomplished without the hard work and dedication of everyone involved.
Posted by at
Labels: , ,

2 comments:

  1. Hi Nate,

    I looked at the IR+ install for your school, and it looks great. I was just wondering if you could elaborate why a new IR was built from scratch vs. sticking with DSPACE. Also, is there a way to get in contact with you?

    Thanks,

    Joshua

    ReplyDelete

  2. There were a lot of factors that played into the decision of building a new product, and it was not taken lightly.

    Here are some factors that led us to our decision:

    DSpace had no concept of a personal file/folder workspace, file versioning, authority name control, data normalization or statistics when we started. It would have required significant changes to make it meet the functionality we wanted to provide, but our user research was showing that these were key functionalities that were necessary for the success of the repository.

    In 2006, when we were ready to commit to this significant functionality change to our IR, DSpace was currently undergoing some major changes with its community and software. It was deciding between a brand new 2.0 version versus going with a slower upgrade path. Without knowing clearly the direction DSpace was going, we felt that building what we really needed was the only self-determinant way forward.

    We felt the best path would be to create a project where we had full control so that we could provide a working piece of software to the community that could be used as a proof of concept for the features that our research found repository users needed.

    We have been very open with our plans and kept the general IR community abreast of our intents. We have talked to both DSpace and Fedora about our project and goals. We’ve done presentations at CNI and I visited RepoCamp at the library of congress in 2008 and told them about our goals. It was actually mentioned in this blog:

    http://pixelatedpete.wordpress.com/2008/07/30/repocamp-library-of-congress-washington-dc/

    We continue to look forward to sharing all of our concepts, research and software that other repository projects feel would be valuable to them.

    If you would like to contact me you can reach me at nsarr- at - library.rochester.edu

    ReplyDelete

Subscribe to: Post Comments (Atom)