Archive for the ‘project management’ Category

Wrapping up

November 16, 2007

As of November 1st the CDI has been flying sans grant money. I was hired onto the project 18 months ago to start up a digital initiatives program and launch a pilot project with a website, preliminary collections and the infrastructure for future growth. It has been a busy 18 months but I think we accomplished a lot, although perhaps not everything that was in the original proposal. This past week has been filled with administrative wrap up details, including writing a final report on our activities during the grant period.  I tend to think more in lists, so here is the wrap-up of what the CDI has gotten done in the past 18 months.

  • CDI office space and digital photography studio designed and built
    • Equipment selected, purchased and installed
  • Backend –
    • Evaluation of available Digital Asset Management Systems (DAMS)
      • Selection of eXist
    • Built metadata administrative interface
      • Dublin Core XForm
      • MODS XForm
      • Solr XForm
      • Authority Control with MADS
    • Creation of metadata workflow (still a bit of a work in progress)
    • Data dictionaries for Dublin Core, MODS and TEI template for transcribed letters.
    • OIA-PMH data provider (with some help)
  • Built web interface
    • Initial design of web interface
    • Site designed and implemented for the EAD collection
    • User testing and design adjustments
    • Implementation of Solr for faceted searching and browsing
    • RSS for news, and search results (still in progress)
  • Steering committee formed
  • Metadata working group formed
  • Content selection committee formed
    • Creation of content development policy and evaluation matrix
    • 6 live collections with around 670 completed records and more than 8000 pages scanned
    • Scanning work on two additional high use photograph collections
  • Several presentations, including an upcoming presentation at the New England Archivists meeting in March 2008
  • A paper in the fall 2007 issue (not out yet) of Microform & Imaging Review

In addition to the final report I’ve been working on getting the MODS editor into production for our next collection. We are trying a new approach to metadata, and bringing in more staff from cataloging to work on metadata creation, so it will be a good test of the forms. I’ve also been exploring what it would take to make the MODS XForm available as open source, it seems to involve some paperwork, some waiting and assurance to the University that the code is not commercially viable. It has been a busy week.

CDI project management

October 3, 2007

I haven’t discussed issues of project management on the blog, largely because I have been spending the majority of my time on programming, and web development. The first few months of work  were devoted to getting a backend in place and setting up workflow to get digitization and metadata work started, after that I moved on to creating the front-end for content delivery. Since the release I’ve been trying to catch up on a lot of back burner and partially complete back-end and front-end projects (MODS/METS editors, authority control, ect.). However, as the period of our initial grant runs out (grant is up at the end of the month) project management issues are becoming much more important, and I’m taking a break from the tech side to work on getting all our ducks in a row for continuing development.

Because so much of the CDI work happens on a distributed bases – content selection coming from Special Collections or the Collection Development Librarian, metadata from the cataloging department etc. – it is becoming increasingly important for me to have a handle on each part of the digitization process, not just the technical side. This includes knowing where the collection development policy is, having ideas about how we can keep content coming in to be digitized even before the collection development policy is completed, understanding what the metadata workflow looks like, and how to keep metadata work moving smoothly, knowing who can do what sorts of metadata work, and how to best utilize the skills of the people involved, and finally, keeping in mind the CDI’s profile in the University with an ear to the ground for any projects that we should be getting involved with.  

For example the library is starting to put into place a programmatic method for accepting and providing access to electronic theses and dissertations. Currently the discussion seems to be revolving around getting a D-space instance to house the project. This is a project I think the CDI could/should play a part in, even if it is only in an advisory capacity. Another example is the digitization of the University Trustees minutes, a project that is being pushed by the University CIO, which Special Collections has been consulting on, and I think the CDI could offer an excellent (and logical) solution to this project.  The trick is how to make sure that the CDI is thought of when these projects come up (which currently it isn’t) and how to make sure the CDI is positioned to take on projects like these. By which I mean that we are staffed appropriately, and that the technical infrastructure is in place so that such projects can be easily folded into the CDI framework.

The most urgent issue the CDI is currently facing is a lack of a collection development policy. The initial grant was fairly specific in its content mandate. We were digitizing congressional materials, which largely came from Special Collections: the Congressional Speeches, Portraits, Letters, and Diary collections.  Although we have a collection development committee, and a policy in the works, the lack of a working policy has led to an embarrassingly long lull in scanning.

The eventual plan is to have a process where collections can be submitted to the collection development team. Proposed collections will be evaluated for appropriateness, and feasibility. Projects will then enter a queue, and scanning and metadata work can happen in an orderly fashion. Currently all of our collection development rests in the hands of one overworked special collections librarian, and when he has other responsibilities the content stops coming in to be scanned.

As an immediate solution to this problem we will start scanning one of Special Collections large historical photo collections of the Burlington era. The collection is a high use collection, and free of rights issues. Because it is such a large collection I think we will use it as a “filler” to be scanned when there is down time, or a lull in other content creation.

My goal this week is to get scanning restarted on the Burlington photo’s, and then move on the working with the cataloging department to get more people involved in metadata creation. Hopefully I can also squeeze in some time to finish up the MODS XForm.

R2 Retreat

August 6, 2007

On Friday the library had a follow up retreat to discuss the R2 recommendations. For some reason the idea of a retreat made me think of this scene in Monty Python and the Holy Grail with the knights all retreating and yelling “run away!”

In reality the retreat was a series of discussions based on a process called “The World Cafe.” There were 10 topics under discussion with three rounds of 45 minutes each. There was one table devoted to the CDI and metadata, which is where I ended up staying for all three rounds. I think I would have enjoyed participating in other conversations (particularly: “Emerging Collections and Services”), but I also think that the CDI currently operates under the radar of many people in the library, and that to have an informed discussion about its future it was helpful to have someone with CDI knowledge at the table. Since Chris was on vacation that pretty much left me.

These were the recommendations for topics to be discussed for the CDI and metadata:

  • Integrate selection for the institutional repository into existing collection development responsibilities
  • Increase expertise with non-MARC metadata
  • Engage UVM administration in a dialogue about university records
  • Enhance OPAC display with reviews and other metadata

I really think the topics should have been rephrased as “what is the CDI’s projected role in the University and how can it grow to meet those needs.” This prevents the CDI from being locked into the dubious distinction of being UVM’s institutional repository, but opens the door to an expanded role outside of Special Collections or the libraries.

In general we had some very interesting discussions. I think because the CDI is kind of low profile we didn’t directly address internal CDI issues much, although we did talk about additional staff, namely a programmer and a metadata librarian. There was a general consensus (unprompted by me, since I obviously have a stake in it) that the CDI should operate as its own department, be permanently funded and be appropriately staffed. There was also some discussion about whether or not the CDI should be a University wide resource, when there are so many library based projects, and the realities of our staffing situation have to be kept in mind.

We also discussed cataloging working with the CDI for metadata, and moving a staff cataloger to the CDI. While moving a staff cataloger to the CDI was an R2 recommendation, it was suggested by the catalogers that staff shouldn’t be moved into the department, rather they should remain part of the cataloging unit with strong ties to the CDI. I think that having staff cataloger working for the CDI but not as a part of the CDI may exacerbate our organizational and management issues, but this could perhaps be handled with better communication, a clearly defined reporting structure, and training.

There was a very interesting discussion about the library cataloging resources that it doesn’t own. Having Collection Development work with faculty to get appropriate web resources into some sort of discovery system (voyager was suggested, but I think as the discussion matured we realized that voyager may not be the only option). There was also a lot of discussion about “hidden” resources on the UVM campus, either faculty research collection such as the Landscape Change project, or data sets like those from the Gund Institute. These resources are hard to find, and the library should be situating itself as a research portal for the University (rather than just an access point to materials owned or paid for by the library) where it aggregates data from or about these resources and helps users find and access them.

This then moved the discussion to the topic of enhanced OPAC display, where I suggested perhaps we should stop thinking in terms of the an OPAC, and broaden our expectations. We should be looking for a discovery tool that intelligent access all of the resources discussed above. The catalog, the 2 special collections databases, the CDI, thesis, campus resources (as appropriate) ect. Even if this tool does not provide access to our article databases it could provide some measure of integration. For example, a result set could be examined, subjects analyzed, and then a list of subject appropriate databases could be presented under a heading “looking for articles? Try these databases.” When a user selects a database their search would automatically be submitted to the database; rather than being dumped out on the start page, they would immediately get a set of results.

There was also some discussion on what kind of service the CDI could be providing the libraries and other university departments and what an expanded role for it could look like. I think we steered away from the term institutional repository, and ended up talking more about an expanded role for the library in general as a “University Community Research Tool.” I think the most interesting discussions at the table happened in a broader sense than just the CDI & metadata issues, and I think these discussions indicate a role that the CDI can play in embracing, encouraging, and providing education about new technologies, and new information management and retrieval possibilities.

One of the ideas not discussed by our table, but that I’m very interested in is the “Discover and Delivery Council,” a recommendation by R2. Unfortunately this idea was generally disparaged, or at least the need for yet another committee was strongly questioned. However, I think that dislike for the idea was most likely a result of a misunderstanding of what this group might do. I see them as keeping up with all these exciting new possibilities that are available in libraries such as: vuFind, PennTags, reviews, tagging, zotero, libraryFind, ect. Figuring out what will most benefit our users and then going ahead and implementing them. I also see the a great need for this group to do outreach and education within the library to make sure our library is educated about what is going on elsewhere in library land and the possiblities available to us, and what implementation would involve.

There was also a comment made during the report back session on how people though that improving the catalog would require too much work, “putting in all those new links and stuff” and we would be smarter to put that work into other areas. This caught my attention because of the way it was stated. I think it shows a general misunderstanding of what it could mean to improve our catalog, and I think it would be really helpful for our faculty and staff to see some demo’s of what these improved catalogs look like, and understand that for some of these open source solutions, much of the hard work has been done. (Yes, I’m still thinking of vuFind, and Antonio’s fairly snappy install time.)

There did seem to be a general consensus (at my table) that voyager may have outlived its usefulness, at least as a front end system. There was also some interesting discussion on changing how we catalog, and perhaps adding more useful information, perhaps gleaned from publisher’s records, such as table of contents, abstracts and such. There are some valid cataloging concerns, but I’m happy to hear this discussion getting started. I think the focus is certainly on the user, and most people are trying as best they can to meet the users needs. We just need more education about how we do that, which I think could/should be part of the Discovery and Delivery Council’s job.

It is unclear to me what happens next, but I think a continuing discussion both face-to-face and virtual will be helpful. Other UVM library opinions can be found here and here, results of the retreat will most likely be posted here.

Project management with Basecamp

July 5, 2007

We have started using Basecamp for project management. Currently there are only two of us using it and we are mostly using it for the to-do lists. So far I like it. It is a helpful way to keep an eye on the CDI as a whole project, sometimes I get caught up in the backend stuff and don’t give the metadata, or collection development the attention it needs. Basecamp also allows us to assign tasks to each other. If for example, Chris runs across a broken feature, or I find some bad metadata, we can just create a new task and assign it to the appropriate person, Basecamp will then send them an e-mail notification that the task has been added to their to-do list.

Project management has been a bit of an issue for us lately. There has been a lack of clarity in who is doing what and how it is being overseen.  I can see how Basecamp can help with some of these issues, particularly in the “who is doing what” arena.  It may also become very useful as we get more people working on the project. For example, it might be helpful in assigning metadata on a collection basis to our cataloger(s)  and also as a way of tracking collections as they move from collection development, though selection and scanning to the metadata entry phase. We could use it to keep track of what projects we have scheduled and who is working on each phase of the project.

If nothing else the to-do lists have been motivational. There is something about seeing all those items with little check boxes next to them that makes me want to get things done.