Unit 13 - Protégé OWL & Conclusion

Discuss your experiences this semester and summarize what you think are the most important concepts and lessons learned.

The Protégé OWL tutorial was a challenging and rewarding assignment. Like many of the units this semester, it made me more confident in the installation and management of my VMware servers. It was finally fun to see a GUI on the server rather than a simple command line. Creating and editing a hierarchical ontology was markedly more frustrating than other tasks, however I believe it had much more to do with the learning curve of the software than the ontology theory.

Like so many units this semester, troubleshooting was as important as learning the software which we surveyed. Rarely do things work exactly like they say in the instructions/tutorials. Having taken some technical writing courses, I can sympathize with the writers of the tutorials, because versions change and steps are changed. In this case, my inverse properties of sibling objects were supposed to automatically update when I created the inverse property of the first object, but this did not happen. I ended up modifying all the objects manually. At least in this instance, it was only a matter of extra work, rather than total malfunction.

My take away from this course is that there are several digital collection management application out there. Each has its own advantages and disadvantages. Some are more socially engaging like Drupal. Others are more focused on one part of the information lifecycle like DSpace is strongest in preservation. All have different documentation and support communities. The only documentation that I hope to never review again is that of JHOVE. No collection management software was without its advantages and any new collection under consideration should be weighed against those pros and cons when selecting software.

VMware, Omeka & Plugins

Although the files are large, it’s possible to create a working virtual machine for download and playback in your Virtualization software (although in practice, it’s almost as much work and sometimes difficult to troubleshoot). In your blog, discuss the possibility of downloading a pre-installed VM versus building your own, from a learning and pedagogical perspective.

I think that Omeka plugins are much more fun than discussing the VM, but I'll touch on both. I think the fact that I preferred to download and install the GoogleTranslate plugin at the command line will give you the idea that I enjoyed setting up and configuring my VM. For known tasks, it is much faster to "wget http://etc" "unzip" and "sudo mv /target/directory" rather than going through WinSCP or other transfer program. By the time I log in to the VM through the program, I could have done it a the command line. Unfamiliar tasks are quite different and in those cases, I would benefit from visually dragging and dropping files (or whatever the tasks might entail). It's just the difference between recalling a command from memory or recognizing it in a list.

Since we did not do our assignments via the download and playback method, it is hard to be fair to both methods. I am happy we built the VMs from VMware and managed them at the command line. I'm no longer intimidated by the command line like I was at the beginning of the summer. The parenthetical, "although in practice, it’s almost as much work and sometimes difficult to troubleshoot" is not selling me on the alternative either.

Unit 11 - IR Home Sites

Harvester and jhove compete for the most mind-numbing software documentation. One long web page of sections and subsections are a hurdle to overcome. Yes the information is there, but most readers do not enjoy a long wall of text. When other software offer at least some screen shots or even video tutorials, these two are falling behind. Omeka is slightly better off in the textual department, dividing documentation into ‘getting started,’ ‘working with Omeka,’ and ‘advanced topics.’ Then there are screencasts that cover approximately the same information, but offers another way to find it for users that might learn better from audio & video. Eprints went the wiki route and features sections colorfully chunked into ‘getting started,’ ‘configuration,’ ‘customization,’ or ‘how-to guides’ and ‘technical reference.’ It is an environment which most users will find familiar and useful. DSpace and Drupal documentation and training are easy to navigate because they generally use a linear list of nodes that can be worked through in order or referenced individually. Drupal’s follows the familiar pattern of guides from installation to theme customization. I appreciate omeka, eprints, dspace, and drupal for all using screen shots that help identify what/where on the software the user is working on. Nearly all of them have mailing lists, discussion forums, or both. I do believe that the usability and currency of all these resources should weigh heavily on the users’ choice of software so they know how quickly or easily they will find answers to the inevitable bumps in the road.

Unit 10 - OAI Harvesters

Try some searches, and also see if you can identify the collections they draw from (there's usually a link somewhere to the list of contributing collections). In your blog, discuss what you think makes a good (useful) federated collection and why and how the service providers you selected did (or did not) create a good (useful) service.

Citebase is seeded by namely the arXiv.org database, a couple of EPrints databases, and some medical databases including Pubmed. The good features of this federated search are that it includes in the results line the number of times the article has been cited. On the detailed item page it lists what other articles cited the current item, easy access to references, and a tab listing similar articles, “cites similar articles to.” Considering its experimental state, the site is in very good condition. The results text could use some formatting to make it easier to read than its current lengthy black and blue text over white background.

8a - EPrints Install

I wish I could have recorded my description of my EPrints install that I told Bethany this week. I was lucky enough to cover a shift at a different library branch and so happened I was able to meet one of my classmates.

As usual, prof. Fulton's notes are very detailed and very precise about how to install step by step. My trouble this week did not come from the installation necessarily, but in the configuration phase. The user profile I created to admin the EPrints repository was without a password. This is documented in the install guide. We were to use sudo su eprints to begin configuration in the eprints directory. Whatever commands I attempted to run, the server kept demanding a password for eprints (which didn't exist). Other students had the same problem (or just confusion?) and documented it in the technical discussion forum. I thank them for that, but where those students had epiphanies, I was still lost. I could understand that eprints is not a sudo user, but that did not solve my trouble that the server insisted on the password. Again, as usual, I decided to start over from scratch the next day to better understand each step I took. Due to luck and persistence, I side stepped the problem and continued configuration per the documentation.

Comparing visual customization against DSpace and Drupal is difficult, since I hardly changed aesthetics on those installations. I can say that swapping the EPrints logo was easy and the documentation on altering html/css in EPrints looked quite straightforward.

Unit 7 - DSpace Cont.

I can see the value that both Drupal and DSpace bring to digital collections. Each has its own drawbacks too. The purpose of a project would determine which one is better suited to that collection. Drupal, for lack of a better word, is more casual. Yes, the administrator can create strict permissions for which users are allowed to post, edit, and delete in each sub-collection. But monitoring posts comes after they have been published to the site whereas DSpace has the ability to configure a structured review of posts before they are made. It is akin to having an editor go over the work before it is posted. This is just one example of how the purposes for which each of these applications were developed differ from one another. Drupal's ability to add thumbnails makes the collections and communities more blog-like. DSpace's workflows, Google interoperability function, and Dublin Core metadata standards which all work "out of the box" remind me more of an online database such EBSCO where context is sacrificed for control.

Since my collection is more like a sample of course documents a French teacher might post for a class and expect students to use as a one-stop place for that course, I can sacrifice the bit-level checksums, file format conversions, workflows, metadata control, and Google scholar "push" cataloging that come with DSpace. Instead, I am gaining a more visually appealing, community tagging, user commenting, and modular application that may take more effort to configure, but ultimately suits my purpose better in this instance.

Unit 6 - DSpace Install

The DSpace install took some time. I was caught off guard by how long the Maven package took to install. I would definitely need to follow a tutorial a couple more times before I felt comfortable doing the install without step-by-step instructions.

I have to say, DSpace was much more functional "out of the box" than I remember Drupal. There was much more configuration to be done with the Drupal install, "go to site building, configure, save, go to site configuration, configure, save, go to site building, etc." DSpace was ready to go with Dublin Core metadata prompts by default. If someone were unable or unwilling to take the time to configure Drupal with all the modules needed to function as desired, then DSpace would be the obvious choice. A user can practically just install it and start adding communities, collections, and items. The new user training units are a great resource for getting started.

Now that I have a small collection of my French lesson recordings entered, I would benefit from going back over the new user training to fill in the gaps that were a little foreign the first time through. Like Drupal, I am afraid that this software takes little time to begin using as an amateur and promises to take lots of time and energy to master. The volume contained in the DSpace Manual is intimidating or it could just be the "wall of text" that is the table of contents.