My final project for KOS is attached in a link at the bottom of this post. It was a really hard course, but I learned a lot, and it was totally worth it.
Something clicked for me in the last week and suddenly possibilities opened up that come with learning about the world of Knowledge Organization that I hadn’t seen until this point. It completely changed my perspective on this class, the field as a whole, and the potential opportunities for really exciting work that can be done with semantics and the next generation of the web. I think I went from believing that this was a really difficult, dry class that I would have to “just get through” and move on FOREVER to more exciting and interesting subjects to developing an interest and passion for the field that I never would have imagined I would uncover in myself when we started this course 7 short weeks ago.
So, here’s what I learned:
- Data is not dry, boring, and difficult to understand. We often miss the human element that links data together into meaningful information that can be used to connect people and research and ideas and causes together in ways that we’ve never thought possible before, and the way to do that is developing a robust semantic web. I have watched Hans Rosling’s TED Talk before (I think it was way back when I first started the UX program last year!), but I didn’t really understand the implications and the underlying significance of what he was saying until we watched it again this week for this class. Suddenly, I realized what he was REALLY saying. And, combined with the second TED Talk we watched, with Tim Berners-Lee all the things that I thought I’d never want to touch again from this class suddenly reminded me of projects I’ve had in the back of my mind and had no idea how to pursue because they seemed impossible to tackle. A web of open, linked data seems to be the answer to make some of those things possible and I can’t wait to discover more!
- I’m not a numbers girl. I love organization and structure, but it’s always been in a more artsy, color-code all my things kind of way. In this course, some of the numbers side of User Experience began to make sense and interest me. Developing a structure knowledge management system for a website requires a more analytic approach than many of the other subjects we’ve covered, but it’s used in a way that makes sense to me. I think this idea is best described by a quote from the course description for module 5: “This approach tells us that we are not limited to causality (the relationship between something that happens, and what causes it); instead we can discover patterns and correlations in the data that offer us novel and invaluable insights. One example related to our current module is association discovery, or revealing the associative relationships between things using Big Data approaches and techniques.”
- We learned about a lot of different ways to map information in this course, and I found that the different structures and ways of presenting visualizations of information became more valuable to me the further I got into the meat of the class. I was particularly fascinated by the Linked Jazz (https://linkedjazz.org/network/) project. I’m a jazz fan. I’ve had several ideas in the back of my mind for projects that would tackle something like this one, but never knew that you could actually sit down and find a way to do it. One thing that I always thought would be fun to do would be to organize as many songs as I possibly could in a linked chain of subject matter/theme. This will reveal a certain level of nerdiness about me, but when I was a kid I was at a pizza place with my family and the jukebox at the restaurant played “Jesse’s Girl” by Rick Springfield, and immediately followed that song with “My Best Friend’s Girl” by The Cars. I decided that these two songs have to be related, and that I wanted to someday link more songs together in a long, giant chain like a story. You could actually do that using an unstructured data-mapping format similar to what we saw on the jazz site.
- Sometimes, your level of experience in a subject is the deciding factor in whether you find a tool useful or not. It was extremely helpful to read other students’ submissions for several of our practice assignments and respond to their work. I didn’t begin to understand the usefulness and powerful potential of the ontology program VIVO until I read my classmates’ submissions, and realized that if you used the tool to search on subjects you already had some familiarity with, the relationships and structures that the ontology made clear suddenly became visible. As a novice on this subject, I may never have realized the opportunities there had it not been for the collaboration with my peers.
- Metadata can be interesting and useful. (Some of you are probably way ahead of me on this one, I realize!) Controlled vocabularies and precision of word choice, how those words are defined and used is something that I hadn’t considered in a thoughtful way before. Our Digital Repository project showed me how these things build upon each other to create a meaningful structure behind the scenes for all kinds of web pages, and the usefulness of metadata, and the kind of beauty of structuring all of those elements well started to resonate with me in a tangible way. I’ve been avoiding developing a robust system of metadata tagging for our projects at work, because I was intimidated by the idea of how it would work, and what we could actually use it for. But now, I’m excited to dive in and make it work for us!
- Creating good search engines is a much more structured, intentional process than it looks on the surface, and that’s probably why people have such a hard time searching for things on different websites. To create a truly robust search function for a site, you actually do have to develop all of the pick lists, synonym rings, authority files, and taxonomies. As we worked on each level of our Digital Repository project from week to week, things that I didn’t understand about the previous level of the assignment became clearer as I built the next level that depended on good work from the first. That was really interesting to me. By the end, I could see how it all worked together.
- Eliminating ambiguity takes practice. It’s not an easy task, and you have to get other people involved in order to do it really well. These tasks are more collaborative than I originally thought when I started the class. I think this relates to many of our other classes, where the concept of “you are not your user” comes into play. You have to consider the vocabulary of your users when you’re building pick lists and synonym rings, and you can only eliminate ambiguity well if you bring users into the process and understand where their experiences will affect how they expect to find information on your site.
- I learned that when you build a database of information the target audience impacts what is included in the data. I was intrigued by the different information we found when creating our Knowledge Graphs. When comparing the data found on Google to the data found on WikiData, you could draw completely different groups of information on the same subject. There were obviously similarities and parallels between the data you could find on a subject, but there was a noticeable “purpose” factor—the reason people search for data obviously impacts what kind of data they’ll find.
- There are great tools to help you get started when you’re creating a good KOS, but they have limits. (Such as Open Calais). Don’t rely on these tools alone. Add your own brain into the mix.
- Knowledge Organization Systems has a human element, and relates to our other courses in a much more meaningful way than I expected. A human must develop the KOS—and each part of the KOS needs a human touch guiding the decision-making and relationship linking in order to realize its true potential as a guiding structure for a robust web. The work that goes into creating a good KOS is much more “alive” than I thought.