This class has been a wild roller coaster. I think the most eye opening aspect of it was realizing how pervasive and wide spread is the world of metadata. Metadata is used to define and control our world through different schemas, but the underlying aspect is the same. It makes me want to start organizing my entire life by metadata and realizing the far flung impacts that metadata has had and is still having on our world. I think that is why I like Dublin Core so much. While it is a simple enough schema to learn and employ it is so customizable it can be tailored to nearly any collection. It was fascinating to see how many times it cropped up in our descriptions of digital repositories or even as some sort of basis for other schemas. Managing metadata is definitely a skill worth learning.
I’m not sure if it is because of my lack of knowledge concerning football, but I found reading the scoring summaries and all that information rather difficult. For instance I was trying to identify a play by the time stamp given in the properties metadata of the photograph. The time given was 9:55pm. So I began searching the summary for a 9:55 time stamp. However, the information I found conflicted with the name and player number of one of the players making an action in the photograph. It took me a while to realize while the information in the photo was a given time, 9:55pm, the times listed in the scoring summary were relative to how many minutes a play occurred into a given period or quarter. After this realization I decided to go with the more concrete information of the player name and number to try and identify the actions in the photo.
As I have currently been spending a great deal of time staring at pictures of football and indexing these images and arranging metadata about them, it got me thinking. I know many sports fans have game updates sent to their phones when they can’t watch a game. It got me wondering about the metadata used in such events. I assume a text is sent every time a team scores and at other pivotal points. Would the metadata include the players name and team number? Is that information governed by metadata schemas created by the leagues such as NFL or by the phone and distribution companies? Are these updates streamed on various television channels websites such as ESPN and is the metadata schema visible there? I don’t know how it would work, but it makes me curious.
I was surprised which elements were the most difficult for me to index properly. When I first looked over the list a couple immediately jumped out to me as stumpers namely the Description, Subject, Format, and Contributor elements. It was interesting as I worked through the process which elements were actually difficult. Elements such as Format and Contributor were actually fairly simple as I just had to locate the correct information from the properties section of the image and copy over the information. Description oddly enough wasn’t as challenging as I expected. I think this is completely due to the fact that I indexed that element after indexing the Subject and Title elements, and Description is basically a combination of those two elements with more detail. The Subject element was difficult although thankfully not as impossible as I feared due to the simple vocabulary provided by my class mate. I was probably the most stumped by the Player Name element as I couldn’t find information for the Southern Mississippi team in 1975 and I had several images from that season. This was information I couldn’t try to intuit and had to end up leaving blank.
As I was working on the indexing project there were several elements that had us enter multiple lines of data for a single element such as Format, Player Name, Type, and Subject element for a few. I was wondering if I missed a memo about the order of information entered within the elements. For some elements, Format and Type, there was a specific order of information for instance file type first, then size, then pixels for the format element. However, when it came to Player Name and Subject there didn’t seem to be any guidance about what order to enter names. It seemed logical to place all names from the same team together and to start with the most important or obvious subject term for the group. But I was wondering if there was an order I missed or if this order is even relevant. I can see it influencing subject terms, but not so much with player names. What do yall think?
When filling out the indexing project it was really interesting to see how the information confirmed itself, for instance I filled the “Date Created” element before I filled the “Temporal Coverage” element. As the “Temporal Coverage” is filled via a simple vocabulary I knew I hadn’t made a mistake for the “Date Created” element when I saw the correct date I had entered in the choices for the “Temporal Coverage”. It is great to see data confirm itself like this as a way of to prevent data entry errors.
I admit when I looked at all the elements to index for our football project I was intimidated. But the more I worked with it I was pleasantly surprised how taking the project on in small chunks made it a lot easier. This is an instance where I definitely think working element by element instead of record by record helped me. I started with some of the easier elements for me (i.e. the elements that I and my group had created) and then worked out to the more difficult elements. Having tackled Player Name element, filling in such information like title and description became much easier.
So Southern Mississippi doesn’t have player numbers or positions for most of their roster for 1975. I’m not quite sure how to figure out this information from a blank roster, and I guess I’ll have to leave most of those names blank. It was also interesting as I went through trying index the title, subject, and description elements how much more difficult it was to define images from Southern Mississippi games since I was missing some key metadata. I feel like those records are a lot more ambiguous or perhaps even have incorrect information entered.
So here I thought that player name element would include the player’s number as that is how the player’s are most clearly identified. I was surprised when I didn’t see that information on where or how to enter the number in the Player Name element instructions. However, after some consideration I realized it didn’t fall within the parameters of the element as having multiple numbers which may include different teams with the same numbers could be distracting and cause confusion.
After getting some good comments I have updated my guide lines for the Element: Identifier
This element specifically and uniquely designates the record to provide disambiguation and exact recall.
Guidelines for Creation of Content
The unique identifier will be built off of the file name of the image. Copy all of the file name except the file type i.e. “.jpg” into the Element and add “_Alabama_vs_” right before the name of the opposing school.
Let me know what you guys think!