I get that metadata is data about data, but the different types of it get mixed up, as do the attributes. I felt the chart describing the different types and attributes was hard to follow and unclear. Metadata management does seem fairly complex. I was surprised to see that a metadata little-known fact is "metadata does not have to be digital." I would have thought that was a given, as data doesn't have to be digital either. Though I suspect that many people don't know what metadata is in the first place, so would therefore not be aware of these little-known facts.
(Mainly, I think most people don't think about metadata much, whatever they call or don't call it-- myself included.) But regardless, metadata is definitely necessary. If we didn't have some way to describe and discuss data, we'd have some trouble managing the enormous amount of information we are faced with. Finding what you need quickly and conveniently is so important.
Databases... The history of databases was somewhat interesting. The article summarizes the different models of databases. I was interested in the information on the security of databases, but I admit I don't understand encryption as well as I'd like to. The article did give me a better understanding of locking. I didn't know there was a difference between shared locks and exclusive locks. There are more database management systems that I knew about before this, though. I wonder what the major differences are, since there are so many.
I had trouble accessing the Dublin Core article, but eventually thought to see if I could access a Google cache of the page, which was successful. It's still annoying that the site has been down all day, but anyway! (For anyone reading this who still can't get it to work, click here) The basic concept here is that there are resources that need to be described-- and there are ways to uniquely identify them. Clarity and meaning are important, and it is crucial that everyone is on the same page when it comes to the meaning of important terms-- the article gives the example of the word "copyright." I agree that with information being passed around on a global scale, precision is definitely important, and we cannot assume that a word means the same thing to everyone.
Subscribe to:
Post Comments (Atom)
2 comments:
Metadata:
I thought that article was rather confusing as well. I do not think they distinguished between different types of data/metadata very well. I think I said in my blog that they seem to confuse the two sometimes.
Databases:
I have had to take classes about/involving databases, and from what I have seen, all the major databases have to support a lot of the same things, like SQL and other standards used to interact with data. I think the main differences are how they do things behind the scenes, and in the specific interfaces that are included with the databases.
Dublin Core:
I thought this article was also hard to follow, not because of a bad explanation but because of all the technical terms. But I think you are right, it brings up how complicated it can be to try to describe things consistently when different people may understand things in different ways, particularly definitions of words. This will be a major issue in our field(s) for some of us, I am sure.
Your remark on getting the different types and attributes of metadata confused reminds me of PittCat. When you apply search limiters to a PittCat search, there are two fields that I just can’t figure out the logical difference between: item type and medium. Examples of item types are books, musical and nonmusical recordings, and maps. Examples of mediums are maps, sound recordings, and text. The difference between item type and medium then? You tell me. Perhaps this is one of those times that we find MARC and AARC2 antiquated, and that we could be better served by XML or Dublin Core.
Post a Comment