1. more on filtering:
Database with well organized metadata is useful if you have a closed set of information that you want to look at in detail. Information about film, movies, directors, producers… IMDB first comes to mind instead of Google. IMDB is a well organized database, it is faceted (sorted by movie, by director, by actors, etc.). if you do a IMDB search on Sally Fields you get a more detailed information, rather than Google, which would find EVERYTHING about Sally Fields and how many Sally Fields are there…
It takes a lot of work to set up a database. When you have someting that is “crowd created” like YouTube, where anyone can do anything. Filters can be used to sort through the information coming in for a better delivery of that information.
In the electronic newspaper International Herald Tribune gives you the filtering options at the very top of its homepage. and to the left side column as well.
RSS Feeds – filtering the news without having to follow a bunch of pages, instead RSS feeds bring the news you want to you. The difault view is headlines – or what’s new. But underneath Google reader subscriptions there are folders that give you a filtering option, just like a bunch of drawers to put “stuff” into.
One can organize and set up a flow of information and filter it by the ways that one chooses convenient to receive the news. We all do this kind of filtering quiet a bit, the tools are becoming more prevalent and the role of us making a decision on how to sort the information becomes more vital to us to stay “on top”
As before we had only a one or two sourses, in today’s postmodern world we have access to a LOT MORE information. The downside is we can not process ALL of the information, so we have to CHOOSE how to have the information delivered to us.
why filtering is important and how does it work (on the way out and on the way in). Hierarchical information.
we can read meta data before we learn to read. A smile, a little sign, an expression, that help us give context or understanding what is going and implicit is made explicit. When we are dealing with a database, that is an explicit metadata. The digital world, the web, the implicit metadata is also there, but it much less controlled. Implicit metadata has a quite a bit of a headroom for interpretation. Implicit can do quiet a bit of help as well as damage. What Weinberger is getting at when we are asked to categorize ourselves, it could be hard, how do you do that well enough to make sure there is enough information about you to make a truthful representation, yet keep your privacy private.
when it begins to be interesting and sometimes frighting is when we can tell implicit explicitly. When you are pulling in this implicit metadata, like amazon for example, they know what you shop for and what you purchased and by running through the algorithms tell you what you may like in the future. But there are a lot more things that are less [anaquis], web history, blogs, facebook, which could be tracked, and potentially become problems. Click stream could be recorded. Google analytic for example tells you a lot about the visitors to your website, but that information, when used in good hands and without harm to the users, can be used for making your website more user-friendly, but how much of a privacy should we talk about is a different subject.
once you have that cloud of information you should be obligated and responsible to use the private data in a non-harmful way;
it is important also to be implicit rather then explicit. Rather than taking an assumption that people are “stupid” we need to make a clear information delivery design-wise, classification-wise. Good way of making a database or a metadata or tagging, is to make the people feel smart, you are doing it well.
Playlist is all metadata and not… It is a lot like filestructure. Playlist is a collection of addresses for songs that rather been played. Why playlist is metadata and what context is to actuall songs, is a good possibility for the essay on a final exam.
Explicit metadata is “Hello, my name is… so and so” the implicit metadata is “non-verbal communication- when I am not smiling, I am not saying that I am happy” The applecup coming up – the explicit is the two colleges are playing they are in the same league and only one that is going to go to the finals, etc. Implicit of the rivalry is West vs. East side of the Washington State. There are economic class divisions, there are geographic divisions. The event is more than just a football state because it is red vs. blue, it is urban vs. rural, it is dealing on the implied level.
p. 171 is a great sum up to revisit:
“in a 3rd order we are externalizing meaning…” the content and the metadata is all digital. Digital miscellany is different from any other miscellany…
Implicit does not always work. Just because poeple who bought Planet of the Apes also bought Martin Luther King documentary, that does not garantee or describe anything. Digital collection of information that we were not able to do before, but there is no garantee that is going to work. Correlation does not equal causation. Implicit metadata allows us to track different connections, but the connections are not final, rationing in the socialist economy is based upon the statistics and tendencies of what people buy, eat and wear, but it does not predict how people are going to feel and in turn what they will choose to do to wear to eat or to buy.