Thursday, February 15, 2007

Lab 5

Unix Commands
The 'whoami' command was one of the commands I learned from lab 5. I think it's important because it confirms who you are logged in as. To go a step further, the 'finger' command, really lets you know that the computer knows who you are. It was fascinating to see all of my information come up that I didn't even know the computer knew. I am not sure if either of these has a counterpart.
'cd' was probably the most important command I learned. It allows you go back to your home directory from where you are. It helped if I messed something up I could just restart in a way. I am pretty sure 'cd' has a counterpart. 'mkdir' was another important command. You can make any directory you want with this command. From there you can put things in files under other commands. 'mkdir' has an equivalent DOS counterpart as well.

Chapter 6
The topic of Chapter 6 of Andy Clark's book was Global Swarming. The introduction analogy of the slug was interesting and sensible. I have experienced the "slug trail" numerous times on the Internet- I am probably the biggest supporter of shopping online instead of going out. And I noticed the feature of "Those who bought this item also bought/looked at.." when it first was introduced and I like the idea behind it. I can't say I've ever bought anything because of it, but I've definitely wasted extra time browsing. Collaborative filtering is an amazing concept too, when you think about it. It's like there is a big chalk board keeping tallies on what consumers buy item A with. Through this method they are able to figure out which items to put on the "If you like this item, you might also like..." list. The best part about these features, is you don't even have to use them. They are really designed to get you to spend more money. It's still has kinks that I think it needs to get worked out. For example, when you sign into amazon, it immediately gives you "personalized recommendations" which are actually just things you looked at before or already bought. It gets really old. Pages and pages of recommendation, depending on how often you shop, which for me is often. You can click 'not interested' on the item, but I think they need to let you clear all previous "trails" if you want.
Overall, though, collaborative filtering and recommendations are a good idea and are more useful than not. I'd give Amazon an A-.

No comments: