1:1 (2)
This project hearkens back to our original idea, or mapping the internet galaxy based on IP address and translating them based on the pages they contain. However, the idea was discarded for several reasons.
1. Content. While some web pages, like Slashdot, can claim an entire IP address to themselves, maybe even two IPs, other web pages, and especially web spaces like MySpace or Blogger would have thousands of web pages to a few IP addresses. Going by domain name gives the system a more accurate balance of content, and a more logical organization.
2. Feasibility. When inputting an IP address into a system, you will more than likely get a page. However, the page may not be representative of everything that is at that one IP address. In order to find out every page that could be found at one IP, you would have to scan every page and read the IP that is attached to it, and build a database off of that. Not only would that be unrealistic, given the resources two New Media students have available to them, but when you're already building a web crawler to figure out IP addresses, why not use the web crawler data?
3. Representation. IP addresses relate to each other in the fact that they consist of numbers - logically, there is a number before and after each number. The contents of two IP addresses could have nothing in common with each other, and an IP-based project would show a relationship between the two just because they happened to be numerically next to each other. However, web pages tend to link to content that relates to their own - for example, graphic designers will link to other designers, and designing resources.
Ultimately, we went with the link-based model for organizing the galaxy, just because it represented the internet better.
No comments:
Post a Comment