Re: .gov crawl

[Carl Malamud <xxxx@media.org>]

here's a statement of work ... let me know if this works for you.   
Basically, we should coordinate on things like what data formats and  
software tools you pick, but the idea is to let you go after the  
Brewster archive as a testbed and see what useful information you can  
generate out of that.  We can then look at it before you write up your  
report and figure out what lessons we can learn from this as far as:

1. do we do this on a crawl or straight on the target web site
2. what kinds of other metrics can we add?
3. is this something we can package up as software or will it be  
strictly alpha geek.
4. what tools do we like, what tools do we wish existed.

Let me know what you think ... (p.s. we just rel

Carl

Attachment: watchdog.net.sow.pdf
Description: Adobe PDF document