pacer

[Carl Malamud <xxxx@media.org>]

So, I haven't heard from any of our metadata dudes, so let me suggest  
the following plan to move forward:
1. I am running an SSN scanner on the data ... that will give me a  
bunch of hits that I have to look at by hand, so it will take me a  
while to scrub it.
2. When the scan is done, I propose we announce the initial progress  
(25% (or more) of PACER created by the pacer p3wner posse).  I will  
announce that as an "alpha" release.
3.  You are free to keep accumulating data ... I want to be > 50% by  
the end of the year.
4. If/when we get metadata advice on how to do the XMP headers using a  
custom rdf schema, I'll stamp the existing archive, and call that beta.
5. Once we have a unique id stamp, we can start pulling in other pacer  
collections, such as my recycling stuff and Tim Stanley's proxy data  
(he uses client pacer id's, bills them for the work, then (with their  
permission) keeps a copy of the files).
I don't want to merge collections unless we can, using a script, be  
able to tell if file a (e.g., document x of docket foo) is the same as  
another file (e.g., document y of somebody else's version of docket  
foo).
Does this sound like a plan?  Our biggest PR boost will be the  
percentage of PACER we've retrieved.  So, if you want to create a  
pacer2 area and fill it, go for it.
One last item ... I plan on giving you guys full credit for your  
work.  So, if you don't want that, better speak up now.  :))
Carl