edfelten / tracking-report-card Goto Github PK

The current formula is essentially GPA - sqrt(numTrackers) + correctionFactor, where GPA is the average grade of the trackers on the site, and the correction factor is set so that the average grade of first parties equals the average grade of the third parties we know about.

The question is whether we want to put different relative weights on GPA vs. sqrt(numTrackers).

Due noon on Saturday.

Problem solved.

Set up ssh keys by following the instructions available at http://help.github.com/win-set-up-git/

Run the fork command:
$ git clone ssh://[email protected]/ewfelten/Tracking-Report-Card.git

Is our output a web site? If so, who is going to build it?

Need to use the latest-greatest version of the input data. Currently using outdated/inaccurate input.

How are we getting the URLs into the crawler? Via HTTP? Via AJAX? Can we use jquery with MozMill or do we do everything without libs?

Clint has examples of this. It's generating weird data downstream.

Currently these are filtered out in Ed's code and ignored. But it indicates a possible error.

Clint is writing the DB analysis phase code. Ed is writing the report-card generation code that processes the output of Clint's code. We need to agree on a format for the output of Clint's code.

update them pls

Need to:
-- pick a domain name
-- register the domain name
-- set up hosting and point domain at it
-- populate host with files

Problem with committing from the commandline

A bunch of the records in input_first-third.json have "UNKNOWN" as the first party domain name. Apparently this is because the URL received from FourthParty can't be parsed by the fourthParty2json code. Can we figure out why, and fix it?