The data collection procedure was based on a model being used by U. of Pennsylvania and U. of Michigan. (http://www.ppehlab.org/what-is-a-datarescue-event)
Locally we downloaded 20 Gigabytes of data in 580 files for archiving.
If you know of data that you would like to nominate for rescue, use this form: http://www.ppehlab.org/datarefugesurvey/
There are several tracks that participants choose.
Seeder sorter: These people nominate page urls to be automatically rescued. There is a list of web sites for participants to begin with and help guide them. https://github.com/datarefuge/workflow/blob/master/docs/seednsort.md
Researcher: Look for ways to download data that wasn't automatically downloaded by the seeders. https://github.com/datarefuge/workflow/blob/master/docs/research.md
Harvester: Working with the researchers, actually download the data that wasn't automatically downloaded. https://github.com/datarefuge/workflow/blob/master/docs/harvesting.md
Checker: Inspect harvested data and make sure it is complete. https://github.com/datarefuge/workflow/blob/master/docs/checking.md
Describer: Create a description for each downloaded dataset. https://github.com/datarefuge/workflow/blob/master/docs/metadata.md
Bagger: Verify that the content is correct and matches what was described, then run a script that packages the dataset with its description. https://github.com/datarefuge/workflow/blob/master/docs/bagging.md
Storyteller: Blog, tweet and otherwise share about the process.