Observe. Theorize. (Repeat as long as it takes to stabilize.)
A great deal of structure has arisen around both of these things. Descriptions of how observations should be carried out, how experiments should be designed to test theories, how results should be reported, how outlying data should be handled, when your data is "good enough" to stop repeating (95% CIs and such), and more.
But our methods for "how results should be reported" were developed in an age when print journals were the main source of scientific reporting. There simply wasn't room for long descriptions of specific setup details, daily travails, etc. There wasn't room for complete data sets of millions of points. Now there is. It's refreshing to see someone taking an approach to science that says "we've got the space to report every detail, so let's do it. Anyone can dig deeper and get the full story if they're interested."
Observe. Theorize. (Repeat as long as it takes to stabilize.)
A great deal of structure has arisen around both of these things. Descriptions of how observations should be carried out, how experiments should be designed to test theories, how results should be reported, how outlying data should be handled, when your data is "good enough" to stop repeating (95% CIs and such), and more.
But our methods for "how results should be reported" were developed in an age when print journals were the main source of scientific reporting. There simply wasn't room for long descriptions of specific setup details, daily travails, etc. There wasn't room for complete data sets of millions of points. Now there is. It's refreshing to see someone taking an approach to science that says "we've got the space to report every detail, so let's do it. Anyone can dig deeper and get the full story if they're interested."