@vineettiruvadi I voted False because for doing any kind of science you need to at least observe something (phenomena, object, etc) or analyze facts. Both of these would require either collecting data or already having the data, right?
Those who voted True: curious to know what type of science doesn't involve observations and facts?
@dlevenstein That's one of the things I'm trying to get at - are the theory folks "doing science" if they never touch data? What part of "doing science" is individual/tactical and what part is community/strategic?
What about if the followup data-folks find that the theory was spot-on right? Together they did science (I think we all agree), but should the "theory" person be funded through traditionally "science" mechanisms? Etc.
Was thinking about Einstein/Eddington/Earman and their order of operations when making the poll (https://royalsocietypublishing.org/doi/10.1098/rsnr.2020.0040)
@vineettiruvadi @dlevenstein
I think the crux is what "having data" means. Think of Van Vreeswijk & Sompolinksy (https://doi.org/10.1126/science.274.5293.1724). Did they "have data"? In the strict sense of analyzing data in their work, certainly not. Did they "do science"? I would say: definitely!
Some might nonetheless say that Van Vreeswijk & Sompolinksy didn't do science. Well, then what about other work that didn't analyze any data, but relied on qualitative observations in other data-driven papers to inform their modelling, or - more loosely - their formulation of new hypotheses? Once data informs their work, does it mean for them to "have data" even if they didn't analyze it themselves?
This quickly leads to the question of what it means to "have data". If we would discard all work that doesn't directly analyze data as not doing "science", almost all of neuroscience theory wouldn't do "science". If we include work that relies on others' data-driven observations, then almost all neuroscience theory is science.
Vineet Tiruvadi
manisha
Dan Levenstein
Antoine Choppin 👺
Jan Drugowitsch