Difference between revisions of "PersonalScienceWiki:Ask questions"
| Line 23: | Line 23: | ||
Experimenting is to formally state spell out and control for as many variables as you can. Science. Because user makes sure to change nothing else during the experiment many issues with analyzing time based relations do no happen. Consequently linear regression, ttest, causal analysis and breakpoint detection work. Maybe some complications in establishing baseline. This needs a page combining (or two and not combining) experimental design and the techniques above. | Experimenting is to formally state spell out and control for as many variables as you can. Science. Because user makes sure to change nothing else during the experiment many issues with analyzing time based relations do no happen. Consequently linear regression, ttest, causal analysis and breakpoint detection work. Maybe some complications in establishing baseline. This needs a page combining (or two and not combining) experimental design and the techniques above. | ||
| − | Second type is observational study. Just monitoring all the variables without a specific structure or interventions. This is what most health trackers do and is useful for many reasons. Unfortunately this is where all those issues I mentioned in [[finding]] and many more in the links all are relevant. | + | Second type is observational study. Just monitoring all the variables without a specific structure or interventions. This is what most health trackers do and is useful for many reasons. Also may remove plcebo effect and stress or nervousness from a more formal experiment. Unfortunately this is where all those issues I mentioned in [[finding]] and many more in the links all are relevant. |
| + | |||
| + | Maybe this comparison may need to be a page too. | ||
== Resolved Questions == | == Resolved Questions == | ||
Revision as of 20:29, 10 March 2022
This is a place to ask any questions related to Personal Science wiki.
Open Questions
Cloning the wiki?
how do I clone the wiki? I need to do it correctly or I could violate Attribution DG (talk)
- DG There is no "easy" export tool baked into mediawiki, but there are some instructions online on how to go about automating the process: https://www.mediawiki.org/wiki/Exporting_all_the_files_of_a_wiki
- It might be that no standard tool exists because people end up having different preferences on "what" they want to export (what they want to do). Also, note that licensing issues can vary according to content. While a lot of content is CC-BY-SA by default, there can be material (e.g. uploaded images) with other licensing, e.g. non-free material that's used as "fair use" in this wiki but may be a copyright violation if used elsewhere.
- I AM NOT A LAWYER AND THIS IS NOT LEGAL ADVICE: Regarding "attribution" requirements, I think it can be tricky to handle (highly visible attribution can disrupt re-use!). Sometimes people try to handle attribution via a link (e.g. on an imported file) to a source URL, and the source URL contains (or links to) information about authorship. If in doubt, you might share your plan and see whether others think it's met the requirement. :)
- Madprime (talk) 19:36, 9 March 2022 (UTC)
- what about just raw "BackupsThere are daily VM backups in Digital Ocean. Should there be other backups? - Katoss" Could you go into detail about "(highly visible attribution can disrupt re-use!)." I was worried the requirement was like on github with the littelest change having attribution. DG (talk)
- There is Special:Export, which provides XML dumps of the full revision history of specific pages and/or categories. This might work for an export of the information you are looking for! - Gedankenstuecke (talk) 19:10, 10 March 2022 (UTC)
Experiment VS Observational studies
Experimenting is to formally state spell out and control for as many variables as you can. Science. Because user makes sure to change nothing else during the experiment many issues with analyzing time based relations do no happen. Consequently linear regression, ttest, causal analysis and breakpoint detection work. Maybe some complications in establishing baseline. This needs a page combining (or two and not combining) experimental design and the techniques above.
Second type is observational study. Just monitoring all the variables without a specific structure or interventions. This is what most health trackers do and is useful for many reasons. Also may remove plcebo effect and stress or nervousness from a more formal experiment. Unfortunately this is where all those issues I mentioned in finding and many more in the links all are relevant.
Maybe this comparison may need to be a page too.