R script to create summary CSV file from Flattr's individual Monthly Revenue CSV files.
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
.Rbuildignore
.gitignore
.travis.yml
DESCRIPTION
LICENSE
README.md
flattr-revenue-201305.csv
flattr-revenue-201307.csv
flattr-revenue-201309.csv
summarize-flattr-reports.R

README.md

Summarize Flattr Reports

R script to create summary CSV files and diagrams from Flattr's Monthly Revenue reports. Example here. You can also flattr this repo:

Flattr Button
Build Status

Usage Instructions

  1. Download and install both RStudio and R.
  2. Go to your Flattr Transactions.
  3. Open each "Monthly revenue"-report (or start with only a few).
  4. Click "Download as CSV" below the table of Things.
  5. Save them to a folder of your choice.
  6. Download this repository to the same or any other folder and unpack the .zip.
  7. Open RStudio, copy-paste this command install.packages(c("plyr", "ggplot2", "scales")) into the console (bottom left) and run it by pressing Return/Enter. Some progress of new packages being installed should be visible. If an error is shown instead, copy-paste the error message into a search engine you don't completely distrust.
  8. Once these packages are installed, open the script (.r file) in RStudio and run it with alt + cmd + R (Mac) or ctrl + alt + R (Win).
  9. Follow its progress in RStudios' Console and Plot tabs. This may take few seconds to several minutes, depending on your number of Flattr Revenue Reports, data points in them, and the speed of your computer. Example: 20 Reports with 200 data points at 2.3 GHz: 10-20sec.
  10. Find the newly generated .csv files and .png diagrams in the same folder as the .csv files you downloaded from Flattr.
  11. Please report back :-) Are the diagrams useful? If not, at which dataset size? Which other summaries, calculations or diagrams would you find useful?

Ideas

  • sort data usefully DONE
  • better file/folder selection DONE
  • automatic download from flattr.com, including withdrawals & deposits
  • draw more useful diagrams
    • coloration of datapoints per Flattr thing DONE
    • auto-determine useful diagram dimensions DONE
    • merge bubble & scatter plots into single diagram DONE
    • improve visualisation for very large datasets
  • summarise Flattr clicks per month, not ordered by thing DONE
  • predictions (anybody knows the statistics behind this?)
  • episodes vs. other things (probably needs reg-ex on slugs)
  • webapp via Shiny that processes given data for a Flattr-click Not a good idea
  • install on server to auto-run & publish diagram

Known Issues

  • monthly_simple_plot contains statistical elements whose computation runs into errors if the number of Flattr Revenue Reports is below 5. Other graphs may also throw warnings. To avoid this while plotting such small datasets, insert # before any line with stat_. Better yet: download more Revenue Reports, see 2.

Contribution Guidelines

I try to follow the gitflow branching model. Therefore, please branch off develop whenever possible, give the new branch a descriptive name and merge it back into develop.

Thanks and Greetings :-)