Playground

Some things get understood by writing them down rather than by running them. The pieces in this group have no notebook and no build behind them. They are reasoning kept in public because the reasoning was the work.

What is in this group

What collaborative filtering assumes about people sets out how user-user and item-item recommendation differ and what each quietly assumes about the people it describes. What a good statistical brief does in four pages reads a statistical brief closely and treats its structure as a format worth copying.

Why these are not called research

Neither produced a result. They clarified a method and a format, which is useful and is not a finding. Calling them research would put them in the same category as work that tested something, and the distinction between the two is most of what the research method is protecting.

What can be expected of a piece here

That it is honest about being an argument rather than evidence, and that it names what would change the conclusion. What cannot be expected is that it has been tested, reviewed, or applied to anything.