aston

You may be interested in this newer paper, reconciling the work of Killingsworth and Kahneman & Deaton. https://www.pnas.org/doi/10.1073/pnas.2208661120

I don't think you want this, but just in case you do :)

    def encapsulate(mod):
        import types

        out = types.SimpleNamespace()

        def replace_global_scope(f):
            # via https://stackoverflow.com/a/1144561, but tweaked for py3
            return types.FunctionType(
                f.__code__,
                out.__dict__,
                f.__name__,
                f.__defaults__,
                f.__closure__,
            )

        for name in dir(mod):
            val = getattr(mod, name)
            if callable(val):
                val = replace_global_scope(val)
            setattr(out, name, val)
        return out

You might like Go-Chainable: https://github.com/neurocollective/go_chainable

gron outputs Javascript!

This is like when you walk into a super complicated legacy code base and immediately have the one magic architectural change that will simplify everything...once you rewrite it from the ground up.

The music metadata situation is pretty bad, but the source of the problem is not really carelessness or greed or avoidance of responsibility (although those are all true). The true source of music metadata complication is the insanely complex copyright regime that music operates under. It's a legacy codebase about a century in the making that is constantly being patched up by congress, mostly by trying to change who is being protected from whom. (Among the folks favored at different times: labels, publishing companies, performing artists, song writing artists, radio stations, streaming music services, live venues, ...).

Perfect compliance with these laws is effectively impossible, so everyone is just doing the best they can. And any attempts congress makes to change how things work end up being gigantic legal battles because it's a zero-sum game and the more money in the "right" hands (e.g. these artists being ripped off) is less money in the other "right" hands (e.g. the unprofitable streaming service we all love).

This is a great example of what Marc Andreesen called the Moby Dick theory of big companies [1]. Paraphrased, the behavior of any big company is largely inexplicable when viewed from the outside. And from the inside, too!

[1] https://pmarchive.com/guide_to_startups_part5.html

Should lawmakers resist? There's no lack of housing and infrastructure projects worth spending on, state-wide.

Blake Ross (co-founder of Mozilla) has a great post about his journey with Aphantasia here: https://www.facebook.com/notes/blake-ross/aphantasia-how-it-...

A nice video explainer involving the authors of this paper: https://www.youtube.com/watch?v=wCrtk-pyP0I

Spotify skipped their lockout period, with pretty boring results.

Here's the simplest way to understand where your money goes (assuming you're a subscriber, otherwise where the ad revenue earned on your eyeballs goes): For each song you stream, that artist gets a fixed fraction of a cent. Unless you listen to an atypically large number of hours of music each month, there will be leftover money. That remaining money goes into a pot allocated across every artist on the streaming service in proportion to their total number of streams.

So, roughly, the policy as it stands pays the biggest artists—or, really, the labels of the biggest artists—more money. Those big artists and labels are powerful enough that the chance such a change will be made is basically nil.

The reality is a little more complicated because the fixed per-stream rate is set based on the total number of streams, but to a first order approximation this is it.

Your dates are sorted incorrectly. W2010 happens before S2010, etc.

The distribution by batch is pretty interesting:

   S2011 10
   W2012 10
   S2012 9
   S2014 7
   W2013 7
   S2013 6
   W2016 6
   W2015 6
   S2016 5
   W2014 5
   S2010 4
   S2015 4
   W2011 4
   S2009 3
   W2008 2
   W2007 2
   W2017 2
   S2006 2
   W2009 1
   W2018 1
   W2010 1
   S2008 1
   S2005 1
   S2007 1

The top three batches were consecutive starting from summer 2011 through summer of 2012. That period happens to also be when the seed and venture capital ecosystem started its recovery from the 2008 crisis [0].

What I wonder is, which way does the arrow of causation point? Did the success of these companies lift the entire ecosystem? Or were macroeconomic factors the dominant driver of capital entering this market, and those three batches happened to benefit?

edit: There's another possibility here, which is that there are two curves that may have maximized for companies around that time period. The first is the batch size, which has increased from ~10 companies to ~100 over the years. And the other curve is that companies take something like 5-10 years to mature. Maybe it's just that the companies of that vintage are just old enough to be really valuable, and that there were enough companies in the batch to push them to the top of this ranking.

My money is on macro effects, though.

[0] https://medium.com/the-mission/state-of-seed-investing-in-20...

Louder is not really worse when it comes to music. In fact, try turning up whatever you're listening to right now--it'll actually sound better!

The loudness wars (mass application of massive compression that reduces dynamic range) are the music equivalent of adding salt to savory foods. People like it better even though it's technically "worse".

Another way to frame this is that it's easier to get rich via your investments than your direct labor. Working at a company that is willing to give you equity is a funny sort of investing: you're investing your time rather than your money.

That the world pays off equity in a company orders of magnitude better than labor for a company is the real thing that will complicate a young person's worldview.