gabelevi

This was just a federal court, not the Supreme Court.

Well this deterrent isn't really working. Fare evasion is currently a big problem for the MTA (http://nymag.com/intelligencer/2019/03/mta-fare-evasion-one-...)

It's also not clear to me that it is ethical to have a punishment be far more extreme than a crime. Sending someone to jail for 40 days for stealing $2.75 seems a little extreme. That could easily turn a poor person with a job into a homeless person with no job.

A fine feels more appropriate. Maybe free rides for those who legitimately can't afford the fare.

Crime in NYC did drop precipitously, but it's not clear that broken windows policing was the cause. It's been studied and debated a lot, but as far as I can tell there isn't a clear consensus. Here are some sources:

* https://en.wikipedia.org/wiki/Broken_windows_theory#New_York... * https://cityroom.blogs.nytimes.com/2007/08/13/why-did-crime-...

I suppose creating an interventionist near-utopia won't be so bad if we don't run into any other intelligent life. The ethics of intervention are complicated enough between humans, so I'm happy to defer worrying too much about the ethics of interfering with other intelligent life until we find it. And other than that, the Culture's post-scarcity near-utopia sounds pretty great, especially compared to the current state of humanity.

The red car had a stop sign too. You can see the back of the hexagonal stop sign.

Yep, you are correct! The hardest parts of getting Flow and Hack to build on Windows are solved. For Hack, there's probably a long tail of small issues before a Windows build can be released, though.

Is Hack on Windows something that you'd use? Or are you asking out of curiosity?

I'm also a fan of namespacing variants, and I used it extensively on an OCaml representation of the SpiderMonkey AST. (Coincidentally, up until writing our own parser we were using pfff's JavaScript parser! Go pfff!). Anyway, here are some other useful things I found in OCaml along the way:

Without namespacing, you might be using mutually recursive type definitions like

  type statement = ExpressionStatement of expression
  and expression = FunctionExpression of statement list

to do this with modules you need to use recursive modules, like

  module rec Statement : sig
    type t = Expression of Expression.t
  end = struct
    type t = Expression of Expression.t
  end

  and Expression : sig
    type t = Function of Statement.t list
  end = struct
    type t = Function of Statement.t list
  end

One of the sucky things about recursive modules is that you cannot omit the signatures, however if your modules only include types then there is a shortcut that looks like

  module rec Statement : sig
    type t = Expression of Expression.t
  end = Statement

  and Expression : sig
    type t = Function of Statement.t list
  end = Expression

Links: Big real world example: https://github.com/facebook/flow/blob/3a5b4040b7d2a648f97a06...

Recursive module docs: http://caml.inria.fr/pub/docs/manual-ocaml-400/manual021.htm...