HN user

hijiri

15 karma
Posts1
Comments17
View on HN

Speaking for (GHC) Haskell:

- There are curly braces available if you really want them, but people almost always use the whitespace-dependent syntax. There are usually a few ways you can organize your code, but I have never seen any "syntax wars" over Haskell code.

- Haskell compiles down to one binary. Foreign C libraries are linked dynamically by default, if that's being counted.

- Comes with (very cheap) green threads and threading primitives, along with useful concurrency structures (like channels). The standard library includes STM for safer shared mutable memory, and many libraries (like async) are available that extend this. I have not done much Go, but I think forkIO is basically equivalent to "go".

- I couldn't find a whole lot comparing Haskell and Go in particular, but here is one blog post where Haskell was about the same speed as Go, in one narrow application: https://togototo.wordpress.com/2013/07/23/benchmarking-level.... The http server (Warp) used by some web frameworks (Yesod, Scotty, ...) has been noted to be very performant.

- Haskell has a strong, active community. It's not super big, but it seems to me the number of open haskell jobs is smaller than the number of qualified, talented haskell users. Recently there was an issue with a company advertising a non-Haskell job on the haskell-cafe mailing list, because they believed they could get some good developers that way.

EDIT: I posted this without realizing I loaded this page yesterday, and that there were already some responses

Reading your and similar comments made me think about it, and I came to the conclusion that the commonality between drugs and guns is superficial.

Starting with what they have in common, I would say the restriction of either creates a black market, so that either allows criminal organizations to make a profit selling them. I think there is a difference even there though, since you can argue that criminals still benefit from loosening of gun regulations, because they would have easier access to guns. The primary buyers of illegal firearms are gangs and similar organizations, so that market is mainly there to support criminals who would not be able to obtain them without it. I don't know if that would change with increased restrictions though - maybe some otherwise law-abiding people would be willing to buy illegal guns for self-defense. On the other hand, I can't see how legalizing drugs could benefit crime organizations, except that maybe they could use their existing infrastructure to pivot into a legal drug business. Maybe that is a significant boon, but it doesn't seem worse than any other "legitimate business" that funds criminal activity.

One of the main arguments in favor of drug legalization is quality control - street drugs often contain adulterants that are more debilitating than the drug itself. From what I know, this can't apply to guns since illegal guns are typically originally-legal guns from ordinary gun manufacturers, and don't have as many quality control issues.

And the way they generate harm is different too - drugs mainly harm the user, and guns mainly harm whoever the gun is used on (may sometimes be the same as the user). So drug legalization might be desirable so the harmful qualities of drugs could be reduced (as earlier), and not restricting guns could be desirable in that people will be able to defend themselves from other people with guns (criminals with guns, an oppressive government). This simplifies the issue a bit, and ignores other factors like drug research or hunting and other non-self-defense uses of guns, but I think it's clear that there is significant room to agree with restricting one but not the other.

My post may reek of bias toward drug legalization, but hopefully you agree that it's not as simple as "If you want to loosen drug/gun restrictions, you should also want to reduce gun/drug restrictions".

The problem is that one language might require less lines of code to achieve something of the same size/complexity of another language, if it's terser.

(Also on github: https://github.com/raymoo/NEET)

I wrote this NEAT library from around a month and a half to a couple weeks ago, and I haven't gotten much review from other people. This is the first haskell library I've written that could have wider appeal, so I don't have a lot of experience behind me. Are there any obvious deficiencies in the way I've coded this? I don't expect anyone to use it heavily at this point, but if anyone does, it would be nice if I could have some feedback on usability too.

Here's an example program I wrote to mess with NEET: https://github.com/raymoo/neet-cave (requires OpenGL, GLUT, GLFW, maybe some others)

EDIT: Neet itself also comes with an example program in Neet.Examples.XOR.xorExperiment. Note that for some reason it freezes for me if I run it from a ghci session in emacs, but works fine if I do it outside my editor.

It might just be that the car traffic signals are patterned to last that long, so the pedestrian crossing signals are the same length to match. But that would depend on the crosswalk.

The GPL does say the conditions are only on redistribution. But I think that's how it works with other derived works as well, by default. The GPL is all about how you can relicense it to other people using only the same license, you're free to do whatever you want if you never give anyone your modified code.