3 ms·

Is this bad since it can't execute a simple command without a library? Or is it good since the error message was so descriptive? If the consensus is "bad": I

Is this bad since it can't execute a simple command without a library? Or is it good since the error message was so descriptive? If the consensus is "bad": I imagine at some point elm made the decision to make things highly modular (something which people usually like) and lazy (another thing people generally like), so why are those things a problem in this case? Update: Mostly what I'm observing here is that the tweet basically implies that the author(s) of elm should be ashamed of themselves for designing (free) software that is so obviously wrong. Publicly shaming projects like this is dumb. I'm not convinced it's ever accomplished anything other than kicking off big flame wars. He should open or contribute to an issue ticket for the elm project. Otherwise, he should just stop using the software if he doesn't like it.

stefano 10y ago
Because those are terrible ideas when applied to the core functionality of an application.
darfs 10y ago
Specifing + in a (the! core)library that has to be loaded(over the network)... You can call it highly modular. Or you call it functionless. IMHO isn't that in any balance.
jack12 10y ago
I think it's possible to view the tweet as general bemusement rather than an attack, though I obviously don't know exactly what is intended. The next tweet is "I'm sure there's a perfectly rational explanation for why this happens, but it's logged in an issue on Github, which I cannot read". I mean, it is a bit amusing to get an HTTP error as a result of asking a local interpreter to perform addition. I'd say it's not unusual for modern languages, particularly those integrating package managers, to lazyily download libraries and/or check for updates on the first run of a REPL. But it is a bit odd or unexpected that elm's REPL requires a network connection for every launch and doesn't fall back to a cached copy of the core library packages from a previous run when the connection fails.
- davesque 10y ago
  Hmm. If it is indeed the case that elm doesn't cache library downloads, that does seem a bit strange. I can't say that it is though since I don't use it.

stefano 10y ago

Because those are terrible ideas when applied to the core functionality of an application.

darfs 10y ago

Specifing + in a (the! core)library that has to be loaded(over the network)... You can call it highly modular. Or you call it functionless. IMHO isn't that in any balance.

jack12 10y ago

I think it's possible to view the tweet as general bemusement rather than an attack, though I obviously don't know exactly what is intended. The next tweet is "I'm sure there's a perfectly rational explanation for why this happens, but it's logged in an issue on Github, which I cannot read". I mean, it is a bit amusing to get an HTTP error as a result of asking a local interpreter to perform addition. I'd say it's not unusual for modern languages, particularly those integrating package managers, to lazyily download libraries and/or check for updates on the first run of a REPL. But it is a bit odd or unexpected that elm's REPL requires a network connection for every launch and doesn't fall back to a cached copy of the core library packages from a previous run when the connection fails.

davesque 10y ago
Hmm. If it is indeed the case that elm doesn't cache library downloads, that does seem a bit strange. I can't say that it is though since I don't use it.