Thoughts on Clojure Hacking | Hacker News Reader

Thoughts on Clojure Hacking

coreyhoffstein.com

47 pointssbochins19 comments

Open articleView on HN

I'm sorry that this might be the wrong place to ask, but why is sbochins's name coloured green?

Is it because I've upvoted him a lot - or he me?

How does it look for other people?

It's a new account. (think "Oh, he's green at that")

Ah, I see. I thought it was a reference to the red-green colour metaphor of + and -, friend and foe, and so forth.

Should we be drawing more or less attention to newcomers? The green is more obvious than the default gray.

I think it's ironic to draw attention to newcomers' submissions - and it might be something that malefactors would look into gaming.

On the other hand, the point of such features has always been to emulate the "I'm new in traffic - cut me some slack and keep an extra eye out for me" bumper sticker. And this outweighs the con, to me.

But it's something that's only been done on forums - bulletin boards if you will. As old as it seems, it looks like a novel approach by pg in the comments site medium.

Have been a long time reader, just signed up a few days ago. Was wondering what the green meant at first too.

Looks green to me

Also, front page? Wow. Must be a slow day on Hacker News, huh? I'm honored, but I don't really think the article has the "substance" to justify a front page ranking.

Clojure articles seem to draw a lot of upvotes by default, to the point of silliness. I love Clojure but it's kind of bizarre how fast they draw upvotes here, since it isn't a Clojure site.

pg is a bit of a Messiah to some Lisp folks. (One of the few who's had commercial success, wrote On Lisp, kept the dream alive...) HN also likes cool new tech. I'm not surprised that people like Clojure, although I agree that some more restraint wouldn't be bad.

HN is a Lisp friendly place and many fluffy Lisp related things make it to the front page on HN, not just Clojure (and this article is not nearly as fluffy as some of the things that make it :)

But I agree, it would be nice if people submitting Clojure articles would submit meatier stuff. Same could be said of the recent Node.js HN contingent. On a related note, one thing I've noticed is that the Node.js troupe seems to flag negative Node.js content - this kind of cargo-cult is a bad sign of where things are going on HN IMO.

Agreed on both counts. I truly love Clojure (and I use your Enlive tutorial any time I need a refresher) but if it's nothing new why does it need a hojillion upvotes.

I'll be curious to see what the effect of pg's changes are on such issues, if anything. I do wonder if seperating saved stories from upvotes might help, since some people may see clojure and use upvote to flag it for later reading.

Part of the issue is that saving an article/site for later perusal is the same as upvoting it. Ideally the two actions could be separated.

How is Clojure debugging now? Last I checked, I couldn't get a reliable stack trace, and the advice I read online said "just step through it" or use somesuch tool, which is presumably not very helpful when you're trying to figure out why an hour-long application unexpectedly crashed. That was enough to scare me away.

I'm not sure what you mean by "reliable stack trace", but I've been doing clojure full time for the past 6 months or so. Our program needs to stay up and it is heavily threaded. In this context I haven't had any problems with stack traces and my app fits your description of hours long run time and some crashes.

There are still thick stack traces by default, but there are tools to help now.

http://clojure.org/libraries#Clojure%20Libraries-Category:%2...

Never knew there was an issue with stack traces in Clojure. I only started programming in it several months ago. From what I've found, its much easier to wade through stack traces in Clojure. You can break out at different points in the stack and have a running repl with that environment. I've found this to be very helpful with debugging.

A minor quibble from the post, but instead of doing this:

  cat $(find . -name "*.clj") > clojure.lisp

you can avoid the temporary file by doing this:

  find . -name "*.clj" | xargs wc -l

wc doesn't leave out comments and newlines.

Oh, I see. He's running sloccount on the resulting file. My mistake.

Reply on news.ycombinator.com