Hacker News Reader

Comment by jbooth | Hacker News Reader

What's the "hacker acceptance factor"? Is that "what do 22yo kids writing webapps and shopping for sailboats to buy after their big idea takes off think about this?" I mean, why should I care about that?

Basically for the niche that Java fits in (networked, multi-threaded data-center applications where performance is important), it hits a sweet spot of ease-of-development compared to performance. C++ is a nightmare to code in, comparatively, and the scripting languages are either dog slow, have poor concurrency support or both. YMMV I guess.

View on HN

What is PG's so-called "hacker radar" if not some analogue to the "hacker acceptance factor" that I tried to allude to?

People think that languages like Python and Ruby are "cooler" somehow than Java - that's all I was trying to get at.

I don't think that I was making any judgments about the validity of that opinion, but I would say that it is fairly pervasive for better or for worse, and for many of the reasons that PG alluded to.

"where performance is important"

Really? I'm so out of touch. I always put Java down as something to use when performance wasn't important.

Hasn't been that way for a long time. Early versions had problems, but they've almost all been ironed out, now. Its speed is similar to C++ for most tasks, faster at some and slower at others. The only real speed problem left is boot time for programs, but that's largely a non-issue for programs that are never closed.

"Speed" is also too generic a term. Startup speed of the jvm is still abysmal, isn't it?

Once HotSpot kicks in, it's a different story.

I was falling for that rhetoric. then I actually tried to do some high performance computing in Java. After a lot of optimizing (and going to non-idiomatic code), it was only x4 slower than the C code it was replacing. The initial "straightforward, idiomatic" code was x10-x50 slower, and taking x5 more memory (which caused it to be totally unusable, as the C code used 1GB of memory, and the machine only had 2).

The computation was really quite simple, but it had to be done a bajillion times. I suspect part of the reason that it was so slow is that the JIT was not inlining something it obviously should have. However, I was unable to find any tools that would help me figure out if that was the case or what the JIT was actually achieving (consulting with Java experts I know didn't give any insight or link to such tools).

If you're telling yourself that "java is good for hi-performance computing", you aren't really concerned with speed or memory.

It really depends on what you're doing. The quirks of the JVM mean there are a lot of hidden things you can do wrong and tank your performance.. JNI calls that you don't realize you're making, or suboptimal use of objects and creating GC debt. Some of the Java math libraries are randomly slow for certain inputs, as well.

That just puts an emphasis on what I said. If there are a lot of hidden things that can go wrong, then in any non-trivial application some of them do, and Java performance is on the average, horrible.

The bigger problem is that there's no reasonable way of figuring out what the problem might be.

And for the record, said code was a tight loop, created no objects, had the same kind of workload of a sort algorithm; The two nontrivial things about it was that it was accessing multiple simple vectors using the same index (non-idiomatic java), and that one of those vectors was a memory mapped buffer (non-idiomatic java, but essential because of the overuse of memory).

Yeah, funnily enough, over the last 15 years, Java's traversed the full spectrum of hater, from "script kiddie language that's too slow for real men's work like C", to "too much static typing and inflexible for rapid app development like Ruby"

No, hacker doesn't mean "web-startup groupie". Hacker is more the opposite of the mindless corporate drone who's coding just for the paycheck. (Note: this doesn't mean hackers never work in the corporate world.)

http://www.paulgraham.com/gba.html

Well, sorry if I insulted anyone, but I'd think worrying about the "hacker acceptance factor" of a piece of tech is awfully non-hackerish.

You can write corporate drone type stuff in any language -- and you can write awesome code in any language. Java with a lot of immutability, closures (anonymous classes) and ExecutorServices is like a syntactically weird multi-threaded lisp. Very elegant. The language itself has something like 15 keywords, total.

As the person said upthread, you can look at some horrible 2002 EJB webapps (and I'm sure you can find their equivalents in rails), or you can look at projects like Hadoop, Hbase, Cassandra, etc. In general, static languages are probably a bad idea for webapps or scripting -- that doesn't mean they're the province of drones.

Nobody is "worrying" about the "hacker acceptance factor". It's just a fact that hackers interested in pleasant expressive coding are pretty unlikely to choose Java for their own projects.

Reply on news.ycombinator.com