Hacker News Reader

Comment by gmosx | Hacker News Reader

Don't judge the whole article by a single line...

btw, I have a special definition of toy:

6. Why not use NodeJS?

Well, NodeJS is a valid, if over hyped, solution. But it only helps you come up with a toy application. With 'toy' I mean an application that runs on your laptop or your staging server. For a production app you need deployment processes, multi version processes, system setup, server tuning, database scaling, backup processes, security policies, monitoring, profiling, admin console etc, etc. App Engine is the easiest way to go from toy to production (though Heroku-Node may provide a viable alternative).

(from http://www.gmosx.com/blog/agVnbW9zeHIPCxIHQXJ0aWNsZRipwwEM/o...)

View on HN

At Plurk.com we use node.js to provide comet updates to over 300.000 active clients at _once_. It's not a toy and it scales much better than most of other web-stacks I have tried. Read more here: http://amix.dk/blog/post/19490#Plurk-Instant-conversations-u...

comet updates are a niche application

inflammatory posts with little to no statistical backing are a niche application.

That said, if you were to flip the post to be "hey node.js is cool, but check out RingoJS while you are already hot for SSJS" Now that would be awesome!

just ignore the nodejs part, then you don't have to like the whole article...

(btw, nodejs and ringojs are on the same side: SSJS)

thanks for the clarification. I was unaware.

The mainstream is composed of a million niches.

Comparing comet performance between node and ringo should be interesting. Ringo uses jetty/cometd, so it should scale pretty decently as well. My guess is that node will use less memory per connection (and thus allow more simultanous connections), while ringo/jetty should have better latency. This is also what this benchmark suggests: http://praxx.is/post/486034949/comet-with-bayeux-node-js-vs-...

Before our node.js solution we actually ran Java along with a highly optimized JBoss Netty comet implementation and it had a hard time scaling since it used _a lot_ of memory. Generally, our Java solution used around 10 times more memory than our node.js solution. Like I note in that post:

"We have rewritten to node.js from Java+JBoss Netty, because the old solution had scalability problems. We have also been able to build a very sophisticated messaging system given JavaScript expressiveness and the simplicity of node.js. Generally, we have found node.js and V8 very impressive - and it uses about 10 times less memory than Java+Netty did."

Jetty Comet uses a lot more memory than a vanilla JBoss Netty implementation - at least based on my benchmarks.

I know, i read that post. I'm planning to do a Node/Ringo comet comparison soon, and memory usage/max connections will be a part of that.

Please do. I don't know why people keep reposting my benchmark, its terrible and really only applied to what I'm doing with Comet. Memory, etc, isn't the biggest factor for us.

> Don't judge the whole article by a single line...

If something is called A vs B I was would hope to see a list of how A is better than B. The only mention of NodeJS is pretty derogatory - it detracts from the promotion of RingoJS (which I found interesting)

The point of the article is to bring Ringo in the spotlight. I strongly believe that the developers are doing a great job and the seamless access to Java libraries is the great selling point.

If the article generates some (well deserved) attention for Ringo, it will be a success (even if some people do not agree with the wording)

This sounds like something Marak would say.

why downvotes? he - again - clearly states his intentions. what's wrong with that.

I guess some times controversy can be helpful. The article did generate discussion on HN. And this discussion led to me taking a look at Ringo (which I hadn't heard of) to find out what the fuss what about. And now I'm considering trying out Ringo for a project I've been thinking about for a while. So, I guess, mission accomplished .

;-)

Ever consider just submitting a talk about RingoJS to "cool marketing" AKA JSConf? We are a pretty open and "underdog-rooting" group. I might know someone up in the decision committee...

Great idea, you should probably ask Hannes (the RingoJS lead developer though)

I submitted a Ringo talk to JSConf.eu 2010, haven't heard anything back from these guys so far. If that fails, I may apply for next JSConf.us. We'll get the word out there eventually :)

clearly points out several advantages, so i think it qualifies as comparison.

  * commonjs conformant
  * mature & stable
  * java access
  * ...

i agree: dislike of nodejs shines through, but that was clear after the first paragraph. hard to talk about ssjs without mentioning node, nowadays.

Reply on news.ycombinator.com