Atlanta Nodejs Developers Message Board › Investigating node.js

Investigating node.js

A former member
Post #: 2
In the spirit of wanting to truly understand node.js from all possible perspectives so that I may make a truly informed decision about it, I would like to request from this meetup group some intellectual responses regarding the decidedly zealous article here:

http://teddziuba.com/...­

I wish he had presented each of his technical arguments in a purely academic way rather than in the incendiary fashion he is apparently famous for, but this is the only article I could find with an opposing viewpoint to proponents of node.js, so this will have to suffice.

I’m hoping to receive some responses that are neither ad homonym attacks on the article’s author nor equally zealous statements. If it’s at all possible, I would like to hear purely academic, scientific responses -- whether in agreement or not -- to each of his arguments.

Any such responses that any of you can give will be most appreciated. Thank you in advance.

Adam Churvis
Rick T.
irickt
Atlanta, GA
Post #: 22
Hi Adam, Could you summarize his arguments briefly? As you indicate, his tone was very unpleasant. I remember scanning his article at the time and I'd rather not read it again. Thanks, Rick
Don B.
donwb
Atlanta, GA
Post #: 3
The gist of his troll-bait post was that by using fibonacci, he was able to show that node didn't scale per the hype and thus node was worthless..

Only problem is he gamed the test to make *sure* node didn't perform. Fibonacci as he implemented it is blocking, which of course in the node world is b-a-d. He could have offloaded the computation to make it work async (and thus behave like proper node modules do), but didn't.. thus the awful performance.

The best thing to do with this blog post is ignore it.
Rick T.
irickt
Atlanta, GA
Post #: 23

Thanks Don,

I remember now.

The key point is that Node has been proved in many production applications - when used as intended. Here's the VP of Engineering of Joyent (biased of course, but as he says they were using Node in production before they bought the project.)

"I think Node is actually far more production-ready than other environments that have been around a lot longer. I would deploy Node way before I deployed Ruby. I would deploy Node way before I deployed Python and I know that with people having aneurysms all over as I’m inciting religious wars, but from my practical experience, we can now determine more about what’s going on for a node program than any of these other services."

http://www.infoq.com/...­
(Listen to the whole interview.)

A former member
Post #: 3
I sincerely appreciate your response, but it would seem that the person you are quoting is making an equally zealous response in opposition, which, though understandable because he owns the company that owns the node.js product, is exactly counter to what I was looking for.

I was hoping to get targeted, science-based responses to the points made by the firebrand in the article I referenced, not a statement that node.js works well or is used in lots of demanding environments. I already know that; it's the reason why I started this thread to get some answers that might illuminate how *real* the pros *and* cons are. I was hoping that you gentlemen and ladies might help me with that, since I have no experience with node.js and you do. Quite a lot, in many cases.

Mr. Dziuba's language is, to say the least, bordering on childishly angry, and so it is difficult to get through while affording him any measure of legitimacy. So when you folks speak on this subject with measured tones and without epithets, I will tend to listen to you much more than I have listened to him.

If anyone wants to make another go at breaking down the claims made in the article I referenced then I will be very appreciative.

Thanks, and I look forward to more feedback.

Adam Churvis
Rick T.
irickt
Atlanta, GA
Post #: 24
Hi Adam,

Please read this HN thread for responses both to the author's tone and substance.
http://news.ycombinat...­

There are two minor straw man arguments: node is a poor http server and node can't handle long computations. The response is that these are irrelevant because in practice node isn't used in that way.

In applications at scale, node is deployed as one component, with other servers, proxies, caches, the host OS, etc. as needed. Node is especially useful at handling I/O bound data traffic.

In less intense applications like simple web serving, node has the performance and ease of use that make it a good choice over other scripting language platforms. Again the arguments don't apply: node is a good enough http server and can offload computations to threads.

It's a lot easier to find articles about successes with node than informed criticism or applications abandoned. There's no doubt a success bias in the reporting, but success is success.

What are your specific concerns?

Rick
Don B.
donwb
Atlanta, GA
Post #: 4

Just came across this. Ryan Dahl, created a proper fibonacci server in node and posted a gist of it..

https://gist.github.c...­

Powered by mvnForum

Our Sponsors

  • BitPay

    They are helping to cover our ATV meeting space costs.

  • Intel

    Help covers our location and any food expenses we have this year.

People in this
Meetup are also in:

Sign up

Meetup members, Log in

By clicking "Sign up" or "Sign up using Facebook", you confirm that you accept our Terms of Service & Privacy Policy