So, kind of the newer versions of Javascript has made this easier. That said, I think Node is not the best system to build a massive server web. I would use Go for that. And honestly, that’s the reason why I left Node. It was the realization that: oh, actually, this is not the best server-side system ever.
I hate when people use Ryans statements like this. It was the same when Ryan stopped working with Ruby and went to work on Node: suddenly people used that to say that Ruby is bad, because Ryan said so. I met Ryan multiple times at EuRuKo and spent many great lunches together and always saw him as a person that sees many facets to a thing. (BTW, these 5 or 6 long discussions are still some of the most important of my career, I was a student back then)
Ryan back then was interested in building async server systems. He started writing ebb in Ruby, just to realize that he’d have to move the whole Ruby ecosystem to async to make that work as cleanly as he’d like. That isn’t possible. Next came Node, which used a programming language that had no IO, so it had no blocking IO. It’s a genius move, if you have no interest in creating a programming language. For some reason, this thing had gaps, so there must be something new somewhere else.
Ryan - to my reading - is driven by very particular challenges that tease him. That’s great, and I learned a lot from him speaking about that. Note how he says “massive”. That’s a very narrow use-case. Most systems are not “massive”. I’m pretty sure, in inquired, Ryan would come up with a couple of hundred places where Node usage is fine.
Nothing you’ve said here disagrees with that excerpt, though. The quote was Dahl saying “hey, this is not the best server-side system”. That allows for your observation of “a couple hundred places where Node usage is fine”.
It might also be instructive to note that Dahl also left Ruby–turns out, perhaps Ruby isn’t a good server language either. I think it’s reasonable to point out that the main benefit of the Rails ecosystem at this point will be legacy consulting work in ten years for those of us willing to put up with it.
I don’t think that Ryan Dahl’s opinion is any more valid than any other developer.
This same logic could be extended to the developers of C, QuakeC, Go, Javascript, or any other oddball platform. I think it is ill-advised to dismiss equally the opinions of people who have actually shipped things and people who haven’t (as your position here seems to suggest).
From the man himself.
I hate when people use Ryans statements like this. It was the same when Ryan stopped working with Ruby and went to work on Node: suddenly people used that to say that Ruby is bad, because Ryan said so. I met Ryan multiple times at EuRuKo and spent many great lunches together and always saw him as a person that sees many facets to a thing. (BTW, these 5 or 6 long discussions are still some of the most important of my career, I was a student back then)
Ryan back then was interested in building async server systems. He started writing
ebbin Ruby, just to realize that he’d have to move the whole Ruby ecosystem to async to make that work as cleanly as he’d like. That isn’t possible. Next came Node, which used a programming language that had no IO, so it had no blocking IO. It’s a genius move, if you have no interest in creating a programming language. For some reason, this thing had gaps, so there must be something new somewhere else.Ryan - to my reading - is driven by very particular challenges that tease him. That’s great, and I learned a lot from him speaking about that. Note how he says “massive”. That’s a very narrow use-case. Most systems are not “massive”. I’m pretty sure, in inquired, Ryan would come up with a couple of hundred places where Node usage is fine.
Nothing you’ve said here disagrees with that excerpt, though. The quote was Dahl saying “hey, this is not the best server-side system”. That allows for your observation of “a couple hundred places where Node usage is fine”.
It might also be instructive to note that Dahl also left Ruby–turns out, perhaps Ruby isn’t a good server language either. I think it’s reasonable to point out that the main benefit of the Rails ecosystem at this point will be legacy consulting work in ten years for those of us willing to put up with it.
[Comment from banned user removed]
This same logic could be extended to the developers of C, QuakeC, Go, Javascript, or any other oddball platform. I think it is ill-advised to dismiss equally the opinions of people who have actually shipped things and people who haven’t (as your position here seems to suggest).
Hm, his opinion for Node does carry a certain weight, but a weighted discussion of Node just isn’t the point of this article.