Video: Ryan Dahl: Node.js
in [JavaScript]
|
Prev: FAQ Topic - Internationalisation and Multinationalisation in javascript. (2010-03-23)
Next: FAQ Topic - Internationalisation and Multinationalisation injavascript. (2010-03-23)
From: Ry Nohryb on 25 May 2010 18:36 On May 25, 9:54 pm, Scott Sauyet <scott.sau...(a)gmail.com> wrote: > > (...) Jorge, next time could you supply a little context? (...) Sure, Scott. Here's some more: http://jsconf.eu/2009/video_nodejs_by_ryan_dahl.html http://vimeo.com/9968301 Enjoy, -- Jorge.
From: David Mark on 25 May 2010 19:06 On May 25, 6:29 pm, Ry Nohryb <jo...(a)jorgechamorro.com> wrote: > On May 25, 11:49 pm, Garrett Smith <dhtmlkitc...(a)gmail.com> wrote: > > > > > The most recent time this came up was when discussing "Non Browser > > Implementations". > > Yes, genius, yes. And 2 months ago, when discussing another of your > screwed-up FAQ entries: "FAQ Topic - How can I access the client-side > filesystem? ". See:http://groups.google.com/group/comp.lang.javascript/browse_thread/thr... > > But, of course:http://google.com/search?q="node.js"+site:jibbering.com > sill yields: Your search - "node.js" site:jibbering.com - did not > match any documents > > Because the FAQ is permanently outdated. Permanently outdated?
From: "Michael Haufe ("TNO")" on 26 May 2010 00:55 On May 25, 1:12 pm, Scott Sauyet <scott.sau...(a)gmail.com> wrote: > If someone is to review this, I hope it is not someone whose main > focus seems to be on browser scripting. > > Node.js is all about server-side JS, using the V8 engine. > > I don't know enough about that subject to offer a review, but would > love to see one from someone who does. There are some surprising > results, including a simple server that outperforms nginx in requests/ > second. > > One interesting thing about the presentation is the argument that > JavaScript is the perfect language for doing IO-centric operations if > only JS supported IO. It's reasonably convincing. Here's my take on it having no knowledge of it beyond the video: *built on V8, hence limited to that version of JS and not whats supported in say Rhino *For the same reason its better than Microsoft's JScript. *Server Speed claim: Needs verification *Consumes a significant amount of memory for what seems to be trivial work. *Uses CommonJS style "Modules". Whether this is a good idea and approach is debateable as can be seen in the es-discuss mailing list. *Concurrency with I/O as presented is far too vague to come to a conclusion. It raises more questions than answers. *All I/O requires a callback or so he claims. This is probably inaccurate at best as I'm sure he doesn't include uses of the Date object for example. Either way, this is pretty inconvenient in comparison to most languages/frameworks. *One of the design goals is to be Low-Level, but:"Threads should only be used by experts.... you'll write it in C". I guess his big ole brain knows better than anyone who plans to use his invention. *He's also agains the use of co-routines because they're too hard, *A design issue that bugs me is the differing patterns for creating objects. Also the argument for the .end() method seems unnecessarily redundant. *overall I think its an immature and premature. If you want JS on the server stick with Rhino or JScript at this point.
From: john on 26 May 2010 02:05 On 25 May 2:54 PM, David Mark wrote: > Ry Nohryb wrote: >> On May 25, 7:52 pm, David Mark<dmark.cins...(a)gmail.com> wrote: >>> Ry Nohryb wrote: >>>> http://developer.yahoo.com/yui/theater/video.php?v=dahl-node >>> Do you read this group at all? :) >>> >>> Stop polluting your brain with YUI drivel. Do I really need to review >>> their "Node.js" >> >> Yes, you ought to. Node.js has nothing to do with YUI. > > Well, that's something. :) it's too bad that the experts/regulars in this group seem to be so unaware of such an interesting project as to not even recognize its name. to a novice (i.e. hobby programmer) such as myself it looks like one of the more interesting additions to the ECMAScript ecosystem in recent times (certainly more so than the seemingly never ending supply of "cross browser" scripting libraries). doing system scripting, HTTP servers, database drivers, web frameworks etc. in ECMAScript would seem like more exciting work than what the typical web application offers. do the people around here with a deep understanding of ECMAScript really not find any interest in the language outside a browsing context? >> Node.js is a >> wonderful thing for any JS programmer, even more so if he happens to >> love unix. > > Whatever. Stupid name though. Sounds like a filename (hence my initial > confusion). certainly no more "stupid" than a library named "My Library." honestly, no offense meant as your library (among others maintained by regulars here) has proved quite educational in my (so far insignificant) foray into browser scripting; but this seems like a really silly criticism in the broad scheme of things. in fact most popular open source projects seem to have "stupid" names. perhaps it's part of a grand strategy :)
From: "Michael Haufe ("TNO")" on 26 May 2010 03:19
On May 26, 1:05 am, john <john.loves.spam...(a)gmail.com> wrote: > it's too bad that the experts/regulars in this group seem to be so > unaware of such an interesting project as to not even recognize its > name. Its an old idea in a new form. node.js is far from the first foray of JavaScript into the server (which was first seen in 1998 AFAIK with Netscape). So is it innovative? No. Is it an interesting project? Sure. Perhaps once it gains a 1.0 version status, and when the novelty wears off it can be looked at in a more objective manner. > do the people around here with a deep understanding of ECMAScript really > not find any interest in the language outside a browsing context? This is a false characterization and exaggeration of the people who visit this group. |