It's a great counter-example to all the recent articles about JavaScript fatigue. It's good for us to see real examples of sites that aren't caught up in the framework of the week hype.
At the end of the day we're trying to make stuff that works. ES5 vs ES6, React vs Angular vs. Ember vs. Aurelia, Angular 1 vs Angular 2, even server rendered templates vs. single page apps - it all matters a lot less than we think.
For a subset of us, framework authors, open source contributors, etc. It's very useful to keep up to date on the latest trends.
For example, the Angular team has borrowed techniques from React. It helps for them to keep up with things.
For people like me though, let's be honest. I don't need to do the latest thing.
for the majority of cases i've seen, these arguments are to save developers time and the company money.
my gut & feelies tell me there is truth to "it doesn't really matter what technologies you use" but my brain and experience tell me that is objectively false.
It matters a lot eventually, but having an "eventually" to worry about is a nice problem to have.
(There's also a big perspective difference between being an entrepreneur writing initial code for a project vs. an engineer hired to maintain & grow that code. Code quality, architecture, and technology choice matters a lot to an engineer who will be working with it directly. It matters a lot less to a business that can swap out engineers until they find ones who enjoy working on that codebase. The fact that you even have money to hire engineers usually indicates that the existing code is getting the job done. The business only gets screwed when the code quality is so bad that nobody can make sense of it, and the environment changes in a way that requires modifications.)
i take your point none the less- there are contexts where it doesn't matter.
crud is often a feature of frameworks. crud should be an external tool like documention tools (doxygen...).
Not exactly any of these technologies, but it is using more javascript at least. I think the example is pretty awesome actually.
Given only the tech stack, it can go either way, depending on whether the implementor takes the easy, lazy route or actually puts some thought and hardwork into it. The problem is, as per good old Sturgeon's law, 90% of sites take the lazy route, and end up with tangled monstrosities that look like offerings to the Flying Sphaghetti Monster.
(Also, it's interesting how quickly 'cool new tech' can become 'Enterprise' shudder when it's widely abused and misused.)
Is this the hipster corollary to coding? Sure, Java was cool when only 100 people were using it...
How is everyone agreeing with you? Is there that much JS fatigue that we're looking at the equivalent of a horse-drawn carriage, and saying wow, I wish things were as simple as they were back then?
There is an increasing trend to move the whole website to the client side to be handled entirely by the JavaScript as a single actual page.
Personally I find most websites work better with a simple round trip to the server to fetch more HTML.
So actually building an entire system in this style isn't out of the realms of possibility at all - you just only write JavaScript for the cases where there is a real legitimate benefit to doing so.
> you just only write JavaScript for the cases where there is a real legitimate benefit to doing so
The snippet above only works because HN is so simple, normally the rule of thumb is if you are going to use JS to change the DOM at all then you need to render it in the first place using JS or else you are double-implementing the render logic and you open yourself up to changing one implementation and not the other. For hiding arrows sure this works but 99.999% of the time it's a very bad idea to use JS to change the dom to a state that the server would otherwise render on a page reload. You may feel that a round trip works better but that's not what users expect or will tolerate most of the time. That kind of thinking feels very close to people who think CRUD is all you need to build an app. Sure you can build an app with just that but thinks break down very quickly.
At home, I wait until lack of a feature (like hot reloading) frustrates me, then block out a couple of hours to do it specifically. Otherwise you've got a flashy stack and no product.
That code is more the equivalent of a digging stick. React is used to build horse-drawn carriages. I'm eagerly awaiting the JS framework that'll let me build cars.
In my comparisons I mention a lot of choices people might make, like Angular vs. Ember vs. React that are ultimately meaningless to the success of your product. I think this is what's resonating with a lot of people.
Make Javascript Great Again! (TM)
Another thing to note is the "RESTful" nature of the links embed in the page. I guess its some form of HMAC with SHA1 + Nonce that authenticates requests for the server. It differs for every href. Elegant.
auth=311110b7ef0f268cafa8616afcbfcec8d977111efunction hide(a){if(a=document.getElementById(a))a.style.visibility="hidden"}function vote(a){var b=a.id.split(/_/)[1];hide("up_"+b);hide("down_"+b);(new Image).src=a.href;return!1};
v and item are merged.
http://closure-compiler.appspot.com/home#code%3D%252F%252F%2...
Original Size: 212 bytes gzipped (318 bytes uncompressed)
Compiled Size: 164 bytes gzipped (182 bytes uncompressed)