Tell HN: Thank you for not redesigning Hacker News

Post by pmarin » Mon, Sep 02, 2019, 6:57 AM UTC

novok wrote: post #147

zeta0134 wrote: post #145

Earlier quoted context omitted.

Okay quick thought exercise. HN doesn't use a react framework and is quite lightweight on clientside JavaScript. How many upvotes does it take using this system before the total data size transmitted (original page + modifications) is, in sum total, larger than a fresh load of, say, Twitter? The reality is that for the bulk of users, votes are an uncommon interaction; most of us are here to read the articles and view…

But they're already using javascript, and the issue is the backend response, so this would only be a backend optimization then and doesn't require any fancy frameworks?

Post by IanCal » Mon, Sep 02, 2019, 7:05 AM UTC

pmarin wrote: post #290

manigandham wrote: post #190

Earlier quoted context omitted.

It's nothing more than an extra header or query param to denote an AJAX request. Then the server can respond with an empty 200/204 status code instead of the entire HTML page. It's an hour of work and would actually save resources. None of this requires a front-end framework.

HN works perfectly with javascript dissabled. It’s a feature

Post by TeMPOraL » Mon, Sep 02, 2019, 7:13 AM UTC

csande17 wrote: post #270

171243 wrote: post #260

Earlier quoted context omitted.

So just to clarify, this behavior seems to be by design. In that case it does seem the original comment is objectively wrong if the code is readable and maintainable and accomplishes the desired behavior.

As parent comments mention, sending back the whole-page response when you vote on a comment is a great idea if you have JavaScript disabled or aren't logged in. But it'd be a quick performance win if it didn't do that when you were logged in with JavaScript enabled. The actual JavaScript part of this is actually already implemented -- that's why the page doesn't refresh when you upvote my comment -- so you'd just nee…

Post by jbverschoor » Mon, Sep 02, 2019, 7:21 AM UTC

samtheprogram wrote: post #250

_ps6d wrote: post #237

Earlier quoted context omitted.

How could it possibly be considered a good thing? I've seen dang say before that they've been running into performance issues, and I bet that they'd see a significant drop in resource usage solely by not generating entire comments pages just to throw them away every time anyone votes on a comment. The response is literally not used at all, they're doing that work and sending that data for no reason. There are multipl…

It's implemented this way for balance of a simple implementation and how lightweight it is. Not to mention, there is at least one benefit: the voting mechanism doesn't depend on JavaScript. And if you are not authenticated, you are instead taken to a login page. Logging in then gives you the redirect and the upvote is registered.

Post by TeMPOraL » Mon, Sep 02, 2019, 7:22 AM UTC

samtheprogram wrote: post #251

_ps6d wrote: post #208

Earlier quoted context omitted.

That frontend code already exists and does that every time you vote (line 27 of https://news.ycombinator.com/hn.js ). The response content isn't used for anything at all.

If you have JavaScript disabled, the response content is used to hide the upvote sign.

  (if *ajax-request*
     (upvote)
     (upvote-and-generate-page))