Hacker News Reader

Comment by godinaa | Hacker News Reader

One thing to consider is that at larger companies, we already have a difficult enough time testing all of our site with JavaScript enabled. Netflix building a version that works without JavaScript is like asking them to build a second website, especially from a QA perspective. When you look at your users and find that .5% of them are not running JavaScript, how do you justify spending the money for that few of people?

View on HN

The NoScript extension is the 5th most downloaded extension for Firefox (https://addons.mozilla.org/en-US/firefox/extensions/?sort=us...). If you want to show a blank screen or a broken website to those users, that's up to you. Some of them may really want to see your site, so they might try to figure out which of the twenty domains they should enable, and which of the thirty they should enable after that one doesn't help them. Most will likely click on the back icon and see what's next on the list.

Yes, and Firefox is the 3rd of 4th most used browser at less than 10% (which is bad). So the subset of actual Firefox users that use NoScript is absolutely not worth supporting.

Sure, but we're not talking extreme niche here or anything. It was downloaded by over two million users.

And the vast majority of those, if they want to use netflix, just whitelist it anyway. The amount of revenue lost must be tiny compared to the costs.

How do you know the percentage. Some stuff you may just never know. I very consciously avoid websites I know use a lot of JS or are bloated on mobile. Mostly news websites.

I often read in subway, where I need to load the page in 3-4 seconds or I'm in the tunnel again without the signal. In some stations I only get 50kbit/s or so.

Most websites don't cut it so I don't even bother. Those who cut it though, I get engaged with more, because I can't go anywhere else without a signal.

In a nice world I would write a scrapper + alternate website that hosts only the content and some simple navigation over plain HTML, preferably with as little CSS as possible. There's the annoyance of copyright though, so no way of making this public.

The reason is mobile. While only .5% of your users may be non Javascript users it's highly likely a significant percentage of your users are on mobile and not optimizing for progressive enhancement means a significantly degraded experience for all of them at various times when it fails to load or loads too slowly on a mobile connection.

I can stand slow loads, my phone has wifi after all. But js bloatware doesn't even allow me to scroll without seconds lags and random misclicks due to delayed event dispatch. Me turned js off not because of security concerns, but because most js-only sites are not scrollable beyond the first screen anyway.

To help our (internal) customers better troubleshoot, we're adding HTML responses to our HTTP (web) services. Add a header, a footer, maybe some HATEOS... Starts looking like a barebones web site.

Netflix is one of those sites that makes we want to run noscript. I regularly makes firefox unresponsive and eventually popup a "this script is using too many resources". When a large tech company can't create working javscript for a minimally interactive interface then something is seriously wrong.

Less javascript would be an improvement for everyone, not just the noscript users.

Try doing some automated testing, it'll take care of most of the non-JS testing and a lot of the JS enabled testing as well. Explority testing is another thing, that QA really does need to be involved with, but if you can't test static content without human intervention, you have other problems that are nothing to do with whether you have JS turned on or off.

They can't justify the cost. I wouldn't want them to, anyway. It would mean fewer improvements for their non-crippled site as resources are diverted for the sliver of people who like to be contrarian.

> fewer improvements

so, win-win?

Reply on news.ycombinator.com