4 ms·

Mainly because updating a page without a refresh was novel. Then developers began exploring how they could manipulate the page in response, using the DOM. The

Mainly because updating a page without a refresh was novel. Then developers began exploring how they could manipulate the page in response, using the DOM. The need for DOM manipulation led to a lot of early JS frameworks-- scriptaculous, then eventually YUI and jQuery afterwards. The explosion of clientside code in the browser also led to the V8 engine; before that, JS was far too slow for web apps to seriously compete with desktop apps. But before all of that, JS was hated by both end users and developers. It did not have much use outside of dropdown navigation, image maps, and plenty of malicious (or at least annoying) client-side hacks (alert box spamming, popups, disabled back buttons, etc). It was not uncommon for users to have JS completely disabled. You needed to design sites knowing some significant percentage of users could not run your JS.

fps_doug 7y ago
The improvements in JS performance over time is crazy. I was an Opera user (started with 5.x IIRC) and its JS engine was pretty slow. I mean so slow that even in the early 2000s some pages with just a few animations were annoying to use and better viewed in IE. Then at some point they rewrote the engine from scratch and also implemented JIT compilation to machine code. Opera might even have been the first browser to do so. After that it was lightning fast, ahead of IE, Firefox and Chrome before Blink/V8. If you use the last Presto Build of Opera today, those pages that still manage to render at all and use heavy JS are painfully sluggish, close to unusable. Opera's JS engine rewrite felt like performance was suddenly IE times 10. Now it feels like Chrome/Firefox are Opera Presto times 100.
dbtx 7y ago
> You needed to design sites knowing some significant percentage of users could not run your JS. Off-topic: if "everyone knows" this is no longer the case, then this statement about the new case is disappointing. It implies that the percentage of users who bother about privacy and security on the web is now insignificant, so nobody actually needs to consider them anymore, and nobody will, and things will just keep on breaking and staying broken.
dasil003 7y ago
There was also a whole scene of demo-like JS for years before XHR and serious use of JS... dig through http://www.ozones.com/ http://www.ozones.com/ for some cool 90s era JS hackery.

fps_doug 7y ago

The improvements in JS performance over time is crazy. I was an Opera user (started with 5.x IIRC) and its JS engine was pretty slow. I mean so slow that even in the early 2000s some pages with just a few animations were annoying to use and better viewed in IE. Then at some point they rewrote the engine from scratch and also implemented JIT compilation to machine code. Opera might even have been the first browser to do so. After that it was lightning fast, ahead of IE, Firefox and Chrome before Blink/V8. If you use the last Presto Build of Opera today, those pages that still manage to render at all and use heavy JS are painfully sluggish, close to unusable. Opera's JS engine rewrite felt like performance was suddenly IE times 10. Now it feels like Chrome/Firefox are Opera Presto times 100.

dbtx 7y ago

> You needed to design sites knowing some significant percentage of users could not run your JS. Off-topic: if "everyone knows" this is no longer the case, then this statement about the new case is disappointing. It implies that the percentage of users who bother about privacy and security on the web is now insignificant, so nobody actually needs to consider them anymore, and nobody will, and things will just keep on breaking and staying broken.

dasil003 7y ago

There was also a whole scene of demo-like JS for years before XHR and serious use of JS... dig through http://www.ozones.com/ http://www.ozones.com/ for some cool 90s era JS hackery.