Hacker News new | past | comments | ask | show | jobs | submit login

This is the problem with what PPK is suggesting.

The Firefox and Chrome developers are mostly certainly professional developers who are well qualified to create a standardized cross browser feature set that will obviate the need for the cruft.

Only recently have we started to emerge from the Silverlight-Flash-JavaApplet-ActiveX-webish ghetto that we were mired in for years as Microsoft refused to add new standard features to IE. His suggestion is essentially to return to that paradigm. He should know better... he built a site called 'quirksmode' after all...




No.

His suggestion is to stop, and wait.

Spend a year on standardizing, fixing bugs (like the bug with chrome where it sometimes just "forgets" float-css rules), a year spent on improving issues and making everything behave consistently between browsers.

A year spent so that we won’t have to use jQuery afterwards anymore. Or underscore.js.

A year spent on fixing all the little bugs, and making a consistent roadmap where we actually want to go. Thinking about stuff like the Virtual-DOM of Angular or React and finding a better solution than that.

This is what we need to focus on. Making stuff behave consistently. And with one year, we can make sure that afterwards most of the market will actually use our new version. We can finally get rid of all the polyfills.


You're missing my point. The world won't stop just because you decide to reflect on the situation and have small groups of people unilaterally decide what will someday maybe be standardized in some browsers. In the meantime the new features will get developed as non-standard features and you'll have even more proliferation of cruft, "polyfills" and whatever else PPK is complaining about.


Why not? Why do we have to push new features every few weeks? The web still works right now, and it did in 2009, too.


Can I sing a few praises for the ghetto for a minute? It had more features. It was friendlier to non-developers because of things like Flash. It had more features. You could directly port native code through things like ActiveX (which we are now only recently getting back through ASM.js, with performance within an order of magnitude). It had more... you know where I'm going with this. The problem with that paradigm is the rise of resource-constrained mobile devices and security issues. But I'm unconvinced that the new paradigm won't just regrow the security issues and the performance problems as the features catch up.




Consider applying for YC's Spring batch! Applications are open till Feb 11.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: