Hacker News new | ask | show | jobs
by romanoderoma 2099 days ago
The argument is not about market domination, but about money power

Apple market is rich, it is the richest tech company, with the highest market valuation ever, one cannot avoid it

So it's gonna drive adoption no matter what

Like it did for Flash (right or wrong, they did it)

I was writing web applications during the 90s, they were called intranets back then, IE wasn't bad, it was simply unavoidable

And being unavoidable halted competition for more than a decade

On another side: having a platform that doesn't obey to common standards creates tech segregation

I remember seeing japanes phones 20 years ago, they were ahead of time,but unusable outside of Japan

Last but not least, having a flag that says "allow me to shoot myself in the foot of I want to" it's less hostile than "you don't know what you're doing let me handle it for you"

1 comments

So then write for all browsers, support what you can for each, and don’t promote Chrome anymore? If you continue to promote a one size fits all browser and the market moves that way then we’ll shortly be having this same discussion about Google. If you were writing websites in the 90s you should know this.
> So then write for all browsers, support what you can for each, and don’t promote Chrome anymore?

WOW

That escalated quickly

I never promoted Chrome, in fact I'm an old time Firefox users and in the 90s I owned a Netscape 3 gold license.

I don't write fronted software anymore, but writing for every browser means adhering to the standard

Apple has steered away more than any other vendor lately

So...

> I don't write fronted software anymore, but writing for every browser means adhering to the standard.

Yes, which you’ll find if you do Safari adheres to quite well. Finding examples where it doesn’t is easy, just like on chrome. Which is why we have media queries and poly fills because it’s simply impossible to have a unified platform without one entity owning it. If you found out Chrome doesn’t render the same on linux as it doesn’t on windows, would you call this not adhering to the standard?

> Yes, which you’ll find if you do Safari adheres to quite well

They don' t

Apple refused to implement 16 web API in Safari

> Which is why we have media queries and poly fills because it’s simply impossible to have a unified platform

And that' s why big players like Apple hat control a closed ecosystem shouldn't be allowed to be assholes

At least on Android I can install other browsers and other browser engines

Do you see the difference?

> Chrome doesn’t render the same on linux as it doesn’t on windows

that' s not true

Windows and Linux that have simply different font rendering engines and different native widgets

That's all

> would you call this not adhering to the standard?

absolutely NOT!

In no part of the standard rendering is formalized

It wouldn't make sense

The way things are rendered is standardized, not the way they should look

https://www.w3.org/TR/2014/REC-html5-20141028/rendering.html

> HTML is intended to apply to multiple media (it is a media-independent language). User agent implementors are encouraged to adapt the suggestions in this section to their target media.

> Apple refused to implement 16 web API in Safari

OK and these APIs don’t stop real work.

> And that' s why big players like Apple hat control a closed ecosystem shouldn't be allowed to be assholes

And putting all that power into the hands of a single browser will not end well, so stop promoting Chrome or saying other browsers should be like Chrome. FF isn’t going to win.

> At least on Android I can install other browsers and other browser engines Do you see the difference?

Yes, next question, do I care?

> that' s not true

100% true, read the dev blogs and understand the WM on linux is fundamentally different than windows or mac, therefore rendering is not exactly the same.

Standards help align, they do not force alignment. If safari implemented a p element as a span, then this is not following the spec. Choosing not to implement a portion of a modular spec, is still following the spec.