3 ms·

The way Google released "experimental" features behind prefixes instead of configuration for years always seemed like a way to get developers to break their sit

The way Google released "experimental" features behind prefixes instead of configuration for years always seemed like a way to get developers to break their sites in non-Chrome browsers as much or more than a way to test features. I assume that Google's takeaway from the Internet Explorer anti-trust case was that they needed plausible deniability for their non-standard browser features.

waboremo 3y ago
Judging by Google's sloppy handling of evidence related to their antitrust suit, it seems your assumption is entirely on the money.
dragonwriter 3y ago
> The way Google released "experimental" features behind prefixes instead of configuration for years always seemed like a way to get developers to break their sites in non-Chrome browsers as much or more than a way to test features. Apple, Mozilla, Microsoft, and Opera all were doing that before Google even had a browser; there never even was a Chrome-specific prefix, all of the features included in Chrome this way were -webkit- prefix and, IIRC, implemented in the WebKit shared codebase.
madeofpalk 3y ago
> The way Google released "experimental" features behind prefixes instead of configuration This was not a "Google" thing - vendor prefixes were a common practice in the industry. Webkit was using -webkit- prefix before Google was using it. Many years ago the harm of this was recognised and browsers did move these to configuration/feature flags instead. https://developer.mozilla.org/en-US/docs/Glossary/Vendor_Prefix https://developer.mozilla.org/en-US/docs/Glossary/Vendor_Pre...
gondaloof 3y ago
Historic tidbit: Mozilla had a slew of "internal" prefixes to be used in their UI and extensions, but Apple really broke new ground and made them mainstream when the iPhone came out and really needed CSS to be useful. That's when stuff like -webkit-transition` and the fabulous `-webkit-box-reflect` were pushed out without waiting for anyone. -ms- wasn't really a thing until IE10 if I remember correctly, when they started prefixing their own IE5-era properties like the DirectX-based `filter` [citation needed]

waboremo 3y ago

Judging by Google's sloppy handling of evidence related to their antitrust suit, it seems your assumption is entirely on the money.

dragonwriter 3y ago

> The way Google released "experimental" features behind prefixes instead of configuration for years always seemed like a way to get developers to break their sites in non-Chrome browsers as much or more than a way to test features. Apple, Mozilla, Microsoft, and Opera all were doing that before Google even had a browser; there never even was a Chrome-specific prefix, all of the features included in Chrome this way were -webkit- prefix and, IIRC, implemented in the WebKit shared codebase.

madeofpalk 3y ago

> The way Google released "experimental" features behind prefixes instead of configuration This was not a "Google" thing - vendor prefixes were a common practice in the industry. Webkit was using -webkit- prefix before Google was using it. Many years ago the harm of this was recognised and browsers did move these to configuration/feature flags instead. https://developer.mozilla.org/en-US/docs/Glossary/Vendor_Prefix https://developer.mozilla.org/en-US/docs/Glossary/Vendor_Pre...

gondaloof 3y ago

Historic tidbit: Mozilla had a slew of "internal" prefixes to be used in their UI and extensions, but Apple really broke new ground and made them mainstream when the iPhone came out and really needed CSS to be useful. That's when stuff like -webkit-transition` and the fabulous `-webkit-box-reflect` were pushed out without waiting for anyone. -ms- wasn't really a thing until IE10 if I remember correctly, when they started prefixing their own IE5-era properties like the DirectX-based `filter` [citation needed]