Re: [compositing-1] please move non-separable effects to Level 2

On 17 Apr 2014, at 22:15, Robert O'Callahan <robert@ocallahan.org> wrote:

>> If they are not moved, we’ll probably put the prefix back on, because we don’t want to ship a non-prefixed incorrect implementation.
> 
> I don't see why this is necessary. If there are values you don't support, so be it ... that doesn't mean the entire property needs to be prefixed*. As soon as those values are present in a level 2 spec, in practice you'll be back to exactly the same situation you're in now.
> 
> * Or better still, disabled. Prefixes hurt the Web.

This list isn't the best place to have such a discussion (I don't know where is, sorry). "Hurt the Web" is something that's a bit difficult to prove or measure. Would a browser prefixing CSS blend modes hurt the Web more than a browser shipping something invasive like Shadow DOM before other browsers agree, and suggesting they will be unwilling to change names? Prefixes are certainly an annoyance to developers in some cases (at least the shrinking minority who do not use a preprocesor), but they could be beneficial to end users (allowing faster iteration with some safety against breakage). 

Anyway, I just said I don't think this is the right forum, and now I've written an email which is likely to get some replies and I'll remember my personal goal of not getting involved in standards email threads :)

> Can you point to an explanation of why Webkit prefers prefixes over the Gecko/Blink approach of user-accessible flags to enable/disable experimental features?

I don't know of one. I think I emailed a summary of a discussion we had at last years contributors' meeting to the WebKit list, but from memory that didn't explain anything. If you can suggest a place to have such a discussion it might be worthwhile. 

For what it is worth, I think prefixes have helped in WebKit. Both the project itself and the Web. We've also always been willing to change behaviour to match the standard and other browsers as we unprefix. 

Dean