Re: CSS Masking

On Aug 15, 2012, at 5:05 PM, Robert O'Callahan <robert@ocallahan.org> wrote:

> On Thu, Aug 16, 2012 at 3:56 AM, Dirk Schulze <dschulze@adobe.com> wrote:
> To your concerns: If you search for '-webkit-mask-box-image' (a property where I thought that it is to specific), you'll find 400,000 search entries at Google.
> 
> Yeah, -webkit-mask-box-image looks used.
> 
> What about mask-attachment:fixed/local? I'm having trouble thinking of a use for it. Webkit doesn't even implement it (https://bugs.webkit.org/show_bug.cgi?id=67137). (However, Google claims 100K hits for 'webkit-mask-attachment', so I think we can't quite rely on those numbers :-).)
I fully agree. Google results are just a hint, but can't give exact results about the usage.

> 
> Looking at the top 50 Google hits or so I don't see any sign of people using -webkit-mask-origin or -webkit-mask-clip either.
I need to investigate more to give a reasonable answer. I will come back to this concern and add issues to these properties that we need to prove the relevance in the meantime.

> 
> OTOH -webkit-mask-position, -webkit-mask-size, -webkit-mask-repeat are all used.
> 
> The last thing is the expectation of web authors. We have '-webkit-mask' properties for quite some time. They are used for mobile applications and websites for WebKit based browsers. You could use SVG mask as fallback, but not a lot of web-authors really do. If we redefine the 'mask' properties we would not match the expectation of web-authors.
> It is also easier for implementors since the same parsing code from background and border-image can be reused for masking. Even the actual implementation is really easy since you can again reuse a lot of code. This makes it more likely that all browsers can implement this feature.
> 
> I think you misunderstood the thinking behind my questions :-). I appreciate the need to be compatible with existing Webkit-prefixed content, which means we're stuck with having 'mask' be background-like whether or not that's a good idea. I'm just wondering whether importing every single feature of CSS backgrounds is necessary or a good idea.
I just wanted to state my thoughts on this and didn't want to interpret your assumptions. 

> 
> I think the case for dropping mask-attachment is pretty strong, given Webkit doesn't implement it, no-one has presented any use-cases, and background-attachment:fixed is a real pain so mask-attachment:fixed probably would be too.
I agree, it also does not make a lot of sense in the context of SVG.

> 
> I think we could also drop mask-origin and mask-clip. OTOH they're not very hard to implement so you could argue we should just keep them for increased consistency with backgrounds. I tend to favour parsimony, so dropping them unless/until there are use-cases, but I wouldn't object to keeping them.
I don't fear that we would break existing content with this change. However, this could be a significant change for the short hand 'mask' and the user experience. To make it compatible to 'background' seems to be less confusing. So at the end it is a consideration between compatibility to 'background' and the sense of certain features. I can understand that you would be in favor for a meaningful feature (I would as well). Please give me some days to find meaningful use cases as well as real usage of these properties.

Thanks a lot for your reply,
Dirk

> 
> Rob
> -- 
> “You have heard that it was said, ‘Love your neighbor and hate your enemy.’ But I tell you, love your enemies and pray for those who persecute you, that you may be children of your Father in heaven. ... If you love those who love you, what reward will you get? Are not even the tax collectors doing that? And if you greet only your own people, what are you doing more than others?" [Matthew 5:43-47]
>

Received on Thursday, 16 August 2012 02:18:02 UTC