The CSS 3 opacity property has been implemented since Gecko 1.7. However, the -moz-opacity alias was never removed. This seems odd, since other experimental properties have been removed after their standard equivalent was implemented. According to MDC, use of -moz-opacity is deprecated since the implementation of the standard property. opacity is perfectly understood by Presto and WebKit (tested Safari 3 for Win). Besides, best practice has always been writing proprietary and standard properties: E { filter: alpha(50); -moz-opacity: 0.5; opacity: 0.5; } So, -moz-opacity should be removed, just like -moz-inline-block was when inline-block was finally implemented. According to LXR some files are still using -moz-opacity, but I think this shouldn't be the problem here. I have not heard of any reasons that speak against this. However, if I missed the reason please tell me.

The reason I left it (in bug 93156) was that a lot of sites were using it. That said, I agree I probably should have removed it (as I have done for later property renames). That said, -moz-opacity is probably the most used -moz-* property on the Web by an order of magnitude. I'm not sure whether I can justify doing it right before a release (increased risk of catching problems resulting) given the rather low benefits, though.

I thought it was a bit too late for doing it now. Hm, just disabling it for Beta 3 is no option? Well, I better do not spam here. But good to know it's on your radar.

http://mxr.mozilla.org/seamonkey/search?string=moz-opacity :|

The benefits of this would be interoperability (for pages using -moz-opacity but not opacity) and not further encouraging code bloat (for pages using both). The reasoning in bug 93156 was good for backwards-compatibility, but that's also four and a half years ago.

(In reply to comment #4) > The reasoning in bug 93156 was good for backwards-compatibility, but that's > also four and a half years ago. Yes. Most users of -moz-opacity also use the standard property as well as Microsoft's filter (which is disabled in IE8's no-quirks-mode to implement real opacity instead). Presto and WebKit both support opacity as well now. There are a lot of things going on this late in the cycle and some of them can break websites way more than this one. Besides -moz-pre-wrap was quickly removed in Bug #418543. This could/should be just as quick. Sorry, I guess this can be interpreted as bugspam, didn't want to waste anyone's time.

(In reply to comment #5) > There are a lot of things going on this late in the cycle and some of them can > break websites way more than this one. Besides -moz-pre-wrap was quickly > removed in Bug #418543. This could/should be just as quick. It wasn't actually.

(In reply to comment #6) > It wasn't actually. Both could be done now, can't they?

Would probably be a good time to do this now before a beta because site developers should be checking out their sites with the beta...at least that is what I think.

I tried to figure out how this can be done. I probably failed because this is my first patch, but the attempt is more worth than another complaint.

And this time it should actually apply. Sorry.

Comment on attachment 339644 [details] [diff] [review] Second Try r+sr=dbaron, though this probably shouldn't land after the first beta

How about removing also the -moz-outline aliases?

http://developer.mozilla.org/en/CSS/-moz-outline: "Starting with Gecko 1.8 (Firefox 1.5), the standard CSS 2.1 outline property is supported as well. Use of outline is preferred to -moz-outline. "

(In reply to comment #13) > How about removing also the -moz-outline aliases? Filed Bug 458588 to achieve this.

The documentation has been updated.