- From: Joergen W. Lang <joergen_lang@gmx.de>
- Date: Wed, 25 Jan 2012 17:17:22 +0100
- To: Lea Verou <leaverou@gmail.com>
- CC: "Tab Atkins Jr." <jackalmage@gmail.com>, jwl@worldmusic.de, Aryeh Gregor <ayg@aryeh.name>, www-style@w3.org
Am 25.01.12 13:19, schrieb Lea Verou: > On 24/1/12 20:37, Tab Atkins Jr. wrote: >> I agree - this seems like a much friendlier way to present the list. >> >> ~TJ >> > > FWIW I think there shouldn't be a list at all. It's a maintenance > nightmare to have to update the spec every time a new property is > defined. As long as it defines how each value type is interpolated, > their combination and potential exceptions, I never got why an all > inclusive list is needed. A list of a few examples, sure, but a list > that has to include every CSS property?! If there are exceptions some kind of list would still be needed. I agree, however, that a complete list (of all CSS properties?) would be very difficult to maintain. Either way, I guess it makes sense to have some way to determine which properties are actually transitionable. If I understand the spec correctly it is actually not the property that is transitioned, but its value(s). Colors, lengths, etc. can be transitioned, keywords can not. So why not say something like: 1. A property can be transitioned if it takes one of the following value types: color, length, percentage, function, [and so on...] 2. If the value is a function (e.g. translate()) the calculated value(s) will be used for the transition. 3. Properties that only take a keyword as a value (e.g. border-style) can not be transitioned. 4. For properties that can take multiple values or value types only those mentioned in 1. and 2. can be transitioned. I'm not sure about 2. since my knowledge about the inner workings of those functions is somewhat limited. Does this help? Best regards, Jørgen W. Lang
Received on Wednesday, 25 January 2012 16:17:55 UTC