0 requires a unit when declaring <time>

[philipwalton]

CSSLint reports that 0s should just be 0, which in incorrect.

According to these sites, time values must have a unit.
http://www.w3.org/TR/css3-values/#time
https://developer.mozilla.org/en/CSS/time

[nzakas]

I'm not sure that's correct. The spec doesn't specifically say omitting the unit is incorrect, and Mozilla's reference page has contradictory information. Even though it shows 0 as invalid in the examples, it also says:

As all units represents the same time for the value 0, the unit may be omitted in that case: 0 represents 0s, 0ms.

[philipwalton]

You're right that the Mozilla site contradicts itself; however, when you implement a CSS transition in Firefox using 0 instead of 0s, the transition doesn't happen at all.

Here's an example:
http://jsfiddle.net/J8NsQ/

Perhaps this should be viewed as a bug in Firefox, but either way, it seems like the CSSLint rules should only test length values until the spec is finalized and this issue is less ambiguous. I'd be happy to submit a pull request if you'd like.

[michalbe]

I wrote a blog post about this and create a bug report about a year ago. The answer was that you can skip the units ONLY when it refers to 'pixels'

[stubbornella]

Wow, this is odd. We need to fix the rule for now. In fact, validation should check that all time measures have a unit.