Re: [css-flexbox-1] Untestable assertions

On Tue, Apr 16, 2013 at 9:28 AM, Sylvain Galineau <galineau@adobe.com> wrote:
>>As I pointed out this was one possible solution, not the only solution.
>>Also without a clause stating that author requirements are not to be
>>included in testing I MUST test them if they contain text pertaining to
>>RFC 2119. There is no question about that from a testing perspective, and
>>it¹s the rule we have always followed.
>
> Actually, I think this very much ought to be questioned from any kind of
> perspective. Maybe we ought to expand a little on overly simplistic rules
> instead of blindly applying them.

Also, what's the "we" you're referring to, Arron?  It can't be the WG,
so I assume it's "the group I work in at MS".

If you'd just told me this was about box-checking on a badly-written
task list, I wouldn't have argued as much.  Problems that can be
solved by boilerplate were never problems in the first place.  ^_^

~TJ