info
Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss
Regarding the CSS env() feature
13 views
Skip to first unread message
Mats Palmgren
Jun 15, 2018, 4:35:54 PM6/15/18
to L. David Baron, Cameron McCormack, Tantek Çelik, Jen Simmons, Emilio Cobos Álvarez, Boris Zbarsky, xid...@mozilla.com
So I think what we (Mozilla) should require before accepting the env()
feature (https://drafts.csswg.org/css-env-1/) is that all <custom-ident>s
MUST be decided by CSSWG consensus using the same process that is used
for introducing any other CSS property value. This is to ensure:
1. that there is a comprehensible specification for each <custom-ident>
so that anyone can implement it without having to reverse-engineer
specific UAs
2. that each value gets proper privacy, security and i18n review
3. the W3C rules regarding patents apply
UA vendors are not allowed to expose new env() features to the web
without going through that process. All of the above should be written
into the env() spec itself as a normative requirement for implementing
env().
Is there anything I forgot that we should also require?
/Mats
feature (https://drafts.csswg.org/css-env-1/) is that all <custom-ident>s
MUST be decided by CSSWG consensus using the same process that is used
for introducing any other CSS property value. This is to ensure:
1. that there is a comprehensible specification for each <custom-ident>
so that anyone can implement it without having to reverse-engineer
specific UAs
2. that each value gets proper privacy, security and i18n review
3. the W3C rules regarding patents apply
UA vendors are not allowed to expose new env() features to the web
without going through that process. All of the above should be written
into the env() spec itself as a normative requirement for implementing
env().
Is there anything I forgot that we should also require?
/Mats
0 new messages