You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I think the Compat Standard is the right place for this kind of work to happen in the near- and medium-term. But IMO in the long-term, if things settle down into a cross-browser equilibrium, we should migrate much of this into HTML (specifically the section on the various navigator properties) and Fetch (to give more guidance on the default User-Agent value).
For example, it would be good to normatively specify in HTML the relationship between navigator.userAgent, navigator.appVersion, and navigator.platform. Today I learned from https://www.chromium.org/updates/ua-reduction/ that appVersion is "everything after the Mozilla/" in userAgent, which is not captured in any spec today. Similarly, I assume there is some relationship between platform and the <oscpu> and/or <platform> tokens in the Compat Standard?
I'm unsure whether we'd want to go further and move the stuff about specific browsers and platforms from into HTML, or leave that in Compat.
The text was updated successfully, but these errors were encountered:
The work in https://compat.spec.whatwg.org/#ua-string-section is very exciting.
I think the Compat Standard is the right place for this kind of work to happen in the near- and medium-term. But IMO in the long-term, if things settle down into a cross-browser equilibrium, we should migrate much of this into HTML (specifically the section on the various
navigator
properties) and Fetch (to give more guidance on the default User-Agent value).For example, it would be good to normatively specify in HTML the relationship between
navigator.userAgent
,navigator.appVersion
, andnavigator.platform
. Today I learned from https://www.chromium.org/updates/ua-reduction/ thatappVersion
is "everything after theMozilla/
" inuserAgent
, which is not captured in any spec today. Similarly, I assume there is some relationship betweenplatform
and the<oscpu>
and/or<platform>
tokens in the Compat Standard?I'm unsure whether we'd want to go further and move the stuff about specific browsers and platforms from into HTML, or leave that in Compat.
The text was updated successfully, but these errors were encountered: