20861 2013-02-04 13:44:02 +0000 Should <keygen> be conforming-but-obsolete 2015-06-17 03:12:46 +0000 1 1 1 Unclassified HTML WG HTML5 spec unspecified PC All RESOLVED WONTFIX P3 editorial --- 1 robin robin hsivonen julian.reschke mike mounir public-html-admin public-html-wg-issue-tracking public-html-bugzilla oldest_to_newest 82517 0 robin 2013-02-04 13:44:02 +0000 It would appear that <keygen> could be a good candidate for filing under that category as it seems clear that at least one browser plans to never have any useful behaviour for it (beyond the required parsing/DOM) so that I don't believe we ever want new content to be using it. 82518 1 julian.reschke 2013-02-04 13:55:59 +0000 But that's because existomg code relies on the UA being *either* IE (-> ActiveX) or having <keygen> support, no? 82519 2 robin 2013-02-04 14:10:07 +0000 (In reply to comment #1) > But that's because existomg code relies on the UA being *either* IE (-> > ActiveX) or having <keygen> support, no? Right, but that changes nothing if the idea is that new content should not use it. Filing it as obsolete doesn't remove the existing support that relies on it, it just makes it clear that new stuff shouldn't use it. 82521 3 julian.reschke 2013-02-04 14:17:45 +0000 So does the spec tell authors what to use *instead*? (/me no crypto expert, just wondering whether we made progress since this came up years ago) 82524 4 mike 2013-02-04 14:29:24 +0000 (In reply to comment #3) > So does the spec tell authors what to use *instead*? > > (/me no crypto expert, just wondering whether we made progress since this > came up years ago) Eventually, the Web Cryptography API https://dvcs.w3.org/hg/webcrypto-api/raw-file/tip/spec/Overview.html though at this point we need to get implemented in more UAs 82529 5 hsivonen 2013-02-04 15:16:16 +0000 I came here to make the comments Julian already made. (In reply to comment #4) > Eventually, the Web Cryptography API > https://dvcs.w3.org/hg/webcrypto-api/raw-file/tip/spec/Overview.html though > at this point we need to get implemented in more UAs That doc says key provisioning is out of scope. <keygen> is used for CA enrollment processes. I think we should not obsolete it without a replacement. 121153 6 mike 2015-06-17 03:12:46 +0000 (In reply to Henri Sivonen from comment #5) > I came here to make the comments Julian already made. > > (In reply to comment #4) > > Eventually, the Web Cryptography API > > https://dvcs.w3.org/hg/webcrypto-api/raw-file/tip/spec/Overview.html though > > at this point we need to get implemented in more UAs > > That doc says key provisioning is out of scope. <keygen> is used for CA > enrollment processes. > > I think we should not obsolete it without a replacement. Agreed, so moving this to resolved=wontfix. There's been no new information on this in 2+ years and it does not seem important enough to merit being kept open for another N years with nothing happening. I can of course be re-opened if anything has changed.