Hi Nils, > Also a lot of user tend to surf with older browsers, if we really want > to build a library which can be used on real life projects we will > have to support as much browsers as possible.
I think it makes sense to implement a fall back. > I am not sure my skill is high enough. And I think it's very trivial to implement (just replace the current "crypto.getRandomValues" calls with a new method and don't break the tests - maybe add some more tests beforehand). I already added the sjcl library as a dependency three month ago. Btw: sjcl is using crypto.getRandomValues too (if possible). Best regards, Alex On 20.04.2012, at 20:18, Nils Kenneweg wrote: > Signed PGP part > Am 20.04.2012 19:07, schrieb Tankred Hase: > > Hey, > > > >> From what I read here, it seems the guys at mozilla are close to > >> getting > > crypto.getRandomValues into FF: > > https://bugzilla.mozilla.org/show_bug.cgi?id=673432 > > > > It may make sense to wait for that patch, rather than add a random > > generator with lesser quality. The less code to maintain, the > > better. The question remains when ie will expose this api... > > As far as I know it with mozilla those guys can look "very close" at > patching something into main for a very long time. > Also I fear IE will take some more years. > Also a lot of user tend to surf with older browsers, if we really want > to build a library which can be used on real life projects we will > have to support as much browsers as possible. > > Just my point of view. > > Regards, > Nils > > _______________________________________________ > > http://openpgpjs.org
