Quoting Steve Langasek (vor...@debian.org): > You're free to fight for this if you like, but this is very much a "make the > mountain come to Mohammed" situation - the number of code monkeys writing > bad web apps is enormous, and all of Debian's users together are unlikely to > make a dent in this problem, even if they all felt it was an important > issue. This doesn't put our users *or* Free Software first; instead you're > prioritizing an ideological dispute the outcome of which will have no > discernable benefit for Free Software, versus simply changing our browsers' > code centrally to spit out the string that appeases the web app writers.
I entirely sustain Steve's (and John Goerzen's...and Joey Hess') position here. Whatever we think about the brokeness of relying on User-Agent for webapps, this is a reality we have to face, and John explained very well why hacks to work this around do not work in the real world of users who don't know what a damn User-Agent line is (or a UA plugin, which was suggested as an alternate hack). Sure, there are tons of our provided browsers that are "broken" the same way on the same websites. But none of them claim to be one of the two most populars browsers in the world. Really, can't we fix that User-Agent string? Does the Mozilla Foundation have a clear policy about this? It's probably not as hard as solving the brand name issue (which I still think is possible to fix....if only we could take time and energy to come back on it....by pyutting together in the same room at the same time a few key Mozilla folks and a few key Debian folks). After all, what does prevent us to just put "Firefox" back in the default User-Agent? We'll see if some Mozilla person raises an RC bug and then we would talk..:-)...That would be the pragmatic approach.
signature.asc
Description: Digital signature
