Chrome "Back" bug breaks HTTP content negotiation; marked WontFix for 1+ years | Hacker News Reader

Chrome "Back" bug breaks HTTP content negotiation; marked WontFix for 1+ years

code.google.com

9 pointsmcgwiz3 comments

It's very troubling, because the decoupling of resources from their representations is so core to HTTP.

A year ago, the maintainers defended Chrome's behavior as consistent with Firefox, and vaguely as consistent with the spec. Since then, more complaints have piled up and those original defenses have been deconstructed (Firefox no longer exhibits this spec-breaking behavior, multiple unmet sections of the spec have been cited). But for the last 9 months, the maintainers have not said a thing, despite doing various house-cleaning chores on the bug report itself.

One can't help but wonder if there's some ulterior motive for ignoring the community and keeping the WontFix status, though nothing obvious comes to mind... The impact this could have and the way they've handled this is reminiscent of how the downfall of a different browser began.

I havent looked at the source code, but it might be a difficult overhaul to fix this issue. There is plenty of theories floating around on how to fix this, but they seem to be pure speculation.

I dont think you can compare this to the downfall of IE, since you yourself could pull the chromium code and fix the issue yourself.... When i see some actual real code fixes in the comments, and the Chromium team STILL hasnt fixed this bug, then i'll call shenanigans.

Until then, everyone in that thread needs to submit patches or GTFO.

If its so easy to fix in theory, then why dont I see someone just submiting a patch, or forking Chromium w/ the correct behavior?

I think its stubborn of them to leave the issue as "WONTFIX" but people whining about it without real-world code example fixes isnt really any more useful.

Nobody wants to spend time researching other people's code and writing a patch only to see that effort go to waste. In this case, the fact that the bug is closed and marked as WONTFIX is a strong indication that patches would not be accepted. Before people can get to code, the maintainers would at least have to say "we've only marked this as WONTFIX because we don't have the resources to deal with this, but we'd actually like to see this fixed if someone else provided a patch". But they haven't said that.

Reply on news.ycombinator.com