Adam Williamson wrote:
On Mon, 2009-11-16 at 00:31 -0500, Gregory Maxwell wrote:
It's not a "perhaps do something stupid": As far as I can tell reliably dies every time during the update and the session saving function fails to do its job. Everything the user was working on in Firefox is lost.
This is not what happens to me. What happens to me is that Firefox continues to run but the search box doesn't work any more.
Perhaps people should recognize that everyone experiences slightly different behaviour in this particular scenario, and that everyone uses their systems differently, and stop extrapolating from their own experience and assume that it's perfectly fine to kill off every running firefox instance on a system when the package is updated. Because that's not something you just do on the 'hmm, well, it feels like it'd work for me' basis.
What are you trying to say. A set of users are saying that current procedures cause data loss for them and are suggesting a procedure that would avoid the problem. You don't experience data loss. That doesn't help them. You are arguing that you don't need their solution and that their solution would inconvenience you. Even though this has not affected me as yet, I think that it is better to inconvenience me than to cause data loss to others. In my world, even denial of service is only a severe error, data loss is a critical error.