And no, my experience on general user behaviour is to expect no backup, but this isn't Auntie Ednas crocheting Facebook group, but Hacker News :)
And no, my experience on general user behaviour is to expect no backup, but this isn't Auntie Ednas crocheting Facebook group, but Hacker News :)
I used Arch briefly and this was not my experience, but at least in that community it's kind of expected that you understand this is a possibility and a tradeoff of running the OS.
I did find the Unity interface more polished however.
It seemed like Canonical did a lot of UX research, to make things like the "power off" button adjoin the corner of the screen, so you could imprecisely flick the cursor and know it has hit the target. It also worked a lot better under old hardware.
I still use Unity with community support. It's a shame, I think as I remembered seeing the Unity interface at work sometimes and thinking that Ubuntu was making inroads.
It is only an impressions, Arch fanboyus will quietly try to fix the mess and blame themselves for the bugs, only some honest users will tell you straight in the face "never update Arch without first reading some news page and never update if you don't hve the time to rollback and fix shit".
Arch is, in my experience, much more stable, and yet you should glance at the news page and run full system upgrades when you could spare some downtime if you had to.
Keep in mind these upgrades are being done to an OS that has a unique history based upon how the computer was used in the past. Issues that did not appear in testing are going to come up after release. Then there is the potential defects in the manufacture of a particular unit or due to how it was handled. In other words, it is legitimate to miss an uncommon fault.
As for Arch, I understand why the warnings exist. That being said, I have found it to be very reliable. I typically attribute it to changes being incremental, meaning that problems are less likely to arise; and due to development being done in the open, resulting in a larger pool of testers before it even hits rolling distributions like Arch (never mind distributions that do their own testing).
Even assuming this is the case, it still doesn't excuse how they weren't able to uncover this in their testing, since they only have to test against their own set of hardware.
Makes you think on the QE that goes into the release.