On Tue, 2013-11-26 at 11:58 +0100, Alberto Ruiz wrote:
> Based on my experience (>10 years of it, with multiple
> OSes), this is an incredibly ambitious goal. It may in fact be entirely
> unachievable as written. I'm not aware of a single operating system in
> existence which actually achieves this.
Citation needed? Windows Mobile, Android, iOS, PlayStation 3, XBox... I
never heard any users of these OSes complaining about how upgrades broke
their system in an ongoing basis.
So, two problems with that:
1) you moved the goalposts. The draft doesn't say 'upgrades should
mostly avoid breaking people's systems', but you wrote "how upgrades
broke their system in an ongoing basis." The draft ties us to a _much_
higher standard than boring old "doesn't break systems".
2) Windows Mobile and Android devices frequently just don't _get_ OS
upgrades, or get them very belatedly. I've seen Android upgrades shipped
that aren't really 'upgrades': you could only 'upgrade' by flashing
clean and starting over. PS3 and Xbox are so different from what we're
doing, plus who knows what the hell is in any of those updates? It's all
secret sauce, all the way down. The only ones that might be somewhere in
the neighbourhood are iOS and Nexus phones, but I don't think even those
hold up to the draft's wording when looked at carefully. It really is
setting an extremely high bar.
Yes, maybe from time to time, somebody
hits a problem there, but the upgrade process in those systems is pretty
robust due to several design decisions.
> Even cellphone manufacturers -
> who have a very clearly-defined single piece of hardware to deal with,
> and a much smaller set of software and use cases to worry about than we
> have - don't achieve this. I'm really not sure it should be front and
> centre in a foundational document without some really convincing
> evidence that it's even vaguely achievable.
Again, I've never had an issue upgrading my PS3, Android phone or my
Again, 'never had an issue' is not the same thing as 'upgraded system
must function precisely like a newly-installed one'.
The fact that we may not achieve this goal in a 100% flawless
doesn't mean we have to give up on it altogether, the room for
improvement here is huge, and anything we can do to make this better is
worth every line of code. This problem is a major scare-away for many
I haven't disputed that, I've just raised a concern that the specific
aspiration written in the draft is an extremely hard one to meet. I
didn't suggest that we should just not care about upgrades, or
Fedora QA Community Monkey
IRC: adamw | Twitter: AdamW_Fedora | XMPP: adamw AT happyassassin . net