On Tue, 2016-11-15 at 10:28 -0800, Chris Murphy wrote: > > I'm not counting change sets. I'm counting user distinguishable UIs, > and those are four distinctly different interfaces and experiences and > they each even have their own names: > oldui > newui auto > newui custom > blivet-gui
I understand your enumeration. However, I disagree that it demonstrates this: > That is a lot of churn in UI/Ux in 5 years Because: 1) You're using a classic bogus statistics trick: choosing misleading boundaries. You can only claim "in 5 years" by *just happening* to set your initial boundary at the precise point where we switched from oldui to newui. oldui had, at that point, existed for, like, a decade. So even accepting your enumeration, it would be equally valid to say 'four UIs in 15 years' (or whatever the precise number is). Which doesn't sound nearly as bad, does it? 2) 'newui auto' and 'newui custom' are really different elements of the same interface. They were designed that way. You can't legitimately count them separately for the purpose of illustrating 'churn', because they can only properly be counted separately for that purpose if we first made one, then changed our minds and made the other. That is not what happened. Both were elements of the newUI re-design, which was a single unitary concept. Both exist right in the initial whiteboards for newUI. It's wrong to count them as two things for the purpose of claiming a total of 4 as an illustration of 'churn'. Just wrong. 3) We have not in fact built an installer interface which involves blivet-gui yet and AFAIK no plans to do so are finalized. So once again, your boundaries are suspect: how can you count something that hasn't happened yet in your 'five year' timeframe? -- Adam Williamson Fedora QA Community Monkey IRC: adamw | Twitter: AdamW_Fedora | XMPP: adamw AT happyassassin . net http://www.happyassassin.net _______________________________________________ devel mailing list -- devel@lists.fedoraproject.org To unsubscribe send an email to devel-le...@lists.fedoraproject.org
