spencerone@???:
>>> sajolida:
>>> Also keep in mind that **none** of these mockups were tested with
>>> users yet. We talked about them amongst ourselves, we reviewed it with
>>> experts, etc. but we never sat with regular users and watch them
>>> interact with those.
>>
>> So I don't want to wait again for months,
>
> No worries. The flow based on the current proposal[0] is here[1]; I
> will try to keep waiting to a minimum. I updated what exists of the
> wizard to match the current proposal, excluding the 'Introduction' page
> (maybe that can be linked to from an information (i) icon).
>
> The only question so far is, should all buttons be aligned to the same
> side? Right now the screens are laid out logically, independently of
> each other, though aligning them may increase visual clarity.
>
> Also, we have solid resolutions for whichever option we choose regarding
> displaying the more technical settings on the first screen.
>
>> which doesn't prevent anybody from organizing a paper test session
>> timely.
>
> I got this, if we feel strongly about testing. I have access to people
> unfamiliar with Tails that would give us what we need, should we ask.
> Questions arise, such as: What are we trying to test? How many people
> should we sample? Specific use cases/threat models should we include?
> And so on. I know we have guides, but are there any specifics not
> addressed here[2], such as when we need this by?
I don't want to plan these testing sessions in detail now but to give
some first rough ideas, I think we need:
- 10 to 20 people both unfamiliar and familiar with the current design
- have people involved in the design and people not involved in the
design process when doing these tests
The calendar mostly depends on when we believe we can have something
ready to test (either good paper prototypes, either with code), then
find availability for some of us to be there, travel, etc.
> If not, no worries, we can rely on logic to make such designations,
> since that's the foundation the brain operates from. We can get super
> close with just thought experiments, tweaking the interface from there
> with user feedback after each release.
I disagree with that. We surely have to do all the thought experiments
that we can but they will never replace real user testing.
