Re: [Tails-ux] [Tails-dev] Testing with openqa?

sajolida:
>>> Not as if we were taking much benefit from Gherkin for
>>> {intra,inter}-team communication yet, but I like it that we're able to;
>>> and I intend to try to use Gherkin more in the future when discussing
>>> changes with UX folks.
>>
>> This sounds interesting. I have read some negative things about this
>> but the argument made was that the language did exactly as it
>> advertised; making humans still have to talk to other humans, as it
>> didn't magically resolve their socially retarded nature :)
>>
>> I am open to poking around with this sooner than later, even in test
>> communication runs, if not with live designations decisions.
>>
>> I have copied the UX list in case people want to nerd out over there, too.
>
> Yeap, I'd like to know more about that intrigeri meant but I guess that
> will be done in due time.

I recommend you tails-ux@ people to look at the features/*.feature files
in Tails' Git. Ideally anyone with domain knowledge about *using* Tails
following the steps in those files should be able to reproduce more or
less exactly what our automated test suite does when following them. No
further explanation should be needed, and in particular one shouldn't
have to look at the code == step definitions.

That said, in many cases this is not true in its current state due to
technical limitations/shortcuts (but there will be quite a few nice
improvements soon, with #6094 being solved). But it would be interesting
to get some help with #10329 from "outsiders" like you tails-ux@ people.
I suppose we, as Test Suite Developers, often think a bit too much about
the implementation, resulting in scenarios where steps are split
according to that, contrary to how a user might think. That may actually
be a good thing (to reduce the code complexity for the step definitions)
but it would nonetheless be interesting to see how they would compare.
Perhaps there is a better balance.

Cheers!