Re: [kde-artists] We need a Vision!

On 13.03.2014 12:40, Aaron J. Seigo wrote:
> On Thursday, March 13, 2014 11:44:38 Thomas Pfeiffer wrote:
>> I agree with Björn here: Just because we currently cannot contact our
>> users easily, we should not just give up and say "Well, let's just do
>> what we think our users need, because we will never find out what they
>> really need".
>
> At the end of the day we will do what we think our users need. The question is
> what will inform that. So "do what we think our users need" is not wrong, nor
> is it related to giving up.

It _is_ possible to find out what users actually need. Of course when 
creating a new UI one first needs to infer the needs from more general 
information (though it can be specific to the task, for example), but 
then with user testing it is possible to find out whether the "guess" 
was correct.

> So what can inform that final position? There happens to be a huge group of
> people who use computers and computing devices on a daily basis. Most of the
> people who live in the same towns and cities as we do, in fact.

Sure. Taking your computer out in a cafe and inviting people to try out 
your software on it is a great way to see whether it works well for 
"normal people", it is what I mean by observing users.

> There is probably no need to reach *our* users to understand what *people*
> need and want. Unless our goal is to make KDE software better only for our
> current users, this is probably more than enough.

It depends on the software. If it targeted at a broad user base, 
gathering information from people on the street, your coworkers or your 
family works great. If it's specialized software, one has to go to 
people who would use it, of course.

> Conversely, trying to go to *our* users will end up with us talking to a
> minority of our user base (the ones we have contact with, and who are
> motivated enough to reach out in return) who, I guarantee you, will not
> represent either our general user base nor the general computing public.

We want our software to work well for our existing userbase, too, don't 
we? Of course they can only be a part of our sample, not the whole sample.

>> As long as we do that, we design and develop based on our
>> own needs and/or guesswork, neither of which is optimal.
>
> Thankfully that's never been what I have suggested, or practiced.

You need to talk to and observe existing and/or potential users, or you 
will end up with guesswork, that's all I said. And I know way too many 
software projects (commercial and community-made alike) that do just 
that. People _think_ they know they potential users, when in fact they 
don't because they incorrectly infer from their own experience, skills 
and preferences to that of their target users.
I don't say _you_ are doing that. I say that without input from 
observation and interviews, you'll end up with that.
______________________________________________________________________________
[email protected] |  https://mail.kde.org/mailman/listinfo/kde-artists