Plasma User Types

Sat Apr 12 15:41:22 CEST 2008

On Saturday 12 April 2008 09:29:21 Michael Rudolph wrote:
> On Saturday 12 April 2008 14:58:26 Celeste Lyn Paul wrote:
> > On Saturday 12 April 2008 08:24:46 Michael Rudolph wrote:
> > > On Saturday 12 April 2008 11:17:23 Aaron J. Seigo wrote:
> > > > On Friday 11 April 2008, Michael Rudolph wrote:
> > > > > ruled that answer out in a previous blog post, Celeste, I
> > > > > think, that "everyone" is the description of the typical plasma
> > > > > user type.
> > > >
> > > > we are creating a framework upon which interfaces get built that
> > > > get used by the "everyone" set, but the group we keep in mind
> > > > while designing things does not need to be the "everyone" set.
> > > > indeed, it can be rather focussed. where that focus gets applied
> > > > is the purpose of this conversation.
> > >
> > > Hello everyone,
> > >
> > > thanks, Celeste and Aaron, for your clarifications. I'm still not
> > > really convinced.
> > >
> > > My point is, that all user types are important and that I don't
> > > really see the benefit of splitting them up. But I see dangers that
> > > this splitting up bears.
> > >
> > > When we start to look isolated at the socialite, for example, we
> > > will come up with personas and scenarios that may very well contain
> > > a twitter applet, some facebook or myspace integration and many
> > > other things. And if we do this, please excuse my strong words, we
> > > would have been better of with just doing a KDE3.6. No one, never
> > > ever, in his right mind never wants to use twitter, ever! People
> > > want to communicate! We have to look at what the user is actually
> > > trying to do. When we look at users in terms of their experience or
> > > their background on Linux or Windows, we are already way too deep
> > > entrenched in all the concepts that make current user interfaces so
> > > wrong.
> > >
> > > My favorite way to clarify this is hanging up a picture. Say you
> > > got that new Picasso for your dorm room and are driving a nail into
> > > the wall as your roommate walks in and asks what you are doing,
> > > standing on a ladder there. What do you say? - I'm putting this
> > > picture up. No one would say: I'm using a hammer. Because a hammer
> > > has a good user interface. The user doesn't even realize he uses
> > > one. He is completely
> > >
> > > concentrated on doing, what he is actually trying to achieve. If
> > > hammer designers went about as we do, they would create a docking
> > > station for the hammer, customizable decals for the shaft and
> > > exchangeable versatile heads, that make the hammer so fragile that
> > > you cannot slam it, but have to tenderly pet the nail. Users would
> > > be more occupied with their hammer, than with their actual task.
> > > You are also not aware of the ladder, you just use it. Because it
> > > has a good user interface. To use a KDE ladder, you'd probably have
> > > to log in, and you'd have to reboot for every rung you wanted to
> > > take.
> >
> > The use scenario is displaying a picture.  The use case is using the
> > hammer and putting a nail in the wall so you can mount the picture on
> > it.  The environment is a college dormitory which could have either
> > cinder block or dry composite walls.
> >
> > Given a different environment, maybe a hammer isn't available and you
> > only have to work with sticky tacks.  That probably happens more
> > often in an environment like a dormitory than a house.  Of maybe you
> > don't have a ladder because you tend to be a tall person.  Or since
> > you are young, you can pick up the 20 Kilo frame much easier than a
> > little old lady.
> >
> > All of these factors are going to experience of the picture because
> > part of the experience is putting it on display.  Now, if you are a
> > college poster manufacturer and 80% of your sales are from college
> > kids, you probably aren't going to worry about how the 1% sales of
> > little old ladies who can't pick up 20 kilos and don't have a ladder
> > are attempting to display your product. Anything you might do to
> > accomodate that 1% customer base might effect your other customers
> > and cause you to loose sales from your 80% customer base.
> >
> > That is why come up with user groups, types, scenarios, and cases.
> > To really understand who is using our product, how, and why.  That
> > way, we can improve the product, reach out to customers we want to
> > include, and understand when we make a change to accomodate one user
> > type, how it will effect other user types.
> >
> > Sure, *anyone* can purchase the poster and frame and put it up on
> > their wall. But unless the manufacturer had a specific interest in
> > reaching out in to a specific customer base, why would they risk
> > upsetting and losing primary customers?  The manufacturer isn't
> > trying to please *everyone*, just their *primary* customers.
> >
> > Also, to further clarify.  This user group information isn't mine.  I
> > didn't make it up.  This is knowledge which exists within the Plasma
> > project.  All I did was go about collecting it in a scientific way so
> > the Plasma developers and sit down and discuss it so they can make
> > use of it in future development of the product.
> >
> > > I'm not sure I can spark a constructive discussion around that
> > > issue, so I'll just let you guys go on for now, but the important
> > > questions we need to ask ourselves right now, have little to do
> > > with experience or curiosity levels, in my point of view.
> > >
> > > michael
> > > _______________________________________________
> > > Panel-devel mailing list
> > > Panel-devel at kde.org
> > > https://mail.kde.org/mailman/listinfo/panel-devel
>
> Hi Celeste,
>
> thank you very much for taking the time to answer, also for writing it
> down in a comprehensible way, like I could not have done it.
>
> The only thing I'm still unwilling to accept is, that one has to leave
> the old ladies behind. (how sad is that? :-) A sign of good design is
> that it caters to 100% (which I like to simply refer to as: humans),
> and only if one fails to come up with good design should one try to do
> bad design, that at least appeals to 80%. That's all.

I don't believe in universal usability, but that is simply a position on 
philosophy.  Even in voting (a research project I work on in my day job, as 
you can imagine it is a big deal in the U.S.) we specify user types because 
it helps us accomodate the users who will have the most trouble or make the 
most mistakes.  

I'm glad my example helped you understand where I'm trying to go with this 
though.

> But I'm very stubborn, so don't waste your time trying to convince me,
> and you seem to be knowing what you are doing, so I will not waste
> mine, trying to convince you, when all it seems to be about is
> methodology. We should continue this discussion when there's a report
> and analysis to discuss. Until then I'm not going to pester you with
> this matter anymore.
>
> So thanks again for your time.
>
> michael
> _______________________________________________
> Panel-devel mailing list
> Panel-devel at kde.org
> https://mail.kde.org/mailman/listinfo/panel-devel

-- 
Celeste Lyn Paul
celeste at kde.org
KDE Usability Project
usability.kde.org