unity-design team mailing list archive

[GonzO <gonzo@xxxxxxxxxxx>]

Evenin'.


> I'm curious if your talking personally or your speaking about people in
> general.

Both.

I was, for a good number of years (and in some respects, still am) a
sysadmin.  I fixed people's computers and networks and then taught
them how to use it all, including customizing their OS to make sure it
could easily do what they wanted it to do.  Then I was top level
software support, and now I design and write the user interface for a
piece of software.

So when I write here, I'm writing about both what I want out of an
OS/GUI, and what my customers / friends / family members have
expressed an interest in having.

*At the end of the day*, it is of course all my subjective opinion...
although I would argue that I base my opinion upon a decently large
sample set witnessed over a decently long period of time.


"To know what apps are open is something that in my workflow only
happens on occasion."

In this particular case, I was talking specifically about the
difference between
1) an icon on the screen that *also* tells you (via a glowing dot or
whatever) whether or not the app it represents is open (i.e. "Old" OSX
behavior)
2) an icon on the screen that represents an app without this indicator
(apparently, "New" OSX Lion, or so I've read)

In either case, the icon is *still there*.  No space is being saved,
but the icon is apparently just made dumber... well, just because.  If
I want to see the at least some of the information it *used to*
contain, I now have to go elsewhere in the OS.


>I think that the direction that Unity is going in
>this regard avoids this issue, you select the application, if multiple
>instances are open you get to choose, otherwise it's launched.

Unity also does not do away with the visual "am I open" indicator,
though.  In fact, it lets you know if its open *and* whether or not it
has the current focus.


>I would judge the later one to be more the realm of
>'power users' and is not a normal behaviour of the average user.

I think we need to do away with this entire concept of "average user"
and "power user".

One problem with these distinctions is that they are fictitious and
apparently apply to nobody; everyone thinks they are the "average
user", and no one seems to want to fess up to being a "power user".

Another big issue is that the criterion and demographics between
"average user," "power user," and "whatever other kinds of users
aren't covered by those two" are shifting.  It is becoming rarer and
rarer for someone to only use their computer to "check email and do
some light browsing" - what I think you refer to when you say "average
user" - because more and more of those users are using their phones
and netbooks to do this.  Which leaves the "average" Desktop Computer
user as either WoW players, or targeted users (schoolwork, graphic
design, music making, programming, etc. etc. - not exactly a "power
user," but certainly not "average" as you seem to envision it).

But the most important problem is that the distinctions between
different sorts of users are: it's just a big red herring.  A UI that
makes sense and is convenient is *not* specific to one's level of
experience or amount of use.  Things either make sense or they do not.
 It is either easy and intuitive to use this tool, or it is not, and
both a rank amateur and a seasoned pro can agree on which is which.

It is for this last reason, especially, that I find the idea of
catering to one crowd or another to be repugnant.  We should be
attempting to design a UI that is useful and intuitive to *everybody,*
not just letting the "average user" mob rule.

(More repugnant than this, however, is that we're apparently no longer
content with stripping down the UI to appease "grandma," leaving us
"power users" to use hidden .gconf/registry hacks to get some
semblance of a usable UI back.  Now the focus seems to be on telling
"power users" that how we use a computer is essentially wrong, and
that we *really* don't need to see that CPU monitor or Clock after
all, and should really just be focusing on the task at hand.  But at
this point, I digress.)


> Hmm. Maybe in your use of a system you need to know the CPU all the time,
> but I know of very very few users who would fine this information useful at
> all, let alone all the time.

I knew someone would pick up on the CPU thing, and then tell me that
this is an uncommon use as a way of both missing and dismissing the
point.

It's not the CPU meter *in specific* that I was pointing out.  Yes,
its one of the gauges that are important _to me_, but the point I was
trying to make isn't about the _specific tool_ I like to see at all
times.  The point, rather, is the *idea* that there are some things
just about *every* user is going to want to be alerted to in as close
to real-time as possible.  What those things /are/, specifically,
differ from user to user, but the need for a GUI to have a way to
notify a user of things is just about universal, which is why all
major OS'n have them.


> Menus, docks, etc are not required
> to be watched all the time to use your computer.

It's not about *watching* the UI elements I was talking about
(although that clock on the wall is only really useful because it is
always visible).  The things I pointed out as "always need to be
visible" are not simply inactive displays.  They aren't useful because
I need to "watch" them, they are useful because I frequently need to
_click on them_, which is much harder to do if I can't _see_ them.

Menus.  Dock apps.  Video controls.  Messaging indicators, clipboard
manager, package updater, menu button.  I *do* use these things very,
very often, and I (and most of the people I've seen and taught how to
use computers) seem to be most efficient and at-ease with the UI when
these important things don't A) disappear or B) move around.


> The tachometer is something
> you can watch if you like when changing gears in a manual car. Most people
> tend to use the sound of the engine (they get used to the sound of how high
> the car is reving to decide if they need to change gears)

Nit-pick from a standard transmission owner and operator: that used to
be true, before engines became so quiet, cabins so insulated, and
radios so loud that you can no longer hear the engine whine until its
in the red, if even then.  And even that point isn't as important as
bringing up people who simply can not hear.


> There is a valid usability based reason to remove data that the average user
> has no use of IMHO.

Again, I think the error here is in assuming there IS such a thing as
an "average user," not to mention pretending to know who they are and
what they want, not to mention using that as an excuse to design a UI
that excludes the other sorts of users.

There is no valid reason I can see to remove the Docks, menus, network
tool, bluetooth tool, or even so help me, the clock - instead of
simply giving everything on screen the ability to be hidden *at the
user's behest*.  I don't see why these very useful tools - all of
which are, IME, best used when they are not invisible or in motion -
need to be wiped from the screen to cater to the shrinking crowd of
those people who only ever turn the box on to click on a YouTube link
they found on FaceBook.  Especially considering the absolutely
*paltry* amount of screen space that we are "saving," - which, in some
cases, is zero - and double-especially since the idea of
"Fullscreening" an app (making it the only visible thing, on top of
all OS chrome) has been around since Windows 95, and now appears to be
a full-time, all-app player in Lion.


> In essence I'm trying to say that IMO the job of the OS should be to allow
> the user to do what they want

THIS user wants to see his mailbox and CPU meter all the time.  No
matter how many times the GNOME and Unity people tell me that
actually, I do not.  ;-)

Funny story: OMGUbuntu announced today the availability of an
"Old"-style menu indicator for Unity.


> But when 80 or 90 percent
> of the time is spent looking at the contents in a window, there is no point
> cluttering up the landscape with irrelevant information.

Which is, frankly, why I like Mac's solution in Lion.  Because it
isn't their solution.  It's the same F11/Fullscreen button we've had
in either this app or that for well over a decade.  It lets the user
*decide* whether or not they want to maximize an app (and still see
indicators) or fullscreen it to the exclusion of any OS chrome.  They
aren't *forcing* anyone to pick one or the other based upon some
"average user" fantasy, and I don't think Unity should go down that
road either.


> I would suggest that people are serious about improving Unity then we need
> to be developing proper use cases to document the different work methods
> people utilize.

Frankly, I think that's aiming so low as to be shameful.

I think we need to be developing a GUI model that is useful no matter
what the work methods.  The only reason we should be finding out how
people work is to incorporate those different work flows into an
overall design that caters to them all, not to say "well, only 10% of
people care whether or not an app is open, so its 90% safe to get rid
of the little dot on the icon".


> To me it seems that a lot
> of the arguments are subjective, some based on an emotional reaction.

As well they *should be*.  The subjective perspective is really the
only one any of us have, and the minute we forget that people stop
using OS's that *make them angry or frustrated* is the minute we lose
all relevant design focus.

IMO, of course.  ;-)

--G