Re: UI design [was Owen Taylor's paper]

On Fri, 2002-09-20 at 11:10, Justin Moore wrote: 
>    If you automatically block them, how do you even know what they have
> to say?  How do you know whether or not they're important?  And what a
> plasma physicist may find unimportant, my Great Aunt Tillie might really
> need to know.

yep your aunt might want to see pop up ads....I certainly don't.  And
like I said..if you are trying to ignore things instead of reading
them...you shouldn't be seeing them to begin with. And in fact a little
inconsistancy in interface helps make people read the information...its
part of what webads call "stickiness". And "important" dialog
boxes...like "warning: about to start WWIII" should use measured amounts
of UI inconsistancy to make people stop and think before clicking
without thinking.  I know this is hard to believe but thinking is
actually a good thing to encourage in users...in human beings in
general.

Take a good long look at pop ads and web services like banks and you see
GLARINGLY HUGE inconsistencies from website to website in how things are
organized.  Sure inside the pages of a certain organization things are
consistent...but from website to website things are very different.  And
yet people can navigate the web pretty handily.  Layout is for
style...and diversity in layout is as much a good thing as it is a bad
thing. But having something akin to a tabbed collection of dialog boxes
that stay out of your way and are still accessible is not a bad idea for
desktop UI.  You don't want to be annoyed by applications giving you
unneeded dialog boxes...have then come up iconified by default.  Don't
give people an excuse to click through on instinct...its a bad habit. 
If you are going to ignore them...then bloody ignore them and don't have
them come up at all or come up iconified.

>    I agree that reading labels is important, but the honest-to-god truth
> of the matter is that 95% of people that use computers don't do that or
> *want* to do that for every single dialog box that pops up.

People don't necessarily want what is best or even good for them.  Don't
pander to  bad habits...encourage good habits by keeping unneeded and
unwanted dialogs out of the way. Make users believe that the dialogs
applications are giving them actually hold information that is of import
and relevant .  You load up applications with dumbed down wizards and
dialogs that users can not control...and eventually they get sick of
them becuase they are unneeded.  Don't give people unneeded information.
Consistant layout just gives people an easy means to ignore the
information presented...if its too much information...or useless
information...the content needs to be fixed or there need to be user
control to limit the flow of that noise. 

>  We should
> take as a given that these people are going to zip through the boxes,
> and plan for that.  
No no...you should plan to give people more control over the amount and
quality of the information the application is trying to present to them.
If people don't NEED the dialogs and are not READING the dialogs...why
in the hell do we have the dialogs there?

>Make the buttons consistent; don't piss your users
> off by putting a "No" box where the "Yes" box usually is and aid them in
> "No"t saving all the changes they just made to their research paper.
If you really believe button placement is MORE important than the button
text...turn off all the labels and icons in your menus and dialogs...and
live in a UI consistant environment...go try to have fun and see how
much quicker you navigate in your positional consistant interface. 
GUI's were never designed with layout placement meant to encode
information...layout is stylistic...not informative.  

> 
>    Someone clicking through a dialog box without thinking is like
> pulling a rolling stop at a Stop sign.  We all know we should stop and
> look and double-check to see what's going on, but in reality we just
> slow down, give it a quick glance and keep going.  That's human nature,
> and inconsistent dialog boxes are just going to piss people off.

And we have LAWS against this kind of behavior!!!!!! Should we pull the
laws off the books..and design our traffic rules around such ignoble
human nature?  Please you really don't want to use this argument...in
fact I'm really disappointed in the quality of your debate.  Designing
for base human nature does not build a better society...nor better
technology.

> The goal: Get 1 million dollars
> Today:    You see a $20 bill sitting on the floor.
> 
> Question: Do you pick it up?  According to you, you wait until you find
> a check for the full amount sitting on the ground in front of you.

What no....you didn't give me a choice of competing options....You
analogy is not analogous to where Redhat finds itself...with a
collection of applications with various features and maturity with
varying UI elements.  Where are the trade-offs between the competing
interests in your analogy?  If Redhat's gaol is to simplify the default
desktop experience...a stronger emphasis on application features and
maturity over UI consistancy is the better compromise in the current
situation.

> 
>    Yes, prioritize, but if there are people that are willing to make the
> UI consistent, but are unable (or unwilling) to write the next great
> Killer App (tm), don't tell them that their work is unimportant.  Just
> because someone is spending time enforcing consistency among a(the)
> desktop(s), doesn't mean they should drop what they're doing and start
> working on the Whiz-Bang CPU-to-cerebellum UI hardware/software.

I didn't say the work is unimportant...its just less important.  And how
do you enforce consistency among the desktops or among websites?  You
can't enforce that...unless you OWN the platform and dictate to
developers.  You certainly can encourage consistency but to enforce it
would imply a level of power over others that OSS and web development
just can't give you.


-jef

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.6 (GNU/Linux)
Comment: For info see http://www.gnupg.org

iD8DBQA9i2R9rWLDmRitRZURAqkwAJ47RNN3+UNyHLayLpXqH1vWSmUrOACePc7o
54Relj9k0jueYEeTqMAS3pc=
=rt9v
-----END PGP SIGNATURE-----