[maemo-users] OS2008 Damn ugly!!!

From: Peter Flynn peter at silmaril.ie
Date: Fri Dec 28 17:24:00 EET 2007
inode0 wrote:
> [...] I'm not hearing very many in this thread who agree with
> the OP's characterization of the new desktop look.

Noted. Some people may not have seen it yet.

>> I agree the original post should have been less inflammatory, but I
>> think it was very valuable nevertheless. Particularly if it helps make
>> the designers think harder next time.
> I don't have any reason to think the designers didn't think hard 

Me neither. On the contrary, I'm sure they did think very hard (you have 
to: design is *difficult*). And sometimes you have to make that leap to 
get rid of an old or outdated paradigm.

Alfredo J. Fabretti wrote:
> Critics are good if they have a basis. Usually I don't listen or pay 
> attention to that kind of comments because they don't have an
> argument which I could use to make a decision or improve a product or
> software.

Well, in this case it did: an allegation of ugliness.

 > Well, sometimes subjectivity shows only a person's taste.

And sometimes not.

> For example saying that "X car is ugly" isn't going to change (or
> help to change) what designers and engineers have planned and
> developed for years, 

Unfortunately :-) The fact that the designers and engineers went ahead 
with the Fiat Multipla doesn't redeem it from being the single ugliest 
car in the world. It has many other good qualities but aesthetics ain't 
one of them.

> time will make people get used to it 

I always worry a little at statements like this as they imply that the 
designer is always right and that "people" are always wrong. I'm pretty 
sure that's not what the author meant. At my age I've "got used" to 
seeing Buckingham Palace, but that doesn't change the fact that it's 
simply not a very good piece of architecture.

 > and understand why things were made that way

This is *much* more important. Good interaction design is 
self-explanatory: the user doesn't have to understand or learn how to 
use it (or only has to be shown once), and it works right first time and 
every time. A good example was the rotary-dial telephone: users did not 
need to know how a Strowger uniselector worked in order to dial a call. 
But interaction design is very different from aesthetic or visual or 
graphic design. With the use of plastics, a telephone could be made to 
*look* like anything you wanted, but its interaction design was the same.

With GUI design we have to be especially careful not to break 
established models without reason: don't suddenly require two clicks 
where one was needed before (or vice versa); don't hide one icon 
underneath another because you think it looks prettier, or because you 
yourself never use the hidden one; don't move important or 
frequently-used functions to a hidden or obscure place several layers 
down a menu; don't promote unimportant or useless functions to desktop 
significance; and so on. People approach the desktop metaphor nowadays 
with considerable loads of background knowledge or even baggage, most of 
which needs to be taken into account, not ignored. It *is* possible to 
ignore it, if you're good enough and big enough -- Apple is one obvious 

 > (obviously things aren't
> perfect and there's always time to change or fix them). Most humans
> reject new things until they get used to them.

Yep. We're a conservative bunch, humans.

>> I agree the original post should have been less inflammatory, but I
>> think it was very valuable nevertheless. Particularly if it helps
>> make the designers think harder next time.
 > Designers need constructive criticism, comments like the original post
 > only hurts their feelings.

I dislike hurting people, particularly electronically, but once or twice 
I have lashed out in public at a particularly unnecessary, careless, 
stupid, or thoughtless action, deliberately to shock the author into 
reconsidering.  Usually it works, but at a heavy penalty, which is why 
the last time was nearly 15 years ago.


More information about the maemo-users mailing list