Re: Clipboard

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On Tuesday 06 August 2002 12:43, Dave Leigh wrote:
> I would give your theory far more credence if it weren't for the fact that
> skinnable applications are so immensely popular. Skins, themes, and
> preferred window managers all argue against you.

skinable apps do decrease usability when the skins are not consistent across 
the interface. this is a known fact in the usability world. people use them 
because they like to customize things that are "theirs", usability be damned 
(and later they complain about inconsistency ;-). KDE (and some other 
desktops) provide across-the-desktop skinnings (e.g. all widgets, all icons, 
etc) which limits the impact of such skinning.

however, we are not talking about *you* skinning *your* apps.
we are talking about skinning *everyone else's* apps.

this is a change we will be imposing upon all those who use kde, whether they 
want it or not. with skinning they have the option of picking what suits them 
best, here we are trying to improve the "one size fits all" suit. this needs 
to be done carefully and is a completley different dynamic, one that can not 
be handled with the same cavalier attitude as creating a new optional icon 
theme.

> Now, that said, I do agree that the replacements should be better. What I
> disagree with is that they need to be *remarkably* better. As I look at the

fine. let's not get into semantics. i'll remove the qualifier so as to give us 
less to debate: the new icons have to be an improvement that makes the loss 
in learned interface worthwhile.

> calendar I find that I've been working in this industry for a few months
> shy of twenty years, as a trainer, programmer, and designer. In that time

length of experience is no sure measuring sitck of anything. so let's not even 
bother going there.

> I've seen users confused about a lot of things, but this simply ISN'T one
> of them. Users just aren't as stupid as you seem to believe. As I've

for the Nth time already: what i'm talking about has nothing to do with 
intelligence, knowledge or lack thereof. i'm talking about how people develop 
patterns and habits. this is completely different from purposeful, conscious 
learning.

> Again, learning the functionality the first time is where the time is
> spent, and something that's more comprehensible and consistent is
> desirable, even if the advantage is incremental. Why? Economics. For an

change your icons to completely different symbols every week whlie keeping 
everything else exactly the same. see how comfortable your already-learned 
system is then. we're talking about habits and comfort which lead to better 
usabilty through both real (quicker responses) and psychological (the user is 
more confident) forces.

> improvement to be a "good deal," the benefit does not have to be delivered
> to the same place as the cost. 

i never said that. merely that there needs to be a net gain. i'd be happy to 
hear where the net gain from these icons comes from, and by how much.

>However, the OVERALL benefit should outweigh the cost.

_thank you_!

> The largest user base is the one you don't have yet. There are
> far more future users than there are established users, and even
> incremental benefits serve to sell the product, even to those users of
> other systems.

completely logical, except you are still failing to answer the eternal 
question (or at least it seems eternal because i keep asking it and you 
aren't answering it):

   are these icons an improvement (support with evidence)?

> The established users are not terribly inconvenienced and
> the future users far outweigh them, both in numbers, and because their
> training costs (where the benefit is primarily applied) outweigh the
> retraining costs of the established base.

the cost to our installed base can not be offset *to our installed base* by a 
future installed base. the costs remain to be paid by them, and they are our 
responsibility as clients of our software, if you will.

and how long until we have an existing installed base that is big enough to 
warrant concern? remember, we don't have to get people to buy a new license, 
so we don't have the economic (mis-)motivations of traditional software devel 
here. we can afford to, and indeed have an imperative to, take care of our 
existing users.

> Remember that the established base of Linux users often chooses to change
> its icons, controls, and UI functionality at will. In the Windows world
> they do choose to do this where allowed by skinnable apps. They would not
> do this if the core of your argument were as valid as you believe. As it
> is, users often choose to change their UI for no logical or functional
> benefit, merely an aesthetic one.

i would challenge this. most of non-hobbyist/power users i have observed and 
read about change their environment early on (if at all, ergo the power of 
defaults) to get it the way they personally like and then keep it there for 
long periods of time because they are comfortable with it. just like the 
couch potato w/his remote. the most drastic changes most average users make 
is to switch their background picture.

> Give your two couch potatoes EACH OTHER'S remotes and ask each of them the
> same question. Unless one has the Magic Mind-Reading Universal Remote
> Control, you'll likely get the same response from both... they each prefer
> their own. One is not more "learnable" than the other simply because you
> learned it first. 

you completely missed the point, which was: given that all interfaces are 
equal, the interface one knows is the interface that is best. therefore to 
change the interface, it should have benefits that accompany that change. i'm 
asking for demonstration of the benefits of these icons beyond "i like them 
better". if that's the extent of the argument, make an icon theme and put it 
on kde-look.org.

> And ONCE AGAIN, it's not the functionality or the shape
> of the buttons that you're talking about primarily, it's the position of
> the buttons. 

it applies to the actual icons as well. people do rely on the icon shape to 
navigate.

> the buttons gives no advantage whatsoever, I'd say don't change it. But if
> there is only an incremental improvement, then I'd say make the
> improvement. The larger number of users benefit with no undue hardship on
> the established user base.

thank you for agreeing with what i've been saying, now can you please 
demonstrate that the icons are an improvement and that the hardship on 
existing users is thereby mitigated by that improvement?

> That's because the premise of the question is off on several points, both
> addressed in this and the preceding posts. For the record:
> 	1. The icons don't have to be remarkably better. Benefits are leveraged by
> the larger new user base and magnified by combining with other incremental
> improvements to give a better overall user experience.
> 	2. The trivial period of time involved is no hardship (any moreso than
> getting acquainted with the controls of your new Lexus after having traded
> in your Redmond Yugo).
> 	3. The gain does appear elsewhere. This, of course, means you have to look
> elsewhere for it, as described above.

these are the assertions you and others here are making. i'm asking for proof 
that these icons actually do this. i'm not saying that if the icons are 
better they won't deliver anything useful, i'm asking "are these icons 
actually better"? personally, i don't think so, but there are ways to prove 
otherwise.

if you choose to reply, please do so with an answer to my question.

- -- 
Aaron J. Seigo
GPG Fingerprint: 8B8B 2209 0C6F 7C47 B1EA  EE75 D6B7 2EB1 A7F1 DB43

New is not beter. Different is not better. Better is better.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.7 (GNU/Linux)

iD8DBQE9UEhg1rcusafx20MRAmpBAJ0bzhkcCPFZtWporisycwX1HrhgMACfX2qp
P7D68PSxL2YdNpZWfIs5T3U=
=sIMK
-----END PGP SIGNATURE-----