On 12/11/2011 12:07 PM, André Neves wrote:
> It seems to me like you're blaming Firefox for Microsoft's inability
> to, or lack of will to, keep the Skype extension up to date.
> Am I wrong?
> I hope so, because this would be a "dick move".
Well, there's no incentive for MS to keep the plugin compatible with FF....
> I also very strongly disagree with your opinion that the upgrades
> above 3.x have zero or negative effects on usability:
> - The area of the browser window for display of webpages was augmented
> immensely;
Two lines is "immensely"????? Anyhow, not an issue, now that all new
monitors are hi-res.
> - The condensed menu speeds access to the most used features;
Menu design, button placement, etc is a perennial issue, and theoretical
talk is pointless. Consider that Apple placed the application's menu bar
on the desktop, because theoretically that would result in shorter mouse
moves. It's one of the dumbest design decision ever made IMO.
There are few logical aspects to menu design - I know, I designed lots
of menus, back when I was writing small apps for myself because no one
made what I wanted. Many of the choices are arbitrary, and for many
their placement is a matter of history. Eg, Options" vs "Preferences"
and "Tools" vs "Edit" for the location of those choices on DOS- vs
*nix-derived OSs.
Any time a design feature is largely or wholly arbitrary, it's
_necessary_ to agree on and adhere to some standard, else people will
make silly (and in some contexts lethal) mistakes. Aircraft designers of
aircraft have figured this out, so why do program developers resist
standards? And even pride themselves on change for the sake of change?
Watch actual users and how they use the menus: Once a person is used to
a particular menu design, they mouse to what they want automatically.
It's like driving a car: over 90% of what we do while driving we do
automatically (and therefore very fast, BTW.)
IME, condensed menus in practice means submenus, and each submenu adds
an extra move.
As for speed of access to most-used features: I've found that typing in
a familiar URL is often faster than finding it in Bookmarks.
> - There is a print preview option;<3
Yup, and NB that was added before 3.
> - There are tab groups;
Tab groups? If you just mean tabs, well, that began with 3.
> - The about window checks, and tells the user, whether Firefox is up
> to date - further offering an option to get those updates;
Providing a program's version number is SOP for many years now. As for
the option to get the latest updates: redundant, since FF checks for
updates by default. But lately, you have go to a sub-menu and uncheck
"Automatically download..." Bah! Double bah! Triple bah! No app should
update by default, not even for, nay _especially_ not for the average
(naive) user.
> - The downloads window no longer jumps on your face when you get stuff;
I don't understand. I want to see the d/l window, so I can decide where
to put the stuff. Which I do 90% of the time. Dumping it all into one
place by default is not a time saver in the long run.
> - etc?
- Changes in names, eg, from Organise Bookmarks to Show All Bookmarks:
why? Users knew what the old name meant.
- Hiding the Menu bar by default. I think having the option to hide it
is fine, but hiding it by default? You should've heard my wife's wails
when it disappeared after an automatic update (now turned off). And my
curses until I found that clicking on an "empty spot" brought up the
menu to unhide it. The "logical"access to that would be right-click on
title bar, right?
> These contradict, IMO, your opinion.
> Many others are too good for me to have noticed, or they feel so
> natural that I can't remember them as differences from 3.x .
IOW, you've habituated yourself to the later versions of FF. Good for
you. But that merely demonstrates my point, which is, that a program
update should require little new learning, and no unlearning.
We've gone back to 3.6.24 on both machines because every change above 3
required unlearning something. Updates shouldn't do that. Learning
something new for a new feature is fine. Unlearning something because a
feature was changed or hidden is bad design.
> And I am not even a usability person, I am just a power user who
> infrequently reads dev-usability.
>
> I'm sorry, I think your contribution lacks solid fundaments.
> Maybe you can review and improve it?
Well, I'm a power user, too (and I bottom post). I'm also a design geek,
who appreciates all aspects of design. That's why I can tell the
difference between an elegant kettle and a usable one, for example. I
also taught logic, so I know the difference between logical and
reasonable. And, as said, I've designed and written programs to suit
myself, during which exercises I found that designing the user-interface
was a knottier problem than designing program structure. Why? Because
any program worth writing/using has several different paths through its
task-space. Setting up a menu system so that it's easy to navigate any
needed path is not trivial. See the Travelling Salesman problem for a
sense of how difficult it is.
The ergonomics of program design are not AFAIK well understood, though
many attempts have been made. But since I was the go-to guy for
computers in my school when we first started using them, I've had enough
practical experience watching how people use computers to tell you that
the first rule of program design is:
Study how people actually do a task; then design a program that enables
them to do it without thinking about how they do it.
Easier said than done.
Have a good day,
Wolf K.