Re: Request for breakage in gnome-panel

[Bryan Clark <bclark redhat com>]

On Mon, 2005-02-14 at 17:23 +1100, Jeff Waugh wrote:
>I don't think it's scary in those terms. It will be removed in 2.12 whether
>the peanut gallery (in this case, me) likes it or not. It was rejected for
>2.10 because we're well enough into freeze that it just doesn't make sense.

I completely understand this and am ok that if we're already too
committed to how things are for the release then we go with that.  It's
precisely why I asked.

>My points were about entirely different issues to that of appropriateness in
>the release process or maintainer responsibility. One of my growing concerns
>is that we are becoming very blas�bout changing things without focusing on
>the costs and benefits of doing so. In this case, I felt that the cost
>soundly outweighed the benefit.

I understand this concern and I agree it is important, and I didn't make
clear in my first message entirely why I wanted to remove the Run
Applications.  I guess I thought I wrote more in my bugzilla comments
and I didn't think that you would see what I was doing as a blas�hange
for no reason.  But I guess if I don't explain it enough then you might
think that. ;-)

>"What are we *giving* the user when making this change?" was a significant
>question when I set about writing the menu change proposal. My objection to
>the removal of the "Run Applcation..." item was inaccurately characterised
>as a fear of change or improvement, which was disappointing. In fact, it's
>quite the opposite: it's a call for measurable user-visible improvement so
>we're not making willy-nilly "fixes" that only serve to piss off our users.
>Let's make *really* good changes, not just ones that appeal to our esoteric
>sense of right and wrong.

Actually I was characterizing any one person there :-)  Not that anyone
actually fears change, but most messages clearly wanted only to keep the
run applications instead of attack the root of the problem or attack it
from a different angle besides shifting it around.

>I very rarely use the "Run Application..." dialogue. I don't particularly
>like having it around. I hope we have a model in the future that makes it
>wholly irrelevant. But I can't rationalise removing it, because I can't see
>any direct benefit to users by doing so, or that it is inflicting obvious
>harm.

Not so much cause harm as it's inflicting pain.  I think I've come to
realize something recently about our community users and that is that we
don't actually inflict enough pain on them.  What we do is leave
band-aids and warts lying around so we ourselves can pretend that our
desktop is good because people are able to get by.  Instead of clearing
those nasty things off the table and actually facing our *real*
problems.  

Swinging the pendulum doesn't mean our users have the greatest
experience, if we think back to when we thought GNOME was most
innovative it also was a pretty rocky experience at the same time.  We
can't have lots of innovation and lots of user cushion going on, there's
just too little resource for that.  I'd rather see more innovation than
more slow and safe changes.  Obviously there needs to be a balance here
and yet everyone has been agreeing that we're on the other side of
balance for a while.

>We are very good at coming up with focused solutions to real user problems,
>and pushing hard past the pain of change to deliver software that "Just
>Works". But sometimes, we're too keen to fix some of the grubby symptoms of
>brokenness - not the real problems causing it - and in doing so, we harm our
>users (and our user appeal).

Yes, this is exactly what I'm talking about but I think we're coming at
it from completely different sides.  I see the way the 'Run
Applications' item is now as a real problem of many things that are
wrong with GNOME.  A problem that hides other problems like
missing .desktop files in other packages, missing menu editors, and
missing "often launched applications" applet just to name a few.

I think we can all agree that the RA (I'm tired of writing it out now!)
is both a band-aid to our problems and a wart on our desktop when it's
out in the open.  However this is not to say it's not useful and we all
have probably made it a part of our lives somehow. But that doesn't make
it right!

When we shuffle some things around and make changes these desktop warts
become exposed to us.  We can choose to continue ignoring them and they
will just return to be a part of us again or we can remove them as
quickly as possible.  

Now, back to user pain.  If we start removing these warts and the
band-aids that cover them, systematically with each release we will
start to really see what's wrong with our desktop instead of keeping
these foggy goggles that make it look ok.  If we stop pretending we're
doing a great job and face the music at least we'll get decent feedback
in bug reports and emails and reviews of where we fail.  When we leave
just enough hacks lying around for people to get by they don't complain
too much they just take it and try to move on.  We cause them enough
pain that they do complain, then at least we know where to fix things
instead of shuffling menu items to make them happy.

Much of our users are community people who will help out if they can via
bug reports or just emailing telling us their pains.  These are the
people we need to cause a little pain every now and then.  Take a crazy
example like: If we took away all access to the file system in GNOME, so
that you couldn't even save a file.  We release this version of GNOME
and everyone and their brother would tell us they need a way to store
their information!  We'd hit ourselves in the head and say duh!  Then we
can start inventing a way for people to save their information, the pain
we caused revealed what they were really trying to do.  Notice how
there's no mention of saving files, just information; we've taken away
the crap we normally provide for saving information and now we can give
them something new.  Something that addresses their problem instead of
sticking with what we have.  [disclaimer] Of course this is a quick and
simple example which is off the wall, nitpick it if you must. :-)

I hope you (somewhat) agree that cutting the cruft on our desktop can
lead to more pain, pain which can lead to more invention (the child of
necessity). I'd even be willing to do a GUADEC talk, I'll call it,
"Causing Our Community Pain So We Can Breed Innovation" we're I'll try
to go in depth on how we need to stop kidding ourselves at the state of
our stuff and just lose the warts and band-aids.  Of course I missed
that deadline as well! :-P

Obviously it's not necessary to cause pain to realize what we need to
do, most people have ideas of what we need to do.  But the pain is about
getting everyone to see where we are failing and focusing what we need
to do to succeed instead of letting everyone find hacks around our
failures.

Trying to address your stuff directly dub man,
~ Bryan

P.S. I'm sorry I want to cause so many people pain, I hope they forgive
me.  It's not much pain, just a pain in the ass.  Hopefully everyone can
do their community part to relieve this pain for others who don't know
how to do it for themselves.