@Matthew (MPT):

Matthew, I'd like to highlight some of the things you've written about user-interface design that directly inform this controversy:

You wrote above:

> Brian, whether an automatically-opening window is 
> from a newly-running application or an already-running 
> application is, ideally, an implementation detail.

It surprises me that anyone with experience or training in user-interface design would write that. The first is a distraction away from one's current intention, and the second is an assist in accomplishing one's current intention. In terms of the user's subjective experience they're opposites. The difference is so great that, far from being 'an implementation detail', as you put it, this surely needs to be a *policy* matter.

Consonant with the belief you assert above, you say at https://lists.ubuntu.com/archives/ubuntu-devel/2005-August/009960.html, 

(Matthew writes:)

> ... even with optimally usable software you'll still need 
> a help function. Just-in-time proactive advice would work 
> better than passive help, and that's true for all kinds of 
> help, not just security. (“Using a table works better than
> lining up text with spaces. To make a table, choose 'Table'
> from the 'Insert' menu.” )

'Just-in-time proactive advice' is a good thing?  Was there *anyone* who didn't disable that annoying-as-hell “proactive” talking paper-clip help feature that was included in Word a few years back? And are there more than three people that can type who don't find the autosuggest word-completion 'feature' many word-processing programs employ distracting and annoying?  

I bring up the reference to point out that your user-interface design philosophy favors lots of active intervention that I think belongs more appropriately in a user-initiated interactive tutorial. You just don't seem to understand how much users hate being distracted from their work, how greatly we resent having our attention forcibly and repeatedly re-directed by programmers.

Further, you referred me to https://lists.ubuntu.com/archives/ubuntu-devel/2009-February/027568.html where you,

( Matthew wrote: )
 
> If your hard disk starts to fail, the system needs ...
> If someone is trying to call you over IM, the system needs ... 
> If you're on the Internet and there's a software update to fix a
> security vulnerability, the system needs ... 
> If three months ago you set a calendar alarm for today, the system needs ...

I imagine you might dismiss what I'm about to say as a nicety of semantics, but it's relevant here to observe that 'the system' does not 'need' anything at all.  Software and hardware don't have needs; *people* have needs.  ( But I think the term 'wants' would be more helpful than 'needs' here. )

Your statement about what should happen when there's a software update available, then, comes down to two things:

(1) What the person who wrote the “listening for updates” program wants, and 
(2) What the user wants.

And that, in my opinion, is the root of the conflict over this issue.

I wrote earlier that developers should at least include a 'Don't show this message again' check box in the window you want to initiate when updates are available, and that if the user checks that box, the system should then employ the Hardy/Intrepid update behavior.  In reply, you, 

Matthew wrote:

> Providing the ability to turn them off would imply 
> either that they weren't necessary in the first place, 
> or that there was a different but similarly effective 
> way to present them. If we thought either of those 
> things were true, we wouldn't be doing this in the 
> first place. 

That's entirely consonant with your distaste for programmers who 'placate people with options', as you put it at http://mpt.net.nz/archive/2008/08/01/free-software-usability.  But your statement above does make the core issue clear, doesn't it?  If we remember that people have wants and computers don't, the issue of whether to include such a check box comes down to whose wants are going to be honored here, re the disposition of the user's attention, yours or those of the user?  

More emphatically, the fundamental question is: 

  >>>  Who has veto rights over how the user's attention is directed?  <<< 

Matthew, you clearly think *you* should.  We users think *we* should: they're our computers and, more to the point, it's our attention.  

It's as if someone who frequently interrupts you wanted to remove the door to your office so you can't close it against him because he's sure he knows better than you do what you need to do or listen to. You're using software to tell users, “I'm going to repeatedly force you to listen to me for your own good, regardless of your opinion as to the importance of what I have to say.”

It really disappoints me that Canonical staff seem incapable of understanding why that makes so many people angry.