I just submitted a paper(44K PDF) to a workshop on situated ubiquitous computing at Ubicomp 2005 in September. I just submitted it minutes ago, so it's hot off the presses and may be totally off-base, but I figured I'd share it anyway. Here's the abstract:
The assumption that the goal of ubicomp is to make technology disappear stems from a Modernist ideal of purely utilitarian design that creates social invisibility. In fact, everyday design is anything from invisible, as can be seen in how furniture and cars are designed and from the hotrod and casemod cultures that modify everyday technological objects. Ubicomp design can learn to understand the design of situated technology from industrial design and from the study of technology modification cultures.
In other words: the way that people choose and modify technology is testament to the fact that they're not interested in having that technology be solely in the background. Understanding the boundaries of choice and the directions in which modification progresses may help us understand how to make ubiquitous computing that feels natural, without necessarily trying to make it invisible.
You maybe already know about his stuff, but Matthew Chalmers' work on 'seamful systems' sounds really apposite ... He's at Glasgow Univ - http://www.dcs.gla.ac.uk/~matthew/ - and works on when systems need to 'show their seams' such that users can constructively engage with it i.e. 'seamful ubiquity' - PDF here: http://www.cogs.susx.ac.uk/interact/papers/pdfs/Theory%20and%20Conceptual%20Frameworks/EqSeamfulUbiquity.pdf
I've used it as a reference in my adaptive design stuff - http://www.cityofsound.com/blog/adaptive_design/index.html - several times, along similar lines you describe - engaging people to adapt by making sure the affordances and seams are clear and malleable.
Danno: I totally agree, when people make casemods, they're definitely trying to make technology how it "should be" in their vision, which is another way of saying that they want it to be awesome--awesomeness is defined by a vision of the ideal. Herbert Simon says that natural sciences seek to describe the world as it is, whereas the "sciences of the artificial" (the title of his book on AI), which includes engineering and design, seek to describe the world "as it should be."
In terms of wanting invisible technology and beautiful design, I assume that you mean that you don't want to have to worry about understanding the technology and you want the thing to look and act how you imagine an idea version of it looks and feels. I define all of that as part of design: design should remove the necessity of understanding the technology (unless you want to) by making the thing look and act as you expect it to.
Ryan: interesting. Thanks! I just read the abstract and I'm not sure I would say that people are externalizing their game experiences onto the game hardware, but I think that both the desire to have a certain kind of experience in a game and the desire to have a certain kind of design for a special object may come from the same place.
You might be interested in Bart Simon's work on case modders [1]. He makes a similar point to yours, contrasting the trend toward "black-box" computing exemplified by Apple and gaming consoles with the counter-trend toward revealing and reveling in the "guts" of computing a la case modding.
[1]http://www.gamesconference.org/digra2005/viewabstract.php?id=52
I think there might be a difference from wanting the technology to be purely in the background and wanting the design of the technology to be purely in the background.
We like things to look cool, to be cool, but we don't want to ever be troubled by them. We don't want the fact that we're using technology to ever itself become a concern.
I mean, people don't do case mods because they like the fact that they're adding plexiglass and cold cathode tubes, they do them because it makes the technology awesome to have near them.
Another tack at the same point might be programming languages: If all programmers were concerned about was functionality, we'd probablly only be programming in either Lisp or C. But, obviously, programmers aren't satisfied with just those two, we want different languages that feel different in the brainpan (but still largely provide the same features as the two old languages).
I want invisble technology and beautiful design, heck, I want the design to transcend design and be art.