Re: [ I know this will raise a storm djf] Hackers make progress towards unlocking iPhone

[David Farber <dave () farber net>]

Begin forwarded message:

From: Dave Crocker <dcrocker () bbiw net>
Date: July 6, 2007 12:44:54 PM EDT
To: bob.hinden () nokia com
Cc: David Farber <dave () farber net>
Subject: Re: [IP] Re: [ I know this will raise a storm djf] Hackers  
make progress towards unlocking iPhone

Bob,


Bob Hinden wrote:
> Very nice comments!  There is a lot more to innovation than just  
> the technical side.  Apple has gotten very good about understanding  
> the relationship between products, markets, and services.  Much  
> better than many of their competitors.

Thanks.

The R&D profession of HCI (human-computer interaction) seems to be in  
poor shape, in my view, with islands of exception.  As a research and  
engineering discipline, it tends to be fragmented into areas of micro- 
focus, with little real progress in broad-scale, integrative  
thinking.  What advances have happened tend to have developed little  
and to be ignored by industry, with some obvious exceptions.

I don't mean that there is no broader thought. Quite the opposite.  
But it hasn't gone very far and hasn't had much effect from what I  
can tell. (I've been tracking the area slightly since the 70's, when  
I was going to specialize in it.  I've renewed deeper interest  
recently, but am still coming up to speed.  There are a couple of  
lines of theory that look promising, such as Activity Theory, but I  
don't see them having much traction yet.)

To that end, the success of Apple (and, umm, Nokia) and some others,  
due to excellent "packaging" -- by which I mean both the raw  
aesthetic but more importantly the deeper issues of HCI (or, really,  
the broader and tonier term of art "user experience, or UX) -- are  
useful.  It has the market being able to give the industry feedback  
about preferences for quality of UX.

The closed aspect of Apple makes industry growth a lot more  
difficult, both in terms of platform development and in terms of  
"social" interaction.  It limits the community of participation.

Where the UX situation really falls down -- and Apple is no exception  
to this -- is with truly distributed systems, nevermind open ones.  
The Internet is used to bring users to a central location, for a  
service, rather than real peer-to-peer distribution.  This isn't  
surprising, since a fully distributed model is a lot harder, as you  
know.  Well, that seems to apply to the UX issues, too.

Worse, it runs smack into serious security problems, since the  
distribution activities tend to look a lot like spam and worms.   
There have been tidbits of work done of security related to mobile  
agents, but it, too, seems to have gained little traction.  My own  
preference is to have systems that receive a mobile agent have  
special application-specific security interfaces registered, each one  
knowing the semantics of a particular domain and offering a  
constrained set of operations to incoming agents.  TCL and Java have  
played with this model.

All of this makes for complex problem and solution spaces.  So it's  
not surprising the pretty much everyone has opted for the simpler,  
centralized models.

"The market" hasn't figured out how to communicate a preference for  
open, distributed UX enhancements.

And the bulk of the market has sufficiently good connectivity so that  
the centralized model seems sufficient to it, even though I sure wish  
it didn't. As nice at it is to focus on better full-time  
connectivity, it will never completely be sufficient and the focus  
cripples lines of innovation.

A market motivated to demand more out of limited devices with  
periodic and limited connectivity would pull much more interesting  
distributed applications out of the industry...


d/
--

  Dave Crocker
  Brandenburg InternetWorking
  bbiw.net


-------------------------------------------
Archives: http://v2.listbox.com/member/archive/247/=now
RSS Feed: http://v2.listbox.com/member/archive/rss/247/
Powered by Listbox: http://www.listbox.com