RE: FYI - "Mobile Web 2009 = Desktop Web 1998"

Eduardo puts it well, as usual :-)

A couple of additional observations:

> -----Original Message-----
> From: public-bpwg-request@w3.org [mailto:public-bpwg-request@w3.org]
On
> Behalf Of Eduardo Casais
> Sent: 05 March 2009 21:15
> To: public-bpwg@w3.org
> Subject: Re: FYI - "Mobile Web 2009 = Desktop Web 1998"
> 
> 
> Nielsen's article is solid. As for the never-ending discussion on "one
> Web",
> here is my take.
> 
> 1) Convergence of mobile and desktop in terms of a few limited (though
> well
> publicized) software characteristics such as the browser is not
> meaningful. A
> real convergence implies closing the gap in processing power, storage
> capacity,
> transmission bandwidth. When looking at the cold figures, both worlds
> still
> differ by orders of magnitude -- fast convergence is unlikely.

I think it needs to be shown that convergence is actually happening. Is
it true that the following characteristics are converging, or are they
in fact diverging?

a) Bandwidth - could it be that techniques that apply to increasing
bandwidth of OTA connections also apply to fixed connections? What are
the relative gains when applied in those scenarios?
b) Processing power and power consumption.
c) Memory and Storage.

> Convergence also
> implies a unification of the context of utilization of the various
> devices and
> applications, and of user interface attributes -- a doubtful
> assumption.
> 
> 2) The "one Web" cannot hide the fact that developing different
> variants of the
> same application is unavoidable. A simple look at various W3C

To my mind there is the "Hard One Web" which states that there is One
True Experience, and there is the "Soft One Web" which says that
bookmarks interwork between contexts, but that the experience differs
between contexts. Of course there is also the "No One Web" position too.

> guidelines (e.g.
> for accessibility) proves the point: such documents are largely long
> lists of
> (justified) exhortations to provide a variety of alternative
> representations:
> for blind users, for deaf users, for browsers supporting frames or
not,
> for
> browsers with or without plugins, for devices with or without pointing
> devices
> or keyboards or touch input, for monochrome or colour displays, etc.
In
> practice,
> this "one Web" is implemented as disparate syntactic sugar to bind
> together
> more or less widely different variants of the same application -- this
> is really
> what for instance alt="...", <noscript>, <noframes>, media="...", and
> @media
> are for.
> 
3) Nielsen's article is based on rigorous studies of actual Web sites
> carried
> out with a range of mainstream devices, ranging from features phones
> through
> smartphones to the iPhone, and that are actually deployed -- not with
> two-years
> away promises from roadmaps. The conclusions might be disquieting for
> enthusiasts
> of the "one Web" -- and of the specialized mobile Web -- but they are
> empirically
> substantiated. Suggestions to rely upon media queries are only partial
> solutions
> (they do not address the issues of page sizes or scrolling, for
> instance), and
> they actually re-state the conclusion of the article: one must prepare
> variants
> of the application (in this case of its styling) for handheld
> devices...
> 

I think this means that there is a place for Server Side Adaptation,
Client Side Adaptation and In-Network Adaptation. Figuring out when and
why to do it in one place or another is hard. But adaptation is
definitely the name of the game.

> 4) We should not forget the dynamic aspect of the entire problem: the
> desktop
> Web is evolving too -- it is a moving target to converge to. The
mobile
> Web
> itself is evolving (sometimes overtaking the desktop Web) in non-
> convergent
> fashion as well: hand-touch devices are introducing changes in the way
> one must
> deal with user interface elements, events, etc -- which differ subtly
> from
> existing constraints and practices but are not relevant for the
desktop
> Web at
> this stage.

I think the evolution of the desktop Web and indeed of the means by
which it is constructed is often overlooked when discussing mobile
evolution. There is something Newtonian at work, like "The Third Law of
Motion" - for every convergence there is an equal and opposite
divergence. (Sorry, Sir Isaac).

It might be interesting to do a historical comparison of the form "The
mobile Web is where the desktop Web was x years ago" and plot it over
time. To be fair to the dynamics you'd also have to capture "The desktop
Web is now n times what it was x years ago".

> 
> All things considered, the equivalence "Mobile Web 2009 = Desktop Web
> 1998" does not seem entirely far-fetched to me. I first encountered
the
> discourse about the mobile-desktop convergence in 1996,
> when the original Nokia 9000 Communicator was launched. I will not
hold
> my breath.
> 
> E.Casais
> 
> 
> 
>

Received on Friday, 6 March 2009 07:43:42 UTC