Kodak and wireless

Curtis Yarvin curtis@buttercat.com
Wed, 9 Jan 2002 22:45:25 -0800


Jim Whitehead writes:

> 802.11x is going to win the wireless standard war. Already we're seeing
> signs of lock-in. 3G will lose, since the platform isn't open.

Bravely spoken, oh mighty warrior!  Why don't you take a few
million bucks from some bonehead VC, max out your credit
line to cover the cracks and put more than three sentences
on that bet?
 
I've worked in the "conventional" wireless Internet industry
for the last five years - currently on my third employer -
if you have a reasonably new cell phone the odds are about
50-50 it has about 10^5 lines of my code in it -

so I hope you will believe me when I tell you, that if you
think the cellular industry is a vast wasteland of bizarre,
proprietary protocols, third-rate software engineering, and
fat-cat old-economy suits, you are absolutely right, and -

unfortunately, it is still not going to lose.  Not to
802.11x, anyway - maybe to UWB, I don't know, maybe someone
knows enough about UWB to comment.

There are three problems.  One, as their software engineers
are bad, their hardware engineers are good.  W-CDMA and the
like, heinous as the protocols are (W-CDMA, which will be
the dominant 3G air interface, is basically the GSM protocol
over a completely new radio layer, except that comparing GSM
to what you think of when you think the word "protocol" is
like comparing your cat to what you'd get if you put it in a
stewpot and boiled it down for about twelve hours - what one
might expect to be discrete parts are not really identifiable;
it's more just a thick, chewy mass) are the product of some
very, very smart people with an enormous amount of experience.
Moreover, at the radio layer, 802.11x and, say, W-CDMA are not
even remotely comparable - saying the one will kill the other
is like saying microwave ovens will kill gas stoves.  You'd be
making more sense to say, maybe, 802.11x will kill PHS.  But
PHS has already done a pretty good job of killing itself.

Two, the software (not just the protocols, but also the
phones themselves) is terrible, yes, but it is not *so*
terrible that it can't be paved over and have useful stuff
built on top of it.  I hate to use that annoying "Internet
routes around X" metaphor, but, yes, the Internet will route
around it - and is indeed in the process of doing so.  It's
taking a while, yes, but stuff does, you know.

Three, you cannot even conceive of how much money and power
these dark empires control.  Yes, you always hear of dark
empires being suddenly toppled by tiny bands of golden-haired
rebels, but the reason it makes the news is it's the exception.
Normally we just beat the golden-haired rebels to a pulp with
our Mag-Lites over back behind the SS7 switches - you ever
wonder why NOCs don't tend to have windows? - then turn 'em
loose with some fresh dentures and a lil' roll of presidents,
and a brand new attitude, I'm tellin' ya.

> When you send pictures over the network, you want them to end up on the Web.
> I.e., your camera is remotely authoring a set of Web pages. WebDAV is a
> widely supported set of extensions to HTTP for remote authoring of Web
> pages. It works just fine for camera images. Hence, the killer camera is one
> with integrated 802.11x wireless, and a DAV stack, the DAV-cam.
> 
> Take a picture, and it automatically gets sent to a DAV server, along with a
> thumbnail, and an update of an image overview page.  From camera to the Web,
> immediately.
> 
> Next, create an Internet picture frame. But, instead of the Ceiva
> phone-home-every-night approach, put a DAV server on the frame.

Sure, maybe, except the frame is more of a client, I think
(the 'caption' being a monochrome LCD window with a URL).
And in the nearer term I think the picture is more likely to
just get emailed to you - network cameras are interesting
and so is public hosting via DAV, but I wouldn't count on
them to chicken-egg each other.

In Japan we're just starting to see a lot of camera-phones,
typicaly more phone than camera - DoCoMo's videoconferencing
FOMA service, of course, but I think many non-FOMA phones as
well.  I really don't know how they're doing in the market,
but I'll be over there next week and might learn something.

Curtis