Untethering and new connections

[David Farber <dave () farber net>]

Begin forwarded message:

From: "Bob Frankston" <Bob19-0501 () bobf frankston com>
Date: November 1, 2009 11:25:18 AM EST
To: "Prof. David J. J Farber" <dave () farber net>
Cc: "'Lauren Weinstein'" <lauren () vortex com>, <nnsquad () nnsquad org>
Subject: Untethering and new connections

Lauren's point about being tethered by the cloud is very important.  
The cloud has its role but we need to remember that “end-to-end” is  
not “womb-to-tomb”. The end points need to take responsibility for  
failure (and disconnection). Navigation needs to work even when one  
isn’t fully attached to the cloud.

This is why building atop reliable systems can be problematic – it  
leads to brittle dependency.

Closely related is our tendency to think in terms of physical objects  
rather than abstractions. Thus we think about “GPS” as a signal coming  
directly from a satellite rather than looking for location  
information. While there is opportunistic use of cellular towers and  
cataloging of MAC address locations why don’t we consider “location”  
as the point of commonality, AKA the API or interface? Why can’t I  
find my location when I’m not tethered directly to a satellite? This  
is not a trivial problem but it is vital for linking abstractions to  
the physical world.

The screen is another example. We already have the idea of a screen as  
an abstraction. So why can’t I use the screen in my dashboard as a  
resource either as whole or windowed. We have protocols such as X- 
Windows and remote desktops but we don’t architect our devices to be  
able to take advantage of these. Given a phone with a Wi-Fi connection  
it should be possible to do this with a software upgrade. Even better  
if the car and phone screens can be used in concert, especially if  
both are touch sensitive. For that matter why can’t we use controls on  
the steering wheel – the ones currently tied to dedicated functions.

Here’s where the concept of “adult supervision” comes to the fore. The  
narrow business needs of each participant makes it difficult to view  
these problems as whole. Perhaps these concepts of resource sharing  
may have a place in an IETF committee but it would require a champion.  
I would be one but for now I have to act as a meta-champion given my  
other distractions. It is in that role that I presenting these thoughts.

Perhaps I’m just impatient. Ideas like XML data formats (and their  
HTML counterparts – microformats) have been slow in getting adopted.  
We talk about specific implementations as with the “smart grid” but  
where is world of smart ABSTRACT things?

I do see some progress. For example some buildings in Amsterdam have  
QR codes that allow you to find out more about them. (http://qr.westergasfabriek.nl/index.php?language=EN 
) In fact the QR codes tell you about an SSID that will give you  
access to a site with more information and a QR reader but, alas, the  
connectivity stops there. Alas, why can’t they refer to “device”  
instead of calling it a “phone” as if you had to get your most basic  
stuff from a provider company whose only goal is maximizing ARPU? This  
is an insidious assumption that keeps us tethered and lame.

As a pragmatic observation while copying the pictures off my SD I’m  
reminded that we still don’t do simple things – like maintaining the  
time offsets in file systems so the time stamp remains constant.  
Instead, in each time after I copy the files in a different place (or  
“D/S”) the time stamps don’t line up.

No wonder we see talk about sharing medical records instead of medical  
information – it’s going to take generations to understand this simple  
stuff. If only I had the patience to wait.


-----Original Message-----
From: David Farber [mailto:dave () farber net]
Sent: Friday, October 30, 2009 15:29
To: ip
Subject: [IP] Why Google Maps Navigation Won't Kill Standalone GPS



Begin forwarded message:

From: Lauren Weinstein <lauren () vortex com>
Date: October 29, 2009 10:01:43 PM EDT
To: dave () farber net
Subject: Why Google Maps Navigation Won't Kill Standalone GPS



              Why Google Maps Navigation Won't Kill Standalone GPS

                  http://lauren.vortex.com/archive/000629.html


Greetings.  There's a lot of excitement over Google's announcing that
mobile Google Maps will be enhanced (at least for Android 2.0) with a
true spoken turn-by-turn navigation system.

It reportedly will be chock full of nifty features, such as satellite
image and Street View overlays, live traffic feeds and other goodies.

I won't even bother digging out all the links where I've spoken
admiringly of Android, praised Google Maps, and even speculated on
issues surrounding turn-by-turn navigation for Android.  I'm looking
forward to playing with Google Maps Navigation as soon as possible.

On the other hand, the breathless assertions from some observers that
this development will destroy the standalone GPS industry seem
exceedingly overblown.  There are several reasons why.

First, the Cloud.  The Cloud is both a strength and a weakness for
Google Maps.  If you're in a strong signal area, and have an unlimited
data plan, loading maps from Google Servers can be entirely
practical -- though the speed of some current phones can make this a bit
sluggish even under good signal 3G conditions.

But users of other "data connection required" cell phone GPS services
know, if you don't have a signal (or you have a poor signal) and you
don't already have the needed mapping data cached on your phone --
well, you're out of luck.  No signal, no maps.

This is problem amenable to technical solutions, but there can be
interesting policy and licensing issues.  Google could cache
significant portions of planned routes, or even entire routes, on
phones so that loss of signal would not cause a sudden lack of mapping
data.  Taking this idea even further, entire cities, states, or even
countries could be cached in advance.

This latter concept is essentially the technique used by the $30
CoPilot Live Android application, which allows the user to download in
advance essentially any part of the desired region (e.g. the U.S.)
that they wish, and keep it resident on their phone.  The amount of
data involved is relatively reasonable, especially for individual
cities and states. CoPilot then provides various live services to
enhance that stored data.

CoPilot displays aren't likely going to be able to match the sheer
classiness of Street View overlays anytime soon -- but knowing that I
always have the entire street mapping database for the state of
California on my phone, even if I'm out of cell range, is a
significant comfort.  Here in the traditional canyons and urban
canyons around L.A., there tends to be an inverse relationship between
cell signal availability and the degree to which I find myself
disoriented in an unfamiliar location.

Google presumably could, if they wished, allow for a very similar
degree of advance map downloading so that dependence on the Google
Cloud in real time would be reduced or eliminated.  But would they?  I
don't know.  I can think of relevant issues both pro and con.

Here's another reason why traditional GPS isn't going away anytime
soon -- it seems unlikely that most drivers who have become used to
large in-dash navigation systems will want to switch to using a small
cell phone display while driving.  Though smartphone displays are
certainly getting larger, there's going to be a limit beyond which
further size increases will be impractical for a "pocket-size" device.
And the smaller the display when driving, the bigger the distraction
risk would seem to be as drivers try to focus on the display.  In
fact, a sort of "reverse flow" could be possible.  Users whose first
GPS navigation experience is on cell phones may be sufficiently
smitten with GPS that they could be more likely to order more
traditional large screen in-dash navigation systems (many of which now
also include various "live" data sources) with their next vehicle
purchase.

Of course, to the extent that drivers depend mostly on GPS voice
directions and don't have to look at the displays, the size of the
displays is of less importance.  And it seems reasonable to assume
that Android-based in-dash navigation units will appear -- an
especially likely scenario given the insane price demanded for annual
data updates to many in-dash navigation systems.  Also, drivers
without in-dash navigation systems at all (the 4-wheeled vehicle I
drive seems almost old enough to require a starter hand crank) may
clearly find simply plopping their Android phone down in a car mount
(as I do) to be a practical alternative.

The next issue may be the toughest nut to crack.  Those of us in the
tech realm tend to live in an environment where it's easy to forget
that not everyone in the world uses smartphones and has unlimited data
plans.  In fact, vast numbers of persons have no data plans at all,
and even if the carriers gave away data for free, many people would be
unlikely to use advanced phone features.

In fact, there's a very significant segment of the cell phone industry
concentrating on what we'd consider to be low-end phones, free of all
advanced features and frills, completely oriented toward basic voice
communications.  Yet users who tend to choose such devices may also
want to avail themselves of GPS navigation, while still likely wanting
to keep their phones small and simple as well.

Such persons -- and I believe they represent a significant and in some
age ranges a growing segment of the population -- would seem much more
likely to go out and buy a under-$100 standalone car GPS unit to clip
on their sun visor, regardless of available phone features.
Inexpensive Android-based "GPS-only" devices are also presumably
possible, but the key point in this regard is that many persons are
looking for basic GPS devices that do not depend on outside services,
and this situation is unlikely to dramatically change anytime very
soon.

The arrival of Google Maps Navigation for Android is a development
many of us have been waiting for, and is a feature that I personally
am very much looking forward to seeing in action on my own Android
phone.  Google Maps Navigation will be immensely useful to large
numbers of persons.  But the observers who seem to already be
engraving tombstones for the traditional GPS navigation companies
are -- in my opinion -- rather seriously jumping the gun.

--Lauren--
Lauren Weinstein
lauren () vortex com
Tel: +1 (818) 225-2800
http://www.pfir.org/lauren
Co-Founder, PFIR
   - People For Internet Responsibility - http://www.pfir.org
Co-Founder, NNSquad
   - Network Neutrality Squad - http://www.nnsquad.org
Founder, GCTIP - Global Coalition
   for Transparent Internet Performance - http://www.gctip.org
Founder, PRIVACY Forum - http://www.vortex.com
Member, ACM Committee on Computers and Public Policy
Lauren's Blog: http://lauren.vortex.com
Twitter: https://twitter.com/laurenweinstein





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



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