(keitai-l) Re: FW: 802.11 for voice

From: cfb <cfb_at_nirai.ne.jp>
Date: 08/12/01
Message-ID: <3B769A9E.D1AD3025@nirai.ne.jp>
Michael Turner wrote:
> 
> One of the most interesting and readable posts
> I've seen on this list recently.  In particular,
> having seen some of the insides of telco network
> management, I found it especially ironic that
> you see the opportunity for a wide network in
> *mismanagement* on the part of others.  So
> much for Benjamin Kowarsch's "oh, just let
> a buncha WLANs gel and we can kiss the
> mobile/wireless operators goodbye."

Apparently you missed the "and hope they don't come after you"
bit.  It's not a viable business plan and I only mentioned it
as a straw man couter point to my own argument.  Anyone stupid
enough to actually go out and invest in the VPN gateway and
bandwidth deserves to fail in the most miserable way and anyone
who would "gel" WLANs in this way really needs to re-evaluate 
their morals, manners and norms.  

Freenets are fine.  If you want to start a opt-in service where
folks "mal-configure" their access point with a known SSID or a
known set of encryption keys, that's another matter.  My argument
is that such a network will *never* be a pervasive as a net that
will be available through user stupidity (or, rather, user 
lazyness (and lazyness isn't always a bad thing)).  Anyone taking
money from investors to pursue either type of network (and 
claiming (hoping) that it will be a pervasive, available, mobil
and usable as a cellular network is commiting an act close to
fraud... and considering how some venture capitalists behave
that might not be such a disagreeable thing).

Rethoric aside, even this straw man idea has many technical 
flaws, most of which go to highlight what a diverse concept 
"internet access" actually is (many access points will be using
NAT or proxy or firewalling or additional authentication/VPNs.
So, even the misconfigured access point won't *gurantee* that
you can obtain unmitigated access to/from the internet.  Besides,
I have yet to see a VPN client with Mobile IP thrown into the
mix... so roaming across the hypothetical misconfigured access
point mesh is right out (and there are only a few 802.11b access
points that allow roaming without all the VPN crud heaped on
top).

The other technical bit that was missed was the WiFi, as a 
hardware standard, requires too much power and space to ever 
replace a cellular network (and the very specifically engineered 
handsets that live on it).  Access point antenna sites are too
poorly chosen (at random), frequency congestion and signal 
propagation limitions are too great.  Also, the cellular operators
are not stupid enough to shoot themselves in the head by offering
data service that could potentially replace voice at a flat rate.
If they were stupid enough, the whole "economic ecology" of the 
network you were depending on for the service would become 
unsustanable. I am of the opinion that probablity of network 
emmerging that can compete will cell nets is very, very slight 
(but there may very well be space for a niche like HAM telegraph 
operators).

I might change my mind when someone points out a cell carrier 
that will let me ping someone's phone without it costing either
of us money (and I get round trip delays that are sub250ms for
a continuous stream ICMP packets of arbitray payload size)...
in which case it's time to start finding out how close the Java
VM on the handset will let you get to the TCP/IP stack and/or 
hacking the firmware on the handset (hardly something that will 
be installed on everyone's phone).  Beyond that, you've got the
whole directory can of worms to open.  I would also point to
P2P as the generic feasability test (I know the current crop of
handsets are a bit too thin to support P2P, but P2P will probably
happen before voice (mainly because P2P is "easy"), if either ever
happen at all).  arbitrary reachability ("an internet connection") 
+ flat rate data + mp3 player + P2P is not something that is just
going to magically appear on cellular networks.

Sorry if it seemed that I was being disparing of someone's ideas,
only to usurp them for myself later in the argument.  I assure you,
I that was not the point I was trying to make... and Curt Sampson is
probably right in saying that I need to get an Air-H card and start
fiddling around with the network from something other than a handset.

[ Need archives? How to unsubscribe? http://www.appelsiini.net/keitai-l/ ]
Received on Sun Aug 12 18:02:12 2001