[LINK] Why the NBN business model is deeply flawed

[Karl Auer]

On Thu, 2012-02-16 at 12:58 +1100, TKoltai wrote:
> We achieved a 98% positive consumer result by using "tricky" diverse
> routing algorithms depending on the packet type.

Will Not Happen with satellite.

> Whatif the Satellites have built-in Storage capacity.

It costs thousands of dollars *per gram* to place a satellite. Storage
capacity on the order that is going to be needed to make any kind of
meaningful dent in the problem is extremely unlikely to be in the
payload.

> i.e.: Latency for the majority of the content is reduced to 170 ms.

Not at all - latency is reduced from horrendous to merely godawful. and
that's only for those protocols where caching makes a difference, and
then only for those items that are actually cached. And all for the
benefit of the few who don't have fibre. And why don't they have fibre?
Because they were not important enough to spend money on - so it seems a
bit unlikely that they'll get engineering's finest working on their side
to improve latency. It's a minority of packets to/from a minority of
users.

There are lots of technologies that twiddle around the edges of the
latency problem (TCP ACK spoofing etc) but the simple fact is that ye
canna change the laws o' physics.

> In other words, the only painful experience will be Voice.

"Only" is wrong in at least two ways - firstly, voice is pretty
important to most people. Secondly, it won't be voice alone, it will be
anything latency-sensitive - gaming for a start, which is a huge
industry. Not to mention people like me, typing into terminal
windows :-)

> Therefore, on current utilisation curves, 95% of all traffic will not
> notice the latency.

Building the limitations of today into the networks of tomorrow is
foolish. We should be doing all we can to remove them. As stopgaps - OK.
As permanent infrastructure - no way.

> At the end of the day, 340 ms single satellite hop delay is an
> acceptable delay for people used to Nextel (PTT) cell phone or CB
> radio type communications.

Tell me you are joking. "Hot nails being driven into the eyeballs is
perfectly acceptable to people used to having their teeth pulled without
anaesthetic."

> Especially if the same communication is accompanied by high speed
> video, graphs and PDF documents to confirm the veracity of the
> communications.

Oh thank goodness, you ARE joking.

> Therefore the real answer is a combination of Terrestrial wireless
> (cellular or Wimax), Satelite Wireless, WiGig Public data showers and
> WiFi in the Home.

Nope - the answer is fibre. A shitload more of it than it seems anyone
is willing to pay for, hence the desperate attempts to convince us all
that wireless is acceptable.

> I don't actually see a future for a cable connecting to any device
> that I use...

Of course not - that's silly. But the devices you use can be untethered
and still have their data travelling everything but the last/first few
metres over fibre.

Wireless for long-range communications is bunk.

Regards, K.

-- 
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Karl Auer (kauer at biplane.com.au)
http://www.biplane.com.au/kauer

GPG fingerprint: AE1D 4868 6420 AD9A A698 5251 1699 7B78 4EEE 6017
Old fingerprint: DA41 51B1 1481 16E1 F7E2 B2E9 3007 14ED 5736 F687
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 230 bytes
Desc: This is a digitally signed message part
URL: <https://mailman.anu.edu.au/pipermail/link/attachments/20120216/b1da28fe/attachment.sig>