LinuxSA Mailing list archives

  From: Alan Kennington <akenning@topology.org>
  To  : John Edwards <isplist@adam.com.au>
  Date: Thu, 28 Jun 2001 21:02:00 +0930

Re: Telstra Broadband & Multi-Users

On Thu, Jun 28, 2001 at 08:14:51PM +0930, John Edwards wrote:
> 
> >I don't actually have any broadband at home, and cannot get ADSL in
> >Prospect. We were going to get BigPond Cable, but then they brought in the
> 3
> >Gb cap. (actually, 2.86 Gb, because telstra say 1 Gb == 1,000 ^ 3, although
> >the rest of the internet says 1 Gb == 1024 ^ 3)
> 
> Actually, the rest of the Internet (where bytes are metered) tends to use
> base 10 accounting. America (the sadly dominant user of the Internet)
> couldn't care less, since they're mostly on unlimited usage systems.
> 
[....]


Here's an interesting link on this subject:

http://physics.nist.gov/cuu/Units/binary.html

They suggest that we call the binary-harmonized versions of
kilobyte etc.:

kibibyte	 = 1024 bytes
mebibyte	 = (1024)^2 bytes
gibibyte	 = (1024)^3 bytes

and so forth.

Personally, I prefer expressions like 20 bMBytes,
meaning "20 binary Megabytes", or 20 bin-MBytes,
but any widely adopted
standard would be better than the complaint "they used
the normal, correct definition of mega/giga etc. - so that means
they're cheating me."

The binary versions are all _slang_. They are slack usage.
They are colloqualisms. They are shorthand.
They are economical (or generous?) with the truth.

======================================================
By the way, on the subject of the ADSL pricing thing,
when you get a hotel room for 2, you generally pay more
than if it's for 1. If you hire a car for 2, you may pay
more than if it's for 1. There are many cases in life where
pricing of a service is not a precise reflection of the
cost of provision of the service but rather the benefit to
the end-user.

If internet access providers charged according to
cost of provision, then the very first customer
woudl have to pay for the entire infrastructure.
Then the next 50 subscribers might pay almost nothing,
and the 51st user might have to pay for the increase of
core internet back-haul bit-rate by a quantum leap.
If you start from the premise that pricing should be
in accord with the cost of provision, then you get
some pretty wierd conclusions from that. So it can't be
correct. Pricing is usually set in terms of an
inferred or assumed or measured demand curve, sometimes
called "how much the market will bear". If there is no point
on the combined demand and supply curves for which the
provider can make a profit, then no profit can be made.

If you are a single person using an ADSL link, then
you should be happy that houses of 3 people are paying
mroe, because they are almsot certainly going to use
more of their monthly allocation than you will on your own.

The _real_ problem with per-user surcharges is
enforceability, which has resulted in nightmares for
both providers and consumers of overseas broadband access
networks. Lawsuits and bitterness galore!
Within a couple of years, enlightenment will dawn upon
the access industry and better pricing concepts
will be implemented.

Cheers,
Alan Kennington.

-- 
LinuxSA WWW: http://www.linuxsa.org.au/  IRC: #linuxsa on irc.linux.org.au
To unsubscribe from the LinuxSA list:
  mail linuxsa-request@linuxsa.org.au with "unsubscribe" as the subject