On 09-Mar-12 03:48, Roland Perry wrote:
In message , at 00:34:27 on Fri, 9 Mar 2012,
Stephen Sprunk remarked:

There's always a small risk, and sometimes the ticket vendor will
make a
[mobile] phone call if a particular cardholder raises suspicions.

If mobile service is available, why not just authorize every card and
avoid the risk of being sued for discrimination--which will cost far,
far more (even if you win) than losing the occasional fare?

Because authorising transactions with a voice call is time consuming,
and may not even be possible as a routine thing.

There is no good reason to do the authorization as a voice call when
there are mobile data terminals that can do it in seconds.

The existing "good reason" is that the mobile data terminals haven't
been designed yet,

They existed over a decade ago.

let alone deployed. And the previous generation has at least 10 years
life left in them.

They were obsolete the day they were purchased; how long they're capable
of meeting obsolete needs before turning into paperweights is not
terribly relevant.

Of course, if there is no mobile data service, there probably isn't
mobile voice service either, so there is no difference in that respect.

Far from it, data service on what's a perfectly good voice service can
be excruciatingly slow. There's vast areas that still only have 2.5G
You'd actually need to use SMS for it to work at all.

The "slowness" of 2G/2.5G is mostly the connection delay while a
separate data channel is set up, which takes several seconds, after
which the data flows reasonably quickly. The main advantage of 3G for
most applications (including this one) is being able to send data
_sooner_, not being able to send it _faster_.

Worst case, the terminal could batch up a series of authorization
attempts for when it next passes into a coverage area; if one of the
responses is a failure, it could notify the conductor, who would then go
back to the customer

Assuming he can find them. It's not uncommon for trains to be standing
room only.

If the train is that packed, how much progress through the train is he
likely to be making anyway?

Note that overdraft (at least in the US) is _not_ guaranteed; the bank
can refuse to honor any debit against insufficient funds at their
whim--but they generally will, since it allows them to charge the
customer massive fees on top of the debit itself.

In the UK you'd only get fees (rather than the agreed interest rate) on
an unapproved overdraft.

Sorry, I don't think the above was clear. Most US banks will _honor_ a
debit against insufficient funds because it allows them to charge the
customer massive fees on top of the debit itself.

For instance, my bank charges USD 35 per overdraft transaction plus USD
5 for each additional day the account has a negative balance.
Obviously, it's in their interests to allow customers to overdraft as
much as they want--as long as the bank can be reasonably sure the
customer will _eventually_ bring the balance up to zero, eg. because
their paychecks are automatically deposited in the account.

I seem to recall hearing about accounts in other countries having a
guaranteed overdraft capability; that would be a "line of credit" in the
US, which is separate from a checking account.

Yes, an overdraft facility in the UK is the same as your "line of
credit", and once set up would have to be specifically revoked with
notice to the accountholder.

As detailed more fully in the part you snipped, that's not how it works
in the US. "Overdraft" on a checking account and a "line of credit"
account are separate services.

Speaking as an employee of a tech products vendor, customers are now
demanding full ROI in 12-18 months, which gives them immediate cost
savings even on a 36-month depreciation schedule.

It's an interesting aspiration, but what happens if such a return is
*impossible*, given the development, manufacturing and operating costs
of the equipment.

If there is no ROI, then customers won't buy it.

There's no *extra* revenue stream here - just reducing
Credit Card fraud a little, and the "acceptable ROI" solution here was
C&P, not "being online all the time".

The return would logically come from (a) not accepting invalid credit
and charge cards and (b) accepting valid debit cards.

I don't understand why the UK hasn't seen the same progression,
especially given you only need to build out one network instead of the
redundant, mutually incompatible networks we had to build.

Read the OFCOM report I linked to a couple of days ago. And by the way,
we have four separate networks (used to be five, but two merged),
because apparently competition and a free market is best (!). Or if you
count GSM and 3G, it's actually seven networks (one is 3G only).

Is there no domestic "roaming" between carriers?

In the US, phones will _prefer_ towers from their own carrier but can
roam to any tower from any carrier using the _same technology_, eg. a
T-Mobile (GSM) phone will connect to an AT&T (GSM) tower if a T-Mobile
(GSM) tower isn't available. When I refer to "redundant, mutually
incompatible networks", I mean that a Verizon (TDMA or CDMA) or Sprint
(CDMA or iDEN) phone is simply incapable of using an AT&T (GSM) or
T-Mobile (GSM) tower. And there are many, many companies that set up
their own towers in rural areas just to handle roaming phones from the
major carriers; they may have no subscribers of their own.

Critically important is that customers who subscribe to a "national"
plan are _not_ charged for (domestic) roaming, so coverage is a matter
of technology and towers, not carrier.

International roaming is another matter, but most US phones (TDMA, CDMA,
iDEN or 1900MHz GSM) don't work in most other countries anyway.
Tri-band GSM phones are the main exception, and int'l roaming for them
is ridiculously expensive--but at least it works. (CDMA and 1900MHz GSM
are also available in Canada, and that roaming is ridiculously expensive
as well.)

S

--
Stephen Sprunk "God does not play dice." --Albert Einstein
CCIE #3723 "God is an inveterate gambler, and He throws the
K5SSS dice at every possible opportunity." --Stephen Hawking