In article <firstname.lastname@example.org>, email@example.com
> In message <firstname.lastname@example.org>
> email@example.com (Robert Bonomi) wrote:
>> There are *very* limited ways for a cashier to themselves steal cash
>> money 'from the store', and approximately _one_ way to steal from the
>> customers, with 'computerized' registers.
>> Doing multiple charges against a card is simply *not* one of them.
>> The audit trail catches it at the end-of-shift, or over-night, at
>> worst. And the records point directly to the perp.
> Sure. However, since the customer can request cash back it's another
> way to steal. The cashier fakes problems scanning an item, punches in
> the amount manually as cash, and gives the customer the item.
> The customer isn't over charged, so has no reason to complain.
> The cashier's till is now over -- Now she just needs to grab the cash
> at some point during the transaction.
Differences in transaction amounts show up on detail reports. I know --
I got really good at spotting manager screw ups for the small retail
chain I used to work for.
Date: Wed, 24 Nov 2004 04:53:04 +0000
From: firstname.lastname@example.org (Robert Bonomi)
Subject: Re: Anyone Having any Luck With Google Ads?
Organization: Robert Bonomi Consulting
X-Telecom-Digest: Volume 23, Issue 565, Message 10 of 15
In article <email@example.com>,
DevilsPGD <firstname.lastname@example.org> wrote:
> technology in place will count as a hit as far as you're concerned,
> but not as far as Google is concerned.
> [TELECOM Digest Editor's Note: This is true. Google does use
> did or did not have on his computer until it was time for the
> been counted, wouldn't it? PAT]
In a word, "NO".
produce it isn't run, and the ad content is thus _not_ downloaded from
their server then the 'impression' is not counted.
'click' on the link *successfully* -- and a 'pass through' hit is
It _is_ possible to get a 'fairly' accurate count of how many times
any ad was actually "seen" by people viewing your page. It *does*
require a fair amount of skill in web-page programming, however.
Starting with reverse- engineering the 'drop-in' code provided by
google. Then you replace their target URL with a reference to a
'unique' URL on _your_own_ server. _THAT_ page does not 'really'
exist. it just returns a 'redirect' to the _actual_ google ad image.
*BUT*, you get a 'hit' in _your_ logs on that dummy page, every time
the ad is pulled up.
Note, _even_ this method will somewhat *overstate* the actual counts.
Not just for Google, but for anybody who in the business of 'serving'
ads -- e.g. doubleclick.com *spit* -- the ads tend to live on
_dedicated_ machines in the source domain. An INCREASING number of
people are using 'router' filters that block _any_ access to those
*specific* servers. As a result, when the web-browser tries to
retrieve the ad, the router blocking steps in, and returns either a
'host not found', or 'page not found' error, and the 'request' _never_
even gets to the server to be counted.
The 'internal redirect' is the _closest_ *you* can get to an accurate
count. But, beyond that, either you trust the ad provider, or you
don't. If you *don't* trust them, you _really_ should be doing
business with 'somebody else'.
The _long-established_ methodology for 'keeping somebody honest',
where -they- are the ones who compute 'how much money _they_owe_you_',
is to: (a) have the *contractual*right* to 'audit' their books and
records, or (b) have such 'audits' done by a trusted third-party --
e.g., the 'Audit Bureau for Circulations' for print publications.
When a gnat is contracting with an 800-lb gorilla, the likelihood of
_getting_ such 'audit' privileges is *damn* unlikely. You either
"trust" them not to screw you; or you go into the deal _knowing_ that
getting screwed is likely, and have decided that the 'other benefits'
are _worth_ the screwing.
[TELECOM Digest Editor's Note: Why couldn't I use a vistor counting
mechanism from another third party such as I used to have on the
old web site which I got from Microsoft? PAT]