[GNC] Reconciliation Report -> Trade Debtors -> pairing transactions

[Michael or Penny Novack]

>
> And in that case of 'breadcrumbs' the entirety of GnuCash is an audit 
> trail, as that is how double-entry accounting works.
>
> If you are looking for some record of exactly who entered which 
> transactions and the date/time stamps, no, that is not available. 
> Technically I think the posting date/time does get recorded, but it is 
> not readily available to users. If you really need it, perhaps start a 
> thread just on that question.
>
> Certainly the 'who' will be impossible, because GnuCash is not a 
> multi-user application. Whomever has access to run the app, and access 
> to the file, can post or edit transactions. 

As somebody who used to do this (for one of the world's largest 
"financials") my comment might (or might not) be useful.

a) The accounting system itself (its own reporting functions) would not 
provide a way of reporting when/who by. It is important to understand 
that this information is NOT part of the accounting system. Accounting 
is not a "real time" process nor would who by (what worker bee) be 
relevant to the question of correctness in an accounting sense.

b) But there are OTHER ways in which general ledger might be audited. 
Internally correct is one matter, but externally correct is another. For 
example,here we have an invoice marked paid by check. But WAS there 
actually such a check, and if not, who (falsely) entered that invoice as 
paid and when. How many of you when first learning to use gnucash 
created a set of "test books" and into these entered a number of "test 
transactions"? Understand? There was nothing wrong internally with this 
set of test books you were using to learn gnucash but they matched 
nothing in the real world.

c) Like I said, there would be no functions PART of the accounting 
system to be used for this sort of purpose. But when the auditors 
(checking against the real world) found something suspicious, this sort 
of problem would come t my desk. "Mike, figure out who entered THIS 
transaction and when". So I would figure out what feed to the system 
would have fed that transaction to general ledger, get the right backup 
tape recovered from the vaults, and with a little "ad hoc" program I'd 
write for the purpose, get that information. It's not as if I'd be 
writing a lot of these ad hoc programs from scratch, just take to one I 
used last time and change a line or two of code.

    THIS sort of "check against the real world" was rare. I [probably 
did only a few times a year. More often it was personal and quickly 
enough after that fact I didn't have to have backups recovered from the 
vault. Thus suppose I was handed a problem from a client area "here's 
what one of our worker bees entered but it didn't "take", figure out the 
bug". Well actually MOST of the time no bug. Just a typo, what was on 
the piece of paper (as what was entered) didn't match what the worker 
bee actually entered at the terminal. We all do typos, and that's how I 
would report the problem as solved. No harm done. I probably did a few 
of THESE per week. It'd have been a waste of programmer time my handing 
off to one of our programmers to try to find a bug that didn't exist.

Michael D Novack