[GNC-dev] tb, gnc or me? my trial balance is wrong and I think it is gnc not me

classic Classic list List threaded Threaded
24 messages Options
12
Reply | Threaded
Open this post in threaded view
|

[GNC-dev] tb, gnc or me? my trial balance is wrong and I think it is gnc not me

GnuCash - Dev mailing list
Background: My gnc TB has been wrong for years.  This hasn't been a
problem for me because I can do my own TB in sql and satisfy myself all
is relatively well with my accounts.  Over the last week or so I decided
to try again and I think the gnc TB report is b0rked.

Looking at the bug list, I'm not making a new observation, the gnc TB is
not a feature of which to be proud but I think I can help.

The problems seem to be centered around two things:

1. using currency other than your home currency
2. related to that, using trading accounts

Guess what?  If it was that simple it would have been solved by now, right?

I have loads of tx in a number of currencies using trading accounts that
don't make the gnc TB throw a wobbly but have identified the first that
makes it go wrong.

So, let's look at a tx that does make it go wrong, maybe someone will
help me to think, it is possible I've looked at this for too long and am
missing something really obvious.

===
Liabilities:Loans:T:T ZAR Dr 7,963.46
Liabilities:Loans:T:T USD Cr 771.22
===

The numbers are real, I've changed the account names slightly, I've
separated this particular tx as the first that makes the gnc TB go wrong
and does actually balance by my understanding of accounting and mathematics.

The prices look right, the actual exchange rate is real, what is wrong?

Things to consider:

If the price to be used is the issue, is the gnc TB using the right
price? My answer may be "yes, usually it does use the right price, I
can't work out why it sometimes doesn't"

Does this raise an accounting honesty issue?

I think it does.

If you need to roll your own to get a TB from a transaction stream
something isn't right.

Alternatively, please help me to see what is wrong with my tx.

have a joke: Trump wants to be the Dutch boy, first he needs a dyke,
then he realises his fingers are tiny :)

--
Wm


_______________________________________________
gnucash-devel mailing list
[hidden email]
https://lists.gnucash.org/mailman/listinfo/gnucash-devel
Reply | Threaded
Open this post in threaded view
|

Re: [GNC-dev] tb, gnc or me? my trial balance is wrong and I think it is gnc not me

GnuCash - Dev mailing list
On 09/02/2019 07:00, Wm via gnucash-devel wrote:
> Background: My gnc TB has been wrong for years.  This hasn't been a
> problem for me because I can do my own TB in sql and satisfy myself all
> is relatively well with my accounts.  Over the last week or so I decided
> to try again and I think the gnc TB report is b0rked.

OK now it is getting weird, this tx involves no fx or trading accounts
but throws the TB report out by a penny.

===
Exp:Week:Rec Dr 4.00
Liabilities:Accounts Payable:Y Cr 4.00
===

I kid you not, the tx is for 4.00 but the TB goes wrong by .01

--
Wm

_______________________________________________
gnucash-devel mailing list
[hidden email]
https://lists.gnucash.org/mailman/listinfo/gnucash-devel
Reply | Threaded
Open this post in threaded view
|

Re: [GNC-dev] tb, gnc or me? my trial balance is wrong and I think it is gnc not me

GnuCash - Dev mailing list
In reply to this post by GnuCash - Dev mailing list
On 09/02/2019 07:00, Wm via gnucash-devel wrote:
> Background: My gnc TB has been wrong for years.  This hasn't been a
> problem for me because I can do my own TB in sql and satisfy myself all
> is relatively well with my accounts.  Over the last week or so I decided
> to try again and I think the gnc TB report is b0rked.

OK, have a think about this.

Having deleted the tx previously mentioned.

If I run a TB report to 2017-12-17 it balances.

If I run a TB report to 2017-12-18 it is out by .01

Problem?  There are no tx on 2017-12-18

The ordinary use of a TB is to help find accounting errors.  If the
report introduces errors I say it is not fit to be used.

https://bugs.gnucash.org/show_bug.cgi?id=797097

--
Wm

_______________________________________________
gnucash-devel mailing list
[hidden email]
https://lists.gnucash.org/mailman/listinfo/gnucash-devel
Reply | Threaded
Open this post in threaded view
|

Re: [GNC-dev] tb, gnc or me? my trial balance is wrong and I think it is gnc not me

GnuCash - Dev mailing list
In reply to this post by GnuCash - Dev mailing list
That sounds to me like it's using a different exchange rate from one day to another, and I'd agree with your assessment in that case. I would have thought that the exchange rate in the transaction would be used.

Mind you, I can't wrap my head around the subtleties that seem to apply on this report.

David

On February 9, 2019, at 5:56 PM, Wm via gnucash-devel <[hidden email]> wrote:

On 09/02/2019 07:00, Wm via gnucash-devel wrote:
> Background: My gnc TB has been wrong for years.  This hasn't been a
> problem for me because I can do my own TB in sql and satisfy myself all
> is relatively well with my accounts.  Over the last week or so I decided
> to try again and I think the gnc TB report is b0rked.

OK, have a think about this.

Having deleted the tx previously mentioned.

If I run a TB report to 2017-12-17 it balances.

If I run a TB report to 2017-12-18 it is out by .01

Problem?  There are no tx on 2017-12-18

The ordinary use of a TB is to help find accounting errors.  If the
report introduces errors I say it is not fit to be used.

https://bugs.gnucash.org/show_bug.cgi?id=797097

--
Wm

_______________________________________________
gnucash-devel mailing list
[hidden email]
https://lists.gnucash.org/mailman/listinfo/gnucash-devel
_______________________________________________
gnucash-devel mailing list
[hidden email]
https://lists.gnucash.org/mailman/listinfo/gnucash-devel
Reply | Threaded
Open this post in threaded view
|

Re: [GNC-dev] tb, gnc or me? my trial balance is wrong and I think it is gnc not me

David Carlson-4
Now if there is an exchange rate entry for 12-19 and 12-16 but not for
12-18 or 12-17-2017 and the report is using nearest in time, that could
cause what you are seeing.



On Sat, Feb 9, 2019 at 7:04 AM D via gnucash-devel <
[hidden email]> wrote:

> That sounds to me like it's using a different exchange rate from one day
> to another, and I'd agree with your assessment in that case. I would have
> thought that the exchange rate in the transaction would be used.
>
> Mind you, I can't wrap my head around the subtleties that seem to apply on
> this report.
>
> David
>
> On February 9, 2019, at 5:56 PM, Wm via gnucash-devel <
> [hidden email]> wrote:
>
> On 09/02/2019 07:00, Wm via gnucash-devel wrote:
> > Background: My gnc TB has been wrong for years.  This hasn't been a
> > problem for me because I can do my own TB in sql and satisfy myself all
> > is relatively well with my accounts.  Over the last week or so I decided
> > to try again and I think the gnc TB report is b0rked.
>
> OK, have a think about this.
>
> Having deleted the tx previously mentioned.
>
> If I run a TB report to 2017-12-17 it balances.
>
> If I run a TB report to 2017-12-18 it is out by .01
>
> Problem?  There are no tx on 2017-12-18
>
> The ordinary use of a TB is to help find accounting errors.  If the
> report introduces errors I say it is not fit to be used.
>
> https://bugs.gnucash.org/show_bug.cgi?id=797097
>
> --
> Wm
>
> _______________________________________________
> gnucash-devel mailing list
> [hidden email]
> https://lists.gnucash.org/mailman/listinfo/gnucash-devel
> _______________________________________________
> gnucash-devel mailing list
> [hidden email]
> https://lists.gnucash.org/mailman/listinfo/gnucash-devel
>
_______________________________________________
gnucash-devel mailing list
[hidden email]
https://lists.gnucash.org/mailman/listinfo/gnucash-devel
Reply | Threaded
Open this post in threaded view
|

Re: [GNC-dev] tb, gnc or me? my trial balance is wrong and I think it is gnc not me

David Carlson-4
Following up on my last email, in version 2.6.17 the Trial Balance report
does default to nearest in time as the default valuation procedure.   I
don't know what happens in 3.4.

David Carlson

On Sat, Feb 9, 2019 at 9:05 AM David Carlson <[hidden email]>
wrote:

> Now if there is an exchange rate entry for 12-19 and 12-16 but not for
> 12-18 or 12-17-2017 and the report is using nearest in time, that could
> cause what you are seeing.
>
>
>
> On Sat, Feb 9, 2019 at 7:04 AM D via gnucash-devel <
> [hidden email]> wrote:
>
>> That sounds to me like it's using a different exchange rate from one day
>> to another, and I'd agree with your assessment in that case. I would have
>> thought that the exchange rate in the transaction would be used.
>>
>> Mind you, I can't wrap my head around the subtleties that seem to apply
>> on this report.
>>
>> David
>>
>> On February 9, 2019, at 5:56 PM, Wm via gnucash-devel <
>> [hidden email]> wrote:
>>
>> On 09/02/2019 07:00, Wm via gnucash-devel wrote:
>> > Background: My gnc TB has been wrong for years.  This hasn't been a
>> > problem for me because I can do my own TB in sql and satisfy myself all
>> > is relatively well with my accounts.  Over the last week or so I
>> decided
>> > to try again and I think the gnc TB report is b0rked.
>>
>> OK, have a think about this.
>>
>> Having deleted the tx previously mentioned.
>>
>> If I run a TB report to 2017-12-17 it balances.
>>
>> If I run a TB report to 2017-12-18 it is out by .01
>>
>> Problem?  There are no tx on 2017-12-18
>>
>> The ordinary use of a TB is to help find accounting errors.  If the
>> report introduces errors I say it is not fit to be used.
>>
>> https://bugs.gnucash.org/show_bug.cgi?id=797097
>>
>> --
>> Wm
>>
>> _______________________________________________
>> gnucash-devel mailing list
>> [hidden email]
>> https://lists.gnucash.org/mailman/listinfo/gnucash-devel
>> _______________________________________________
>> gnucash-devel mailing list
>> [hidden email]
>> https://lists.gnucash.org/mailman/listinfo/gnucash-devel
>>
>
_______________________________________________
gnucash-devel mailing list
[hidden email]
https://lists.gnucash.org/mailman/listinfo/gnucash-devel
Reply | Threaded
Open this post in threaded view
|

Re: [GNC-dev] tb, gnc or me? my trial balance is wrong and I think it is gnc not me

GnuCash - Dev mailing list
In reply to this post by GnuCash - Dev mailing list
Ok. I'll preface my comments with the note that every time I try to understand how the trial balance report works, someone on the list points out how misguided my understanding is....

That being said, I don't understand how a trial balance of completed transactional history could change value based on a change in price history settings. The euros I bought last week might be worth more this week, but I'd expect the report to be based on what was real, versus the imaginary value those euros have gained or lost since then.

David T.

On February 10, 2019, at 2:40 AM, David Carlson <[hidden email]> wrote:

Following up on my last email, in version 2.6.17 the Trial Balance report does default to nearest in time as the default valuation procedure.   I don't know what happens in 3.4.


David Carlson


On Sat, Feb 9, 2019 at 9:05 AM David Carlson <[hidden email]> wrote:

Now if there is an exchange rate entry for 12-19 and 12-16 but not for 12-18 or 12-17-2017 and the report is using nearest in time, that could cause what you are seeing.




On Sat, Feb 9, 2019 at 7:04 AM D via gnucash-devel <[hidden email]> wrote:

That sounds to me like it's using a different exchange rate from one day to another, and I'd agree with your assessment in that case. I would have thought that the exchange rate in the transaction would be used.

Mind you, I can't wrap my head around the subtleties that seem to apply on this report.

David

On February 9, 2019, at 5:56 PM, Wm via gnucash-devel <[hidden email]> wrote:

On 09/02/2019 07:00, Wm via gnucash-devel wrote:
> Background: My gnc TB has been wrong for years.  This hasn't been a
> problem for me because I can do my own TB in sql and satisfy myself all
> is relatively well with my accounts.  Over the last week or so I decided
> to try again and I think the gnc TB report is b0rked.

OK, have a think about this.

Having deleted the tx previously mentioned.

If I run a TB report to 2017-12-17 it balances.

If I run a TB report to 2017-12-18 it is out by .01

Problem?  There are no tx on 2017-12-18

The ordinary use of a TB is to help find accounting errors.  If the
report introduces errors I say it is not fit to be used.

https://bugs.gnucash.org/show_bug.cgi?id=797097

--
Wm

_______________________________________________
gnucash-devel mailing list
[hidden email]
https://lists.gnucash.org/mailman/listinfo/gnucash-devel
_______________________________________________
gnucash-devel mailing list
[hidden email]
https://lists.gnucash.org/mailman/listinfo/gnucash-devel

_______________________________________________
gnucash-devel mailing list
[hidden email]
https://lists.gnucash.org/mailman/listinfo/gnucash-devel
Reply | Threaded
Open this post in threaded view
|

Re: [GNC-dev] tb, gnc or me? my trial balance is wrong and I think it is gnc not me

Adrien Monteleone-2
I’m in agreement on that note. I understand why someone who doesn’t close books would like a report that shows a current position. But the Trial Balance report is really only useful for the close books crowd. It should only reflect actual transactions in the various registers. In the case of assets then, it shouldn’t be estimating based on any user choice of average, nearest date or such, it should be taking actual purchase price just like you would if you did it on paper.

Regards,
Adrien

> On Feb 10, 2019, at 2:47 AM, D via gnucash-devel <[hidden email]> wrote:
>
> Ok. I'll preface my comments with the note that every time I try to understand how the trial balance report works, someone on the list points out how misguided my understanding is....
>
> That being said, I don't understand how a trial balance of completed transactional history could change value based on a change in price history settings. The euros I bought last week might be worth more this week, but I'd expect the report to be based on what was real, versus the imaginary value those euros have gained or lost since then.
>
> David T.
>
> On February 10, 2019, at 2:40 AM, David Carlson <[hidden email]> wrote:
>
> Following up on my last email, in version 2.6.17 the Trial Balance report does default to nearest in time as the default valuation procedure.   I don't know what happens in 3.4.
>
>
> David Carlson
>

_______________________________________________
gnucash-devel mailing list
[hidden email]
https://lists.gnucash.org/mailman/listinfo/gnucash-devel
Reply | Threaded
Open this post in threaded view
|

Re: [GNC-dev] tb, gnc or me? my trial balance is wrong and I think it is gnc not me

Geert Janssens-4
Op zondag 10 februari 2019 09:56:51 CET schreef Adrien Monteleone:
> I’m in agreement on that note. I understand why someone who doesn’t close
> books would like a report that shows a current position. But the Trial
> Balance report is really only useful for the close books crowd. It should
> only reflect actual transactions in the various registers. In the case of
> assets then, it shouldn’t be estimating based on any user choice of
> average, nearest date or such, it should be taking actual purchase price
> just like you would if you did it on paper.

In Belgium you have the choice (or at least used to have - I haven't checked
lately):
We can/could report the balance using purchase values or using current values,
with current values meaning current at the report date.

Regards,

Geert


_______________________________________________
gnucash-devel mailing list
[hidden email]
https://lists.gnucash.org/mailman/listinfo/gnucash-devel
Reply | Threaded
Open this post in threaded view
|

Re: [GNC-dev] tb, gnc or me? my trial balance is wrong and I think it is gnc not me

David Carlson-4
Lets all step back a second and consider what the report is supposed to
show before deciding whether it is always correct.

The report is supposed to cover all of the assets and liabilities that were
selected when configured only over time interval which was also selected
during the configuration. It is supposed to use a method of valuation for
accounts not in the base currency which is also selected during the report
configuration.

Over the last several years there have been a number of discussions about
whether the various valuation methods work correctly, with respect to
report dates vs dates available in the data file  and even whether the
values should be displayed or calculated with exact i.e. fractional numbers
or in floating point.  All of that is in the infrastructure, not in the
actual report code.

I believe there were some code changes implemented over the life of the 2.6
code series, and there were more changes in the transition to 3.x.

I think the bottom line is that yes, different code releases will give
different results with some data even if the report configurations are
scrupulously checked to be identical.

Now, where do we want to go from here?


David Carlson

On Sun, Feb 10, 2019, 3:17 AM Geert Janssens <[hidden email]
wrote:

> Op zondag 10 februari 2019 09:56:51 CET schreef Adrien Monteleone:
> > I’m in agreement on that note. I understand why someone who doesn’t close
> > books would like a report that shows a current position. But the Trial
> > Balance report is really only useful for the close books crowd. It should
> > only reflect actual transactions in the various registers. In the case of
> > assets then, it shouldn’t be estimating based on any user choice of
> > average, nearest date or such, it should be taking actual purchase price
> > just like you would if you did it on paper.
>
> In Belgium you have the choice (or at least used to have - I haven't
> checked
> lately):
> We can/could report the balance using purchase values or using current
> values,
> with current values meaning current at the report date.
>
> Regards,
>
> Geert
>
>
> _______________________________________________
> gnucash-devel mailing list
> [hidden email]
> https://lists.gnucash.org/mailman/listinfo/gnucash-devel
>
_______________________________________________
gnucash-devel mailing list
[hidden email]
https://lists.gnucash.org/mailman/listinfo/gnucash-devel
Reply | Threaded
Open this post in threaded view
|

Re: [GNC-dev] tb, gnc or me? my trial balance is wrong and I think it is gnc not me

John Ralls-2
In reply to this post by Adrien Monteleone-2
The "Average Cost" price source creates a price based on the actual transaction prices and is designed for the TB report, see the thread beginning with https://lists.gnucash.org/pipermail/gnucash-devel/2008-July/023297.html. I broke it for 2.6.13-2.6.21, see https://bugs.gnucash.org/show_bug.cgi?id=775368. The "Average Cost" and note that even 2.6.21 didn't fully fix it; that didn't happen until 3.2.

Average cost is the *only* price source that works for the Trial Balance Report. In particular the "Nearest in Time" and "Most Recent" sources will take prices from the price database that are unlikely to reflect the actual prices for the transactions and will vary depending on the date of the report. I agree with Adrien that we should consider removing the price source option from that report and hard-wire it to average cost.

Regards,
John Ralls

> On Feb 10, 2019, at 12:56 AM, Adrien Monteleone <[hidden email]> wrote:
>
> I’m in agreement on that note. I understand why someone who doesn’t close books would like a report that shows a current position. But the Trial Balance report is really only useful for the close books crowd. It should only reflect actual transactions in the various registers. In the case of assets then, it shouldn’t be estimating based on any user choice of average, nearest date or such, it should be taking actual purchase price just like you would if you did it on paper.
>
> Regards,
> Adrien
>
>> On Feb 10, 2019, at 2:47 AM, D via gnucash-devel <[hidden email]> wrote:
>>
>> Ok. I'll preface my comments with the note that every time I try to understand how the trial balance report works, someone on the list points out how misguided my understanding is....
>>
>> That being said, I don't understand how a trial balance of completed transactional history could change value based on a change in price history settings. The euros I bought last week might be worth more this week, but I'd expect the report to be based on what was real, versus the imaginary value those euros have gained or lost since then.
>>
>> David T.
>>
>> On February 10, 2019, at 2:40 AM, David Carlson <[hidden email]> wrote:
>>
>> Following up on my last email, in version 2.6.17 the Trial Balance report does default to nearest in time as the default valuation procedure.   I don't know what happens in 3.4.
>>
>>
>> David Carlson
>>
>
> _______________________________________________
> gnucash-devel mailing list
> [hidden email]
> https://lists.gnucash.org/mailman/listinfo/gnucash-devel

_______________________________________________
gnucash-devel mailing list
[hidden email]
https://lists.gnucash.org/mailman/listinfo/gnucash-devel
Reply | Threaded
Open this post in threaded view
|

Re: [GNC-dev] tb, gnc or me? my trial balance is wrong and I think it is gnc not me

Alex Aycinena-2
In reply to this post by GnuCash - Dev mailing list
>
>
> ---------- Forwarded message ----------
> From: John Ralls <[hidden email]>
> To: Adrien Monteleone <[hidden email]>
> Cc: "[hidden email]" <[hidden email]>
> Bcc:
> Date: Sun, 10 Feb 2019 08:54:03 -0800
> Subject: Re: [GNC-dev] tb, gnc or me? my trial balance is wrong and I
> think it is gnc not me
> The "Average Cost" price source creates a price based on the actual
> transaction prices and is designed for the TB report, see the thread
> beginning with
> https://lists.gnucash.org/pipermail/gnucash-devel/2008-July/023297.html.
> I broke it for 2.6.13-2.6.21, see
> https://bugs.gnucash.org/show_bug.cgi?id=775368. The "Average Cost" and
> note that even 2.6.21 didn't fully fix it; that didn't happen until 3.2.
>
> Average cost is the *only* price source that works for the Trial Balance
> Report. In particular the "Nearest in Time" and "Most Recent" sources will
> take prices from the price database that are unlikely to reflect the actual
> prices for the transactions and will vary depending on the date of the
> report. I agree with Adrien that we should consider removing the price
> source option from that report and hard-wire it to average cost.
>
> Regards,
> John Ralls
>
>

It is possible that Wm is noting a problem in gnucash that I'm trying to
address with my 'Book Currency' enhancement (unfortunately, a bit delayed).

For most users who deal in a home currency most of the time and
occasionally buy a foreign currency, say on a trip, and spend it on
expenses, this deficiency won't show itself. But for people who deal in
multiple currencies often, with complicated transactions, it may.

Consider the following scenario:

1. A user is based in Europe and considers their home currency to be Euros
2. The user uses Euros to buy multiple lots of GBPs at different times. The
transactions each have different implicit exchange rates in the individual
splits, but gnucash doesn't do any automatic lot tracking. Some of the GBPs
are used for expenses expressed in EURs. The splits associated with these
expenses also have implicit exchange rates, but they don't have any
relationship to the purchased GBP's costs unless the user makes carefull
off-line calculations and enters the right amounts.
3. The user then uses left-over GBPs to buy USDs. The split entered into
gnucash has an implicit exchange rate of USDs to GBPs but nothing expressed
in Euros.

If you want to run a report representing these transactions in Euros there
is no way to do so unless you use an externally supplied exchange rate
(e.g., from the price db) because the splits themselves don't have all the
required information.

If you want to run a report 'at cost', you also can't do this because item
3, above, doesn't contain the right information (so you have to 'fudge it'
with an amount from the price db). This can be overcome procedurally in
gnucash by using the trick of entering the #3 transaction in a register of
an account denominated in Euros even if that account isn't involved in the
transaction. One split will sell the GBPs in EURs, the other will buy USDs
in EURs and as soon as you hit the enter key, the transaction will
'disappear' from the register it was entered in (since neither of the
splits were for that account). The transaction, however, will show up in
the registers for the accounts involved and they will contain the implicit
exchange rates that were missing above (but not necessarily with any lot
tracking and still requiring a lot of off-line calculations to figure out
the right numbers to enter into the splits). Now a report 'at cost' could
be run, but only if the trick was used procedurally for every transaction
not involving the home currency. Of course, this can't be assumed to be the
case.

The 'book currency' feature is intended to deal with this by, if the 'book
currency' feature is selected, forcing every non-book-currency split to be
denominated in book-currency (i.e., like the trick, above, but without
having to use a third account register) and enforcing lot tracking for each
of these transactions (to get rid of all the off-line calculations), thus
providing a basis for tracking cost and eliminating the need for an
external price reference (unless you want to see an estimate of current
value).

I don't know if this is related to the problem Wm is seeing, but it might
be.

Alex
_______________________________________________
gnucash-devel mailing list
[hidden email]
https://lists.gnucash.org/mailman/listinfo/gnucash-devel
Reply | Threaded
Open this post in threaded view
|

Re: [GNC-dev] tb, gnc or me? my trial balance is wrong and I think it is gnc not me

GnuCash - Dev mailing list
In reply to this post by Geert Janssens-4
On 10/02/2019 09:14, Geert Janssens wrote:

> Op zondag 10 februari 2019 09:56:51 CET schreef Adrien Monteleone:
>> I’m in agreement on that note. I understand why someone who doesn’t close
>> books would like a report that shows a current position. But the Trial
>> Balance report is really only useful for the close books crowd. It should
>> only reflect actual transactions in the various registers. In the case of
>> assets then, it shouldn’t be estimating based on any user choice of
>> average, nearest date or such, it should be taking actual purchase price
>> just like you would if you did it on paper.
>
> In Belgium you have the choice (or at least used to have - I haven't checked
> lately):
> We can/could report the balance using purchase values or using current values,
> with current values meaning current at the report date.

that would be for a balance sheet, p+l or similar public facing
document, not a trial balance

a trial balance is an internal document or work sheet to help figure out
(usually) human type accounting errors like something being put on the
balance sheet side rather than the p+l side or similar (e.g. an expense
as a liability or some such, these are quite easy mistakes to make if
you don't have formal accounting knowledge or even if you do, happen to
be busy and think "I'll just post it and sort it out later")

some people actually stop formal accounting training at trial balance
stage in the UK, I am not joking.  if you can bring a set of accounts to
trial balance you will probably find work.  the balance sheet and p+l
are the easy parts if someone else has done the work :)

if the value of a tx changed because of a share price rising or falling
or an exchange rate changing it doesn't affect the trial balance at all,
it is the numbers that we are seeking to balance, not their values

in english language accounting we call it the trial balance because it
is for testing the stuff we produce later (p+l, bs, etc)

--
Wm


_______________________________________________
gnucash-devel mailing list
[hidden email]
https://lists.gnucash.org/mailman/listinfo/gnucash-devel
Reply | Threaded
Open this post in threaded view
|

Re: [GNC-dev] tb, gnc or me? my trial balance is wrong and I think it is gnc not me

GnuCash - Dev mailing list
In reply to this post by John Ralls-2
On 10/02/2019 16:54, John Ralls wrote:
> The "Average Cost" price source creates a price based on the actual transaction prices and is designed for the TB report, see the thread beginning with https://lists.gnucash.org/pipermail/gnucash-devel/2008-July/023297.html. I broke it for 2.6.13-2.6.21, see https://bugs.gnucash.org/show_bug.cgi?id=775368. The "Average Cost" and note that even 2.6.21 didn't fully fix it; that didn't happen until 3.2.
>
> Average cost is the *only* price source that works for the Trial Balance Report. In particular the "Nearest in Time" and "Most Recent" sources will take prices from the price database that are unlikely to reflect the actual prices for the transactions and will vary depending on the date of the report. I agree with Adrien that we should consider removing the price source option from that report and hard-wire it to average cost.


 >

Why average cost?  A TB should use actual numbers.  We are meant to be
checking what happened rather than the value of what happened.

There are separate reports for that, we call them the Balance Sheet,
Equity Statement, Income Statement and similar.

The TB is internal.

At the risk of pushing too much burden onto you, JohnR, for which I
apologise, I'm not at all sure gnc's TB should be there as it stands
*unless* one rarely uses commodities of any sort.  If you're a home
currency only person that thinks stock markets are the devil's own work
it will probably be good enough.  Otherwise I think it is time we said,
not fit for purpose.

The logical problem for me is that the SQL is so simple and the Scheme
involved in a new report when it is being deprecated is largely pointless.

--
Wm

_______________________________________________
gnucash-devel mailing list
[hidden email]
https://lists.gnucash.org/mailman/listinfo/gnucash-devel
Reply | Threaded
Open this post in threaded view
|

Re: [GNC-dev] tb, gnc or me? my trial balance is wrong and I think it is gnc not me

David Carlson-4
Wm,  Where does unrealized income fall into tb if there is no price in the
database on the closing date for one or more security or currency accounts
because it was not sold?

David Carlson

On Sun, Feb 10, 2019 at 4:21 PM Wm via gnucash-devel <
[hidden email]> wrote:

> On 10/02/2019 16:54, John Ralls wrote:
> > The "Average Cost" price source creates a price based on the actual
> transaction prices and is designed for the TB report, see the thread
> beginning with
> https://lists.gnucash.org/pipermail/gnucash-devel/2008-July/023297.html.
> I broke it for 2.6.13-2.6.21, see
> https://bugs.gnucash.org/show_bug.cgi?id=775368. The "Average Cost" and
> note that even 2.6.21 didn't fully fix it; that didn't happen until 3.2.
> >
> > Average cost is the *only* price source that works for the Trial Balance
> Report. In particular the "Nearest in Time" and "Most Recent" sources will
> take prices from the price database that are unlikely to reflect the actual
> prices for the transactions and will vary depending on the date of the
> report. I agree with Adrien that we should consider removing the price
> source option from that report and hard-wire it to average cost.
>
>
>  >
>
> Why average cost?  A TB should use actual numbers.  We are meant to be
> checking what happened rather than the value of what happened.
>
> There are separate reports for that, we call them the Balance Sheet,
> Equity Statement, Income Statement and similar.
>
> The TB is internal.
>
> At the risk of pushing too much burden onto you, JohnR, for which I
> apologise, I'm not at all sure gnc's TB should be there as it stands
> *unless* one rarely uses commodities of any sort.  If you're a home
> currency only person that thinks stock markets are the devil's own work
> it will probably be good enough.  Otherwise I think it is time we said,
> not fit for purpose.
>
> The logical problem for me is that the SQL is so simple and the Scheme
> involved in a new report when it is being deprecated is largely pointless.
>
> --
> Wm
>
> _______________________________________________
> gnucash-devel mailing list
> [hidden email]
> https://lists.gnucash.org/mailman/listinfo/gnucash-devel
>
_______________________________________________
gnucash-devel mailing list
[hidden email]
https://lists.gnucash.org/mailman/listinfo/gnucash-devel
Reply | Threaded
Open this post in threaded view
|

Re: [GNC-dev] tb, gnc or me? my trial balance is wrong and I think it is gnc not me

GnuCash - Dev mailing list
In reply to this post by David Carlson-4
On 10/02/2019 12:40, David Carlson wrote:
> Lets all step back a second and consider what the report is supposed to
> show before deciding whether it is always correct.

I think the report is conceptually broken.

> The report is supposed to cover all of the assets and liabilities that were
> selected when configured only over time interval which was also selected
> during the configuration.

It doesn't do that as far as I can tell.

> It is supposed to use a method of valuation for
> accounts not in the base currency which is also selected during the report
> configuration.

It doesn't do that, actually.

My reading of the code (and I am an acknowledged fool when it comes to
scheme) is that it uses a balance sheet type approach rather than a p+l
type approach to time.  If I am right then the report is wrong to begin
with, because a trial balance can be for either,

a P+L type report (i.e. tx from begin date to end date)

OR for

a Balance Sheet type report (i.e. start at the beginning up to a point,
what have we got?)

> Over the last several years there have been a number of discussions about
> whether the various valuation methods work correctly, with respect to
> report dates vs dates available in the data file  and even whether the
> values should be displayed or calculated with exact i.e. fractional numbers
> or in floating point.  All of that is in the infrastructure, not in the
> actual report code.

That is incorrect.  A trial balance is a report, it shouldn't be doing
valuations at all *except* right at the end, maybe.  If it is going to
do valuations they should be presented separately from the account balances.

Remember flox, at trial balance (and it can be from any date to any
date) you have something like

100 in your own currency
1 in your fantasy currency
10 shares (or similar) in investment A
11 units (or similar) in investment B
etc

*that* is what a trial balance should report, the *money* value of
investment A, Investment B and the fantasy currency is what the Balance
Sheet, etc is for, the TB is a different animal.

> I believe there were some code changes implemented over the life of the 2.6
> code series, and there were more changes in the transition to 3.x.

I don't think we're getting anywhere (I have read the bug thread)

> I think the bottom line is that yes, different code releases will give
> different results with some data even if the report configurations are
> scrupulously checked to be identical.

That's scary.

> Now, where do we want to go from here?

I think we need a new TB.  The problem is it can't be implemented
however good it is because we're stuck with Scheme reports.

--
Wm

_______________________________________________
gnucash-devel mailing list
[hidden email]
https://lists.gnucash.org/mailman/listinfo/gnucash-devel
Reply | Threaded
Open this post in threaded view
|

Re: [GNC-dev] tb, gnc or me? my trial balance is wrong and I think it is gnc not me

GnuCash - Dev mailing list
In reply to this post by Alex Aycinena-2
On 10/02/2019 19:46, Alex Aycinena wrote:

> It is possible that Wm is noting a problem in gnucash that I'm trying to
> address with my 'Book Currency' enhancement (unfortunately, a bit delayed).

I am certainly interested.  I went to the end of your message and read
that just in case you thought Brexit was a good idea:)

I'm also not convinced that tricks are good and certainly wasn't
expecting what I think you are proposing, Alex.

> For most users who deal in a home currency most of the time and
> occasionally buy a foreign currency, say on a trip, and spend it on
> expenses, this deficiency won't show itself.

I think gnc is not a one currency one country application any more and
as a result I think the reporting needs to catch up.  Small expenses or
income in another currency are normal as people trade on ebay and
similar.  It seems to me the only people that don't realise this are the
very particular small minded whites that voted for Trump and maybe some
weird religious people in other parts of the USA.  Almost everyone I
know has had an international transaction of some sort in the last year.

> But for people who deal in
> multiple currencies often, with complicated transactions, it may.

I will see what you say below, I think the accounting basics are largely
correct.  I wouldn't be here if I found them wrong.

> Consider the following scenario:
>
> 1. A user is based in Europe and considers their home currency to be Euros

following

> 2. The user uses Euros to buy multiple lots of GBPs at different times. The
> transactions each have different implicit exchange rates in the individual
> splits, but gnucash doesn't do any automatic lot tracking.

Are you sure gnc doesn't do lots, Alex?  My needs are modest, heh, maybe
I'll look at lots when I've finished with this trial balance crap.  I'm
sure gnc tracks my fx tx, Alex.

If anyone is trading intra day they're on their own, I don't do that and
I am certain if someone is taking that risk they should be able to bear
the costs them self from available funds and never say to their partner
or parents or anyone else, "I fucked up".

Having said that, I'm benefiting on EUR:GBP because I'm in London and we
are screwing Brexit up badly.

> Some of the GBPs
> are used for expenses expressed in EURs. The splits associated with these
> expenses also have implicit exchange rates, but they don't have any
> relationship to the purchased GBP's costs unless the user makes carefull
> off-line calculations and enters the right amounts.

that is what my northern relatives would call a red herring

actually, I haven't phoned my aunt in Norway, but you're presenting a foil

the used currency is between you and your tax people (presuming I
understand your question).  nothing to do with gnc.

> 3. The user then uses left-over GBPs to buy USDs. The split entered into
> gnucash has an implicit exchange rate of USDs to GBPs but nothing expressed
> in Euros.

this is sounding like a 1970's communist theoretical play to me, have
you been reading either Chekhov or McCarthy? :)

Ummm, basically the current gnc TB has no clue about any movements, it
makes a guess at best.  It is functionally broken because it starts as a
balance sheet rather than a trial balance and cannot be fixed until our
seniors kill it, replace it or otherwise retire Scheme as the main
reporting sub optimal system.

We all know it is wrong, no one can fix it!

Yay :(

> If you want to run a report representing these transactions in Euros there
> is no way to do so unless you use an externally supplied exchange rate
> (e.g., from the price db) because the splits themselves don't have all the
> required information.

Nope.  The right way to do it is to use the tx values and ignore prices
altogether.

What you end up with is amounts of stuff in your accounts without the
report forcing the idiocy of a price change.

If you want to see the value at a point in time?  That is called a
Balance Sheet.

A trial balance is a different thing.

> If you want to run a report 'at cost', you also can't do this because item
> 3, above, doesn't contain the right information (so you have to 'fudge it'
> with an amount from the price db).

Ah, I think I see your point now

> This can be overcome procedurally in
> gnucash by using the trick of entering the #3 transaction in a register of
> an account denominated in Euros even if that account isn't involved in the
> transaction. One split will sell the GBPs in EURs, the other will buy USDs
> in EURs and as soon as you hit the enter key, the transaction will
> 'disappear' from the register it was entered in (since neither of the
> splits were for that account). The transaction, however, will show up in
> the registers for the accounts involved and they will contain the implicit
> exchange rates that were missing above (but not necessarily with any lot
> tracking and still requiring a lot of off-line calculations to figure out
> the right numbers to enter into the splits).

Alex!  You are a very naughty person!  I think if we ever meet in person
I will hug you if only because you are doing sums about and even within
gnc and I sometimes do that too.

> Now a report 'at cost' could
> be run, but only if the trick was used procedurally for every transaction
> not involving the home currency. Of course, this can't be assumed to be the
> case.

I think if we took that to a logical end it would be too expensive for
most people, to the extent they'd stop using gnc.

> The 'book currency' feature is intended to deal with this by, if the 'book
> currency' feature is selected, forcing every non-book-currency split to be
> denominated in book-currency (i.e., like the trick, above, but without
> having to use a third account register) and enforcing lot tracking for each
> of these transactions (to get rid of all the off-line calculations), thus
> providing a basis for tracking cost and eliminating the need for an
> external price reference (unless you want to see an estimate of current
> value).
>
> I don't know if this is related to the problem Wm is seeing, but it might
> be.

Alex, you're addressing different stuff.  I am interested but, to be
honest, I think the chances of gnc implementing your ideas soon are
remote.  I'm struggling to get the gnc team to get a basic accounting
report done.  I'm struggling to get gnc to get one of their actual
reports fixed or removed.

Oddly enough very few people seem to mention accounting these days.

Wonder why that is?

--
Wm





_______________________________________________
gnucash-devel mailing list
[hidden email]
https://lists.gnucash.org/mailman/listinfo/gnucash-devel
Reply | Threaded
Open this post in threaded view
|

Re: [GNC-dev] tb, gnc or me? my trial balance is wrong and I think it is gnc not me

David Carlson-4
Wm,  before you run off at the mouth with all your innuendos, look at facts.

Did you try running a test on one of your backups from around June 2016
using Gnucash 2.6.12 or earlier?  I think that if you do you will find that
the Trial Balance Report is correct if you scrupulously check the settings
to use the Average Price valuation method and check your price database to
be sure that the transaction generated prices are present.   John Ralls
alluded to that earlier today.

In fact, I bet you could try re-running the Trial Balance Report in release
3.4 on a current data file using the Average Balance method of valuation.
There was a period when GnuCash was not entering transaction generated
prices into the price database as Alex noted, but that was corrected some
time ago and recent prices from recent transactions are all entered in the
price database.  Granted, it will be difficult to find many of  those old
missing transaction generated prices that involve the base currency but
maybe they might get added back by using the trading accounts feature.  I
have not tried that, but it might be worth a try.   As Alex suggested,
transactions that do not involve the base currency would need to be dealt
with some other way.

David Carlson

On Sun, Feb 10, 2019 at 7:03 PM Wm via gnucash-devel <
[hidden email]> wrote:

> On 10/02/2019 19:46, Alex Aycinena wrote:
>
> > It is possible that Wm is noting a problem in gnucash that I'm trying to
> > address with my 'Book Currency' enhancement (unfortunately, a bit
> delayed).
>
> I am certainly interested.  I went to the end of your message and read
> that just in case you thought Brexit was a good idea:)
>
> I'm also not convinced that tricks are good and certainly wasn't
> expecting what I think you are proposing, Alex.
>
> > For most users who deal in a home currency most of the time and
> > occasionally buy a foreign currency, say on a trip, and spend it on
> > expenses, this deficiency won't show itself.
>
> I think gnc is not a one currency one country application any more and
> as a result I think the reporting needs to catch up.  Small expenses or
> income in another currency are normal as people trade on ebay and
> similar.  It seems to me the only people that don't realise this are the
> very particular small minded whites that voted for Trump and maybe some
> weird religious people in other parts of the USA.  Almost everyone I
> know has had an international transaction of some sort in the last year.
>
> > But for people who deal in
> > multiple currencies often, with complicated transactions, it may.
>
> I will see what you say below, I think the accounting basics are largely
> correct.  I wouldn't be here if I found them wrong.
>
> > Consider the following scenario:
> >
> > 1. A user is based in Europe and considers their home currency to be
> Euros
>
> following
>
> > 2. The user uses Euros to buy multiple lots of GBPs at different times.
> The
> > transactions each have different implicit exchange rates in the
> individual
> > splits, but gnucash doesn't do any automatic lot tracking.
>
> Are you sure gnc doesn't do lots, Alex?  My needs are modest, heh, maybe
> I'll look at lots when I've finished with this trial balance crap.  I'm
> sure gnc tracks my fx tx, Alex.
>
> If anyone is trading intra day they're on their own, I don't do that and
> I am certain if someone is taking that risk they should be able to bear
> the costs them self from available funds and never say to their partner
> or parents or anyone else, "I fucked up".
>
> Having said that, I'm benefiting on EUR:GBP because I'm in London and we
> are screwing Brexit up badly.
>
> > Some of the GBPs
> > are used for expenses expressed in EURs. The splits associated with these
> > expenses also have implicit exchange rates, but they don't have any
> > relationship to the purchased GBP's costs unless the user makes carefull
> > off-line calculations and enters the right amounts.
>
> that is what my northern relatives would call a red herring
>
> actually, I haven't phoned my aunt in Norway, but you're presenting a foil
>
> the used currency is between you and your tax people (presuming I
> understand your question).  nothing to do with gnc.
>
> > 3. The user then uses left-over GBPs to buy USDs. The split entered into
> > gnucash has an implicit exchange rate of USDs to GBPs but nothing
> expressed
> > in Euros.
>
> this is sounding like a 1970's communist theoretical play to me, have
> you been reading either Chekhov or McCarthy? :)
>
> Ummm, basically the current gnc TB has no clue about any movements, it
> makes a guess at best.  It is functionally broken because it starts as a
> balance sheet rather than a trial balance and cannot be fixed until our
> seniors kill it, replace it or otherwise retire Scheme as the main
> reporting sub optimal system.
>
> We all know it is wrong, no one can fix it!
>
> Yay :(
>
> > If you want to run a report representing these transactions in Euros
> there
> > is no way to do so unless you use an externally supplied exchange rate
> > (e.g., from the price db) because the splits themselves don't have all
> the
> > required information.
>
> Nope.  The right way to do it is to use the tx values and ignore prices
> altogether.
>
> What you end up with is amounts of stuff in your accounts without the
> report forcing the idiocy of a price change.
>
> If you want to see the value at a point in time?  That is called a
> Balance Sheet.
>
> A trial balance is a different thing.
>
> > If you want to run a report 'at cost', you also can't do this because
> item
> > 3, above, doesn't contain the right information (so you have to 'fudge
> it'
> > with an amount from the price db).
>
> Ah, I think I see your point now
>
> > This can be overcome procedurally in
> > gnucash by using the trick of entering the #3 transaction in a register
> of
> > an account denominated in Euros even if that account isn't involved in
> the
> > transaction. One split will sell the GBPs in EURs, the other will buy
> USDs
> > in EURs and as soon as you hit the enter key, the transaction will
> > 'disappear' from the register it was entered in (since neither of the
> > splits were for that account). The transaction, however, will show up in
> > the registers for the accounts involved and they will contain the
> implicit
> > exchange rates that were missing above (but not necessarily with any lot
> > tracking and still requiring a lot of off-line calculations to figure out
> > the right numbers to enter into the splits).
>
> Alex!  You are a very naughty person!  I think if we ever meet in person
> I will hug you if only because you are doing sums about and even within
> gnc and I sometimes do that too.
>
> > Now a report 'at cost' could
> > be run, but only if the trick was used procedurally for every transaction
> > not involving the home currency. Of course, this can't be assumed to be
> the
> > case.
>
> I think if we took that to a logical end it would be too expensive for
> most people, to the extent they'd stop using gnc.
>
> > The 'book currency' feature is intended to deal with this by, if the
> 'book
> > currency' feature is selected, forcing every non-book-currency split to
> be
> > denominated in book-currency (i.e., like the trick, above, but without
> > having to use a third account register) and enforcing lot tracking for
> each
> > of these transactions (to get rid of all the off-line calculations), thus
> > providing a basis for tracking cost and eliminating the need for an
> > external price reference (unless you want to see an estimate of current
> > value).
> >
> > I don't know if this is related to the problem Wm is seeing, but it might
> > be.
>
> Alex, you're addressing different stuff.  I am interested but, to be
> honest, I think the chances of gnc implementing your ideas soon are
> remote.  I'm struggling to get the gnc team to get a basic accounting
> report done.  I'm struggling to get gnc to get one of their actual
> reports fixed or removed.
>
> Oddly enough very few people seem to mention accounting these days.
>
> Wonder why that is?
>
> --
> Wm
>
>
>
>
>
> _______________________________________________
> gnucash-devel mailing list
> [hidden email]
> https://lists.gnucash.org/mailman/listinfo/gnucash-devel
>
_______________________________________________
gnucash-devel mailing list
[hidden email]
https://lists.gnucash.org/mailman/listinfo/gnucash-devel
Reply | Threaded
Open this post in threaded view
|

Re: [GNC-dev] tb, gnc or me? my trial balance is wrong and I think it is gnc not me

GnuCash - Dev mailing list
In reply to this post by David Carlson-4
On 10/02/2019 22:52, David Carlson wrote:
> Wm,  Where does unrealized income fall into tb if there is no price in the
> database on the closing date for one or more security or currency accounts
> because it was not sold?

Essentially it doesn't occur, it is just a number of something if you
aren't using Trading Accounts.

The TB isn't about the *value* of stuff, it is about what you paid for
it at a time and how much of it you have at a point in time or over a
point in time.

The Balance Sheet is about the value of stuff at a point in time.

The Income Statement is about how you got to a value over a specific
period of time.

The TB is worthless or pointless if it "double dips" into either of
those major reports remits.

===

Consider this tx

2010-01-01
Assets:Cash -100
Assets:XYZa +100

Cash is real money in your home currency
XYZa is an unlisted investment

What your TB should show wrt XYZa is that you have 100 home currency
units (usually called money) worth of it.  That never changes unless
something else happens.


Consider this tx

2010-01-01
Assets:Cash -100
Assets:XYZa +4 shares @ 25 each

Unless someone else is buying and selling those XYZa shares they are
only worth what you paid for them at best, your TB should reflect that
you have 4 XYZa shares not guess what their value is.

Am I explaining this well or not?

The Balance Sheet allows you to put a value on stuff

The Income Statement says how you got there

The Trial Balance should say what you actually have and what actually
happened rather than what you think you have and would like to have
happened.  Why?  Because it is good accounting.

--
Wm















_______________________________________________
gnucash-devel mailing list
[hidden email]
https://lists.gnucash.org/mailman/listinfo/gnucash-devel
Reply | Threaded
Open this post in threaded view
|

Re: [GNC-dev] tb, gnc or me? my trial balance is wrong and I think it is gnc not me

GnuCash - Dev mailing list
In reply to this post by GnuCash - Dev mailing list
On 09/02/2019 13:03, D via gnucash-devel wrote:
> That sounds to me like it's using a different exchange rate from one day to another, and I'd agree with your assessment in that case. I would have thought that the exchange rate in the transaction would be used.

You are correct.  I have an amount of something (shares, foreign
currency, whatever) worth (nominally) what I paid for it.

>
> Mind you, I can't wrap my head around the subtleties that seem to apply on this report.

There should not be any subtleties in a TB, it is an internal document
not a public facing one, the tax people don't ask for your trial balance
in any place I know unless they think you are  bad person in the first
place.

If I don't buy or sell some of whatever it is, I just have what I have
for TB purposes.

If I want to place a value on whatever I put a price in the price
database on a date and use a Balance Sheet report.

The TB is about *actual* tx not values at some putative point in time later.

--
Wm

_______________________________________________
gnucash-devel mailing list
[hidden email]
https://lists.gnucash.org/mailman/listinfo/gnucash-devel
12