tax software (was Re:Speech on Linux...)

george fiala fiala-WCaKCDwya6ZYzD5mSbZInQ at public.gmane.org
Fri Aug 13 12:50:51 UTC 2004


I have been in touch with CCH.ca (they are the ones who write CANTAX)
apparently they are working on an OpenSource/Linux version, but it will
not be ready for this upcoming tax season.

This is from the President and the dev. manager.

George.



On Thu, 2004-08-12 at 22:49, Christopher Browne wrote:
> > On Thu, 12 Aug 2004, James Knott wrote:
> > > > ...I think you'd need a program, not just a set of rules. 
> > > > I doubt you could write a rule-interpreting engine that could do the job
> > > > and wasn't essentially an interpreter for a programming language.
> > > 
> > > No, not the program, just the rules for running one.  As it stands now, 
> > > it publishes the relevant acts for accountants and whoever else can be 
> > > bothered reading them.  How's that different from publishing tax rules 
> > > in a "computer freindly" format, that apps could use?
> > 
> > The problem is that the rules are so complex and so unsystematic that a
> > "computer friendly" format almost certainly would have to *be* a program. 
> > I think you're grievously underestimating the complexity of turning the
> > current RevCan pronouncements into something that's easily analyzed and
> > unambiguous.
> 
> CRA pronouncements aren't everything, either.
> 
> 1.  What they document is the forms that _they_ have set up as _their_
> interpretation of tax law.
> 
> 2.  Real tax law is much messier still.  It represents the combination
> of:
> 
>   a) The Income Tax Act, along with its numerous amendments over the
>      years, as acts passed by Parliament;
> 
>   b) Interpretation Bulletins, written by CRA, that provide CRA's
>      interpretation of how to handle various issues;
> 
>   c) Precedents set by tax cases decided by the various courts in the
>      country.
> 
> These are the "authoritative" sources for tax law, and they don't
> forcibly have to agree with one another.
> 
> Note that the bulk of the law has to do not with determining tax rates,
> but with deciding how to recognize activities as being events that
> are either:
> 
>   a) Taxable transactions, under one section or another, or
> 
>   b) Not taxable, which might also imply things like "not deductible,"
> 
> as well as determining _who_ would be assessed taxes on the transaction.
> 
> A good tax lawyer spends their time figuring out how particular
> transactions are to be taxed, which has pretty nearly nothing to do with
> how tax returns are prepared.
> 
> The fairly enormous complexity is all about there being a whole lot of
> "corner cases" that someone found a loophole for that
> Parliament/CRA/Courts decided needed to be closed up due to some notion
> of "unfairness."
> 
> What this implies is that typical proposals to somehow magically
> "simplify" tax law are pretty much doomed to failure.  Those that
> propose them seem to fall into two categories:
> 
>  1.  Those that don't understand why their idea is doomed, which implies
>      that they are in some fashion "too stupid" to deserve to set tax
>      policy, or
> 
>  2.  Those that _do_ understand that what they're pretending to do won't
>      address the real complexities, which implies that they're lying
>      scoundrels that are _morally_ deficient and so shouldn't be
>      "guarding any cookie jars."
>  
> > > > (Say you've got things running great, including the complications of
> > > > self-employment... and then RevCan decides that most small businesses
> > > > henceforth have to use the calendar year as their fiscal year...
> > > 
> > > Will off the shelf tax programs handle that?
> > 
> > Can't say for sure -- haven't used one -- but any program which can
> > handle self-employment *must* handle this, and more people than you
> > might think have some amount of self-employment income in addition to
> > a job.
> 
> Self-employment is the "big punt" where some complexity leaps in.
> 
> > A generally useful program can't just say "hire an accountant" when it
> > hits anything even vaguely complicated, because there are too many
> > such things, and too many people to whom they are relevant.
> 
> I set up tax forms last year in Prolog that know how to cope with, hmm,
> let's see...
> 
>  - Schedules 1, 3, 4, 5, 7, 9, and 11.
>  - T1
>  - Ontario and Federal LSIF forms
>  - ON428 and ON479 forms to calculate Ontario credits and taxes payable
>  - T2038, T2204, T4s, T626, T691 (Alternative Minimum Tax).
> 
> That more than covered the forms I was concerned about, with roughly 400
> Prolog rules in about 100Kb/13K-lines of code.  Come next year, I'll add
> in 2004 rates, and update it for new 2004 rules, and probably toss in a
> few more forms.
> 
> Throwing in "self employment" income involves building an income
> statement, which amounts to throwing in a spreadsheet that ends with the
> line "Net Income" that turns into one line on the tax return.  There may
> be a host of complexities to that, but in most ways they have relatively
> little to do with the complexity of a tax return.
> --
> wm(X,Y):-write(X),write('@'),write(Y). wm('cbbrowne','cbbrowne.com').
> http://www3.sympatico.ca/cbbrowne/freetaxsoftware.html
> You don't *run* programs on Ultrix.
>                 - Mark Moraes
> Right, you chase them.
>                 - Rayan Zachariassen
> --
> The Toronto Linux Users Group.      Meetings: http://tlug.ss.org
> TLUG requests: Linux topics, No HTML, wrap text below 80 columns
> How to UNSUBSCRIBE: http://tlug.ss.org/subscribe.shtml
G E O R G E F I A L A
Director
IQ Partners, Inc
99 Spadina Avenue, Suite 650
Toronto, Ontario
M8V 5P3

--
The Toronto Linux Users Group.      Meetings: http://tlug.ss.org
TLUG requests: Linux topics, No HTML, wrap text below 80 columns
How to UNSUBSCRIBE: http://tlug.ss.org/subscribe.shtml





More information about the Legacy mailing list