Mailinglist Archive: opensuse (1696 mails)

[Dan Goodman <dan.goodman@xxxxxxxx>]

Boyd Stephen Smith Jr. wrote:
> In <200907161322.00011.rschulz@xxxxxxxxx>, Randall R Schulz wrote:
>
> > Nor would I want to trust an amateur's encoding of the tax
> > laws to guide the creation of my tax return.
>
> While there are amateurs involved in open-source projects, there are also
> plenty of professionals.
This may be true, but as to the feasibility/probability of getting it
right, and the consequences thereof:

My wife has prepared both individual and business taxes for clients for
many years, both before and after she passed her CPA exam, and has tried
many of the "industry standard" solutions, both those for professionals
and those for the general public.

And every year she finds that she must wait til near the end of the
yearly cycle to file for her clients, because the software is still
being corrected/updated right up to almost the deadline. Often they are
subtle bugs, but often, too, they involve significant errors or changes.

And US tax law makes you, the taxpayer, ultimately liable for any errors
in your tax submission. And the preparer as well, if a professional is
paid to prepare it.

In light of that, and the fact that the big players have individuals
dedicated fulltime to just tracing/testing and verifying last minute
changes to the tax code, there is just no way to get volunteers to meet
or beat that level of effort to obtain accuracy, quality and ease of use.

There is simply no way to ensure that an opensource project could or
would be able to muster that kind of deadline-oriented, detail-oriented
effort every year. And there is little likelihood that any preparer,
whether a business, an individual, or a tax preparer, would accept the
liability for any errors that crept in due to the use of software that
had not gotten those last-minute bugfixes and/or code changes.

Added to all that, every few years, the IRS waits until almost the last
minute, due to Congressional constraints beyond their control, before
publishing the final version of the tax computations for that tax year.

As much as I like Linux, I would either run a commercial package (where
at least you can hide behind the idea that lots of people got a similar
buggy program), or would hire a professional to take either my paper
financial data or my computer files of my financial data, and compute my
taxes from that data.

The only way I see this working in Linux is either a government
initiative that explicitly is required to be open source, or someone
leveraging a Linux based tax solution into a commercialized version of
opensource, similar to Tripwire, or perhaps Mozilla, where revenue for
staff can be obtained on a regular basis.

Because without individuals dedicated to both taxcode interpretation
(the formulae per se are never published, just the legalese that was
mandated by Congress), and other individuals dedicated to either some
form of input table maintenance or code maintenance, it just won't "gel".

Besides the inherent complexity of a wiki-based solution for tax
formulas, what about the issue of how conflicting wiki entries will be
resolved?

Imagine if using Wikipedia had major financial impact on users, and
there was no dedicated Wikipedia staff to research and resolve issues of
accuracy, etc. Would you trust it at all under those circumstances?

And do you trust Wikipedia (as opposed to using it as a form of
sophisticated knowledge search engine that has to be verified
independently), even as it is today?

Build a Linux-based tax product that can be sold as an appliance,
similar to firewall appliances, and that is cheap and easily networked
for small offices (without technicians!) and if you make it cheap
enough, perhaps, a very BIG PERHAPS, enough tax preparers would buy into
it, once it reached a critical mass and was able to support a research
and update staff similar to, e.g., TaxAct.

And I can't imagine anyone being able to raise enough VC to fund the
liftoff to that level of critical mass, not in today's world.

To steal from Archimedes, there is no royal road to simple, easy to
build and use, opensource tax software that is also accurate enough to
avoid penalties and or lost deductions at a very high percentage of
success.

And if it isn't at least 99.99% right, with a bound on the size of
errors as well (an impossible condition in itself), it is not a smart
move. After all, when your competitive alternative can deliver this sort
of thing for $20 to $40 dollars, and a tax preparer can and will do a
decent sized small business for a few hundred dollars, where does the
big savings come from that makes people want to switch? (And most of the
cost is usually in the hand-holding and "shoe-box" organizing a tax
preparer must do -- and that doesn't go away with Linux -- so the actual
dollars of cost you could reduce are minimal.

Sorry, but I think that that is one field of dreams that you could
build, and still very few would come...if any, other than those
committed to the project.

Dan

Notice: This communication, including attachments, may contain confidential or 
proprietary information to be conveyed solely for the intended recipient(s). If 
you are not the intended recipient, or if you otherwise received this message 
in error, please notify the sender immediately by return e-mail and promptly 
delete this e-mail, including attachments, without reading or saving them in 
any manner. The unauthorized use, dissemination, distribution, or reproduction 
of this e-mail, including attachments, is strictly prohibited and may be 
unlawful.
--
To unsubscribe, e-mail: opensuse+unsubscribe@xxxxxxxxxxxx
For additional commands, e-mail: opensuse+help@xxxxxxxxxxxx