Y2K may have already caused personal property tax problem

greenspun.com : LUSENET : TimeBomb 2000 (Y2000) : One Thread

In Miami County Kansas a man got a personal property tax bill for his stock trailer based on an appraised valuation of $5,221,800 rather than its true value of $2,218. This caused more problems than just someone getting a bad bill - the school district, the county and other jurisdiction budgets were already set based on the assessed valuation of $1.5 million that really wasn't there!

It's already started folks...

-- (victoria@oz.sunflower.org), December 06, 1998

Answers

Odd that no one involved in the budget process noted the discontinuity in total valuation.

-- Tom Carey (tomcarey@mindspring.com), December 06, 1998.

Doesn't surprise me too much.

Tom, I once headed up a group of engineers, all young (except me), who swore that if it came out of a computer it was correct. It took me a couple of years to correct this notion.

I had to keep telling them that if the results violated the Law of Least Astonishment they'd better check things out.

Maybe these people have never been exposed to garbage results before and just accepted them.

-- rocky (rknolls@hotmail.com), December 06, 1998.


I've been of the opinion that the IRS is already toast but it isn't public knowledge. They spent, what, $40 Billion and 4 years trying to upgrade and it ended in failure? Their CIO quit. Their systems have to be suffering through some serious fatigue already.

Mike ==============

-- Michael Taylor (mtdesign3@aol.com), December 06, 1998.


Just be glad we don't get ALL the government we pay for!!

Bill

-- Bill S. (Bill_S3@juno.com), December 06, 1998.


Got my automatic upgrade to TurboTax last night for 1998 tax year. Their instruction claim Y2K compliance, and tell how to enter either 4 digit or 2 digit years, and what defaults are used.

Said very clearly that the output reports are two digit years 'by law" and that the program is matching IRS requirements for 1998 reporting. Means that IRS is screwed up at least through mid-April (thought testing was to finish in January ????) by receiving non-compliant dates.

They (the IRS) can convert, but this shows me that they are planning their future correctly (efficiently).

It's starting.

-- Robert A. Cook, P.E. (Kennesaw, GA) (cook.r@csaatl.com), December 08, 1998.



Robert, I believe you are suffering from a common mis-conception regarding the Y2K problem. In your last post, you said the following:

"Said very clearly that the output reports are two digit years 'by law" and that the program is matching IRS requirements for 1998 reporting. Means that IRS is screwed up at least through mid-April (thought testing was to finish in January ????) by receiving non-compliant dates."

Remember, software systems don't actually use dates but rather representations of dates. As long as the actions that are performed on those representations are the ones that are desired, the system is compliant. The use of four digit years in representing date values does not equate to compliance, nor does the use of two digit years automatically represent non-compliance. It is all in whether or not the system in question correctly recognizes the dates represented as they are meant to be represented and correctly processes them. Four digit years could still be processed incorrectly (although it would be a screw-up of major porportions) and two digit year representations can be processed correctly.

Using a system called 'windowing,' many systems out there will interpret the value '01/01/00' as meaning January 1, 2000. A subset of this technique is known as 'sliding windows,' where the interpretation of a date value is based not only on the two-digit representation but on the current date as well. As a specific example of 'sliding windows,' the payroll system used by my employer will interpret a couple of two-digit date representations in the following way:

Date of Birth: Always defaulted to be the most recent year in the past for which the representation is applicable. Today it interprests 3/4/80 as March 4, 1980 and 3/4/00 as March 4, 1900. In December of 2001, it will still interpret 3/4/80 as March 4, 1980 but would interpret 3/4/00 as March 4, 2000.

Date of Hire: Assumed to be up to 1 calander year in the future, otherwise it is in the past. Today, a hire date of 3/4/80 is March 4, 1980 and 3/4/00 is March 4, 1900. On December 8, 1999 it will treat 3/4/80 as March 4, 1980 and 3/4/00 as March 4, 2000.

This sliding window approach isn't perfect, but in many cases it is good enough to be a practical solution so long as there is some sort of 'manual override ' process to handle the exceptional cases. Certainly the use of four digit year representations is preferable, to be blunt, it isn't always practical, especially as the time left to make corrections becomes vanishingly small.

-- Paul Neuhardt (neuhardt@ultranet.com), December 08, 1998.


Paul - you're right - two ways. Windowing is part of how turboTax is interpretting "98" type dates. Their instruction manual covered that.

Also, I mis-typed my comment about the IRS - it should have been "inefficient" instead of "efficient"- rather than correcting even the simple "received dates" created by the electronic exchange industry, they (the IRS) are postponing even that level of compliance by demanding two digit dates.

But, they can correctly process process the "xx" dates.

-- Robert A. Cook, P.E. (Kennesaw, GA) (cook.r@csaatl.com), December 08, 1998.


Moderation questions? read the FAQ