These numbers aren’t right

I was close to releasing the code – and that’s when I looked at the numbers. And that proceeded to change my whole sunday morning.

In early 2000, I bought my first home – through a mortgage broker, who promptly sold the loan to Wells Fargo. And somewhere along the way there was an error. When Wells Fargo sent my first statement, the interest was normal, the principal was normal, the PMI was normal, but the property tax escrow was close to 10 times what was needed. My tax bill was supposed to be about $1,500 (Wake County + Garner tax rates were a little over %1 at the time)- and instead of taking the ~$125, they were billing me $1,300.

I called Wells Fargo, got the front-line support, and the first representative told “Well, we collect your property tax for escrow and it’s based on your local tax rates”

Yes, I know that. But it’s a 10 times more than it’s supposed to be. That’s when I got the killer response – that I remember verbatim to this day:

“Well, this is what it says on my screen. We’ll have to put in a research request, that will take up to 6 weeks.”

It says this on my screen

I asked for supervisor escalation, and get a very polite supervisor that is saying the same thing. “it’s based on your tax rate… we’ll have to research it”. And I’m like “Ma’am – There is NO place in the entire United States with a Property Tax rate that high. It’s kinda of interesting that it’s approximately 10 times more. Do you think that the decimal place is off? “Well, so it does seem to be. Let me check on that and be back with you shortly”

And without a research request, without up to six weeks, when the brain finally engaged – I got the issue corrected.

See that’s what happens. Somehow all the brains get turned off with what the “computer says” – and then suddenly when the faith in the almighty computer is shaken, it affects the whole system – because no one gets the technology. It’s magic somehow (and the IT staffs are well to blame for this, because we think the magic helps keep a segment of our ranks employed).

Two weeks ago, I was merging some code that I had been writing for reports and number summaries back into the code base for our directory/workstreaming application, and was looking at the published item and edit counts for our applications (summarized out of the activity streams). And the numbers were good, great, through the roof.

And not right.

Not that we won’t have those numbers, but it just didn’t make sense. And it turned out that I had logic errors in the code that forgot about passing object variables around in Ruby and just because it wasn’t in the object didn’t mean it wasn’t being passed by reference. And a few hours later that Sunday morning, I had the problems fixed – before the numbers got out there, and before others started trusting was on their screens.

My lesson – and the ongoing lesson for all us in systems administration, programming – finance and anyone using tools we put together.

You have to have a feel for the numbers and whether the numbers are right. You have to question how they got there – especially when the numbers confirm your own opinions.

And when the numbers aren’t right – you have to check the numbers.

And that makes all the difference.

Here’s an idea

Okay, okay, I know that I said no more politics posts, but this one is slightly different.

I’m not exactly sure where to find this data, but I’m under the (perhaps mis-) impression that some campaigns end up with a surplus once the election is over. Particularly the national committees. (I’ve always heard anecdotes about “campaign coffers” and “war chests” in off-election years).

On November 5th, any and all surplus funds that the Obama/Biden and McCain/Palin campaigns have should be turned over to the treasury to finance the bailout. Any and all surplus funds raised within any of the U.S. House and Senate races get turned over to the treasury to finance the bailout. All money currently raised by incumbent Senators not up for re-election this year, gets turned over.

The campaigns and the national committees will be allowed to hold a portion of funds for operating expenses. A volunteer independent panel of American citizens that do not have family or direct business ties to the national committees and campaigns will be selected, similar to how juries are selected, to review and make decisions on the “fair wages” (operating budgets) of the staffs of the campaigns and committees.

For the next year for all House candidates and those Senate candidates up for re-election in 2010, and for the next two years for all others, including the DNC and RNC – they will solicit contributions to the campaigns as normal, but all funds, minus operating expenses, will go to the bailout.

All public financing is cut off and that money shifted to the bailout.

Those campaign bailout investments, once recouped from those institutions that the government is investing in, will be directly returned to the campaigns, along with all investment earnings (calculated as a flat percentage across the board of all investments – making sure that it’s fair to all parties), minus administrative overhead. The more you put in, the more you earn out – provided of course, that the investments bear a return. For those investments not yet coming to fruition, and for campaigns that are not needing the liquidity for a few years, they get “shares” in the ownership stake that the government gets in the institutions that it invests in.

Yes, of course, aspects of this are an absolutely ludicrous idea. Yes, of course, we aren’t talking too much money, percentage wise to the overall bailout monies, maybe a few million dollars, or slightly more across all of Congress and the committees.

But can you think of a better way to get the campaigns personally invested in seeing that the oversight on the bailout is done right?