December 2004 Archives

Blegging For An iPod

| No Comments

OK, I've checked it out and decided that this 'free iPod' thing is worth a shot. Here's how it works: you register at the site and sign up for one or more of several offers through a referral link. Once you do that, get five friends to do the same thing via your own referral link. Once they do, you get a free iPod. At press time, you can choose between a 20GB iPod or a 4GB mini iPod (in true Apple style, you get your choice of color). Yes, it's definitely multi-level marketing, but it ain't like Amway where you have to have a pyramid of 64 people extending eight levels below you to make any money at it. One offer, five friends at one offer each (and most of them are cancellable after freeipods.com credits you for completing the offer but before you're actually charged any money), and you're done. I chose Blockbuster Online because I've been thinking about trying it out anyway.

You can read more about the offer on wired.com, engadget.com, and geek.com. And I'll make the same offer that my referrer made to me - once I get my iPod, I'll change my referrer link to yours (one referral at a time, first come first serve).

Have I Mentioned That Proposal Work Sucks?

| 2 Comments

A couple of years ago, I did a tiny bit of work on this, and I got a whiff of the shit sandwich known as Proposal Work. Last year, I spent a couple of weeks writing technical verbiage for another proposal, and I got a small bite of that shit sandwich.

For the last two months, I've been getting a shit sandwich jammed down my piehole faster than I can swallow it.

But you know what? The next time I get asked to work on a proposal (and it will happen; a corrolary to Carter's Trap all but guarantees it), I'll say "Yes, sir, may I have another?"

Beyond "Don't Bring A Knife To A Gun Fight"

| 8 Comments

Once you've got that squared away, it's time for Lesson Two: "Don't Bring A Rifle To A Tank Battle."

This guy must have skipped muj class the day they discussed that:

Launch in external player

No Pressure, Right?

| No Comments

So we're banging away on our proposal, trying to find that happy medium between "so expensive they'll never go for it" and "so cheap we'll lose our shirts trying to do it." As you might expect, a lot of this involves pushing numbers around on spreadsheets and playing 'what if.' And it's not like the formulae involved are real complicated, but there are an awful lot of them, which is why I worry about things like this:

A simple spreadsheet error cost a firm a whopping US$24m.

The mistake led to TransAlta, a big Canadian power generator, buying more US power transmission hedging contracts in May at higher prices than it should have.

In a conference call, chief executive Steve Snyder said the snafu was "literally a cut-and-paste error in an Excel spreadsheet that we did not detect when we did our final sorting and ranking bids prior to submission," Reuters reports.

Maybe TransAlta is an extreme example, but this kind of thing happens all the time. I especially like this one, which had the potential to be an order of magnitude worse than TransAlta:

In about 30 minutes, officers from my own company would meet their counterparts from another multinational company to negotiate a multi-year cooperative marketing agreement. Breaking the news was not going to be easy.

My own boss would be a key player at the table. He was well briefed on the business case from our analysts, showing that we could improve profits by $200 million at a net cost of $20 million. ROI for the other company also looked good.

I didn't look forward to telling him, now, what I had just discovered: our gains are really projected as $25 million--not $200 million. We just lost our negotiating elbow room.

Why didn't we know this sooner?

TO ERR IS HUMAN
The business case came from some very competent analysts. They had checked and re-checked input data and assumptions continuously during several weeks of case building. Certainly the Excel-based financial model at the heart of the case projected $200 million in gains and $20 million in costs. What I found at the last minute, was that some very long spreadsheet formulas had parentheses out of place. When I put them where they belonged, our projected gains fell by seven-eighths.

Everyone on our side was relieved, I think, when the negotiations fell through later that day. We had learned the truth about spreadsheet errors, in spades: They don't reveal themselves.

To paraphrase Barnum, there's a spreadsheet screwup born every minute, and two to take advantage of it. First, this relatively harmless case:

By way of example, in November last year we bought ScS Upholstery (a specialist sofa retailer) at 172p after it warned on profits. A spreadsheet error had caused the wrong cost to be input into the system, with the result that their gross margin had slipped without management knowing. We knew that this would quickly be rectified, and sure enough the share price bounced back rapidly.

Although we sold at 224p in March (strong results last week sent the price above 240p), we were very happy to take our profit once the short-term mis-pricing had been eliminated.
Although I have to wonder how the investment firm knew about the spreadsheet error.

And second, this case where ENA ate ISIS' lunch:

Education Networks of America (ENA) operates the Internet network connecting schools across Tennessee. ENA twice won multimillion-dollar state contracts with bids that appeared higher than others. Jacqueline Shrago, who was the Education Department's project manager, said the ISIS 2000 bid appeared cheaper until costs for all three and a half years were computed. Looking more closely, it appeared that ISIS had made an error on a spreadsheet and that the bid actually was higher than ENA's, she said. 'We asked questions in the protest period, and they would not give an answer,' Shrago said. She said ISIS never explained, either, to federal agencies, which fund a large part of the schools computer system. So ENA got the contract.

How Not To Fix College Football

| 8 Comments

Bob Davie is, as they say, one of the best X's and O's coaches in the business (and I'm not just saying that because he wrote this epic singing the praises of Braylon Edwards). I especially like his Football 101 series.

But he's not a big-picture kind of coach, which is one reason he didn't last a long time at fND (lack of talent development, then lack of recruiting were two others). That flaw shows itself here, with what I think is a pretty dumb idea on how to 'fix' college football:

Everyone agrees that the current BCS system is headed for a potential nightmare if USC, Oklahoma and Auburn remain undefeated.

. . .

I offer a simple solution -- put it in the hands of the 117 Division I-A coaches and let them decide their national champion. With so much at stake, let's not look back after one team is left out and say, "We should have tweaked the system."

1. Let all 117 Division I-A coaches have a vote
Currently, 61 coaches get to vote and it's human nature that there is going to be some bias and some coaches will have an agenda. With 117 coaches voting, those biases or agendas will be diluted.
Recall that the BCS was invented to solve the problem of split or disputed 'titles' (I'm using scare quotes with malice aforethought here - just because a group of writers and/or coaches say you're the best doesn't necessarily make it so). I suppose only using one poll instead of two eliminates the split poll issue, but that's like saying you'll reduce wear on your car tires by only driving on two wheels.
2. Provide the coaches with game tapes
Let the coaches have each team's final three game tapes to evaluate. The reason I say give them the final three tapes is because it should be determined who is playing the best football late in the season regardless of who their opponent is. Coaches are experts at analyzing and breaking down opponents' tape. The problem in the current system is that it is virtually impossible for the 61 coaches voting to get to see all the unbeaten teams.
So how many teams do you evaluate? Who compiles the game tapes and sends them out? And most importantly, who pays for it?
3. Give the coaches two days to evaluate the tapes
Given time, the coaches will do a thorough job. If we have to, let's put a moratorium on recruiting and make this the coaches' No. 1 priority. I have great confidence that the coaches realize what is at stake and with no one being able to gain a competitive advantage in recruiting, coaches will dedicate the time and provide a thorough evaluation.
I'll give Davie credit for addressing one of the major problems with the coaches' poll: a coach knows his own team best, his opponents' teams next best, and everybody else not well at all. Forcing the coaches to spend two days looking at game films could solve this problem.

But.

I am as confident as Davie is, but in the other direction. I can't imagine ALL 117 COACHES being willing to take a two-day hiatus from the business of their own team to decide who 'wins' the 'title.' And just because they couldn't recruit doesn't mean that they necessarily would spend the two days reviewing the tapes (there are lots of other things they could be doing that don't involve recruiting), so there goes the throrough evaluation Davie is expecting. And that doesn't even count the schools who cheat now - they'd DEFINITELY figure out a way to gain an unfair advantage over those two days.
4. Eliminate the media and computers
With no disrespect to the media or those running the computers, this is the coaches' national champion. Coaches make their living by evaluating strengths and weaknesses, so let them do what they do best.

And I'm sure the media would just roll over and let this happen. Right after monkeys fly out of my butt.
Why this will work
Obviously, there is no perfect way to decide which two teams should play for the national championship. Under this plan, we can keep the integrity of the system and simply tweak it if three teams remain unbeaten. Obviously, one team will still be left out, but I think USC's Pete Carroll, Oklahoma's Bob Stoops and Auburn's Tommy Tuberville would agree that their peers are the most qualified to make the final decision.
No. The teams are the most qualified to make the final decision. On the field, the way they do in every other college sport - INCLUDING THE LOWER DIVISIONS OF FOOTBALL! C'mon, Bob; this ain't brain surgery - take the top 12 teams (hell, use the current BCS formula for it; that'll be good enough. Or use a selection committee, the way they do in basketball) and move the top 4 second-tier bowls to the week before New Years to match up teams 5-12 (give byes to teams 1-4). Those winners go on to the current 'big four' bowls on New Years. Create two new bowls for those winners the next weekend, then one more for the championship game the next weekend.

Nobody plays more than three games more than they otherwise would, and the season's still over before the Super Bowl.

Instant Computer Museum

| No Comments

Got 300 grand? Want your own computer museum? Go see eBay.

Pages

Powered by Movable Type 4.34-en

About this Archive

This page is an archive of entries from December 2004 listed from newest to oldest.

November 2004 is the previous archive.

January 2005 is the next archive.

Find recent content on the main index or look in the archives to find all content.