The inside was no less elegant. The dark wood motif continued to the sink area and the border of the mirror. The brass door handle and plaque became brass faucet handles. "There's just something about a men's room with wall-to-wall carpeting," thought Bill as he took a quick peek at himself in the wood-framed mirror. Everything was perfect. His pin stripes were not too wide, not too narrow, his tie was dark blue, his shirt a glaring white, and not a hair on his head was out of place.

Satisfied, Bill moved to the business end of the men's room and was immediately presented with a ghastly problem. The company CEO was occupying one of the four urinals. And furthermore, he had selected a central urinal -- number 2 - as opposed to the usual choices, number 1 or number 4, which left the maximum space for the next occupant. "A bold choice," thought Bill admiringly. "No wonder he's the CEO."

Quick thinking was called for. Urinal number 1 was out of the question. That would give the impression of cozying up to the big guy or, perhaps, something even more unthinkable. Urinal number 4 was the conservative choice, but what about number 3? Was this an opportunity? A powerful statement of comradeship without the unthinkable alternative interpretation? Or would it be an unwelcome intrusion on the CEO's personal space? Bill's felt a chasm opening under his feet. His career hung by the slenderest of threads. For a split second he considered diving into a stall and defusing the whole dangerous situation, but finally he seized his opportunity and bellied up to urinal number 3.

"Bill," said the CEO. "I understand you're giving us a little talk on that new system for your department. I hope it's going to solve those customer reporting problems. We can't lose customers like that. Word gets around..."

"I think you'll be pleased with what we came up with," said Bill. "The reporting should be excellent, and we estimate quite a head count reduction as well."

"Looking forward to hearing about it," said the CEO as he zipped up and headed out.

Alone at urinal #3 Bill exhaled the breath he had been holding and was finally able to complete the function he had come in to perform.

During his presentation Bill had noticed that one member of his audience had been squirming around in his seat and looking uncomfortable. It was George, the token geek on the committee whose function was to provide the technical point-of-view on new projects. As soon as Bill opened the meeting to questions his hand was up.

"Sounds like a very, er, promising project," said George, "But I've got a few questions on the implementation."

Bill had only heard a portion of what George had said. He had been distracted by the tabs on George's shirt collar which turned up slightly at the ends. These people shouldn't be allowed to wear suits, thought Bill, they should be issued some kind of coverall with their name on the pocket. Still, George was dangerous; he was the one person on the committee who might ask questions that Bill couldn't answer with generalities and vague promises.

"First, you didn't mention a database product. Which one are you planning to use?" said George, "and then I've got some questions about things that seem to be missing from your fixed costs section."

Now what was that database product he'd heard mentioned, thought Bill frantically. was it SAGE or FREEBASE? Suddenly Bill had an inspiration, "Well, George, we'll use the company standard product of course," he said with just a hint of approbation in his voice.

"That's fine," said George, "But now about those fixed costs... I don't see data conversion mentioned, and what about interfaces to the mainframe, and have you considered..."

"Nice presentation, Bill," said the CEO, tiptoeing into the conversation like a moose into a church social, "This will give me some ammunition to use when I go talk with Graham from United."

Relief! Elation! Not only did he now not have to answer the troublesome questions that George was in the process of asking, but also, the CEO's approval meant that the project was a go. There was no way the committee could reject it after that statement had been made. At the same time, however, the mention of "Graham from United" referred to an embarrassing incident that had recently occurred in his unit, where a customer had been so upset with the reporting Bill's unit provided that he had stopped purchasing from Bill and gone to a competitor. That part of the statement made him feel a bit like the family dog that had just pooped on the oriental rug and was being scolded by his owner. The elation elevator going up and the poop elevator going down passed each other at high speed, cycled back and forth, and finally came to rest in the form of a lump in his stomach.

"Yes, sir. I think this new project will address all of his concerns."

You're a US senator attached to a budget committee. A scientist makes a presentation to the committee for funds to build a powerful new cyclotron. He explains that this device will allow his group to smack two gold nuclei together at almost the speed of light. This, they hope, will result in the creation of a new particle more massive than any seen previously which will immediately decay into a swarm of quarks, gluons, muons, photons, left-handed pi mesons trying in vain to shake hands with right-handed neutrinos, and a host of other rare and tiny creatures. Analysis of this data will allow the scientist to figure out why things weigh what they do and, in nine gazillion years, when the earth has been destroyed and our sun is a burned out cinder, whether the universe will be getting a little bigger or a little smaller.

So how do you vote? Fund the new cyclotron or not? Bear in mind that you didn't understand a word the scientist said and you have absolutely no idea why anyone would want to know the things he's trying to find out. The only parts of this that you do understand are that he's asking for a whopping big amount of money and that the people in your home state -- who you assume are glued to their TVs watching you on CSPAN -- need almost that same amount of money for new schools and highway improvements.

If that was the way it worked, the only science projects that would be approved would be ones to determine the most soothing colors to use in decorating day-care centers. So how does it work? An intermediary agency helps smooth the way. The National Science Institute produces a Research and Development Budget which includes all the outstanding R&D items as line items. "That's more like it," says the senator. He can approve the R&D budget as a whole on the grounds of "maintaining America's leadership position in technology" even though he might have to disapprove every individual item within it. Do you remember "Don't ask, don't tell," that ridiculous method of dealing with gays in the military? Perhaps this was its true genesis. The senator won't ask about the line items unless the budget has increased dramatically from the previous year or he's desperate to find budget cuts, and the National Science Foundation that produced the budget isn't about to explain in detail what the senator has just approved.

It works about the same way in business. New data processing projects come out of a budget with a name something like the Capital Expense Budget. Like the R&D budget, by the time this item comes up for approval it consists of a large number of line items scattered over every department in the company, and no one really wants to take the time or expend the effort to understand each one.

All good so far, but a mystery remains; where did Bill get the Cost and Benefit numbers he presented to the committee? No analysis has been done. The task of doing an even slightly reasonable analysis job is a project in itself that would require its own approvals and the assignment of a resource. So what Bill did was wing it. On the Benefits side, he merrily made assumptions about staff cuts and new customers that were based on nothing much more than the good mood he was in, and then converted them to dollars and made slides out of them. The cost side was a bit trickier. He began by taking the costs from a system that a fellow department manager had recently installed, and then modified them for his own unit. His modification technique consisted of eliminating a great many items from the other system that didn't apply to his system, and adding in a few items from his system that weren't part of the other. He was pleased to find that the costs for his new system would be dramatically lower than that of his fellow manager.

What will happen to these numbers after the meeting? Surely no one can take them seriously. Regards the Benefits numbers that's true. No one will remember them after the meeting is over and the go/nogo decision has been made, and no one will ever go back years later and check to see if any tiny fraction of those expected benefits were realized. The costs are a different story. They will become part of the company's budget for the next year and, largely will become the new projects budget. They will have an enormous influence on how the project turns out and even on what it contains.