Benjamin Graham in his book Intelligent Investor advocates an approach to investing called value investing—buying stocks below their intrinsic value. No one internalized Graham’s value strategy better then Warren Buffet. Buffet knows the returns come not from buying the undervalued stock but rather what happens when the market realizes its true value.

But nothing gets Buffet more excited than if the company also increases its earnings. In this case, not only will its earnings per share increase but the PE ratio also—growth companies have higher PE ratios—causing the stock price to balloon.

Just as Buffet’s dedication to value investment separated him from other investors, businesses can separate themselves from other businesses by delivering their products and services faster. In a world where companies compete selling similar products, the ability to deliver faster will generate not only more business but also higher prices. Like the earnings per share and PE ratio double effect Buffet delights in, businesses can delight in greater market share while reaping higher margins.

The book Competing Against Time illustrates the dual benefits of delivering faster. An example highlights Atlas Doors which makes industrial doors. In an industry growing at 5 percent with deliver dates between 12 to 15 weeks, Atlas takes only 3 to 4 weeks to deliver a door while growing at 15 percent. Not only were they growing faster, their profit margins were 5 times higher than the industry average.

It would be hard to find a more dedicated management team. When a customer calls to request a door they might have to wait as long as a week for a response. Atlas automated its order system linking engineering, pricing, and scheduling. Now 80% of order inquires can be answered while the customer is on the phone. The factory machines were upgraded to reduce changeover times along with a custom system to track the parts in production so they arrive at the shipping docks at the same time. While competitors were saying their customers will wait for their doors, Atlas was busy stealing them away.

But if machine setup times and supply chains plague manufacturing then group organization and workflows plague the software industry. For example, many software shops struggle to fix customer found bugs in a timely manner. It may take 15 weeks to deliver a door but it can take years to deliver a requested bug fix. So few organizations have solved it that many just expect to lose a certain percentage of customers due to unresponsiveness.

The truth is more complicated. Most organizations are set up in vast pyramids with each group reporting to the group above it. Communications between groups is minimal at best. Some reported bugs simply never get routed to the proper group and the ones that do may not get prioritized and scheduled correctly. Worse, some bugs need collaboration from multiple groups to fix.

Beyond the communication issues lies process problems. Most companies incorrectly try to fix problems by working a new process into the existing organization rather than reorganizing the people to work around the new process. In Competing Against Time they introduce the concept of closed-loop teams. A closed-loop team includes everyone needed to handle a process. In banks instead of passing the loan application through different departments with different processes and lead times, one group is formed including credit analysis, asset appraisal, and senior bankers for fast response times. All the decisions for the approval of a loan can be done in one small effective group instead of three uncommunicative ones.

No where have I seen the concept of closed-loop teams applied better in the software industry than at National Instruments. So important is fixing customer issues fast that they staff actual engineers, called application engineers, on the front lines. When a customer calls in with an issue they won’t be speaking to a generic customer service representative but to an actual engineer who can usually solve their issue on the spot. Putting engineers on the front lines costs money but the fast customer turnaround creates loyalty and willingness to pay more for products.

As vertical software segments mature and competing software packages converge to the same features, competing on time may be the only way to differentiate yourself. Customers might not tell you they value time but make it a priority and you will be rewarded.

Walk into any nice golf course in Las Vegas and upon paying for your round you will be asked one question: whether you are a local resident. What seems like a simple question has great impact on how much you pay.

Most golf courses know that local golfers are a bunch of price sensitive whiners. Increase the price of a golf round too much and they will disappear. Decrease it and they will play more. You couldn’t pick a better demonstration of how prices influence demand. Although most golf courses have only local customers, Las Vegas courses have an additional visitor: the tourist.

While the locals are price sensitive the tourists are not. They have come long distances to play golf and asking them to pay more will not derail their objective. If the locals are a bunch of price sensitive whiners the golfing tourist is a price insensitive high roller.

Golf courses have realized this long ago. Why charge only one price? Unlike most things in life, they can have the best of both worlds by charging locals one price and tourists a much higher price. So different are the prices that tourists commonly pay hundreds of dollars more. Still, they enjoy their experience while golf courses enjoy the higher profits.

Golf courses may price discriminate against tourists but Amazon—in their early days—tried charging different prices for the same book. Through the use of internet browser cookies, they tracked how much you were willing to spend for a book. As good as the profits where for Amazon‘s bottom line, it turns out that while customers tolerate group pricing like senior or student discounts they don’t like hearing their neighbor paid less for the exact same book ordered on the same day. Amazon has since reversed course and charges the same amount for the same book—at least we think they do.

Coffee shops identify those that can afford to pay more not by charging different prices for the same cup of coffee but by charging different prices for nearly identical cups of coffee. It cost next to nothing to add whip cream or chocolate and those that are less price sensitive usually opt for the more luxurious cups. Coffee shops make more money by having a fuller menu than by having just one type of coffee with one price.

Like the coffee business, the tech business offers different tiers of products. IBM’s LasterWriter E—a low end laser printer—had the exact parts as the high end LaserWriter model except it had a special chip to slow it down intentionally. While nothing drives engineers more crazy than watering down an existing product, having two products prevents those that can afford to spend more from buying the cheaper version. By having two products, the rich will always buy the more expensive fully featured version. Short of asking their net worth, nothing is more effective at smoking out the rich.

Printers are just the beginning. Intel designs different speeds of processors from the same chip design. Microsoft offers its operating systems in home and professional versions. Much to their delight, most businesses buy the more expensive professional version.

From the food business to the tech business, the game is getting those that can afford to pay more to do so. Next time you see different priced items ask yourself, will I pay more? The businesses certainly are.

Ben and Jerry’s Ice Cream, better known for their hippy ways than their ability to win fights, unexpectedly found themselves in the ring with a heavy weight called Pillsbury. There was no bigger opponent.

Pillsbury in 1984 was a 4 billion dollar conglomerate who acquired Haagen-Dazs for 80 million in 1983 from the feisty founder Reuben Mattus. Haagen-Dazs could be found in every supermarket and controlled 20% of the ice cream market. Ben and Jerry’s in contrast had only conquered the small towns of Vermont, and had just recently launched into the Boston market.

Despite their small size, they caught the attention of the Haagen-Dazs executives when their products started showing up on the shelves of Boston’s supermarkets. Even more concerning to them was they were selling. Paul Tosi, who ran Paul’s Distributors and owned the Ben and Jerry’s distribution contract for Boston, was stunned in March 1984 when the inquires of Jim Richards, a Haagen-Dazs sales manager, turned into a nasty ultimatum. Stop distributing Ben and Jerry’s ice cream or lose the right to sell Haagen-Dazs.

The threat was real. Grocers usually only worked with one distributor for their frozen desserts to reduce traffic to their stores and billing headaches. Grocers wanted Haagen-Dazs and if Paul couldn’t supply it, they would work with another distributor. If Pillsbury and Haagen-Dazs intended on throwing the first punch, they wanted to make sure it landed.

Ben and Jerry, not known for their athletic prowess, felt it. Despite being better known for peace and love, they knew they could win a court battle but they also knew a long court battle would only favor Pillsbury. They understood that if they were in the courts, their products wouldn’t be on the shelves. While Pillsbury was focusing on the ensuing street fight, Ben and Jerry decided instead to focus on the minds of the customers.

In an ingenious maneuver—instead of playing defense like most companies—Ben and Jerry went on the offensive. They reasoned that if the customers of both Haagen-Dazs or Ben and Jerry’s knew what Pillsbury was doing, things would work in their favor. Pillsbury may own the courts but Ben and Jerry would own the media.

They came up with a campaign around the phrase: “What’s the Doughboy afraid of?” They issued a press release stating they were no threat to Pillsbury’s business. Jerry himself flew to Minneapolis and picketed in front of Pillsbury’s world head-quarters with a sign “What’s the Doughboy afraid of” and pamphlets explaining the story of Ben and Jerry’s ice cream. They said: “Do you think the Doughboy is afraid of two guys working with twenty-three people in four thousand square feet of rented space? Do you think the Doughboy is afraid he’s only going to make 185.3 million in profits this year instead of 185.4 million. Do you think that maybe the Doughboy is afraid of the American Dream? We only want to make our ice cream in Vermont and let the people of Boston make their choice in the supermarket. Next time you’re in your local market, pick up a pint of Ben and Jerry’s and give it a taste. Because that’s what the Doughboy is really afraid of.”

On the back of the flyer was a coupon for a write-in kit that included letters of protest to the FTC and William Spoor, the chairman of the board at Pillsbury. The persistent could also get a t-shirt. They not only attacked in Minneapolis but in Boston as well where billboards on highways, buses, and even airplanes made appearances.

Worse than the billboards and signs for Pillsbury was its effect on their public image. The media attention was trashing it. Pillsbury caved and the final agreement was signed on March 6, 1985, only a year after it started, removing the threat. What would have squashed any other company didn’t squash Ben and Jerry. By controlling the message not only did they save their company but they created publicity worth millions that typically takes companies years to establish.

If Ben and Jerry taught us anything, it’s that the real battle is not about the products but about how people perceive them. Ben and Jerry established Pillsbury as the big evil corporation trying to crush the little guy. They were in control of the message and winning the ground war.

Too many companies forget to play offense allowing others to control their message. When this happens they spend most of their time responding to the claims of their competitors towards them. Like an army under attack, they run around the field doing damage control. And like war, which is only won by attacking, businesses only win by attacking. If they don’t establish their message in the minds of their customers others will.

In 1878, Thomas Edison become interested in the light bulb. During the time open arch lights—strong electrical currents that jump the gap between two carbon rods—were starting to light the streets of France. The holy grail of lighting, however, remained the incandescent bulb containing a thin metal element encased in a glass bulb. After decades of experimentation by William E. Sawyer, Joseph W. Swan and others before them only poor results where obtained.

But Edison wasn’t like most inventors. He was no theorist instead favoring experiments to fancy equations. When a bright mathematician named Francis R. Upton came to work at Menlo Park to assist Edison in his light bulb adventure, Edison assigned him the task of calculating the volume of a pear shaped glass bulb. Upton carefully drew the exact shape on paper but before he could bust out his calculus skills, Edison became inpatient demanding an answer. To prove a point, Edison simply filled the bulb with water and poured the water into a measuring device to achieve the answer instantly.

Edison wasn’t always so brash but he demanded hard work. People who applied to work at Menlo Park always wanted to know two things: what was the pay and the hours. Edison's common response: “We don’t pay anything, and we work all the time.” Edison wasn’t joking. He worked around the clock taking only small cat naps. When others tired at the lab, he could be seen at his desk meticulously recording the results of his experiments.

And with that work ethic and Menlo Park's sizable lab, Edison publicly announced he would do what others couldn’t: make an incandescent bulb. Despite Edison’s track record of success, many in the academic community scoffed at his boastfulness considering the task too complicated for an experimenter with no formal education. Influential investors in New York felt otherwise and financed Edison creating the Edison Electric Light Company.

The investors made a good bet. No one in the world had more experience working with electric circuits than Edison. His knowledge convinced him of the importance of parallel circuits to subdivide the current to many bulbs but it was his realization that a high resistant lamp was needed instead of the low resistant lamp others were trying that pushed him on the right track.

But the real obstacle was finding an element that could glow brightly with enough resistance and wouldn’t burn out—something that eluded others for more than a generation. Carbon, Platinum, Boron, Chromium, and virtually every type of metal was tested with each substance carefully measured for endurance and resistance. He also experimented with different types of gases in the bulb all the while trying to get a better air tight vacuum. The dizzying amount of experimentations and failures would have overwhelmed others but still he plowed on.

Edison, who often said “If you come across anything you don’t thoroughly understand don’t rest until you run it down”, was running down many things he didn’t understand. But by taking careful measurements he was able to compare results which allowed him to eliminate possible elements. Platinum, for example, which others had experimented with extensively and brought some success in his experiments was abandoned.

During his experiments in late October 1879, Carbon was combined with tar and tried with surprising results. Later a cotton thread was carbonized and lasted 40 hours before burning out. On the right track, Edison tried other vegetable fiber materials like fishline, flax, cocoanut shell, hickory, twine, and cedar shavings. Nothing was left off limits. Even human hair was used. The winner was simple Bristol cardboard that lasted 170 hours. Eventually, however, bamboo was used in the bulbs and last 1200 hours. Edison would go on to light up entire neighborhoods in New York and the tiny Edison Electric Light Company would turn into General Electric.

What separated Edison from other inventors was not his genius smarts but his relentless dedication to his work. Edison himself understood this when he exclaimed: “The trouble with other inventors is that they try a few things and quit. I never quit until I get what I want!” Not only did he work hard but he was continually learning. When he wasn’t experimenting, he had his head buried in a book or technical journal.

In his book Outliers, Malcolm Gladwell argues that outliers in a field are not geniuses but rather overly dedicated individuals. Few would argue that Edison wasn’t an outlier but Gladwell would just say he was more dedicated. Edison, who made his own telegraph at age 15, was simply able to get more practice in before others his age.

Gladwell in his book introduces the notion that ten thousand hours of practice is needed before anyone can become great at a profession. He sites a study of young violinists all starting out at the age of 5. He begins:

“In those first few years, everyone practiced roughly the same amount, about two or three hours a week. But when the students were around the age of eight, real differences started to emerge. The students who would end up the best in their class began to practice more than everyone else: six hours a week by age nine, eight hours a week by age twelve, sixteen hours a week by age fourteen, and up and up, until by the age of twenty they were practicing—that is, purposefully and single-mindedly playing their instruments with the intent to get better—well over thirty hours a week. In fact, by the age of twenty, the elite performers had each totaled ten thousand hours of practice.”

He continues:

“The striking thing about Ericsson’s study is that he and his colleagues couldn’t find any 'naturals', musicians who floated effortlessly to the top while practicing a fraction of the time their peers did. Nor could they find any people who worked harder than everyone else, yet just didn’t have what it takes to break the top ranks. Their research suggests that once a musician has enough ability to get into a top music school, the thing that distinguishes one performer from another is how hard he or she works.”

Too many people think that those at the top are naturally gifted and talent alone got them there. Gladwell’s insight not only dents that thinking, it squashes it altogether. Anyone, no matter how stupid they may seem, can become an outlier if they work hard.

When I started my first job, I worked under Elton Wells an amazing Software Engineer. He was shockingly productive, thought of good designs, squashed bugs like an overzealous exterminator, and flat out got things done. Even more important than any natural talent Elton possessed was he simply out worked you.

You see most Software Engineers go to work do their thing for the minimum eight hours and then go home and do nothing more. The more I studied Elton, the more I came to understand he didn’t do that. He read MSDN articles, technical books, and coded on the weekends. While other programmers were trying to avoid anything related to their profession after hours, he was actively strumming his violin.

Through Elton, I learned that successful programmers continually learn and practice just like Edison. As I have progressed through the software world, that has become more apparent. Practice is not something you do once you become great but rather practice is something you do to become great. After all it was Edison who first said: “Genius is 99 percent perspiration and one percent inspiration.”

Jones Beach, just 25 miles from Times Square on the southern end of Long Island, serves 6 million visitors a year. So beautiful are the white sands that on a summer day one might think they are closer to Jamaica than New York City. While considered one of the best beaches in the world, it was only sunken marsh land in 1922 when Robert Moses first laid eyes on the property.

The township of Hempstead would have liked to keep it that way. They were intent on pulling oysters from the bottom of the bay rather than drunken New Yorkers. As Parks Commissioner, Moses could buy, condemn or appropriate land owned by private individuals but land owned by townships could only be acquired through town votes. The votes were not forthcoming.

Part of the reason was Moses himself. He would not reveal the actual route of his Meadowbrook Causeway upon which thousands of vehicles would travel. Knowledge of future road plans were greatly coveted because, if known before hand, land could be acquired at pre-parkway prices. No doubt as demand for stores and homes close to the parkway increased, land prices would follow shortly behind.

If buying land was coveted, building contracts were more so and no one loved them more than Nassau County GOP leader G. Wilbur Doughty. A corrupt old school boss with close connections to construction companies in the county, he ran a tight ship. Many citizens owed their livelihood to the construction jobs those companies produced. What Doughty said was law and Doughty wanted nothing to do with the plans of Moses.

The idealist Robert Moses fresh out of Yale and Oxford scoffed at back door dealing. Even the Robert Moses of 1925 tried many times to persuade Hempstead Township to give him Jones Beach. Doughty even went so far as to place the proposition on the ballot in November 1925 but the vote was 12,106 to 4,200 against.

In those dark hours, Moses thought he had lost Jones beach and with it his elaborate park plans for Long Island. Whether someone convinced him to deal or he thought of it himself we don’t know, but later in the year we find Moses in private meetings with Doughty. Of course, no contract or proof of any deal surfaced but certain things were agreed to. Among them were contracts for the construction companies connected to Doughty along with Hempstead Township having full access to the plans for the Meadowbrook Causeway.

In November 1926 the proposal passed 18,872 to 5,076 in favor. The park would be built after all and Doughty would be the first to spread his beach towel upon the white sand.

Lucky for Moses he learned to compromise early in his career. He went on to build many of the bridges and roads in New York City. For a man with strong opinions, Moses knew when to compromise in order to get things done. No doubt, the lessons of Jones Beach often played out in his mind.

The lessons of compromise are not limited to beaches. In the book Making it Big in Software it talks about the idea of an emotional cache. An emotional cache is the amount people feel invested in your ideas. Including ideas of others in architecture decisions or proposals while acknowledging their contributions helps build support. The further you advance in your career the more your success depends on the ability of the team to get things done.

Unfortunately many programmers, to their detriment, are often uncompromising on technical matters all the while alienating teammates. It is often said that what matters is not how much you know but how much you do. Individually, you can only accomplish so much. The great leaders who get things done know how to give in order to get.

For all the theories explaining the stock market not one has gained more traction than the Efficient Market Hypothesis. Coming to prominence in the 1960s the theory states that given many intelligent individuals with equal access to information, stocks that are underpriced will be bought and stocks that are overpriced will be sold thus creating accurately priced stocks.

Nothing sums up the theory better than a conversation between a professor and his student. Upon the student noticing a ten dollar bill on the ground the professor quickly denied it stating it was impossible because if it were really a ten dollar bill someone would have picked it up already.

But if the group philosophy helps the theory it also hurts it. The thing with groups is they sometimes behave like herds and go trampling off in the weeds. No one can say that markets had assets priced correctly during the end of the 1920s or during the top of the dot com boom. Yahoo, for example, was priced at $237 on January 2000 but only $11 dollars a little over a year later.

And it wasn’t like Yahoo’s earnings changed drastically during that time. Instead, the herd mentality took over and a bubble formed. Eventually someone noticed they couldn’t see out of the head tall weeds and the sell off began. If markets were run by robots, the Efficient Market Hypothesis would be well efficient.

But markets are not run by machines but humans complete with greed, pride, and the competitiveness to beat their neighbor into submission. It can be shown that markets are rarely priced perfectly but instead follow a pattern similar to a clock pendulum in that most of the time they are either to the left or right of the correct price. It is like listening to your sister recall her last date. Either things are going really well or they are going terribly. People tend to think the same way about stocks.

Boone T. Pickens was never one to follow the crowd. Having successfully created Mesa Petroleum and then taking it public, he became fascinated with the relationship between oil company’s stock prices and the underlying values of the oil and gas reserves it owned. In 1969, his mind perked when he noticed that Hugoton Production—a larger company with extensive gas reserves in southwestern Kansas—had a stock price much lower than what the gas reserves it owned would fetch if sold. He launched a hostile takeover of Hugoton and merged it into Mesa creating an even more powerful independent oil company.

Pickens didn’t forget about Hugoton. In the early 1980s, he again noticed that oil company validations were not priced correctly. He moved quickly targeting Cities Services which was three times the size of his company but its stock was priced at only a third of what its oil and gas reserves where worth. Mesa acquired blocks of stock in Cities Services and though it lost the battle for the company it made a 30 million dollar profit on its shares. Further profits of 500 million were made in much the same way with Gulf Oil in 1983 when their stock went from $41 a share to $80 a share.

If there is one thing Pickens proved it’s that real money can be made by those who take undervalued assets and make them valuable. And like undervalued companies, big money can be made by employees who take undervalued projects and make them successful.

In sports, for example, coaches rise to fame not because they took a winning team and kept them winning but because they took a bad team and changed them into a winning team. There is very little prestige initially coaching for a bad team but turn it around and people notice. Make no mistake, the quickest career advancements happen because of the ability to turn undervalued assets into valued ones.

People in their career tend to make the same mistakes as people who buy stocks that are pitched by CNBC. They buy the stocks expecting big returns but by the time the market has realized the true value of the company the big returns have already materialized. The people who invested before CNBC made the stock newsworthy are the ones truly rewarded.

Likewise, joining a company that has already hit it big greatly decreases your potential to hit it big. People who join these companies expecting to make big gains will be disappointed. The hard work of turning the undervalued products into valued products has already been done and the employees who did the work have already been rewarded.

Let’s be clear, there is nothing wrong with joining a successful company. It is just like investing in index funds. With index funds, you are guaranteed to get an average return. Just don’t expect to beat the market.

When Alexander Fleming returned to work from an August vacation in 1928, he began the monotonous task of going through his staphylococci cultures—bacteria that cause boils, sore throats, and abscesses—housed in Petri dishes. While examining each dish he noticed something unusual. One culture was contaminated with a mold, which by itself isn’t that unusual, but around this mold the colonies of staphylococci that surrounded it were killed.

Realizing the unique properties of the mold, he spent weeks growing mold trying to determine the type. After consulting with a Mycologist, they determined the type to be a rare strain of Penicillium notatum. Fleming named it penicillin.

The interesting part of the story is not the discovery of the disease killing mold but Fleming’s curiosity and desire to learn more. Of all the qualities a scientists needs, nothing is more critical than curiosity peppered with desire. Other scientists might have seen the same mold phenomena but their lack of curiosity prevented them from moving forward.

Fresh out of college, I started my career at National Instruments. Among all the employees that walked through the doors of the Mopac C building, none had a greater influence on the software direction of the company than Jeff Kodosky. He invented the software language LabVIEW that powers everything National Instruments does.

I sat close to him for a couple of months and among his duties was to take pictures with campus visitors. When visitors weren’t clambering to see him, he would take trips to various places. After he returned, he often sent an email attaching the photos from his latest adventure. More interesting than the pictures, however, was the quote seen at the bottom of all his emails. It said something to the effect of: “Great discoveries are started by saying ‘that is weird'.”

What Kodosky understands is that every discovery is made by first observing something acting differently than what we expect. More importantly, he understands that programmers make discoveries because the software doesn’t act correctly. Programmers are computer scientists after all, and the effective ones are curious about odd behavior. Kodosky expects this quality from the hundreds of programmers hired to work under him.

Great programmers are like Fleming in that when they see a program behave in ways they don’t expect the curiosity overwhelms them and they can’t help but relentlessly track down why that is. In the process they kill bugs that plague the product and produce higher quality features. Make no mistake, these types are the ones that keep the toilets flushing and are the foundation for stable products.

As important as these types are, the real problems occur from programmers who overlook the mold in the Petri dish rationalizing odd behavior away as chance occurrences not worth their time. As behavior is overlooked, the defects sink deep into the software like two week old Chinese food making its way through your digestive tract. Every good software manager knows the longer a bug remains in the system the harder and more costly it is to fix.

Even more frustrating is trying to tell what type a potential hire is during an interview. Like a heard of wilder beasts, programmers all look the same but after a couple of chases by hungry lions the weak ones become apparent. But unlike wilder beasts, where strength and skill are paramount and curiosity gets you eaten, curiosity must be one of the top things you look for. New frameworks and languages can be taught. Curiosity and drive cannot.

Chef Gordon Ramsay has a show called Kitchen Nightmares where he attempts to turn around troubled restaurants. In each episode, Ramsay begins by sampling a lunch service which usually ends with him masterfully controlling his gag reflex.

More troubling than the food—Ramsay finds—is the attitude of the owner toward it. In all cases, the owners stubbornly defend their food quality arguing its merit. In efforts to prove them wrong, Ramsay can be seen making a break for the kitchen finding rotten food in the fridge, old frozen dishes, and if he is lucky dead mice.

Worse than the dead mice is the owner’s dead leadership. Many have bad chefs and fail to hold them accountable. Worse—in an effort not to make waves—the owner rationalizes their performance encouraging slothfulness all around him.

Nothing does more harm to an organization than failure to hold poor performers accountable. Wimpy leaders, rather than confront bad performers, lower the bar allowing the limp to clear it. But far more damaging than the lowered bar is its effect on high performers. Quality performers want to work with quality performers but when accountability is lacking they get discouraged and will either leave or stop performing themselves.

Ramsay knows effective leaders don’t lower the bar but expect all to rise to it. Quality restaurants have high standards which employees are held to. Strong leaders spend more time rewarding effective employees and less time dragging along the bad ones.

Ramsay can be accused of many things but no one is better at holding people accountable. Far from the frozen dishes and rotten food that plague many restaurants, his restaurants have a reputation of excellence that any master chef would jump to work for. By demanding high standards and taking quick action when those standards are not met, he has created an organization of high performers and quality food.

Just like restaurants depend on good chefs to make quality food, software companies depend on good programmers to make quality software. Unfortunately, the same problems of leadership accountability that effect restaurants haunt software shops. Littered around the information highway are failing software companies that have created a culture of mediocrity.

And like frozen food that is heated before service, customers can tell. Although customers don’t eat the software, they know when it causes a bad taste in their mouth.

When the incumbent Mayor of New York John Mitchel lost the election in 1917 by a wide margin nobody was more shocked than Robert Moses. For the last 4 years Moses toiled on the mayor’s Municipal Civil Service Commission working on a report recommending salary and performance reforms for public city employees. Now Moses was out of a job.

Despite these bad circumstances he had a lot going for him. He worked hard and was passionate about civil service. His two years at Oxford convinced him of the superior efficiency of the British civil service system. His thesis: The Civil Service of Great Britain was widely respected.

But if the civil servants were hard working and efficient in England they were notoriously corrupt and slow in New York. Instead of focusing on better ways to do their jobs most spent their energy buttering up the Tammany political machine. After returning from England—the taste of Yorkshire Pudding still clinging to his tongue—he applied to The Bureau of Municipal Research.

Of all the places an aspiring government reformer could begin his career in the early 1900’s none was better than The Bureau of Municipal Research. Founded in New York by William Allen, Henry Bruere, and Frederick Cleveland in 1907, the Bureau focused on eliminating government corruption through hard core journalism. If the citizens knew how much the government was paying for bags of cement, for example, they would be more inclined to induce change.

The Bureau and Moses’ mission couldn’t have been better aligned. At first, Moses took to the work. He researched, wrote memos, but before long the slow pace began to nag at him. But it wasn’t only the slow pace; Moses was beginning to realize that the only way to get things done was not to publish research reports but to be in government itself.

His chance came with the election of John Mitchel to Mayor of New York. Mitchel, a lawyer, had connections to the Bureau of Research where he assisted in the investigation of well known corrupt Manhattan Borough President John Ahearn. Mitchel, running on the progressive platform, was intent on replacing the existing Municipal Civil Service Commission which had a reputation for handouts and issuing inflated salaries. Moses—well known at the Bureau—landed a job on the new Civil Service Commission.

From the beginning at the commission, Moses was intent on grading performance of all public employees from the executive to the custodian. He came up with a complex grading process whose numbers were plugged into an algebraic equation giving each employee an overall grade. Moses not only pushed his performance scale but also salary cuts in a number of divisions. Moses was confident in his plan but what he didn’t understand at the time was that the 50,000 public employees, who owed their paychecks to the Tammany machine, were not interested in limited salaries. Although Moses plan may have been misguided, his bosses at the Service Commission applauded his work.

But what happened next in Moses’ career would launch him on a path of successes. After a year of working humbling jobs after Mayor Mitchel’s loss, he got a call from Henry Moskowitz’s wife Belle. Henry Moskowitz had been the chairman of the Municipal Civil Service Commission and worked closely with Moses. No doubt he mentioned the quality of Moses’ work to his wife for she was now a close advisor to Alfred Smith—the new Governor of New York. Smith, a Tammany man, wanted to reorganize the state’s social welfare programs and turned to Belle to make it happen. Moses, Belle thought, would be perfect.

But the most impressive thing about Robert Moses is not that he went on to build much of New York’s roads and parks but how effortlessly he secured future work in those early years. Unlike bad rumors which spread through the streets of New York on a daily basis, the rumors spread about Moses were good ones. While many of his peers were looking for work, Moses’ past bosses were busy securing his work.

It is easy to dismiss this story as not relevant to software developers today but to do so is a serious mistake. Every day you show up to work is a day your reputation is defined. Like the government sector in Moses’ time, the software industry is closely connected.
 
Showing up to work you can fake, passion you can’t.

Pilots pay attention to lots of things but nothing focuses their attention more than an airplane fresh out of maintenance. A Boeing 747, for example, has 6 million parts including linkages, fuel lines, bolts, wires and countless gages. Even the most safety conscientious mechanic can forget to tighten a bolt or secure a fuel line. At no time during the lifetime of an aircraft is it more vulnerable than in the safety of a hanger under the comforting sounds of hammers, wrenches, and the occasional cussword.

Take an Airbus A330 departing Vancouver International Airport on November 6, 2003. Shortly after takeoff, the control tower informed the pilots that smoke and vapor were pouring out of the #2 engine. Although the pilots made a safe landing after shutting down the engine, further investigation revealed why the airplane looked more like Superman streaking across the sky while taking an in-flight bathroom break than a normal Airbus flight. Mechanics had unnecessarily removed the low pressure fuel line while fixing another issue and failed to reconnect it back properly.

If the most well intentioned mechanic can cripple an airplane, bug fixing programmers can cripple a program. And just like mechanics who make mistakes on big complex airplanes and little airplanes alike, all but the smallest trivial programs are targets for mistakes. But unlike airplane maintenance, software maintenance has a little dirty secret: fixing one software bug somehow creates two more.

And if fixing one bug creates two bugs, it is easy to see that no piece of software is bug free. Most complex software has outstanding bug counts measured not in hundreds but thousands. Despite escalading bug counts, software gets away with it because unlike airplanes—where everything must be functioning perfectly or disaster will strike—all but the most mission critical software can function with bugs because most are either never hit or infrequently hit.

With large bug counts, deciding what to fix is non trivial. To mange the bugs, it helps to rank each one according to two variables. The first—perhaps the easiest—is how many customers are affected? If usage stats are gathered this becomes easy; if not more subjective. The second is how bad is it. Does it crash the program every time or is it a misspelled word at the bottom of the page in 2 point font that no one reads?

The graph below shows an example. Here we have plotted the bugs with the percentage of customers affected on the horizontal axis and the severity on the vertical axis. As we move to the right, bugs become more visible while towards the top more severe.

Looking at the graph above, we might expect to fix all bugs above and to the right of a imaginary line running from the top left to the bottom right. But in reality, the bug fixing boundary follows more of a y = 1/x curve than a linear one as shown in the graph below. This is because the less severe and less visible bugs congregate towards the left and bottom edges of the graph. All other bugs are either severe enough or visible enough to be fixed first.

At first glance of the graphs above, it might appear that only excluding bugs that are close to the left and bottom edges will result in most bugs being fixed. But like middle school students at their first dance, a surprising number of bugs cling to the edges of the graph. This can be attributed to internal opinions or small one off requests from individual customers that produces bugs that turn out to be either not important or visible enough.

But it’s a third parameter not the other two which is often overlooked: the risk of making a fix. How complicated is it and will it break other things resulting in more than two new bugs? Imagine this parameter being the z index of the graphs above springing out of the page towards you.

With this system, you don’t have to be a wizard to determine what to fix and what not to.