Something recently reminded me of this article I wrote back in September 1994. For you historians, that would have been the month that the Intuit purchase of Parsons technology was finalized. It was revised in 2001. So if any of it sounds dated, it’s because it is.
I assume this would’ve appeared on my original blog back before they were called blogs, but that sounds too early. So I’m not sure why I wrote this.
Management Principles
September 24, 1994
I’ve always been struck by the fact that great people seem to operate on a small set of principles which they consider inviolable. These principles, right or wrong, guide them in all of their decisions. Their decisions, though sometimes wrong because of the wrongness of their principles, are quickly made and internally consistent. Collectively they form the “conscience” or “personality” of the organizations these people inhabit.
I believe it was Ross Perot (or perhaps it was Tom Watson — it was one of those IBM guys) whose management style was described by an employee as “Management by Hanging a Sign.” Mr. Perot would hang signs around the office declaring some axiom of life as only Mr. Perot could have imagined it, and he expected the employees to follow.
Within the religious community, great men of faith are often known by their basic principles. A church I used to go to counted Dr. Bob Jones Sr. as one of their heroes. I can’t count the number of “as Dr. Bob always said…” illustrations I’ve heard in sermons.
And then there’s the Ferengi Rules of Acquisition.
You get the point. I’m not claiming that what follows is anything special, or that I’m the next Bob Jones, Ross Perot, or Grand Nagus Gint. But it seems that there is some value in setting these things down and trying to follow them.
While these principles are not strictly taken from the Bible, I hope they’re at least biblical. I’m a little concerned about “It’s easier to get forgiveness than permission” though I would contend that statement is at least true from a biblical point of view. It may not be the best policy, but it’s true.
This is just a list of those aphorisms I’ve found that I live by in business. Names in parenthesis are the source when I remember where these came from. While the ones I’m claiming as mine are things I really think I came up with on my own, many are pretty simple, and they’re undoubtedly based on principles I’ve heard or read about. When I’m pretty sure I read it somewhere but can’t remember where, I’ve left it unattributed.
Decision Making
It’s better to make a decision than to make a right decision.
Making a decision permits you to move forward. Waiting for all possible information to come in so that the right decision can be made often results in no decisions happening for a long, long time.
This is a good principle to teach to those who report to you, and for you to keep in mind as you report to your boss. A manager doesn’t have time to consider the implications of every decision that has to be made in the organization. That’s why he or she hires people. Those people need to be willing to make decisions and not continuously bring them to the boss. The quickest way to prove you’re dispensable is to be good at gathering facts but slow at applying those facts to daily decisions without the boss’s OK.
Our failures form the basis for our success.
(Russ Novotny)
There’s no such thing as bad decisions and failures. All “wrong paths” taken in the course of getting something done contribute to the overall success of the project. If you hadn’t gone down that “wrong” path, you wouldn’t have learned all those lessons.
Programs which have suffered no false starts nor setbacks tend to be fragile. Making a mistake helps us to identify where the weaknesses are in a plan or design. Once we’ve seen how we can fail, we can take steps to make sure we don’t fall into the same trap again. Not only will the current project benefit, but future ones will as well when we have learned from experience what to avoid.
It’s easier to get forgiveness than permission.
This is an old standby. It basically says that when you’ve got something to do that is risky, it’s better to do it (perhaps hoping you won’t get caught) than to try to get permission. Getting permission involves someone else (usually higher up in the organization) in the risks. It’s likely they will be reluctant to go along, and you won’t be able to get permission to do something you know needs to be done. If you simply do it, you might not get caught. If you do get caught, and the project was even somewhat successful, you come off well by looking like a person willing to take risks to do the right thing. If you get caught and the project failed you’ll probably get by with a reprimand. Worst case is you’ll get a chance for a change of scenery.
Note that this isn’t a license to do all manner of evil. But rather, it’s a way of saying that sometimes you just have to do what’s right even though nobody would agree with you if you tried to explain it to them.
People Management
The three-sided manager: Planner/Cheerleader, Bridge-builder, Wall-builder.
(Craig)
I’ve come to believe that managers (at least of software projects — you’ll have to decide if this applies to you) have three roles in their organizations:
- As a planner/cheerleader you’re responsible for looking to the future, bringing in new work, getting people excited about what’s coming up, and keeping them excited about what they’re doing.
- As a bridge-builder you’re responsible for connecting your people to the resources they need to get their jobs done — whether it’s equipment, furniture, training, assistance from others, books, or whatever.
- As a wall-builder you’re responsible for protecting your people from interference from upper management, policy police, fellow employees, etc.
Interestingly enough, none of these roles is the traditional authoritarian “boss”. People who spend their time telling their people what to do are limiting the possibilities of what their group can do. By making all the decisions for them and enforcing your will on your employees you destroy morale and run the risk of making poor choices because you haven’t considered the opinions of the people who really know the job — your employees.
Deal with problems as they’re encountered.
(Craig)
We all prefer to ignore problems, especially people problems. As a manager I’ve had to deal with an incredible range of really nasty situations, from people using their business computers for “inappropriate Web browsing” to people I had to recommend for counseling. It would have been much easier to ignore these problems and hope that they went away. But doing so would only let the problem get worse.
A simple application of this principle is this: There should be no surprises at an employee’s annual performance review. An employee shouldn’t learn of significant deficiencies for the first time at his or her review. If it’s serious enough to merit mention in the review, then it must be important enough that you should have dealt with it when it happened.
Any eight hours will do. Any forty hours will do.
(Craig)
If at all possible, implement “flextime” within your organization. Flextime says that an employee can come in at anytime in the morning and leave at anytime in the afternoon as long as they’ve put in their eight hours. It permits an employee to take a half-hour lunch or a one-hour lunch. It should also permit occasional afternoons off if the time can be made up within a reasonable period of time.
This is especially important in engineering and creative disciplines. (It’s not very functional in sales or customer service organizations.) Flextime recognizes the employee’s ability to manage himself. It creates a very stimulating environment in which people will be free to leave the facilities even, if that’s what it takes to do their job. It’s not unusual in my group for a couple guys to head out to the software store to see what the competition is up to or to evaluate a new idea by looking at what’s already available.
Flextime also permits an employee to arrange his family time and work time in a way that meets his needs. Some employees like to spend time in the morning with their kids. Others prefer to get home right away so they can pick up kids from school or get a head start on the evening’s activities. By implementing flextime, you earn the respect and appreciation not only of the employee, but of his or her family.
Vacation time is a right, not a privilege.
(Neil Ennis)
Many employers make vacation time something you have to beg for. As long as you’re not in an organization that requires a certain staffing level each day (telemarketing, restaurants, etc.) you should always honor every vacation request. The employee has earned it.
Many managers fear employees will take time off right at the critical phase of a project. But no responsible employee will ask for vacation at a bad time. Even if they do it’s better to let them take it. Adding to the stress at work by adding stress at home isn’t going to get the project out the door any faster.
Promote employee ownership of the product.
(Dave Theilen)
Tell them everything we know about the competitor and the marketplace, let them make decisions about implementation, and believe their schedules. As a result, they turn out a better product and do it on time without forced overtime.
Let them work it out
Take the time to help an employee who is having difficulty with an assignment to get the resources he needs to accomplish the task. Let people make mistakes, then help them figure out what they did wrong and how to avoid the mistake in the future. You don’t need to give them the answers. They can figure it out if you give them the opportunity.
Don’t answer the phone when you’re in a meeting with an employee.
(DeMarco and Lister?)
There’s no better way to tell someone you don’t care than to take a phone call in the middle of a meeting with them. This goes for retail sales people who answer the phone while you’re checking out, too. Just ignore the phone or forward it to voice mail. If it’s important they’ll call back.
Project Management
Your employees know more than you do.
Let them do their jobs.
There’s no such thing as overtime.
(DeMarco and Lister)
Overtime hours are less productive. Employees make them up after the project is over through decreased enthusiasm and burnout. Overtime robs employees of their families and makes you the enemy.
Our programmers are our first customers.
(Craig)
Our programmers should be familiar enough with the application domain that they’re using our products to solve their own problems in that area. As a result, they know what users want because they are themselves users.
WDYR? (What do you recommend?)
(Gus Pagonis)
Don’t tell me about a problem unless you have a solution. Similarly, don’t tell your boss about a problem unless you can offer a solution.
Software Development
This program works.
(Craig)
Programs should be implemented incrementally. Each increment should be proven to be correct. At every stage, the program should work.
Debuggers don’t.
(Craig)
Debuggers give you a micro-view into your code. Most coding problems are logic problems which are macro-problems. If you think about your code while you’re implementing you don’t need a debugger.
There’s no such thing as too many comments.
(Craig)
The quality of internal documentation is the mark of a great programmer.
There’s no such thing as reusable software.
(Craig)
It is possible to create small utility functions that can be reused in many projects, even over a long period of time. I had a collection of BIOS display functions that I used for about six years when I was writing DOS programs. It was a very simple set of functions but it performed some very useful tasks and saved me a lot of time.
But when people refer to software reusability they usually mean creating large scale libraries of functions or classes which can be used on many projects. They seek to standardize on these common functions with the idea that once this code is written they’ll never need to do it again.
I’ve found that none of these efforts produce anything that’s usable for longer than a year. Technology changes, and the solutions that seemed right a year ago are useless now. Instead of trying to constantly position for the future, it’s best to follow the next axiom.
Implement today’s programs with today’s tools to today’s standards.
(Craig)
It seems there’s always some Great New Thing coming out in three to six months that’s going to revolutionize the way you write software. But three to six months from now there will be some other Great New Thing coming three to six months from then.
I’ve found that it’s best to use the most promising tools available today to solve today’s problems. Half-way through the project you’ll be tempted to change your implementation plan to take advantage of some new technology. It’s almost always better to continue with your original plan. Let someone else debug this new stuff while you get your product out the door. Then come back and take advantage of the new stuff on version 2.0.
Working with Other Companies
Ask for more than you need.
Buy yourself some negotiating room. You can always give some back. Applies to a lot of areas; similar to “under-promise and over-deliver”.
Use voice mail.
Leave detailed messages. Don’t just say “call me back.” If you leave a message it should convey new information.
Let 1000 flowers bloom.
(Guy Kawasaki)
The little guy you’re not interested in today could save your company tomorrow. Don’t burn your bridges. Leave your options open.
Customer Service
Under-promise and over-deliver.
(Guy Kawasaki)
Don’t make a promise you can’t keep. In fact, if you make a promise, keep it sooner and better than you promised.
Hiring
Interviews take at least four hours and often eight.
(Craig)
It takes about an hour to and hour and a half to get to know the candidate’s capabilities. It takes a half hour to an hour to tell about our company. The rest of the time is for the candidate to ask questions.
Hiring a new employee is a bigger decision for him or her than it is for you.
(Craig)
You’re asking the applicant to completely change his life. All you’re doing is choosing one person from a list of several potential candidates. As a result, give the candidate adequate time to become familiar with you, the job, the company and the city.
Let the candidate talk to his potential co-workers.
Involve other members of the team in the interview process. Then believe what they tell you about their impressions.
Life Principles
Think about it overnight.
(Bob Parsons)
Before sending a scathing e-mail or letter, let it sit overnight at least. The recipient will still be there tomorrow. I wish I could remember to do this more often.
Pray for your competitors.
(The Bible)
For several years I used a competitor’s mouse pad as a constant reminder to pray for them. I’ve never felt bad about steering a customer to a competitor’s product when they have something we don’t. I think in the long run it builds goodwill with customers and your industry peers.
Say you’re sorry.
(The Bible)
Admit your mistakes. It disarms people.
Send your wife flowers half way through a week-long trip.
(Craig)
I reserve this for really long trips but it really helps.
Always be reading a good book.
(Kurt Hansen)
It keeps your brain from decomposing.
Read about people whose lives you admire and want to emulate.
(Rush Limbaugh)
Have you noticed that squirrels seem to never learn from their friends’ experiences while crossing the road? Humans have the benefit of being able to learn from other peoples’ mistakes so they won’t have to make them themselves. They also learn from others’ successes. Don’t be a squirrel: Read a book.
Tip well.
(Craig)
Be loyal to businesses that treat you well. They’ll treat you better each time you come in.
Keep a to-do list.
Craig: Fill in this paragraph.
If you didn’t write it down, it didn’t happen.
(Cliff Stoll)
I’ve always wanted to keep a diary, but I’ve never managed to muster the discipline. But at work I try to keep a phone log and a brief diary. It makes it a lot easier to remember what happened when and to remind you of your accomplishments.
During the development of one of our software products we wanted to make use of some data, the copyright status of which was in question. I contacted the owner and was told that I didn’t need to ask their permission to use this data — my usage was not inflicting on their copyrights or trademarks. I made a note in my log of the date and time, along with the name of the person I talked to.
About four years later, the president of the company which owned the copyright on this data contacted me and wanted to collect back-royalties on our usage of the data. When I told him I had verbal permission from one of his employees, he didn’t believe me. But when I produced a copy of my dated phone log, there wasn’t much he could say. While he probably could have challenged us in court, he chose to honor his employee’s decision and let us off (he would have lost in court anyway).
If you have a legitimate complaint, mention it when you check out.
(Craig)
Mention one or two things you had trouble with every time you check out of a hotel. More often than not you’ll get a free night.
People don’t expect enough from hotels. Remember that the purpose of a hotel is to provide a comfortable place for people to sleep. Anything they do that undermines that purpose shows that they don’t understand the needs of their customers.
A classic mistake seems to be renting out large public areas for private events. We were staying in one of those large hotels with the big central atrium one time when they had scheduled a rock concert in the atrium. While it was great fun for those who attended, it wasn’t much fun for our family. On another occasion, we discovered the hotel rented its pool to a water aerobics group each evening during prime guest pool usage time. Needless to say, I mentioned both of these incidents to the respective managers and received credit for at least one night’s stay in each case.
Always leave the bathroom cleaner than when you entered it.
(Craig)
While the principle applies well at home, it’s also very suitable for public restrooms. Imagine if everyone just cleaned up after themselves in men’s rooms. The world would be a nicer place.
Make sure the slots in the screw heads line up in the same direction.
(Dad)
In other words, pay attention to the details. Or, “a job worth doing is worth doing well.”
A few years ago I was mounting an outlet cover plate on the wall after doing some wallpapering. I noticed that I subconsciously was making sure the slots in the screws lined up in the same direction (in this case, up and down — not sideways). I realized that I had picked this up from my dad years before. Then just a month ago I noticed that the electricians doing some remodeling for me did the same. I was pretty impressed there for a while… until I stepped back and realized they had mounted the outlet box at a slight angle.
I guess some people can learn principles but not quite fully apply them. It’s nice to have those screw heads straight, but I’d rather have the outlets square with the rest of the world.
Essential Reading
Certain books have been critical to shaping the way I think about management and product development. I consider these to be important to people managing groups of engineers or people involved in creative endeavors.
Zapp! The Lightning of Empowerment, Byham
Peopleware, DeMarco and Lister, Dorset House
The Bible, Various
The Macintosh Way, Kawasaki
“Contributors”
Dad — The best thing my dad ever did for me as far as preparing me for doing things myself was help me build a Soapbox Derby car. In the course of that project I learned a lot about doing a job and doing it right.
DeMarco, Tom and Lister, Tim — Met one of these guys at a Software Developers conference. Authors of Peopleware, one of the definitive works as it relates to my style of management.
Ennis, Neil — My supervisor at Rockwell. My first boss. An incredible role model and probably the source of more of this material than I realize.
Hansen, Kurt — My first close co-worker at Parsons. Rejuvenated my love of reading though he probably doesn’t realize it. Kurt once read through the first seven volumes of the encyclopedia. (OK, it would be more impressive if he had read the whole thing, but… seven volumes! That’s pretty good.)
Limbaugh, Rush — Radio host, author, consummate conservative.
Kawasaki, Guy — Author of The Macintosh Way and Selling the Dream. Offers a fascinating view of the computer industry and software development.
Novotny, Russ — Team leader on my first project at Rockwell.
Pagonis, Gus — Author of Moving Mountains. Managed logistics during the Gulf War.
Parsons, Bob — Founder of Parsons Technology. Madman and mentor. Probably also contributed to more of these aphorisms than I’ve acknowledged.
Stoll, Cliff — Author of Cuckoo’s Egg. Astrophysicist and techno-philosopher.
Theilen, Dave — Author of No Bugs. Saw him speak at a Software Developers conference. First software manager I heard who thought the same way I did about managing programmers. Again, many of the ideas I claim as my own are probably Dave’s, though I think I had many of my concepts nailed down before I first started reading his material.
I recently ran into a situation where I wanted to significantly customize the behavior of a scrollbar in a Cocoa 
Java, C++, Objective-C, and C# all use braces ( 