Bolg - The Chris Blanc Weblog

Microsoft’s Bill Gates spills some new strategy:

We’re hard at work, I would say, on the next version, which we call Windows 7. I’m very excited about the work being done there. The ability to be lower power, take less memory, be more efficient, and have lots more connections up to the mobile phone, so those scenarios connect up well to make it a great platform for the best gaming that can be done, to connect up to the thing being done out on the Internet, so that, for example, if you have two personal computers, that your files automatically are synchronized between them, and so you don’t have a lot of work to move that data back and forth. ^{^}

And:

Microsoft is set to announce Tuesday that it is launching a “public preview” program for two server products based on its Windows Server 2008 operating system.

The products, one aimed at small business and the other at midsize firms, combine the server operating system with Exchange Server and other software into a bundle designed to cost less and be easier to install than acquiring the products separately. ^{^}

Microsoft is going to push Server 2008 instead of Vista toward business, while refining Windows 7 as a dual attack: sharing of data across multiple devices, not SaaS, and making a light and fast operating system because, among other reasons, portable devices like the Asus Eee PC are redefining how we use computers, watch television, and communicate.

They are listening to their customers in Redmond, and have come up with a mature strategy.

Many managers observe that some tasks happen quickly and effectively, while others (usually the least exciting) drag on into infinity. What’s most interesting is that these tasks are approached different ways, and that can make a huge difference in what work needs to be done. If there are 100 tasks needed to be done on a project, but ten of those tasks constitute 90% of its functionality, tackling those ten in a dynamic matter might liberate the rest to be done later or make them easier, since the framework will already be built.

Numerous studies have found 10:1 differences in productivity and quality among individuals and even among teams. The original study that found huge variations in individual programming productivity was conducted in the late 1960s by Sackman, Erikson, and Grant. They studied professional programmers with an average of 7 years’ experience and found that the ratio of initial coding time between the best and worst programmers was about 20 to 1; the ratio of debugging times over 25 to 1; of program size 5 to 1; and of program execution speed about 10 to 1. They found no relationship between a programmer’s amount of experience and code quality or productivity. ^{^}

It’s not just programming: the best workers in all fields know how to separate the consequential from the inconsequential.

Q: Do workaholics accomplish more than people who work fewer hours?

A: Often, they don’t. That is because, as perfectionists, they may become so fixated on inconsequential details that they find it hard to move on to the next task, [Psychiatrist Bryan] Robinson said.

As Gayle Porter [a professor who has studied workaholism] put it: “They’re not looking for ways to be more efficient; they’re just looking for ways to always have more work to do.

{ deletia }

The person may look like a hero, coming in to solve crisis after crisis, when in fact the crises could have been avoided. Sometimes, the workaholic may have unwittingly created the problems to provide the endless thrill of more work. ^{^}

The management science involved is knowing the design of the task you’re attempting to accomplish, and finding core tasks that are necessary, then postponing others so they don’t get in the way. This keeps people from inventing work, or from getting stalled on relatively inconsequential tasks when bigger tasks need doing.

Another view of the same:

Sarah and I just got back from a talk at Haas about “deliberate practice” as it relates to excellence. The idea is that how good (or expert) you become at a skill has a lot more to do with how you go about doing your work than it has to do with merely performing the skill a large number of times or over a long length of time.

{ deletia }

This all reminds me of an old study of what differentiated classes of swimmers, The Mundanity of Excellence (it seems to be readable through Google book search). The researchers found that swimmers who moved up in class did it almost entirely by how they went about performing their practice. It was the quality of their work, not the quantity of their work that mattered. Moving up in class could be as simple as changing the way you cupped your hand during your swim stroke, as long as you were willing to practice that improved stroke during every lap of every practice. ^{^}

And still another take from one of the great success stories of the last century:

But if Toyota doesn’t look like an innovative company it’s only because our definition of innovation — cool new products and technological breakthroughs, by Steve Jobs-like visionaries — is far too narrow. Toyota’s innovations, by contrast, have focussed on process rather than on product, on the factory floor rather than on the showroom. That has made those innovations hard to see. But it hasn’t made them any less powerful.

{ deletia }

So how has Toyota stayed ahead of the pack?

The answer has a lot to do with another distinctive element of Toyota’s approach: defining innovation as an incremental process, in which the goal is not to make huge, sudden leaps but, rather, to make things better on a daily basis. (The principle is often known by its Japanese name, kaizen — continuous improvement.) ^{^}

Clearly defining important tasks, focusing on what they are as actions and goals, and then streamlining to make those goals occur more reliably and efficiently is a cornerstone of good management. This applies whether you’re talking about programming, dog walking, rocket science or impressionistic painting. It’s good discipline — good psychology — for succeeding at anything. The core of it is filtering the fringe tasks from the core.

And then there’s this 1994 book: Office Biology or Why Tuesday Is Your Most Productive Day and Other Relevant Facts for Survival in the Workplace. There is a whole book about how Tuesdays are so productive. How could I have missed this incredibly important fact for all these years? ^{^}

It’s not that Tuesdays are magic, but the position Tuesday holds in the week. Monday is getting re-acquainted with what will get done that week, and getting a huge list of tasks as everyone meets. Tuesday is when you cut that list down to what can be reasonably accomplished and pick the most important parts.

Again, it’s defining task, by what is most important, in order to stay on task.

When the company brain trust repairs to the Ballmer Bunker to chew over its next big idea, post-Yahoo, I’ve got a suggestion: how about doing something to deal with e-mail and its discontents? Something grand–like bringing Microsoft Outlook into the 21st century. I don’t mean a tweak here and there; I’m talking about a top-to-bottom overhaul.

The product debuted in 1997 and has improved very little since. Given the absence of real competition for most of the last decade, you shouldn’t be surprised at the glacial pace of improvement. ^{^}

John C. Dvorak refers to himself as a “cranky geek,” but the word he’s really looking for is cynical. Cynicism means you know that people are out in the world acting selfishly, and you see the results, and then realize it’s very little — another two hundred person hours, another 10% of budget and time — that separates the mediocre standard products we have now from ones that are actually a joy to use like the applications that inspire people to go into computing.

I agree with the authors above that Microsoft Outlook is a piece of junk, although it has some surprisingly flexible features, and that its interface is as god-awful as Windows 3.1. However, I have to ask: why is it so hard for humanity to produce good versions of its most basic software needs, like browsers, email clients, even internet forums and blogs?

I haven’t yet found an email client on Windows that I like. Pegasus I can respect, but there are parts of its interface that are so brick-stupid it’s impossible not to scorn it and some point, and although it functions well in most cases, under heavy load in some areas it snaps like a twig. Thunderbird? Junk. Opera’s built-in mail? OK, with some glaring oversights. Eudora? Mostly, except it has always been feeble like most Macintosh software. It’ll do OK if you have very basic needs, but put it under pressure and it doesn’t even make it to twig strength.

The same could be said of browsers. Opera is all-around the best, but even it collapses when Flash plugins tax it. Firefox is good, but often has inexplicable problems and seems to crash quite a bit, a good deal more than IE and much, much more than Opera. IE… well, Microsoft had a good five years without competition and did nothing impressive, and now it’s fairly big and some of its standards interpretations are weird. Safari? Who needs Safari? It’s like a Firefox clone with bad JavaScript.

I could go on, but you see the point. The everyday apps are the ones that people assume have no glory, so they get ignored. If they’re not ignored, the high competition means no one is going to jump into the fray and try to shave off some percentage of a massive userbase. Let’s not forget that Firefox inherited Netscape’s userbase, and Opera has forged a might 1.69% after ten years of existence. Uh, yeah.

I wish the Open Source community would tackle more of these fights, because what needs to be done is clear. Outlook could be replaced by a program that took Eudora’s approach to interface, Pegasus’ approach to file formats and data handling, and Lotus Notes’ approach to calendaring and encrypted, revisioned mail. It really isn’t that hard. Just go through what exists, pick the best features, find a way to make them work together, and then code something as light and fast as possible. The users would benefit.

Why even bother with candidates at this point? Elections have come to resemble fund-raisers or sports events, and I can’t tell which.

When we go to vote, hand us a budget with all known expenditures listed, and the ability to pencil in a percentage of our nation’s budget to each expenditure. At the bottom, let us write in future programs we’d like and how much of our budget they should get.

Throw away all ballots which do not add up to 100%.

Then average the results, and have a second election for the stuff we penciled in.

We’re about at that stage anyway.

Neat stuff during a busy time.

Lots of good things these past few weeks, and it’s hard to pick just a few.

The old, Google model: lots of little servers in a network designed for redundancy in case of failure.

The new old IBM model: a few centralized servers, but they’re more efficient and reliable and engineered to not only not waste cycles, but to last for decades. However, they’re designing them around an old mainframe strength — internal bandwidth speed — and using it as the basis for a new mainframe computing based heavily on flexible parallelism, or “cell” computing.

“We have been running multitenancy [running multiple customers on a single machine with a single application instance] for decades and decades,” he told me.

“It’s a mainframe model where things run together but in isolation. The issue is whether the machines will bear up under the load of diverse work or will they grind down and you’ll need to provision another machine. You need reliability, security, auditing, privacy, data integrity, automation and full isolation. You need to have a lot of layers in the environment.”

In 2000, IBM resurrected the mainframe by bringing Linux and WebSphere to the platform and lowering the price of entry, according to William Zeitler, senior vice president of IBM’s Systems & Technology Group. “You can build out a thousand smaller servers that need to be connected to ports and a fabric. You end up with a complexity crisis that has to be rationalized,” Zeitler said. ^{^}

The more small servers you have talking to each other, the more the communication becomes complex. Like in CPUs, the real question is how fast can they move information, not how fast they can calculate. Calculation speed is the model of old personal computing, from the early 1980s. Now we’re talking about moving massive amounts of data around and avoiding latency and internal correction that slow the process down.

A UC Berkeley paper [PDF] recently submitted to the IEEE International Parallel and Distributed Processing Symposium manages to highlight two common and seemingly unrelated themes that have come up a number of times over the past few years in my reporting on the high-performance computing (HPC) space: 1) IBM’s Cell is really good at HPC workloads when you invest the time to write custom code for it, and 2) Intel’s Xeon platform is perennially bandwidth starved and not very power-efficient. ^{^}

IBM’s solution: Use processors that emphasize moving information between each other and working collaboratively, like the Cell. Build a few giant boxes and over-engineer them so they’re reliable and efficient. This is in contrast to the PC/server market, where a new design comes out every six months and is under-engineered to avoid introducing potential conflicts and to get it out the door on time. They also have found out that these products are nice and green, since it’s easier to constrain efficiency on a few specialty designs than impose it on general purpose ones.

International Business Machines Corp (IBM.N) on Wednesday launched tools to reduce computer energy consumption as IBM hopes to boost its business of selling power-saving technologies.

{ deletia }

“Energy efficiency has become a critical business metric, like product reliability and customer satisfaction,” William Zeitler, head of IBM’s systems and technology group, said in an interview with Reuters. ^{^}

After having seen too many servers fail over the past few years, and having heard service technicians refer to two-year-old machines as “antiques,” I think this is a positive possibility. We’ve reached a possibly temporary plateau in processor power; programmers are still finding ways to take advantage of multiple cores, and it’s likely we’ll need to redesign how we write code and operating systems. Then again, the mainframe guys have been doing it all along.

The conventional dogma goes this way: people need money, so they’ll take a job they really love, and then they’ll become team players and we’ll all be happy. The reality of the situation is that there are few good jobs, and endless menial clerk-like ones, especially in IT, and that most of us work to pay for our families and don’t really like what we do.

What we do like is being effective, which is a way of having power without having to control other people except as is natural in pursuit of a goal. Having cake, eating too: power without being a jerk. It’s a form of creative power. When we have a job, we either feel that power or have a nagging suspicion that we’re wasting our time and will eventually incur the displeasure of others.

There is an anti-pattern that I’ve seen in large organizations which I have come to call “the Dead Sea effect”. The Dead Sea, of course, is a large body of water between Israel and Jordan, located well below sea level. The Jordan River empties into it; water leaves only by evaporation, which means that over the eons, the Dead Sea has become very salty (e.g., 8x saltier than the ocean). As such, it is generally unable to support life, except when spring floods temporarily lower the salinity.

Many large corporate/government IT shops — and not a few small ones — work like the Dead Sea. New hires are brought in as management deems it necessary. Their qualifications (talent, education, professionalism, experience, skills — TEPES) will tend to vary quite a bit, depending upon current needs, employee departure, the personnel budget, and the general hiring ability of those doing the hiring. All things being equal, the general competency of the IT department should have roughly the same distribution as the incoming hires.
^{^}

What happens in this pattern is that people with skills and knowledge and intelligence keep moving on, and they leave the average people behind. This presents a huge problem to a corporate IT department, because it means that all of the knowledge capital built up goes out the door and what is left are the people who are everyman cases. They follow instructions. They are competent at a limited range of tasks. They are oblivious to how to make things better and will move at a microscopic, dead-end pace without achieving any great output. It’s questionable how valuable they are except as warm, trained bodies who run at the machine guns when the enemy is at the gates.

In contrast, there’s a new pattern: the skilled leave work (for various reasons, although maternity’s a good one) and will come back if they’re given a chance to jump into the action, feel that creative power, get things done quicker and better than others, and then go back home:

Lots of employers would like to be able to hire cheap, temporary teams of seasoned pros with experience managing $2 billion investment portfolios, running ad campaigns or earning Ph.D.s in neuroscience.

But few know the secret to finding temps of that caliber: Look on playgrounds and at PTA meetings.

The decision among some highly educated women to stay home with children is sparking a countertrend: The rise of the mommy “SWAT team.” The acronym, for “smart women with available time,” is one mother’s label for all-mom teams assembled quickly through networking and staffing firms to handle crash projects. Employers get lots of voltage, cheap, while the women get a skills update and a taste of the professional challenges they miss.
^{^}

As mentioned before on this site, it’s good to have people with interests outside of work, especially family. This keeps them from, like the Dead Sea people, hanging around at a job and making small tasks into ongoing, unending labor to justify their own position.

However, once an employee shares all of his external knowledge, learns all that there is to know about the business, and applies all of his past experiences, the growth stops. That employee, in that particular job, has become all that he can be. He has reached the value apex.

Skilled developers understand this. Crossing the value apex often triggers an innate “probably time for me to move on” feeling and, after a while, leads towards inevitable resentment and an overall dislike of the job. Nothing – not even a team of on-site masseuses – can assuage this loss.

On the other hand, the unskilled tend to have a slightly different curve: Value Convergence. They eventually settle into a position of mediocrity and stay there indefinitely. The only reason their value does not decrease is because the vast amount of institutional knowledge they hoard and create.
^{^}

If you are unskilled in a job, you take the negative aspects of the job in exchange for it being there as a source of income. One group that appears unskilled but is not are The Unmotivated, which seems to be most of the bright liberal arts graduates in America who take slacker jobs and do poorly at them but not poorly enough to get fired. They recognize this system is slanted toward the unskilled, because although they are mediocre, in any industry that is not aiming for radical growth they are tractable, obedient employees.

As a manager, your job is to get the unskilled into clerkship positions and to get the skilled into positions where they can exercise creative power, or you’ll have the worst of both worlds: clerks in middle management and skilled people long gone and now seated in positions of more power at your competitors.

A long time ago, I wrote a story called Glitter Gold about those who huff paint and what it does to them. In it, one detail was that gold paint gets paint huffers the most intoxicated. Ever since then, reality has been imitating fiction:

According to a Bellaire Police Department report, Tribett’s pupils were constricted and he replied slowly to their questions. Oh, and “officers observed the paint on face and hands,” as can be seen in the below mug shot. ^{^}

One point of the story was that humans had to make life hopeless for paint huffing to seem attractive (as is the case with many intoxicants). You don’t need an escape valve until you so screw up the situation that people are desperate for escape. They don’t even want to enjoy life — they just want to check out.

In surveying the park, the officer noticed a man sitting in a lawn chair outside of a residence. He asked if the man had been huffing paint and the man said no.

However, when the officer approached and shined his flashlight toward the man, he noticed what appeared to be “fresh, gold-colored paint clinging to his nose and cheeks.” The officer also noticed paint in the man’s facial hair. ^{^}

One disturbing aspect of checking out is that once you’ve been out, you don’t want to be back in. Literally, you’ve seen a world where you don’t care about a damn thing except your bag full of paint. Why would you go back, to mortality, wars, corruption, pollution, Schadenfreude and bad TV? Inhale. Check out. Repeat.

In June 2006, Wheeling police said they found Tribett on 16th and Main Streets intoxicated and covered in paint. He was charged with public intoxication.

A week prior to that arrest, police found Tribett huffing paint under the Interstate 470 bridge. Police said when they found him, Tribett looked right at them but continued huffing. ^{^}

As much as the story shows its age, or rather my lack of experience at the time in getting said what I needed to say, its premise still rings true. People lock themselves into mazes of “can’ts” and the messy control issues of others, and finally, it all culminates in either total checkout or a conflagration.

Sloan based this on management principles. But of course it is the first lesson of political theory and political history. Authority without responsibility is illegitimate; but so is responsibility without authority. Both lead to tyranny. Sloan wanted a great deal of authority for his professional manager, and was ready to take high responsibility. But for that reason he insisted on limiting authority to the areas of professional competence, and refused to assert or admit responsibility in areas outside them. Peter F Drucker, Adventures of a Bystander

I’ve worked in a number of roles from development to copyrighter to documentation to glorified administrative assistant. I’ve been on the creative side, the technical side, and the demi-management side a project manager inhabits. I’ve done everything from sweep floors to define department policy.

One constant has been that people who are closest to directly performing work, such as art directors and developers, often complain about management. A frequent sentiment: executives don’t do anything, project managers are glorified clerks, and upper management has no idea what we’re actually doing.

All of these are true.

Just like most corporate IT departments hire randomly, retain people haphazardly, are managed badly and end up as IT ghettoes where anyone with a better option has moved on, most managers are either on their way to something good or stranded in a job that’s above their ability and for that reason, they can’t give it up and become defensive.

An enduring truth of life and business is that most things are done badly and most roles are poorly performed. Is the solution to curse the role, and say all executives are bicycles for fish?

Whether you work as a developer, or an art director, or any other role, you try to do a good job. That’s great until you run into a command from above that limits how effective you can be. Maybe it’s a simple request that you spend more time on the TPS reports than on writing code, or worse, it’s a silly direction to take the product or a command that contradicts all sense and knowledge.

This is why the study of management, and finding and keeping good managers, is essential. Someone competent in that role can make all the difference, while someone below that role can’t fix the damages of incompetent demands from that role.

Drucker’s statement reflects this truth. Authority without responsibility is illegitimate; but so is responsibility without authority. Both lead to tyranny. In this case, the developer has responsibility without authority, and if management is making bad decisions, they’re acting as if they have authority without responsibility.

Here’s a small, silent plea to managers of managers: know your people. Invest in their education not just in their job, but in the jobs of those that they manage. You don’t expect executives to be C++ programmers in their spare time, but if the product is written in C++, how horrible is it to insist they know something about the topic? A community college class at night is cheap and effective in this regard, and will help them understand those who they must manage.

When people have other obligations outside of work that they actually care more about than your probably-not-so-world-changing idea, the crutches are not available as an easy way out, and you’ll have to walk by the power of your good ideas and execution or you’ll fall fast and early. ^{^}

The hardest part of any project is knowing the mountains from the molehills. When we calm down, and stop trying to act like management or look like we’re heroes working late into the night, we can see what actually needs to get done, which is often a very small subset of what people belabor. Do the minimum, but make sure it’s the minimum required to achieve the end goal. You wouldn’t break out a slide rule to sweep a floor, so don’t make mountains out of molehills.

At the same time, good managers need to know how not to go into denial and make molehills out of mountains. Too often the big task is so daunting, and everyone so unprepared, that they spend their time doing everything but what actually completes the job. This process happens organically, so a manager without much control over the situation wakes up to it happening around them, but it doesn’t need to.

Some will say I’m a dreamer, but I think as our belts tighten, the concept of intelligent management is finding full flower. It’s time to be more efficient in what we do, and to eliminate the politics and infighting and confusion that take good people, with good intentions, and have them make bad products. If you want to know one ultimate way to “go green,” it’s to not make unnecessary activity, to not produce reams of unwanted and irrelevant reports, and to make products that are good and long-lasting so they don’t get discarded so quickly.

More Druckerisms:

Knowledge has to be improved, challenged, and increased constantly, or it vanishes.

Rank does not confer privilege or give power. It imposes responsibility.

Effective leadership is not about making speeches or being liked; leadership is defined by results not attributes.

The most efficient way to produce anything is to bring together under one management as many as possible of the activities needed to turn out the product.

Executives owe it to the organization and to their fellow workers not to tolerate nonperforming individuals in important jobs.

It’s this kind of thinking that keeps us all from being trapped on dead-end projects with frustration vibrating through our nerves.