Monday Nov 26, 2007

Two Great Linus Quotes

Here are two Linus quotes from a recent article in Informationweek -- Torvalds On Where Linux Is Headed In 2008:
And not only do we tend to support many different models of virtualization, but one telling detail may be that I am personally so totally uninterested in it, that I am really happy that I have almost nothing to do with any of them.

And I mention that as a strong point of open source! Why? Because it actually is a great example of what open source results in: one person's (or company's) particular interests don't end up being dominant. The fact that I personally think that virtualization isn't all that exciting means next to nothing.


But in the end, a lot of this is just a huge amount of individually small changes that may not be even interesting on their own - what is then really stunning is how big a difference all those small not-so-interesting changes make when you put them all together.

In other words, I'm a huge believer in the "99 % perspiration, 1% inspiration" rule. It's a lot of hard -- but happily, mostly interesting -- work, and there is seldom, if ever, any single big silver bullet.

Even though I took these two quotes out of the context of the original article, they illustrate two points I agree with: (1) one person or company shouldn't control the entire community, and (2) the real value of community development comes over the long term and results from many small contributions, not one big one. To me, these are two issues facing the OpenSolaris community right now. The small things matter a lot. They add up. And I think we miss that many times because we are too focused on "Sun" in the community on the one hand and the so-called "lack of leadership" on the other. There are no silver bullets for OpenSolaris. And there's nothing to replace the perspiration of doing the hard -- but interesting -- work.

Tuesday Nov 20, 2007

Do. Don't Talk

Today Patrick Finch directed me toward a great open source community presentation from Ted Leung (blog, slides). I agree with many things in this preso, but slide 14 really jumps right out at me. The headline reads: "Build a talk-o-cracy not a do-ocracy." That's obviously what you should not do. Then the two bullets read: "Doing is more important than talking. Talker-not-doer's are a DOS against the community." That's so true. Doing is better than talking generally, and as long as you are transparent in your doing, it's simply the most effective way to get attention on OpenSolaris as well.

Wednesday Oct 03, 2007

Cutting Through Noise

Here's a jet quick way to cut through noise in any community: "Having your voice listened to is a privilege, not a right, and it's a privilege that's earned in proportion to the contribution level, not volume level." -- Alan Burlison. I rather like that concept. It may not get you listened to by the noisy ones, but it will surely connect you to other contributor-oriented people in a community. Those voices come through clearly. The other side of this concept is expressed by Bryan Cantrill, who talked about expressing ideas in code. Both tap into the notion that the talking comes after the doing. Or, perhaps more accurately, talking without any doing is meaningless. Something like that.

Monday Sep 24, 2007

Opening from Closed: 3 Quick Steps

Over the past four years I've been trying to boil down to just a few steps just exactly how we are building community on the OpenSolaris project. After all, we are opening a previously closed project that already had significant operations, so there has to be some sequence here, right? Well, it may not seem like it at times, but there is. And it's complex. But I think I can fit just about everything into one of these three big buckets right here:

3 Steps to Opening from Closed:

  1. Move your own operations outside in stages. Begin with the obvious stuff such as lists, people, and conversations. Then move to code and infrastructure and tools and such. Absolutely nothing will happen outside, however, unless you first start talking outside and building a body of content outside that everyone can see and participate in. It doesn't matter that there is no community yet. It will develop over time. Just close the internal lists, open external ones, and move yourself outside. You may have to keep some closed lists to discuss the proprietary stuff you haven't opened yet and you may have to live in two worlds for a while too. But this should be temporary.
  2. Engage external people as insiders. This is an attitude shift. You have to see external developers as your peers. They are not a market and you are not building a corporate program. They are your peers and you are building a community of equals. Everyone's an insider and everyone's an outsider. Next you start working and collaborating on the code and infrastructure you've been moving out as well. Over time, a new dynamic will form where you create stuff outside, but first just get the closed bits out there so you have something to work on together. You can talk forever. Eventually, you have to work on something.
  3. Build community and leadership. I've always believed that the building of community occurs as a result of talking outside and working outside. It's really very simple. You will attract people if you are open and if there's something for everyone to do. And you have to talk to each other to get anything done. Now, over time you will have to enable people to grow and lead projects and such, so you have to let go. You are no longer running a closed project or corporate program. You are now part of a community, and you have to earn your way just like everyone else under whatever system the community creates. This last step of letting go is just as importing as the first step of moving our own butt across the firewall. Without it, nothing happens.
What do you think? And better yet, how many more big details can you add to the three buckets? Each step has a million details, of course, and many steps happen simultaneously and repeatedly as you open -- especially on large projects involving millions of lines of code, thousands of people, and processes that would fill a library. It takes time. Get used to that idea right up front. It's not quick.

Monday Feb 19, 2007

A way of Thinking

Another interesting article on Toyota -- From 0 to 60 to World Domination. This is a really long piece, over 8,000 words, but it's really nicely done.

Toyota employees think long term. They invest heavily in R&D -- much more than their competitors. Goals of quality and efficiency pervade the organization in engineering and marketing and manufacturing and pretty much everywhere else. Serving customers and building great products while not simultaneously hurting the environment (or at least not making it any worse) don't seem contradictory to these guys. They skip the utterly obscene executive pay packages common in the U.S. Unions are not present, nor are the American-style health care costs. They value evolution, not revolution. They prefer long-lasting and well-researched yet flexible strategies over short term sprints based on fads or whims. Their engineers very clearly lead and do significant -- at times obsessive -- field research first hand behind the wheel all over the world. Marketing is both traditional and grass roots and apparently quite simple and effective. They learn from their mistakes. They are remarkably open about their processes, but they also keep secret some of their innovations just as any smart company would. They are a culture built on top of Japanese culture, for sure, but they are by no means exclusively Japanese. They evolved based on the personal experiences of a unique group of people who dealt with the challenges of a country destroyed by war in a particularly innovative way. They are not perfect and don't lead in every market, but they are certainly on a roll in the biggest market and are delivering one body blow after another to the U.S. auto industry. Very interesting story.

There are a lot of great quotes in this article, but this one just jumps off the page:

Toyota spends $20 million a day ... on research and factories. "They are outspending G.M. in R.&D., product development and capital spending," says Sean McAlinden, an economist at the Center for Automotive Research, a not-for-profit consulting firm in Ann Arbor. "If that trend continues, we're dead. The problem is, suppose we made a car" as good as a Toyota. "Then we only have a car as good as they do. It's not just about catching up, or getting into the game. You’ve got to get ahead somehow. But how?"

So, even though the Toyota Production System is open, and even though this article makes it clear that Toyota "has never really caught the Big Three by surprise," people are still asking "how" they do it. Fascinating. Just having access to an open process will only take you so far, I guess.

Further down in the article you'll find the bit that helps explain why so many miss this point:

Management theorists who study Toyota's production system tend to say that it is difficult to replicate, insofar as the company's methods are not simply a series of techniques but a way of thinking about teamwork, products and efficiency.

A way of thinking. That's tough to copy. Even Toyota formally teaches the system to employees now since the company is growing so rapidly outside Japan, and they are concerned about quality in some markets. I'd like to take that class, actually. Wouldn't you?

Saturday Dec 23, 2006

The Toyota Way and Open Source

More news of Toyota taking out GM -- Toyota’s Sales Projections Show It Surpassing G.M. And more analysts are pointing to the famous "Toyota Way" business processes the company uses as the critical factor. From the Times article:

Toyota’s rise would also prove a victory of sorts for its unique corporate culture, the so-called Toyota Way, which is rooted in an obsession with craftsmanship and constant improvement, or "kaizen." Analysts said the Toyota Way would likely become enshrined as the industry’s gold standard, and the model to mimic or surpass for new challengers from South Korea and China.

"Enshrined as the industry’s gold standard, and the model to mimic," eh? That sounds like open source coming to the auto industry. After all, Toyota's processes are open, aren't they? But the notion of simply mimicking someone else's processes sounds trivial. The implementation is just as important as the source or specification of any business process. And that's much more difficult to mimic because what makes an implementation special is buried deep within the culture of every person doing the implementing. It's not necessarily secret, but it's oftentimes incomprehensible.

Tuesday Sep 12, 2006

Toyota's (Open)TPS

Really long and comprehensive article in Baseline on Toyota -- What's Driving Toyota? Nice piece, actually. I don't know where to begin since so many bits in the article interest me. How about starting with some basic business assumptions and how the Toyota Production System is defined. To quote the article:

The engine behind its success, say insiders and outsiders alike, is the Toyota Production System (TPS), a set of principles, philosophies and business processes to enable the leanest manufacturing.

And behind TPS is information technology -- supporting and enabling the business processes that help Toyota eliminate waste, operate with virtually no inventory and continually improve production.

Technology does not drive business processes at Toyota. The Toyota Production System does. However, technology plays a critical role by supporting, enabling and bringing to life on a mass scale the processes derived by adhering to TPS.

"What strikes me about Toyota is, if you were to ask them if they have a technology strategy, they would probably say no, we have a business strategy," says Philip Evans, a senior vice president at the Boston Consulting Group who has studied Toyota. "They have a very clear understanding of the role technology plays in supporting the business."

This strikes me, too, because I'm so used to focusing on the technology that the business case sometimes gets lose or the technology ends up driving the business case or obfuscating the business case. The sequence, however, seems quite clear at Toyota. But reading only this far, I thought for sure that this TPS thing must certainly be based on Toyota proprietary IP, right? It appears not. Later in the article you'll find this:

Unlike the formulas to blend Coca-Cola or the latest blockbuster drug, there is no veil of secrecy behind the Toyota Production System. In fact, Toyota openly invites general visitors and competitors alike into its plants to observe its operations and manufacturing techniques.

In 1992, it opened the Toyota Supplier Support Center in Erlanger, Ky., about an hour's drive north of the Georgetown plant, to teach other companies the principles and concepts behind TPS and to help implement TPS in their own operations. To date, it has worked with more than 100 companies as varied as office furniture maker Herman Miller, seat manufacturer Trim Masters and several hospitals. The supplier center now operates as an independent consulting firm.

It even created a joint venture with GM in 1982, taking a plant that was to be closed in Fremont, Calif., and reengineering it into a lean manufacturing facility based on TPS. That plant, renamed New United Motor Manufacturing Inc. (NUMMI), quickly surpassed all of GM's plants in North America in productivity, quality and inventory turns. NUMMI became a living laboratory for hundreds of GM executives and now manufactures Corollas, Tacoma pickup trucks and the Pontiac Vibe.

Toyota is open with the strategy behind TPS because it wants to raise its North American suppliers up to its own level of efficiency and quality, Liker says. At the same time, it can afford to be open with its competitors because Toyota is constantly raising the bar. By the time they copy its current processes, Toyota will have moved on.

So, the business model comes first at Toyota, and technology supports the business model -- not the other way around. Then both are packaged and implemented via the TPS, which is open and enables others to benefit while Toyota profits and drives its thinking deeper into the market. And Toyota is not worried about opening up its production processes because the company is confident it can out innovate competitors and, actually, the company would like suppliers to come up to its standards. Talk about confidence. My goodness. I think I'm going to cite this example the next time someone is worried about opening their code. This has an open community dynamic to it of sharing as well as competing.

The article goes on to explain the core elements of the TPS: Just-in-Time, Jidoka, Kaizen, Andons, Poka Yokes, and Genchi Genbutsu. To me, that last one in the list is the most interesting. According to the article,

The literal translation of this term is, "Go and see for yourself." Rather than hear about a problem, Toyota requires its workers, team leaders and executives to go and see a problem directly and to work collectively on a solution.

Interesting. So, there seems to be a community dynamic occurring internally as well as externally. There are several other examples of this in the article. If Toyota were a software company, I bet they'd participate in open source, don't you think?

Back to the article ...

Together, the elements of TPS form the basis for a system of business process management that allows Toyota to continuously look for ways to optimize its operations and put thought into action. Sounds simple, but it requires a basic cultural change in an organization, and that, according to Gary Convis, can be the most difficult challenge. Convis, chairman of Toyota Motor Manufacturing Kentucky, oversees the company's manufacturing plant in Georgetown, Ky.

By the way, after his promotion, this guy Convis moved his office from the administrative building to the factory floor. The chairman. On the factory floor.

So, why can't the American car companies tap into a system like this? It obviously works pretty well since Toyota just picked off Ford and is on its way to taking down GM. Why are those guys doing so poorly while Toyota (and Honda as well) are doing so well? Especially, when at least Toyota opens its processes? I think I'm just now beginning to understand what the answer is and why Toyota isn't afraid of opening those processes. Ok, the answer is obviously massively complex -- especially when you consider American union, health care, pension issues, and missing market shifts - but perhaps a few of these elements are involved as well: (1) openness can help build markets, (2) those who open some of their stuff end up leading within those markets, (3) and the culture of pervasive quality is almost impossible to copy because it leads to unique value every time it's implemented. And by "culture" I don't necessarily mean Japanese vs American. Yes, I think the Japanese notion of quality and service is somewhat higher than what most Americans can even imagine, but it doesn't have to be that way and it wasn't always this way in the past.

Now, the article is not all rosy for Toyota. In fact, the company has actually had some tough times lately with re-calls and quality issues. But most analysts feel that things are turning around, they have confidence in the company, and they still put Toyota quality way above that of their competitors. Nevertheless, "[a]t a news conference in July, Toyota president Katsuaki Watanabe bowed deeply and apologized for the recall troubles. 'I take this seriously and see it as a crisis,' Watanabe said at the conference. 'I want to apologize deeply for the troubles we have caused.'"

Bowed deeply and apologized. So, perhaps I should add a touch of humility to my little list of elements to consider.

Saturday Jul 08, 2006

Connected Capitalism

I love the term "Connected Capitalism" coined by Simon Phipps to describe how "open source works by everyone contributing what they want without compulsion and using what they need without restriction -- as a counterpoint to people who try to call open source 'communism'. Think Benkler."

That's a jam packed Simon quote from a comment he left to a recent James Governor blog about press coverage on Simon's keynote at OSBC a couple of weeks back. Ben Rockwood also has some interesting and valuable thoughts on the subject. Ever since I tripped over open source here at Sun about six years ago, I've been fascinated -- mostly because the culture reminded me of things I had seen in the past but couldn't really fully participate in. There is so much to talk about, but I'll just carve out my favorite little bit from Simon's comment -- connected capitalism.

I see a nice consistency between capitalism and open source. If open source were really about communism, as some detractors assert, I'd dump it pretty quickly. Not because of any political belief (I don't waste time on such issues), but because it would be incapable of providing me enough value so I wouldn't want to contribute to it. I'm looking to earn a living with multiple and diverse streams of income but in a way that contributes to the community (whatever community), not detracts from the community like the robber barons of times past (and some present, I suppose). If I don't have the ability to get something out if it, I'm gone. It's that simple. Food is important. So is health insurance. I have a significant amount of experience with the medical community, and I'm determined to have enough money to pay for as many circumstances I can imagine. And a steady flow of cash well into retirement should come in handy as well. So, my perspective is especially economic and I'm getting more and more focused on that every day. It's my primary goal, in fact. But I've never been a believer in the zero sum game, either, so that's why I like the concept, culture, and dynamic of open source. It's the perfect solution.

As an economic system, capitalism has done a lot of good and certainly provides some pretty massive incentives for growth. It has some pretty big holes, though, and many people react negatively to the term. Perhaps because used alone it can sometimes connote "big" and "exclusive" and "exploitive" and the connections supporting it generally pervade insiders -- the special ones, the privileged ones, the rich ones, the ones who control things like Big Oil, Big Unions, Big Education, Big Agriculture, Big Banking, Big Construction, Big Shipping, Big Government, Big [insert your favorite big thing here]. So, if that's the reason someone doesn't like the term, I can understand it. And, for the most part, I agree. Those things bother me, too. Breaking into some of those entrenched industries without paying off the controlling parties is challenging. Those industries are not communities whose members openly welcome new contributors, that's for sure.

I tried to break into the construction industry in New York as a small business owner, and, boy, did I learn a lesson in, ah, capitalism. My goodness. It had nothing whatsoever to do with open competition or competing based on talent, better pricing, better service, better equipment, better ideas, or better innovation. Instead, it had everything to do with paying to play in a controlled market, and it represented an absolutely stunning destruction of innovation and inspiration. I had to pay to enter the kingdom of those who had gone before and who were carefully guarding the gate to their paradigm -- at all costs. The controllers viewed their game as zero sum for sure, and I didn't fit in very well at all. In my case, I ran into several powerful construction and trucking unions (I was non-union), dozens of well-connected contractors (I didn't have their money), many government agencies (I didn't have political connections), and a few other rather strange characters (I'd rather not talk about). At times, the lines supposedly separating these groups blended all too closely, which was confusing and unsettling at best. The experience was both disgusting and exhilarating, and everything I believe today about economics and politics I learned from those early battles in the construction business. Back then I worked with some well-meaning business people -- true entrepreneurs -- and I learned all about what creative, innovative, talented individuals could build and how easily a powerful, centralized, controlling group could take it all away -- sometimes violently -- because the leaders felt threatened. I came to believe that open competition terrified people who contributed so very little..

I've always wondered about this though. In capitalism, why can't the individual and the community (or company, government, or union, or school, etc) benefit simultaneously so the cycle is self referral? To me, open source goes a long way to solving this problem because the culture of that system openly welcomes new contributions from creative, innovative individuals anywhere -- and the more the better. This is not necessarily exclusive to open source software development; it's probably present in other engineering and scientific disciplines as well. I sure saw it at Tufts University where I worked for a few years with physicians, veterinarians, and a variety of researchers. Heck, I bet any true craftsman or artist in any field would get this concept of individual, contribution, community. In other words, the community benefits but so does the individual contributing. In fact, the more one contributes, the more one benefits. It's that whole "you get out of it what you put into it" thing, and it supports the notion of enlightened self interest, which Simon rightly explains can be a problematic term but one I have no problem with (other than the fact that I need to assert it much more often). So, you enrich the commons and all the individual people within the community managing the commons benefit as well, and those especially hard working people have a limitless opportunity at their fingertips. Opportunity is there for all but it's proportional to how hard one works and how much one contributes. And around it goes. Generally speaking, of course. And by "enrich" I don't necessarily mean only in pure monetary terms. Rich means many things to many people: money, reputation, connections, access to shared resources, conservation, safety, insurance, learning, skill development, contribution, feeling of well being, donations, participation, desire to help others, etc. Simon explains all this much better, of course, but I'm trying to understand it based on my past experiences and on my future plans.

Now, elements of this absolutely remind me of capitalism -- but a very, very special form of capitalism. It's called entrepreneurialism. Small capitalism, I guess. Capitalism for the little guy. Capitalism the way it should be. Capitalism where everyone is welcome to participate and dare and risk. Capitalism that the old capitalists and the old communists would both have a hard time dealing with. That's the kind of capitalism I like. I just call it entrepreneurialism because my perspective starts with an independent individual being able to take care of him/herself so as not to be a burden but to do so by being connected so contributions benefit all. I think small entrepreneurs understand this more than big capitalists do (yes I'm splitting hairs a bit here) because the entrepreneurs usually start small and depend heavily on connections to others for resources, whereas those big capitalists tend to rely more on buying their way around since they already have lots of resources. It's certainly not true in all cases, but that's what I've seen along my way.

So, when Simon puts "connected" in front of the term "capitalism" it gives the entire principle a new, de-centralized, individual, empowering feeling and that's something every entrepreneur can understand. Hard work, perseverance, talent, skill, opportunity, luck. Not for Big Enterprise, but for the individuals who can't help but dream of what's possible. In this context, the term "connected" is critically important because it implies rather directly that for all this to work one has to connect to the community and help manage the stuff in the commons. Connected also implies responsibility, and if you are connected to people rather than capital it's more important that you do the right thing, not necessarily the most profitable and selfish thing. It's the connection to the community that creates the opportunity to create individual value, and that's what has hooked me. It solves my problem. I can be entrepreneurial without being a robber baron. The elite wouldn't understand this because they are not connected to anyone other than elitists like themselves. Whatever economic system they use in whatever society they live tends to exploit the commons, not contribute to it, and over time that pisses people off on all sides of the political, economic, and social fence.

"Connected Capitalism" works well for me. And I've yet to see a more powerful expression of the entrepreneurial spirit than the dynamics of a thriving community. Have you?

Friday Sep 23, 2005

Community Reactions

Sometimes people ask me how they should interact with the OpenSolaris community and how I think the community will react to various issues. This is difficult to predict, as I've come to learn (and am still learning). Many times it's just a person who has worked internally and wants to get used to the free flow of an open community conversation. That's pretty easy. But the challenge comes when people internally need to make difficult decisions that they feel the community many not agree with. What to do about that?

First, there's no need to pre-judge the communities reaction because you really don't know unless you directly and personally engage. Second, we need to remember that Sun is part of the OpenSolaris community now. True, we are leading OpenSolaris and sponsoring the project, but the process of making decisions within a community is quite different than making decisions in a closed system. The process of opening should be to engage with the external community and consider their ideas and needs right along with ours here internally. Now, we're the ones doing the opening (and there's a lot to open), so a lot of the discussion internally is based around how to open our own processes, which is understandable. But in general, if we are serious about building a community -- and we are -- more and more of our decisions should be made by involving the community -- internal and external -- as much as possible. You can see this beginning to happen on multiple lists, and over time it will only increase as more of what's in Solaris is opened.

So, this is what I tell people ... if someone at Sun has to make a business, legal, or engineering decision that the "community out there" may not like, then that decision should be made after engaging in an open conversation on the project's lists. There are exceptions, of course, but that's the direction we're moving. I'm finding that open conversations can actually move faster than closed conversations because the group self selects and there's less politics.

In general, we are trying to publish documents early to engage more of the community at Sun and outside Sun. Some well meaning people internally are sometimes concerned that this may bind the company to something or mean the company has to adopt every change the community (internally or externally) wants. This is a legitimate concern but fairly easily addressed. My feeling is that documents are published as drafts intentionally, so community members can participate in the creation of those documents (or processes or decisions or whatever) as much as possible. So, it's a process of leading an open discussion to come to a better, more inclusive, more focused decision. People leading the discussion need respond to feedback, consider all reasonable opinions professionally, and actually engage in an iterative process. In other words, emails going back and forth on list for everyone to see. If you have to do something that may be unpopular even after this process, at least everyone was involved and had a say up front. In other words, you have to talk to people and consider their views and articulate yours. Hopefully, learning goes both ways when this works. You can't fake it, by the way. It's pretty easy to spot. But as a practical matter, what usually happens when you engage the community early is that the initial idea is flushed out more thoroughly and everyone wins because everyone participated.

We have early drafts of the Governance, Charter, and development process definitions, as well as various development meetings (here, here) out in the open for feedback and participation. Anyone in the community at Sun or outside Sun can comment and be heard. It's the only way to be heard on these issues, actually. Now, there's a great deal of work to do on these governance and development processes documents, but when they are completed and implemented, it will be the OpenSolaris community that was responsible for the decisions. That's what we are shooting for.

Saturday Jul 30, 2005

The Art of Managing Projects

Nice little podcast here with Scott Berkun, the author of The Art of Project Management. It's really short, so give it a listen if this stuff interests you.

What's nice about Scott's take on project management that it seems quite realistic. Which is refreshing. He takes things like politics, power, execs, and egos seriously when he talks about managing projects, developers, schedules, budgets, and everything else that comes into play. So much to learn. He's got some interesting stories from Microsoft, too. :)

Scott's web site is full of resources for project managers. I'm participating in his PM Clinic forum now. It's interesting to exchange ideas and information with other PMs across the industry.



« May 2015

No bookmarks in folder