Top selected pieces
Brains, Banks, and Factories
The economic model we live by grew out of humanity's effort to scale up its supply of basic needs, giving communities the means to survive, raise their standards of living, and escape poverty. Successive generations embraced and evolved methods of planned and mechanized production, with each growing its output, while preserving a tradition of asset ownership and seasonal productivity. The result today is a model of economics that strives for short-horizon profit. Now that technology makes the goal of plenty attainable, a new model may be needed for a `smart world', one that redefines the goal of the economy from growth for its own sake to something more invested in social cohesion. Read more
Can we make progress in system administration education?
Invited comment to JESA journal.
Artificial reasoning in non-intelligent systems
Wed Jul 27 12:27:27 CEST 2016
Smart materials, smart houses, smart cities: are these things related to artificial intelligence? I argue that they are, but that our notion of AI, shaped by the desire to ape human capabilities, is currently too narrow in scope and scale to understand the connection. Studying systems that we don't normally consider to be smart is the route to understanding what smart means, and understanding how intelligence scales is the key to understanding how its scope and limitations.
The Semantic Spacetime Project
Mon Jun 27 15:28:49 CEST 2016
Theoretical physicists have owned the term spacetime since the beginning of the 20th century. Today, it conjurs popular figures like Albert Einstein, Relativity, Stephen Hawking, and Black Holes; but, it would be a mistake to dismiss it as only of interest to theoretical physicists and astronomers. Spacetime is far from a theoretical fancy. It is both everywhere and all of the time. What could be more practical than that?
Brains, Societies, and Semantic Spaces
Sun Mar 27 12:25:33 CEST 2016
Minor revisions 5 June 2016
Comparing what we mean by `smart' with the technologies and processes that matter to us, now and in the future. We might need to rethink what we believe smart is. If smart is simple and generic, can we engineer it into everything? What would that mean? Could we live in a truly smart world?
Fri Feb 19 11:04:45 CET 2016
In information technology (IT), we think we know a thing or two about scaling systems. Recently I stumbled across some fascinating work started at the Sante Fe Institute, on how the properties of cities scale with population. This grabbed my attention immediately, because it deals with many issues for which we have no data in IT. It was a chance to apply promise theory to exactly the kind of problem it was intended for.
The End of Sympathy?
Thu Sep 17 21:59:03 CEST 2015
Giving up our world to the robots, and why it bothers us... What gets overlooked, in the polemics about how humans will be replaced by `robots and A.I.', is that the arguments we use to express fear of automation apply not only to information technology, but to every kind of delegation of work, whether it be to a machine, an animal, or to another person.
The Cyborg Compulsion
Sun Jun 7 13:04:59 CEST 2015
Robots and artificial intelligence remain widely misunderstood, and a prime target for fear-mongering about the future of human jobs. I am not particularly afraid losing control to a machine intelligence, or even of losing my job to a machine. What we might lose control of, on the other hand, is our notion of civil society, as factions within us increasingly want (not to be replaced by, but) to actually become the machines.
Laugh IT up --- the Internet is just a gas!
Thu Apr 16 13:04:19 CEST 2015
As we redefine space to be increasingly functional, actionable, and with agency, then we have to be able to locate and reach functions and services, as well as understand what they will do for us. This is the world of pervasive computing, and smart environments.
In Search of Science Behind `Complexity'
Tue Mar 3 09:49:12 CET 2015
Complexity theory, as it is often called, has caught the popular imagination in recent years, with many authors claiming to write about the subject from a layman's perspective, applying it to all manner of problems from business to system design. But is there a science behind the bravado?
SDN: software defined networking, ... or small distributed namespaces?
Sat Jan 18 12:36:57 CET 2015
Software Defined Networking (SDN) sometimes seems a little bit like Las Vegas. The shiny virtual facade defies the shaky foundations it's built on, and there seems to be a song and dance (and a certain amount of gambling) involved to come out of it richer on the other end. This essay sketches out ideas about networking that I think are increasingly obvious to users in the datacentre, but which also needs to be applied to a larger Internet of Things. The key is all about tenancy. Within a truly shared infrastructure, the straw building up on the camel's back will be the container.
Why I stopped caring about peer review, and learned to love the work
Wed Dec 10 19:19:14 CET 2014
Peer review is often upheld as one of the revered checks and balances that ensures some kind of quality in science; but, in an increasingly competitive world, peer review in journals has become ineffective way of getting to the best science, and it is not always used with honorable intent. Today, if the proliferation of thought and ideas is your goal, the best chance you have is simply to publish work openly online and wait for peer review to come.
Virtualizing virtualization to scale infrastructure to an Internet of Things
Thu Nov 20 16:14:14 CET 2014
I propose a new angle for exploring infrastructure abstractions, building on all of the idea of an abstract material or space. This offers a foundation for scaling, and unifies the different aspects to address a number of basic questions for networking, computation, storage and knowledge.
Infrastructure management timescales
17th November 2014
A brief sketch of the current timescales we deal with in infrastructure management. Without a proper understanding of time and how it is used, we end up wasting a precious resource as well as looking foolish.
The Forsaken Art of Pedagogy
25th September 2014
Are we losing the ability to communicate?
14th August 2014
Trapped by failure.
The Gamesters of Transmogrification
31st July 2014
Are APIs opening up IT services to a free and open dialogue with users, or are we on a collision course with complexity?
The Brain Horizon
22nd July 2014 (revised 22 December 2015)
After years of investing in distributed approaches, IT Server Management has begin to return once again to that age-old idea of centralized control; now also Software Defined Networking has proposed the same -- surely unthinkable. Even Google is going out and supporting it. Surely centralization doesn't scale? What does this mean? Personally, I like to think about this in terms of brains and societies. Read more
The Making of a Software Wind-Tunnel
Could we make the analogue of a wind tunnel for scale testing of software?
10 April 2014Part 1 Part 2 Part 3
Automated EBook Publishing for Kindle from Latex source with CFEngine
9th March 2014
As someone who uses Latex for its superior print quality, I was disappointed (but not surprised) to learn that the tools for publishing on Kindle do not include Latex support. So here is how to build Kindle from Latex using CFEngine. Read more
Configuration Management for Continuous Delivery
25 February 2014
Continuous delivery of software has been championed lately, and made popular by Thoughtworks philosophers, most notably Jez Humble and David Farley, amongst others of Martin Fowler's group. Their excellent book Continuous Delivery lays out the reasoning. I am a great supporter of this manifesto. Today, we are being challenged to apply it to all manner of information driven processes, not just software building.
The evolution of configuration management thinking as a business problem (in 2 universes)
13 August 2013
With the help of the Quantum Food Guide and Devomonitorama Brain Extender, we are able to overhear the thoughts of two parallel instances of An Ops Guy in parallel universes contemplate the meaning of configuration management in their businesses... In universe #1, GuyOps runs an online business; in universe #2 he runs a hotel/restaurant business. His thoughts allow us to see how CM is really a knowledge management issue.
"20 years of CFEngine: design promises"
31 Jan 2013
2013 is the 20th anniversary of the CFEngine project. Sometimes I get asked what makes CFEngine different from other software. Here is my answer: design principles that get the basics right. Whatever your aesthetic preferences, design princples have kept CFEngine reliable and secure for 20 years, under a welter of dynamic challenges.
"We need to trust trust"
Although I have never properly met Bruce Schneier, and we have only exchanged a handful of words over the years, I have often viewed him as a kindred spirit, handling the topics he discusses just as I might have. So I suppose it is no surprise that I am going to recommend his latest book.
Science is Culture, Technology is Art (reprise)
22 Nov 2012
In 1994, I produced a CD-ROM for Norwegian schools, touting the slogan (in Norwegian) in the title of this essay. Over the years, I've repeated this phrase on many occasions, in different contexts, as I believe there is a lesson here that the gatekeepers of learning seem to be unable or unwilling to preserve. I was reminded of this just the other day, while talking to a friend, caught between these worlds in another way, and it reminded me of the discussion about DevOps culture that is being played out on the net today.
Deconstructing the `CAP theorem' for CM and DevOps
9th August 2012
In this essay (in two parts), I want to explain the issues around CAP more carefully and add a broader perspective from the viewpoint of time. The aim is to talk about its potential impact on the world of infrastructure operations. I shall try to pick apart what CAP is about, what it does and doesn't mean, and what is important to understand in the discussion for DevOps.
The Promises of DevOps
23rd June 2012
DevOps is evolving from what started as little more than a rally cry, into a culture for continuous delivery of IT services. As more pieces of the puzzle fall into place, we understand that DevOps is primarily a matter of human collaboration -- assisted, but not encompassed by, certain tools and technologies. This essay is my take on DevOps, and why I think it is about embracing distributed, autonomous operational methods that support scalable human-computer processes.
The Scaffolding of Knowledge (Beyond desired state configuration management, for the Third Wave)
5th April 2012
What if I don't know the answers? What if I won't know what to do? A common nightmare surely everyone has had at some time -- being immobilized before an impending crisis, helpless, unable to move or do anything about it. The nightmare is only a dream fantasy, but the reality is lived out by someone, somewhere in a workplace in every generation of change. The dark force of this immobilization? Insufficient knowledge. What if that happened in a mission critical IT system? How could we get smart fast enough?
Break the mirror: Amortality
5th April 2012
Some comments on Catherine Mayer's book Amortality: the Pleasures and Perils of Living Agelessly.
Clouds: I've looked at life from both sides now...
10th March 2012
Is there an academic divide -- a chasm between business and academia? Should there be a divide and, if so, should it be bridged? Cloud, in IT, is a technology that purports to be about bridging the business-IT divide -- it allows evolutionary improvement of IT-based activity, which is a perfect platform for society to advance. Surely there are analogues worth learning from.
When and Where Order Matters
26th December 2011
Recently I have been hearing the resurfacing of an old debate: whether order matters in the execution of configuration management policy. The debate started years ago when CFEngine introduced a declarative approach to configuration in order to separate the WHAT from the HOW -- and move to a more knowledge-oriented form of management. Alva Couch, my friend and colleague, likened CFEngine to Prolog.
3 Myths and 3 Challenges to bring System Administration out of the Dark Ages
7th December 2011
There is a very clear change taking place in the way that system administration is being done in a wide range of organizations - it is being driven by the needs of businesses in modern society -- the need to support the wealth of freedom that we as end users enjoy thanks to mobile devices. We see a proliferation of IT that has already swept across the western world and shows no real sign of abating -- and that is driving maturity as well as big changes in the way that sysadmin is done.
Striking The Balance Between Man And Machine In IT Management
27th October 2011
How shall we use humans and technology to manage IT resources? During the past twenty years there have been large shifts in the technologies used to manage IT systems, but very few shifts in the underlying principles of management. I believe that we need a better understanding of this collaboration between man and machine to understand management in general. In this keynote, I will use the theme of human knowledge and comprehension to drive a picture of what is needed of IT management in the future.
17th July 2011
A talk presented at Amsterdam University on 20th June, in honour of Jan Bergstra's 60th birthday. It discusses how a search for a solution to division by zero confirmed some theory on which CFEngine is based.
Proof that rollback in system admin is a "total" fiction
24 June 2011
Research has shown several times that transactional rollback is not a reliable or meaningful operation in IT management. Indeed, it can cause as much harm as the original error one attempts to back out of. Referring to the work of Bergstra and Tucker, this work done with Alva Couch proves a link between "undo" and inverses of operations that have incomplete information. In the case of a policy enforcement, a "rollback" corresponds to division by zero. One can nevertheless consistently recover from a rollback operation by repairing the system forwards in time usin convergent operations. This does not change the fact that such operations should be avoided if system predictability is to be assured. Thus strategies based on rolling back should be treated with the utmost suspicion. See this paper written in honour of my friend Jan Bergstra on his 60th birthday.View
Change = mass x velocity - and other laws of infrastructure
15 June 2011
Keynote held at Velocity 2011. The key challenges for infrastructure designers and maintainers today are scale, speed and complexity. Mark Burgess was one of the first people to look for ways of managing these issues based on theoretical analysis. Much of his work has gone into the highly successful software Cfengine, which is still very much a leading light in the industry. In this session, Mark will ask if we have yet learned the lessons of infrastructure management, and, either way, what must come next. View
Forgetting how to think
16th February 2011
Today, a growing proportion of the populace aspires to be on a par with experts, and sees real experts as suspicious or even bogus figureheads, like the rich landowners of the industrial revolution who kept the general population poor with malice and greed. I now see this kind of challenge quite often when people don't like the conclusions experts present: a head-in-the-sand rejection of knowledge in what is now seen as a kind of open market for being right. Read more
LISA Knowledge Management Workshop
13th December 2010
I recently chaired a Knowledge Management workshop at LISA 2010. I have said on several occasions that Knowledge Management is one of the central challenges of the next decade. Below is a brief summary of the workshop.Read more
The CMDB Imperative
23rd November 2010
Glenn O'Donnel and Carlos Casanova have written perhaps the first worthy analysis of the success and future of the CMDB, with a title befitting a good spy drama. Read more
Universality and IT management
30th October 2010
The death this year (14 October 2010) of the Polish-French-American mathematician Benoît Mandelbrot, best known for introducing the idea of fractal dimension to mathematics, reminded me of issues that I have moved away from over the last ten years, but which are every bit as important now as they were back then. We must learn to appreciate that scale is not merely a design issue in IT management. It might hold some surprises for us. Read more
The trouble with physics
9th October 2010
What has beloved physics turned into? Certainly not science, that much is certain. It has become a vanity fair and media circus, often little better than pulp fiction, which overshadows the work of real science. String theorists especially have stolen far too much attention. Read more
Three myths holding system administration back...
9th September 2010
Hierarchy, rollback and ordering ... three words that have haunted me like a sour reflux for the past 15 years. It's time to put pen to paper and explain why I believe these notions are harmful the IT industry. Read more
7 July 2010
June 2010 has seen the passing of one of Norway's most important composers: Arne Nordheim. He was a central figure in Norwegian music, who wrote the opening music for the Lillehammer Winter Olympics, and who has stirred controversy as a modernist. Read more
What is the value of System Administration to Business?
8 May 2010
Many, if not most, business leaders see IT services as a cost centre, a sump for funds to disappear into. They tend not to see it as a strategic tool for supporting business growth or developing new opportunities. They need a trusted interpreter. Knowledge Management is a central pillar to developing and measuring system administration value: System Administrators must become teachers. Read more
Made in Japan
1 May 2010
Japan is the most alluring of worlds. Old meets new without obvious disdain, and there is no shame in visual expression, as long as it is tempered by ritual modesty. I spent only 10 days in Japan, travelling with friends between Tokyo, Kyoto, Osaka, Naoshima and Kinosaki Onsen, but this hardly seems to scratch the surface of what must be the most intriguing place I have ever been. Read more
Promises of Virtualization (shapes in the clouds)
29 Mar 2010
Cfengine today has developed a broad range of virtualization techniques that simplify the packaging and configuration of computer systems. Still, there is a lot of confusion about what virtualization is and is for. Promise Theory (on which Cfengine is based) could be seen as a form of virtualization itself, so how does Promise Theory help us to understand and manage virtualization and its Cloud Computing dreams? Read more
The first worthy documentary about the Internet?
28 Mar 2010
Bless the BBC for still making high class documentaries, second to none. Much as I despair the demise of Horizon, there is no shortage of quality documentaries being made. I have recently been watching Coast, about Britain's fabulous and historic coastline, and this past month I have been watching Alex Krotoski's excellent series Virtual Revolutions. A lot of fluff has been presented about this Internet. This series presents an excellent and balanced summary, in my view, of what is going on in the world today by a compelling presenter.
7 Mar 2010
I have expended lot of effort since 2004 on understanding the concept of promises (or Promise Theory, as it has come to be known). Last year, I decided to put together everything I had written on the subject into a book, to be written with my friend and collaborator Jan Bergstra. The story of promises has not fit snugly into the precisely measured and artificial 10 page chunks that journals mandate, so putting this into a book, the whole picture will make the task of explaining Promise Theory much simpler and clearer. Of course, writing a book can take a long time, so I am posting drafts so that others do not have to wait for completion. Download
Offline, from Carlos Zafon to Lady Gaga
27 Feb 2010
I take great pleasure in the enterainment industry when it succeeds in lifting up genuinely talented individuals, who herniate through the often pathological ceiling of formulaic production values in music and film. I don't find stuff I like every day, but I am fortunate to have broad tastes. Read more
Travelling `light' into the future
13 Feb 2010
I have recently been moved to think more about the efficiency of technology... Anyone who has used Windows along side Linux on the same PC, knows about inefficiency from painful experience. Software engineering teaches programmers to be resource hogs. Read more
The Business Value of System Administration
1 Jan 2010
The status of system administrators as experts is at stake as both technology and businesses evolve. To evolve in step, professionals need to become more business aware.
How do IT departments impact on the businesses they support? How might we align tools and resources to more closely support an organization's primary goals? These are important questions that have often been glossed over when embracing Information Technology (IT), but for the past few years researchers have indeed attempted to answer them, groping for some sort of models and metrics to measure success. Here, we attempt to summarize the discussions of Business alignment that took place at the LISA/BDIM workshops over the past two years, and place the points in the context of the wider view. Read more (general version) [Short version for sysadmins]
Goldsmith on LB4-26
October 25 2009
This year is the 30th anniversary of Ridley Scott's film Alien, and in commemoration, there is a finally a digitally remastered re-release Gerry Goldsmith's brilliant film score for the movie.
The new release contains much music that was omitted previously. For me, this is one of Gerry Goldsmith's finest scores, and Goldsmith himself was one of the finest composers in the 1970s, and surely one of the most imaginative film composers of all time. Here we find him at the height of his powers, leading the way with an ensemble of biazarre instruments to top even his own brilliant soundtrack from Star Trek: The Motion Picture. No, no one is paying me to say this -- I am just thrilled by this release. And it does not disappoint -- it is just perfect for those long, dark winter nights here in Oslo, with a tube of silver toothpaste.
I have been waiting for this literally for 20 years since acquiring the original release CD, which was badly sampled and distorted in the final climax. Several conductors have attempted to re-record this music, but all of them have failed to capture the ominous power of this original recording. If you are a Mozart lover, you probably won't like this -- but if you love symphonic electricity to dazzle even Schostakovich, then you must have this in your collection.
The Nightmare of Knowledge
September 27 2009
Knowledge management in organizations is a serious and pressing issue today. As the pace of our world increases and people move more quickly from task to task, or even job to job, knowledge that is locked inside people's heads tends to remain there. Organizations spend a lot of money and resources on retraining new employees because their core processes were not well documented and could not easily be transferred to newbies. Dreaming on the other hand crosses knowledge boundaries, building linear narratives from complex networks of things we know. It is effortless and imaginative. Could the answer to Knowledge Management lie in dreaming? Read more
Running a business is a PhD
PhD students learn unique skills in order to develop an overview of their subject: they have to understand the breadth and depths of their topic, embrace common sense, and learn to defend against every kind of criticism. To succeed they have to learn time management and task prioritization. These are exactly the skills needed to run a business. Why is it then that business and academia are such unwilling bed-fellows? Read more
Earlier this year my friend Steve Pepper (of Topic Map fame) pointed me to one of the most fascinating and beautifully written books I have read in a long time: Guy Deutscher's The Unfolding of Language. Steve is a compulsive linguist. In this book, Guy Deutscher (whose name is even indicative of his multi-national life and learnings) describes the state of modern linguistics with enviable skill. It makes both fascinating reading and for a rare appreciation of the linguistic diversity around the planet.
Deutscher describes how two processes have taken us from cave-man grunts to modern day grammars: inflation of importance and erosion of accuracy. Without wanting to give away too much from his story telling (you should immediately buy the book), the answers struck me as being satisfyingly universal. Most processes, from natural evolution to IT management are similar equilibria between the powers of erosion and growth, but the we often forget about the extraordinary transformations of structure that take us from amorphous goo to specially compartmentalized: what might popularly be called self-organization. I have written on this theme several times in research and popular writing, in particular that system configuration is about grammar, because grammars and patterns are the same.
This is a topic worth thinking more about; so let me not spoil it for you by unfolding my own thoughts on the subject. Suffice it to say that much more is going to be written on this topic.
A couple of years ago I came across Richard Layard's book on Happiness. This book describes the research that has been done on what makes people happy. It is a fascinating discussion of the human condition using the tools of good science. Who ever said that science is cold?
Layard is an economist from the London School of Economics. He demonstrates that the happiness of society does not necessarily equate to its income. He is best known for his work on unemployment and inequality, which provided the intellectual basis for Britain's improved unemployment policies. He founded the Centre for Economic Performance at the London School of Economics,
I was indeed happy to see a conclusion that I have realized myself for some time: we are not here to punish ourselves, fulfill others' expectations, or work exclusively for pay without gratification. We only get one chance, so we should strive to be happy. It's not really very hard to understand.
Selling research by the pound
When I look back to my student days, I recall thinking that there was something odd that happened to research during the 1980s. If I was looking for answers, i.e. real understanding, I would have to go back to papers written from the early parts of the 20th century up to about the 1970s. But after that -- chaos, or `The Rest is Noise'. Research papers seemed to disintegrate into gobbledygook or self-aggrandizing wish-wash. It was as though people had stopped trying to make their message clear, and had begun faking it.
It was unclear to me then exactly what happened or why, but today I see things more clearly. Research has become distorted by a number of forces, but increasingly by government policies and the associated bureaucracy. Research has been dumbed down -- and while science itself is guilty of allowing it to happen, (clichés aside) governments really are to blame.
In modern society, academics are not valued very highly; research is often seen by many as a way for unstructured layabouts (i.e. `academics', a dirty word let's face it) to get free money for doing something that no one understands (and is therefore highly suspicious). Money spent on research is little better than social security hand-outs for people who should really get real jobs. Obviously, these people need to be reined in and controlled! (Think: What did the Romans ever do for us?)
I observe that government ministers around the world usually have relatively little schooling, and many seem to think that this means no-one else should have any either. Governments around the world have therefore tried to address this by bringing in `Quality Assurance' methods from the world of bureaucracy. ISO-9000 overdrive is now used to hold researchers to ransom.
It goes like this: value for money means Quality Assurance, and Quality Assurance means bureaucracy -- the only instrument available to bureaucrats let's face it, so naturally what they turn to when commanded to control the rebels. We try to turn research (an inherently risky and uncertain endeavour) into something like a production line, so that no matter what happens, something predictable will come out of the conveyor belt.
So, if you play the government game, churn out typically 100 pages of documentation in pre-planning, explaining exactly what you are going to find out (use crystal ball, or other technology to short circuit the process, or simply choose something trivial and unimportant to `discover'), bringing together a politically correct consortium of representatives from various countries, and your research application survives a process which is designed to preserve the political status quo of the haves and have-nots in research, and you agree to write `deliverables' (homework assignments written for committees, explaining what you did at research camp) every months or so, and attend all the meetings, and deliver audited financials -- then, and only then will funding agencies pay you a pittance to work on an approved project. You are literally guaranteed to discover nothing new, because all possibility of working on something unknown has beem abolished.
The result is a farce. Research has been mostly forced into a pattern of form without substance. People go through the motions, attend the conferences, write the papers (a minimum of 5 per year if you want to keep your money), but what comes out of it? In most cases, we've seen it all before. Research now attracts people with management skills instead of ideas. People with real ideas are rejected for not following suit. It takes real perseverence and a bit of luck to bring about anything new. And the few who dare are never heard, because the level of junk quota-papers being presented drowns out any interesting message that might be lurking in the background.
Reasearchers are writing too much and thinking too little, and the reason is: funding policy and the devaluation of academic pursuit in society.
In 2007 it reached the point where I never had time to do any honestly innovative work because of homework from an EU project. Research is supposed to be about risk. I believe absolutely that less money would be wasted and science would be served better by blindly giving away money to academics on trust, without all these overheads. Because even if only 10 percent of the money resulted in something useful, that is surely better than spending money to blow original thoughts away, clog real thinkers' heads with dehumanizing procedures, waste committee's time and feed paper mills.
After some soul searching, I decided: to hell with it. I don't need the kind of approval that is ransomed on a poverty of ethics. I will find another way to do my work -- by starting a company. No doubt government workers will jump with glee: this is exactly what we wanted all along! Now you are doing `applied research'! But the truth is government bodies have contributed nothing but dispair to this decision. By creating a funding process which favours those who like bureaucracy and insisting on maximizing publications, they have only succeeded in turning research into a garbage dump -- isolating innovators and promoting practitioners of little talent who can play the game of recycling ideas into quota publications. This costs taxpayers more than before for less result. Imagine every radio station turned to noise with a barely audible tune playing, and you will get some idea of what scientific journals and research have become.