Arjen's blog

The Last HOPE

Just finished my talk at the Hope conference about the national Dutch open standards / opensource policyplan. Hope stands for Hackers On Planet Earth and has been a bi-annual event since 1994. HOPE features presentations on a range of (IT) security related topics but also hosts all kinds of techno-creative workshops and lectures on the social relevance of technology. For a full listing see the conference site. Video's of the talks will become available in the future.

Because the mid-town Manhattan hotel just across from Penn station in New York may be demolished next year it is unclear if there will be (a) hope after this one. Because of this, and a certain facination with morbid themes, the conference organisers are calling this 'The Last Hope', refering both to their conference and the state of (digital) civil liberties in the US.

Cnet has the first coverage on the event.

My slides are here in ODF and PDF. For more background links and video see the 24C3 page (Dec 2007 talk in Berlin) and my blogpost about LinuxTag (May 2008 Berlin).

Workplace 2028

After a brainstorming session with about 50 employees of Brunel Engineering last week I was asked to write a piece on the workplace of the future.

Office 2028
How are workplaces will look like in 20 years depend on two basic factors. One factor is the development of technological instruments to aid us in our work. Think all those IT-related things such as the speed and capacities of computers, networks and data storage systems. Scientific insights are put to work to make these instruments easier and more effective to use. Improved understanding of human cognitive and neural processes allow for better information-processing tools. Then there is the physical location (slowly but surely becoming less important); new materials make for better chairs, desks and ultimately an entire building specifically designed to facilitate mental work.

The other factor is us. Our (in)ability to actually use all these tools effectively my become the major bottleneck in improving the effectiveness of office workers. Techno-scientific developments tend to out pace the ability to change of most organisations (and the people in them) by at least a decade or two. There will be major difference between and within organisations. Age, education, upbringing and attitude will all be factors determining a workers ability to absorb the skills required to use new tools to do things better or to do better things for the organisations they are a part of. The type of work is also relevant. Graphics designers, architect and other creatives usually want and need the best possible tools and know how to use them. Someone writing a policy document can probably make due with the same word processors we use today. Some people may perform relatively simple administrative tasks in an organisations with a low level of automation while another performs complex, creative and knowledge intensive work were the limits of technology are sought. Thé office of he future therefore does not exist.

In 2028 many officeworkers will still be behinds desks wile others will be plugged into the Internet from somewhere on the planet. Some of them will work alone but more will work in teams assisted by intelligent systems that take most of the analytical and non-cvreative work out of their hands. Education, training, good leadership and access to communities of knowledge will be the determining factors of their productivity and creativity.

Will Apple be the next Sony?

I normally don't review products or specifically recommend them but the difference between the success of the iPod and Apple TV teaches some important lessons to anyone with ambitions in the digitized media markets of the 21st century.

Apple TV's competitor is not the video rental store on the corner, it's not even online companies like Netflix. It's The Piratebay and sites like it. And it will be hard to re-learn new customers to buy media for watching though devices like Apple TV when they've been using somewhat comparable devices for  freely watching media for a few years.

I own both a Mac laptop and an iPod and use them both to listen to music, podcasts, recorded lectures, audiobooks and watch movies on planes & trains. Even though my 60GigaByte iPod is two-thirds full I've never bought anything through iTunes. Why would I? Firstly I'd have to pay for it while there's more than plenty music freely available (downloading is legal in the Netherlands and some other EU countries as long as you're not making a profit from it by reselling or renting). Then there's the fact that Apple ties the music it sells to me to their device. And while I like my iPod a lot I want to retain the option to switch to something else in the future (maybe my new Nokia phone that holds 8GigaByte of data). My downloaded MP3's do not restrict me in any way. Oh, and did I mention they were free? It's a pretty easy choice.

The iPod is successful because it does not restrict me from playing a relatively open (and widely available) format such as MP3, downloaded or ripped from CD's I owned. Apple TV on the other hand comes with all kind of restrictions built in. While I could buy (or rent) movies from Apple and watch them on my TV through the Apple TV device this is a big limitation compared to being able to choose the source of media myself, choose to payment level and method myself and then do with the file as I please (transfer it to a non-Apple device for instance). While the Apple TV has a nice Interface to control it from the couch so do various other devices at a lower cost. These also do not restrict me in my sources of media like Apple TV. So why would I buy a 300 euro Apple device that ties me down to the Apple media store when I can get a better device for under 200 that does not?

Apple seems to be making the same mistake as Sony did some years back. You remember Sony don't you? They used to make these things called Walkmans fthe late seventies - I bought several of them. Then when customers started demanding MP3 players Sony did not deliver because it's management feared that would hurt their CD business. Apple did not have a CD business and felt no such limitations. They blew Sony's 25-year legacy of building  the hottest mobile music devices out of the water in in few short years.

So now there's Apple trying to tie us to their storefront through a device while other device makers don't. Its pretty obvious what is going to happen. As long as Apple (or any other manufacturer) tries these kind of tie-ins I'll stay away. And so will the dozens of people who rely on me to make tech-purchasing decisions for them. Instead I'll move to a device that gives me the freedoms I demand from a media center and I'm taking a buss load of potential customers with me. So will most of my tech-savy friends. If Apple does not want to be the next Sony the need to shape up and let me choose freely the source and file-format of my digital media. Only then will I consider taking my mac-ownership to the next level.

(this brainwave was created over Portuguese chicken with Aldert - thanks man!)

Desktop mono-culture for the Dutch

The Dutch national government is proceeding with its new implementation of 'GOUD' (Gold), its new plan for a 'standardized desktop' for ministries and other governmental institutions. After the much publicized new Dutch National IT-policy of last autumn this might sound wonderful. It's not.

The problem with the 'standardized desktop' is that it uses very few real standards. Instead it uses products and pretends they are standards. The difference between 'standards' and the idea of 'standards products' is widely misunderstood, often confused and this is where projects such as 'GOUD' go off in troublesome directions.

Standards are things like Open Document Format, HTML (the programming language of web pages), LDAP (a protocol for defining access-control on a network) or TCP/IP (the low-level network protocol of the Internet). What all these things share is that they are not owned by a single entity. They are a formalized form of knowledge about how to achieve certain technical objectives. This knowledge can be used by programmers to build tools that actually do these things for us. This means that there will be many different, competing, implementations of the standards that can inter-operate. This diversity fosters a working market and ups the quality while lowering the price (often down to zero).

'Standard products' on the other hand are, by definition, products and generally owned by private corporations with a profit motive. Standardizing on a product is essentially creating a mono-culture of one type of product, making yourself totally dependent on the vendor of that product. In the case of desktop computers this is extra problematic because there exists a de facto monopoly in this market. Creating a mono-culture in a monopolized market actually re-enforces the monopoly, the exact opposite of the aims of the adopted policy plan.

The Dutch government (or any other for that matter) should not work towards a 'standard desktop' but towards open desktop standards. This is precisely what the Dutch parliament unanimously requested from the government in 2002. Open standards, such as the ones mentioned above, allow computers of many different types and the people using them to communicate.

The Internet is a network of billions of computing devices made by thousands of different vendors and these can all communicate through open standards. Because of this I can e-mail other people without having to know anything about the computer they use. As long as we all agree to adhere to the standards information sharing is possible with a variety of systems and products. Open standards make this possible and should be the cornerstone of any governmental IT-architecture.

I've had many conversations with civil servants on this issue over the years and I conclude from these that the vast majority of them have no conceptual understanding of the difference between (the use of) standards and product mono-culture. In 2005 I was speaking at the Ministry of the Interior to a civil servant who enthusiastically talked about the 'standard' desktop they were developing for the entire government (350.000 desktop in the Netherlands). When I asked him what standards they were using he responded with a list of products: 'this version of an operating system, that version of a word processor', and so on. I tried not to cry. I really did.

In the 'Gold'-documentation (yes, the project name is a gift to punsters) the same way of thinking is pervasive; in order to collaborate, all civil servants need to use the exact same software products. I call bullshit. One of the primary recommendations of a counter-expertise report was to replace product-specific standards by open standards. This would enable interoperability with a variety of systems in the future and thus free the government of its dependence on a single vendor. This remains one of the primary stated goals of the national policy set by the government and applauded by parliament.

I can imagine how 15 years of vendor-lock-in-by-standards would lead people to believe computer systems can only share data if they are all supplied by the same vendor. This is however not a law of nature but a symptom of a badly functioning market. It is a problem to be solved by making a policy plan and sticking to it. The national action plan obliges the government to actively seek solutions to the problems of interoperability by the mandatory use of open standards.

In a letter to parliament the responsible minister for the 'Gold' plan claims two things:

1. With the new 'standard' desktop civil servants will be able to collaborate better. It is not clear why they cannot collaborate now. Don't they all have e-mail and a phone? What 'collaboration' and 'better' means in this context is never defined so it's kind of hard to verify this claim.

2. By implementing a mono-culture of Gold PC's the cost of systems maintenance and administration will be reduced to a level that will compensate the cost of procuring and implementing them in the first place. Not that this claim requires a detailed understanding of the existing level of cost.

MPs I work with will be seeking freedom-of-information acces to all relevant documents pertaining to this project from 2006 onwards. I'm looking forward to being proven wrong but I'm not holding my breath.

7-year olds get school laptops... in Peru

Wouldn't it be great for education if all school kids had their own laptops? Well, in Peru this is happening. And in Uruguay, Mongolia, Libya and Ethiopia. The laptops are designed and developed by a non-profit foundation called One Laptop Per Child (OLPC) started by Nicolas Negroponte, formerly of the MIT Medialab.

The OLPC is a technical miracle. It was in fact considered impossible by all the major commercial manufacturers and software builders like Intel, Dell, IBM and Microsoft. That did not hold back the OLPC team though and they build something not only affordable ($150-$180) but also rugged, maintainable and very, very eco-friendly. My Mac laptop eats 30-80 watts, the OLPC does 1-2 watt. That matters when your primary source of electricity is a $10 solar panel. If the world switched to OLPC's from their current systems the savings on power cost (even excluding the environmental cost!) would be sufficient to buy 50 million OLPC's per year for the third world.

On the software side the OLPC foundation chose to use open source software wherever possible, both to keep the cost to their clients as close to zero as possible but also to allow local communities to adapt the software to their needs, wishes, languages and cultural conventions. All the specifications of the laptop (both hard- and software are available for public scrutiny and improvement from the OLPC wiki. This guarantees the independence of nations basing their education on OLPC devices, they will remain in control over the content and methods of education because they have full control over the technology.

But the technical brilliance is not the most important part of the story. The way digital teaching aids allow for new ways of disseminating knowledge is. When every student has a networked digital device that can store and display digital information, schoolbooks can be fully digitized. That gives opportunities to free the ownership of the information in them from the classic model of big publishers that exert control over what goes in the books. A schoolbook is then just a part of a wiki-type system where teachers, experts and other interested parties can collaborate on educational materials that is then shared freely with the world. Some of these contributers may be financially rewarded for such work in the same way we now financially reward publishers. The output however would belong to all of us and we would not have to buy the same knowledge over and over again.I've written more about this here.

The interesting question for western countries is whether they will lag behind places like Peru and Mongolia in modern education. Western European countries typically spend 150-300 euros per child per year on schoolbooks so there's no question that the budget is there. It just needs to be spent slightly differently.

Much has been made of the fact that teachers would not be able to handle the technology. But teachers don't teach kids these laptops, they mostly teach each other. But don't take my word for it. Listen to the 9-year old expert:

Some interesting ideas from Robert X Cringley on the future of education if all kids have full access to online information resources. As often with these type of articles there are as many good ideas in the comments as in the article itself