Status-Q

I’ve posted a picture of the innards of the next Ndiyo Nivo on the Ndiyo blog.

Some good press coverage this week for some projects close to my heart:

• Business Weekly on Ndiyo
• ZD Net UK on Ndiyo
• Business Weekly on DisplayLink

Duncan Goldie-Scot pointed me at John C. Dvorak’s article: The $100 Laptop: What went wrong? Anyone familiar with JCD will recognise that the dismissive past tense for a project that is just warming up is simply part of his normal marketing style. He raises some good questions, though…

…I myself have moaned about the details of this One Laptop Per Child scheme as folly or idealistic. The basic argument is that with $100 you could almost feed a village for a year, so why waste that sum on a laptop? What are they thinking?

But Zachary has a more profound point: “The fact that these people need electricity more than they need a laptop is only part of the problem,” he says. “The real problem is lost mind share. The people are harmed because these sorts of schemes are sopping up mind-share time of the people who might be doing something actually useful.”

To summarize, there are only so many hours in the day, and we should not be wasting them on this kind of naïve feel-good showboating. Let’s face it: These high-tech gems are a laughable addition to a mud hut.

Classic Dvorak. (I’m very skeptical, in passing, about $100 being enough to feed a village for a year. I read a report on OLPC recently which said that average governmental spending in the countries concerned is around $200/head, so $100 spread over the laptop’s expected 5-year life sounds much more reasonable.)

However, I also don’t buy his larger argument, for various reasons.

The first is that different people are inspired to make the world a better place in different ways, and resources, innovation and people can’t be shuffled from project to project in quite such a simple fashion – inspiration is not subject to the rules of double-entry bookkeeping. Besides, this is an old story: In the seventies the big argument was that we shouldn’t be making weapons when there were people needing ploughshares. Today, people like me probably shouldn’t pay a premium for organic food when there are others going hungry in Africa. These are both true, but they are pie-in-the-sky arguments because nothing is that simple in reality. If my hotel decided not to have such a nice carpet in its lobby, would that mean that there were fewer homeless people on the streets? Of course not. And if the OLPC project weren’t there it doesn’t mean that the resources that RedHat, say, are putting into it would automatically be diverted to water purification.

The second is that it can be rather too easy for the wealthy part of the world to decide the priorities for the less wealthy part. Perhaps the best example that I’ve come across in my work is the spread of mobile phones. Did you know that over a third of the world’s population have a mobile phone now? The number is growing by more than a million users per day, and it will be less than five years before half the world have cellphones. “Why on earth”, we might have asked ten years ago, “would somebody in an African village with a poor water and electricity supply want a mobile phone?” But that’s because we take for granted things like the ability to travel, or the reliability of postal services, and forget that the simple ability to transmit and receive information over a distance is incredibly empowering. Mobile phones are transforming millions of lives. We don’t know exactly how people around the world would use a PC if they had access to one, but it is one of mankind’s most sophisticated and flexible tools, and I don’t think it’s our place to deny it to others.

I have many questions about OLPC, some of which we’re trying to address at Ndiyo. I doubt that it will succeed on the scale that its founders hope. And I think there’s a good argument that, for example, Dean Kamen’s latest initiative around water purification and electricity generation is rather more important than both projects.

But I have huge admiration for people who dream a big impossible dream and work hard to make it a reality. Give me the option of a world with more well-intentioned visionaries, or a world with more armchair cynics, and I know which one I’d choose!

Hamachi looks interesting. It allows you to create a VPN – a private network – between arbitrary groups of friends or co-workers, and join or leave it simply by clicking a button.

Normally, I detest VPNs. They’re a good idea in theory – make you look as if you’re on your office network when you’re somewhere else – but my experiences of them at three different organisations has not been good. Fortunately I’m geeky enough to find alternative solutions using things like SSH, VNC, SFTP and WebDAV, which generally have performed much better. These days, though, if you have many mobile workers, it can be better to put your company services out there on the internet (with appropriate security measures, of course) where they can be accessed from your own network and from elsewhere.

However, more peer-to-peerish solutions like Hamachi could be very useful if, say, you want to back up your mother’s PC or print on your brother’s printer. As well as Windows support, there are console versions for Mac and Linux, and a new version with a GUI for Mac OS X.

Disclaimer – I haven’t tried any of these. Just an interested observer.

Ndiyo, and Yours Truly, are mentioned in this week’s Economist.
John has an extract.

We’ve been playing with a new way to make a single PC into a multi-user machine. It’s based around the idea that a simple USB hub could be the basis of a thin-client terminal. We call it Hubster.

More information on the Ndiyo site.

Have posted some thoughts about this over on the Ndiyo blog.

I’ve started a new blog for more informal news about what’s going on at Ndiyo.

Please take a look at it and subscribe to the RSS feed if you’d like to keep in touch!

Status-Q will, of course, continue as normal…

As of tomorrow, there will be a free version of VMware server available.

This is good news – VMware have a great product – and it’s particularly interesting for us at Ndiyo. We’ve been starting to plan some experiments using a single machine to serve terminals with a mix of Linux and Windows – an organisation where one user really has to stay on Windows software could then still make a switch to a predominantly Linux system. Another option is that somebody with a Windows machine might be able to run a virtual Ndiyo server, and support multiple Nivo-based users, without having to pay any more licence fees.

It’ll be interesting to see whether any of this is viable on a modestly-priced PC.

Paul Theroux makes a comparison between Malawi and Ireland in this Sunday Times article.

There’s a very nice post by Nicholas Carr about Newnham Research and Ndiyo.

A key part of our message at Ndiyo has been that the traditional model of ‘a PC under every desk’ was a good one in the 80s and 90s, but it will never be sustainable on a global scale. (We call it the SUV Model of Computing). So we came up with a model that lets us share a PC between several people for a much lower cost than buying one PC each. And because the PC is running Linux, there are normally no extra software licensing costs to be paid when you add extra users, unlike proprietary software, where the licensing costs now often exceed the hardware costs.

The further we move towards web-based services and applications, the less dependent people are on any particular operating system and, as Nicholas points out, the more scope there will be for alternative hardware models. It’s also good for most software companies, too; I recommend Paul Graham’s article The Other Road Ahead if you haven’t read it.