I love Linux User Groups (LUGs). For those of us that live on the 10,000 foot level in corporate IT or work in technology services firms and spend less and less time in front of hardware and the console, LUGs are an excellent way of learning about what is being used in the trenches, and how regular everyday people are using Linux to solve everyday and creative problems.
One thing I’ve learned about LUG members is that they are frugal as hell — most of these guys don’t have multi-million-dollar IT budgets or even budgets in the thousands. They keep their PCs for years at a time, and some of them are genuine system pack rats and dumpster divers, who will build Mad Max-style boxes or “Frankenputers” using whatever parts come their way.
One PC packrat I met at a recent NYLUG meeting is named Ghassan, an Israeli-born Palestinian academician who works at a well-known New York City-based university. He asked me for advice how one might best set up labs in the Holy Land for Palestinian and Israeli children (Jews and Muslims) so they can collaborate on educational and cross-cultural projects to promote peace and understanding.
And he wants to do it using re-purposed, outdated machines what you and I might consider to be junk: four- and five- and seven-year-old Pentiums that would have problems running even the latest Linux distros, let alone Windows XP. If this wasn’t a calling from a Higher Authority to perform my quarterly mitzvah and stack a few points on the Karma Airlines frequent flyer’s program, I don’t know what is.
My first reaction, other than, “How can I help this guy?” was to think about what Open Source technologies he had at his disposal. Certainly, Edubuntu with its built in LTSP server, combined with a new $500 Taipei special with two Ethernet ports, a $50 Ethernet switch, some Cat-5 and a dozen of those outdated PCs and monitors with the hard drives torn out and PXE boot turned on, along with a SOHO router and cheap broadband would get him off the ground. That’s what we’re planning to do as the prototype in New Jersey, where we can test it out on Jewish and Muslim kids at a local community center.
But we need to scale much beyond some suburban town in New Jersey. If we want to do this all over Israel and in the Palestinian territories, we’re gonna need a much more inexpensive solution unless we get a ton of donated equipment. And while donated garbage PCs are certainly welcome, they do have the disadvantage of being power-hungry and need to be maintained or replaced: they have maintenance overhead. Additionally, not everyone can hook up a switched Ethernet network, or even hook into SOHO broadband via cable or DSL. Although Israel itself is a very modern country, there are parts of the Palestinian territories where the fastest net access is still dialup, or 56K GSM cellular data if you are lucky.
So I also searched around and found more Thin Client stuff. I looked of course at the usual suspects, such as DisklessWorkstations.com and Symbio.com, who sell $200 LTSP terminals. That’s great for corps, but that’s gonna blow our budget even for our test LAN. But then I remembered Ndiyo.
Based in the UK, and founded by Quentin Stafford-Fraser, a computer scientist who worked on the VNC project, their aim is to create a cigarette pack-sized device, dubbed the Nivo, that is completely solid-state and diskless, and connects to an Ethernet LAN and a mouse, keyboard and monitor. They even have one that can use USB as the network interface thru a USB hub instead of Ethernet, for smaller five or six station networks. It uses an embedded thin remote console protocol combined with a session manager which can run any Linux desktop GUI on it.
With a network of Nivos, a few extra monitors, keyboards, and a modest PC to act as a server, a whole classroom of kids can work on a whole slew of software. They’ve even done field tests in third-world nations with just a cell phone as the Internet link, and it worked amazingly well — you can see a video on the Ndiyo site and see for yourself. With economies of scale and manufacturing, the target price of a Nivo device is slated to be around $50, so you could even build this into any PC monitor for marginal added cost, say for an additional $25-$30.
What do basic 15-inch LCD screens cost now, $120-$140 wholesale? And with OLED technology, that’s likely to drop in half within the next two years. Unfortunately, right now, because of limited production runs and limited funding, the development versions of the Nivos cost $400 apiece. So it will be a few years yet before we can think about deploying them abroad, or even in Jersey. The LTSP Mad Max configurations will have to suffice, for the time being.
I keep thinking of the underpowered OLPC XO and what kids in 3rd world nations and in under funded inner-city schools in our own country could do with a full-blown, Open Source desktop OS. Sure, the XO is a good idea, and I am sure lots of kids will benefit from it– but I have to think all those $200 donations could have been better spent on getting real computing in the hands of the masses instead, using proven off-the-shelf and sensible technology instead of reinventing the wheel.
A dozen Nivos in every classroom. Can we make that a reality? Please?
Loading ...
