Slashdot Mirror
OLPC Developers Boost Security
eldavojohn writes "The developers of software for the One Laptop Per Child (OLPC) initiative are redefining security for the personal PC. Since the laptops have the potential of communicating with any other laptop, the developers have a unique opportunity to implement both virus protection on the kernel, master boot record and also the way in which the laptops deal with security and 'code-sharing.' The developers are currently seeking outside counsel from security experts and if you're worried about these security schemes posing only problems to the children, 'these security measures can be turned off by the PCs' owners. To protect against that leading to disaster, the laptops will automatically back up their data up on a server whenever the machines get in wireless range of the children's school. If a child loses data, the files can be restored by bringing the laptop within wireless range of the server.'"
This sounds like a really clever idea. I'm suprised there's not more demand for it in school/colleges now, and it has taken a project like this to actual take steps to do it.
That being said, this project has been in the news for a LONG while now (at least it'd seem), are they any closer to actually reaching their goal?
> but it strikes me that they are implementing many unique (not "proprietary", but same difference) technologies
;). Anyway these laptops are going to kids who have never owned a computer, so they do not have to be compatible with all the legacy crap that we own here in the rich west.
./configure && make && make install
This is a "good thing", this will be the first ever laptop designed around the needs of education. This is also the first ever laptop designed just for Linux, loads of legacy hardware (BIOS, Serial cable) is not needed.
The hardware, software and bundled books and homework ('content' if you like) are all being specifically re-designed.
>The Sugar UI, networking, now security, and probably a bunch of things I don't know about.
I am sure there are many good things that some people do not know about
I think the Sugar UI is great. All the current desktop UI's, e.g. Windows Vista, OS X, Gnome, etc have the same metaphors as 'Lisa' in 1983. Now there is a UI designed for the Internet age, it has messenging and collobration as the core of the system.
>Where will they find apps for this platform?
It is still a Linux Distribution.I assume you are a Windows user (sorry if you are not), but most if not all the important sourceforge.net apps will work, because they are 'source', they will be recompiled for this Linux Distribution, not just by anyone, they have Red Hat behind them.
To get most apps working will just take:
>Is there an IDE for OPLC? Other tools?
I think the plan is for as much as possible to be in Python,so you would just use the normal Python IDEs and tools.
> * Will OLPC supply all apps themselves?
Redhat and chums will provide the inital batch. Python apps will probably 'just work', some other apps may need some tinkering if the laptops do not have all the most famous libraries, for example, I cannot imagine them having both GTK and QT.
>Perhaps this has been long ago addressed, or I over-estimate the problems or challenges, but I would be interested in the answers.
I think you are overestimating a bit. It is Fedora Linux tweaked a lot, but it is still Linux. I think the users will run out of space on the device long before they run out programs.
My little Linux and tech blog
Any computational system which is fertile for infection will eventually attract viruses. It's like a law of nature.
If this initiative is successful, Linux could become the most widely used operating system in the world, and would therefore be the biggest target for infection. Virus writers would turn their attentions from Windows to Linux. Our experience with Firefox has shown the falsehood of confident expectation that open sourced software would be immune to malware, and it's only got 10% of the market. Any system with the infection potential of these machines is going to be targeted and exploited.
Imagine a 0day exploit for the wireless driver like we saw a few weeks ago. These things hook up into mesh networks. You could start with a single point of infection which would spread from machine to machine, cover a whole city and then a whole continent within hours. The virus potential is extremely serious and deserves careful attention in the software design.
I live and work in Melanesia, I can say for a fact that the cargo cults (and their mentality) are alive and well here. The impact of this mentality on development is significant. It's quite common for people to, for example, request a truck - or a computer - from a donor and then run it into the ground. No maintenance, no care taken whatsoever for its sustainability. Why? Because when it breaks, all they have to do is ask for another one.
It's pretty frustrating, to say the least. The attitude extends from the most remote islands (where some groups still dress in mock-US Army uniforms and parade in order to induce the heavenly powers to bring back the largesse that accompanied the Allied presence in WWII) to the highest levels of government. People here have come to expect easy profits from the developed world, and for its sins, the developed world complies.
That said, there's a real need for improved access to information and communications. People like me and my colleagues have been working together for years to improve the situation, and one of the ways to address the cargo cult mentiality is to stop giving things to people and start getting them to pay. Cheap laptops are a perfect vehicle for this.
USD 100 represents about 2 weeks' wages here (for the minority who are employed), and that's a pretty ideal price for something like this. It's an investment that parents are willing to make in their children's future, one which has enough value for them to insist that their children actually use them.
There's a tremendous interest in computers here. It's universally recognised that IT-related work is something that can create real economic wealth without as many problems as commercial agriculture, resource extraction and tourism. People can compete on the world market for employment and bring real prosperity into their communities for the first time.
The OLPC doesn't fill in every piece of this puzzle, but it's a damn good start.
Crumb's Corollary: Never bring a knife to a bun fight.