Darren Waters

First impressions: Classmate 2

  • Darren Waters
  • 17 Apr 08, 15:19 GMT

id="showplayer"> type="application/x-shockwave-flash">

I'm writing this post on one of Intel's new generation of laptops designed for the developing world - the Classmate PC - and it's quite a task, mainly because my fat fingers are too large for the keyboard.

This is the machine that arguably ended Intel's relationship with One Laptop Per Child (OLPC), makers of the XO laptop. According to Intel, OLPC wanted the firm to stop making the Classmate. Intel refused and pulled out of their deal.

Whatever the truth - the Classmate is definitely a rival to the XO's ambitions in the developing world generally and in education in particular

I've been playing with the Classmate for a few days and my first impression is that the two laptops herald from polar opposite philosophies about how to end the digital divide.

We've written lots about the XO and the first Classmate, so I won't repeat myself.

Certainly, the Classmate 2 is exactly the sort of machine you would expect a commercially driven firm to make for the developing world.

It's a cut-price, cut-down laptop that runs XP moderately well, and connects to the net without a hitch. In fact, it accomplishes most tasks thrown at it without a hitch and its underpowered processor only really struggles when it is attempting to multi-task.

It has been designed with education and the developing world in mind, says Intel, and yes, there are a small number of unique features. It has mesh networking - so that laptops can piggyback their wireless and create ad hoc networks.

But this feels like a feature pilfered from the XO and added only because it was such a glaring oversight in the first machine.

When you first connect to a wireless network it searches for other teachers' laptops on the network. It also comes with a nifty handle velcro-ed on to the machine for carrying and has a "water resistant" keyboard.

But that's about it, to be honest. It's small, well-made, quite rugged, with a decent keyboard and decent screen, albeit in only 800x400 resolution.

The key difference between the Classmate and XO is the different approach to software and operating system.

The XO comes with a variation of Linux installed - a user interface designed specifically for education, and applications built to support the educational environment.

The Classmate machine I'm using comes with XP loaded as standard - but there are options to have Linux.

And while there are plans afoot to port XP to the XO machine, it remains by and large a Linux machine.

There is educational software pre-loaded onto the Classmate machine and a teacher can monitor the work of children from a host machine.

The use of Windows gives the Classmate a recognition and sense of security to governments and educators looking to buy laptops for schools.

After all, Windows is the world's most dominant operating system.

So what else has changed with the second generation Classmate laptop?

Not much: the second gen laptop now has two models (7 inch and 9 inch screen) and has a built-in webcam.

It's clear the re-design of the Classmate is more about making the machine more design friendly to consumers and educators in the developed world than improving the internals of the machine.

Last year nine-year-old Rufus Cellan-Jones gave his impressions of the XO laptop. We've given him a 7-inch model and expect his comments some time next week.

The machines go on sale for between $300 and $500, which is a lot more expensive than an XO, more expensive than an Asus EEE PC and probably a lot more than many educators can afford to pay.

Nicholas Negroponte's dream of one laptop per child - whether an XO or a Classmate, is still some way off.

PS: The BBC has now upgraded its blogging software so that, amongst other things, it better withstands the volume of comments we get. From today if you want to comment on any BBC blog you will need to register first. You can read more about this on the Editors' blog.


  • Comment number 1.


    Have to disagree with you about the comment on Mesh.

    Proprietary mesh solutions have been around for a while as have open source solutions based upon AODV and OLSR (the two base research approaches to mesh routing). The problem is Mesh networking doesnt really deliver or scale unless you move to multiple radios or are providing backbone infrastructure... there are ALOT of issues!!

    Intel folks correctly did not provide mesh networking because it cant deliver on the promises that OLPC folks knowingly misled on. You need only look at OLPC's experiences in Mongolia with mesh networking or elsewhere to see that they still have not solved the issues.

    "We have painfully discovered the limitations of the mesh and current collaborative software in Mongolia, where the convolution of the number of laptops with bugs #5335 (more mDNS traffic than expected) and #5007 (mesh repeats multicast too much) make the perfect storm, which prevents anybody from using the network. "

    Mesh networks do not scale in the manner required to be used inside schools or even by 100s of kids in an area. Not yet anyway! So Intel took the smart approach. Many of us had told OLPC folks that these issues existed. They chose not to listen.

    Intel folks have taken the standard based approach and are following the IEEE 802.11s standard - something OLPC folks did not do. Ask yourself which is more responsible? Introduce a technology too early in emerging markets and see it fail, or take the time to ensure it works first before your preach?

    As a mesh networks researcher, I use to cringe everytime I read OLPC stories about Mesh networking or Negroponte's "lily pads and frogs".

    I totally agree and aspire to the vision, just not OLPCs approach to pushing the technology as the solution to digital inclusion etc. That was purposely misleading!

    Sadly it leads to comments from writers like yourself with respect to Mesh Networking addition in Classmate PC - "But this feels like a feature pilfered from the XO and added only because it was such a glaring oversight in the first machine."

    Why is this a "glaring oversight"?
    Show me where Mesh technology is succesfully being deployed and used in OLPC deployments to breach the digital divide or provide mass connectivity without infrastructure?

    It simply isnt.

    Will it get there eventually? I hope so... but it isn't there yet...

    If you would like to discuss this more, please feel free to email me.



The BBC is not responsible for the content of external internet sites