OLPC Products: XO, Server, Charger, Antenna, & Repeater


In this weekend's , OLPC VP Walter Bender casually drops an OLPC product lineup bombshell:

We are shipping five products this fall: (1) the XO laptop; (2) a school server; (3) a multi-battery charger; (4) an active antenna; and (5) a solar-powered WiFi repeater. Much of the emphasis has been on the laptop, but a push from Quanta this week has resulted in firmer plans for the other products.
gang charger
OLPC multi-battery "gang charger"
While I am glad that we're starting to see all the OLPC peripherals, Walter Bender's list seems a little ad-hock. We are all intimately familiar with the OLPC XO, but what are the other four "products" that Water speaks of?

The last time we checked, school servers were still very much an idea or barely Beta, and now they're going to be production ready? The multi-battery charger, or "gang charger" is even more mysterious. There is a basic Wiki entry, and now a photograph, but not much else.

Walter tells us about the active antenna for the first time in the very same update:

Thanks to John Watlington and the team from Cozybit, we have out first working "active antenna" prototypes. Attaching them to an XO lets you optimize the placement of the antenna: use with a mesh portal will double the network throughput. They can be used on the school servers or attached a 5V power supply to build a stand-alone WiFI repeater.
Luckily, thanks to Aaron Kaplan, we now know more about the solar mesh repeaters, but that was only last week.

I would've hoped we could have a little more detail on what each product is - spec sheets, photos, even developer chatter. An open peripherals development process to dovetail with the generally open laptop development and very Open Source software development. But maybe I am just too demanding.

human power
OLPC human power generator

I don't think its too demanding to wonder what will be the cost of all these products that OLPC will ship in addition to OLPC XO's. In essence, how much will a full OLPC implementation cost? If its $176 per laptop, then how much is it for laptop + four more products? Maybe that's why the original MOU with Libya (before they dropped their order numbers) came out to $208 per laptop?

And will children still be the OLPC maintenance plan once school servers or tree-mounted repeaters are included? Both are more complex to repair and/or have a greater impact if the repair isn't successful, than a single laptop. Not that single laptops aren't Humpty Dumpty all by themselves.

Last but not least, what every happened to human power generation? Where are the yoyo's in the OLPC product mix and how much will they cost? Or is OLPC relying on grid powered gang chargers and long battery life to illuminate households at night?

Related Entries



I seem to have difficulty in interpreting you post.

Do you think this line-up is a setback for the OLPC, or will it make the XO more functional?

Is the availability of peripherals that can, but must not, be bought detrimental to people who don't need them?

Is the availability of the school server good or bad for the project?

Although these questions seem simple, I cannot see what you think of them. You seem to have an opinion on their effect on the future of the OLPC, but it escapes me what that opinion is.

(maybe it is partly due to my second language mastering of English, but I am not alone here in having learned English only in and after high-school)


I have to agree w/ Winter. Walter announced these products in his weekly e-mail to enthusiasts (like you and me). It's a little early in the game to demand extensive documentation. When Negroponte announces it at the G8 Summit, then we should expect a whole lot more documentation.

I think its great that OLPC is looking beyond the laptop at all the peripherals required for a successful implementation: School servers, for example, are integral to storage & connectivity for the OLPC XO's. The WiFi repeaters are brilliant to expand connectivity. But if mass production & distribution is scheduled for September, I would've expected the products to be more baked by mid-June.

No matter their production status, I'm more concerned about the cost. If each school requires a server + every class a gang charger + how ever many antennas and WiFi repeaters, what will the total country bill be, in addition to $176 per child?

Will these peripherals add another $10, $50, $100 per child? Or $10, $50, $100 million per country? It's not like $176 x every child was affordable for most target countries to begin with.

Also, the addition of these peripherals makes the continued "$100 laptop" marketing even more egregious.

The peripherals are essential for the OLPC ecossystem, not different than the iPod ecossystem. Again, just like apple I doubt they will be required to make the XO work, but MIT is just offering options of great products (Negroponte-branded), instead of a generic taiwan server/antennas and etc.

But then again, if MIT is taking pages from appleĀ“s book, they need a steve jobs keynote to announce their products..

No need to worry: that's just more make-believe crap from the OLPC's quarters. They will NOT ship anything in September 2007.

But then again, the word "ship" means, to Negroponte and followers, "sending yet another prototype to my PR person in Brazil or one of the other 4 shill countries".

Kids in K-6 (the OLPC's target group) DO NOT need PERSONAL laptops. They need better classrooms, more food, more teachers and a single computer lab room where they can be taken every day for one hour to learn basic computing skills under SUPERVISED conditions.

That's the way it is done in wealthy nations. Even in the USA, where the government can spend any amount of money on laptops or anything they can think of - good or bad - , laptops are not given to kids, because they would become simple toys, devoid of any educational value. That's why Negroponte CAN NOT sell the idea in his own backyard. His gringo friends are not stupid, so he goes to the little "uncivilized lands" where he can get gold for his little mirrors.

Let's make a rough cost estimation:
Target countries typically need per one million XOs:
1'000'000 x $176 = $176'000'000 for the XOs themselves
5'000 x $300 = $1'500'000 for school servers
100'000 x $100 = $10'000'000 for gang chargers
5'000 x $20 = $100'000 for repeaters + solar panel
20'000 x $10 = $200'000 for active antennas
1'000 x $500 = $500'000 for satellite dishes
200'000 x $10 = $2'000'000 for power yo-yo

This amounts to $190'300'000 for 1 million XOs or $190.3 per XO.
Here are not yet included distribution and installation cost and teacher training and cost for digitizing school books and new teaching content. Remember, from the receiving harbor to the schools the countries are responsible for that. For all this there are about 18$ per laptop left (in case of Libya). And all this is just purchasing cost. The running cost for internet access, power, maintenance is yet another contribution to TCO.
This estimation may, of course, be wrong by a factor of two. And I think it is on the optimistic side.

I don't understand how the gang charger will be applied. The picture shows it has 10 large slots big enough to take 10 complete XOs.
So when the XOs are in the charger how long will the kids have to wait to get the XOs back. I remember another picture from one of the pilots, I guess, where many individual power plug chargers were put in one bar with 10 sockets. So the kids could use their XOs while charging.

What's the use for those gang chargers then? Is it to give the teacher an excuse to separate the kids from their laptops for a while?
Will the XOs be delivered with the individual charger plugs also for home charging? Or is the idea to force the kids to charge only in school enforcing attendance?

Roland may have something with the concept of gang chargers to immobilize the XOs during class time. Very perceptive!

Roland's number of one yo-yo per 5 XOs is a little low, given the claimed 10:1 run:charge ratio. Such a ratio of chargers to laptops would mean that each yo-yo would be in use 50 percent of the operating time of the laptops. I would be concerned for the lifespan of the yo-yo in that case, not to mention the physical condition of the string-pullers.

This, of course, assumes that all charging is done by yo-yo, but there are many communities where mains ("line" to Americans) power is not available at all. I would set the ratio of yo-yos to laptops at no less than 1:2.

Oh, and Wayan, "ad hoc" is not spelled with a "k", even though it sounds like it.

When you think again about the gang chargers you find that they allow the school/teacher a higher level of control over the usage of the XOs. (Of course only if no home chargers are handed out.)

Ways of control in school:
Kids want to charge for their spare time. So they pay for it by some school time without XOs.
In order to charge they need to show up in the classroom every day.

Ways of control out of school:
The time the XOs are allowed to charge in school determines how long the kids can play in their sparetime. Power is becoming a method of reward/punishment.
Interactive gaming uses up power faster than reading. This encourages also to read not only to play.
Handing out (and collecting back) home chargers for limited or continued time can be a condition to do some group project work or a general reward/punishment of any kind.
The battery capacity limits the computer time over the weekends providing time also for other things than computing.

Is this the reason why W. Bender reports in item 8 of his Laptop News from 2007-06-09 that they have reduced the capacity of the XO-batteries?
The question remains whether this is good or bad. It depends on the application of these new ways of control.
And it could create a market for third party home chargers.

Then again when you think harder about those third party chargers you may find even more pieces to the puzzle of XO gang charging.

The kids could try to circumvent the school's control over power by hooking up non-OLPC power sources to the power contacts of the complete XO. Or they could take out the battery pack and try to charge it alone.

How could kids be prevented to do this? By mechanical coding of the power receptacle and the contacts of the battery pack of the XO? Or by electrical coding?

I guess the charge controller is on the mainboard. And the battery pack has at least a temperature sensor.

Electrical coding during charging of the whole XO would be easy since the charge control electronics could easily test for quite sophisticated properties of the original power source. However, how would you prevent charging of the battery pack by an external charge controller? You would have to add a very simple and cheap blocking electronics that only allows charging the battery when it is hooked up to the original XO charge controller. Also this could easily be possible e.g. on the same miniboard of the battery's temperature sensor.

Both protections could be achieved by hardware. Software control is not as safe because the kids are expected to learn to rewrite XO software. The bar lies much higher if a soldering iron was necessary to cheat.

This leaves some more questions:
a) is this just speculation or is it true?
b) What sort of control will be executed via internet access?
c) Are there other sorts of control execution? Maybe remotely switching the device off via the anti-theft security mechanism?

Has anybody supporting or denying information?

Actually the charging protection supposed above would simultaneously serve as a anti-theft protection. If the kids are not handed out home chargers the XO is useless without access to the gang charger in school.

Actually, most if not all laptop bateries include a power monitoring microprocessor (it's what tells you the charge status of the battery), communicating to the laptop using a standardized protocol.

In theory you could add code to it which would require a password from the charger to operate, however the problem of individually coding each processor with a different password would make the whole process unworkable, especially for a project predicated upon Soviet-style mass production at low cost.

More likely is that one fixed password would be encoded into the processor, which would require that thieves also steal a gang charger to circumvent the protection. Since this would be quite easy, I conclude that the protection would not be encoded into the battery.

The XO's typical running power needs escape me, but I recall it was just a few Watts at perhaps 12V. As 12V 1W amorphous PV "trickle solar chargers" globally abound for ~US$15,I can't really see why folks are so focused on the pull cord & solar issues/offerings above, as almost anyone handy could soon rig up a DIY PV to suit an XO. Numerous design variations on such panels exist => Google. A good 8 hours of sun would provide enough charge for maybe 4 hours of XO running, & down trodden parts of the world are certainly not short of a solar resource...

Stan, you wrote:
"I can't really see why folks are so focused on the pull cord & solar issues/offerings above, as almost anyone handy could soon rig up a DIY PV to suit an XO."

I think the reasons are of non-technical nature.
a) When you need millions of chargers DIY method seems not very efficient. If you had to supply all the parts and tools logistics would get complicated and rather expensive. And the recipients are 6-year-olds.
b) As we suspected above maybe charging an XO with any arbitrary power source was intentionally prevented. Charging might only be possible with original OLPC chargers in order for the schools to execute some control and for theft protection.
c) Maybe in some countries the pull cord yo-yo is the only charger provided outside of the schools. And its tediousness might be intentional. First to still motivate kids to show up in school for charging and second to make it unattractive for thieves.

Roland- sorry but those reasons don't really stack up!

a) If millions of chargers are indeed needed,& someone handy can rustle "one offs" 1W PVs for US$20, then mass production should be able to do it for US$10 as you well know! There are numerous such handy solar "pocket chargers" already on the retail market, widely used for LED lamps,cell phones & AA batteries -just Google. Of course rugged,roll up flexible PVs are the way to go,but these are hugely costly for their power output, appealing mostly to the boating wealthy & the like!

b)Like most engineers,I'm dead against specialized charging sockets, as when they fail the entire appliance may become useless. I've had this happen in the past with IBM ThinkPads, Ericsson cell phones & specialised Yaesu radio gear. GRRR ! Of course, given the XO's widely promoted "26 screws to totally disassemble" feature,any one with a screw driver could readily access the internal batteries for DIY charging anyway. Naturally they may wreck the XO while tinkering around,& also supply it with unsuitable charging power (raw AC etc).

c)Yikes! Showing up at school just to charge batteries is akin perhaps to having to walk into town just to have your shoes polished! Even if the XO runs 6-12 hours between charges,there will be bad weather,holidays & w/ends when the school will be closed or kids remain at home- with flat batteries. I've visions of 6 yo. waifs trudging thru' the tundra & savanah with their dead OLPC, only to wastefully stooge around at a near empty school while awaiting a charge...

More OLPC gang charger news:

The industrial design for the multi-battery charger has been approved. A full mechanicals package has been created and is off to tooling vendors to get some quotes.

Gecko had a stereolithographic (SLA) model of the front panel build with five battery slots in order to test the battery retention system, which is designed to allow battery removal with one hand. We should have a mechanical model in Cambridge soon for evaluation.

Flextronics finished up the thermal simulation of the inside of the charger. We are looking at a 15-degree C rise over ambient temperature. Simulations are showing that we are well within the operating range of the components while operating at our target of 50-degree C ambient temperature.