Cancel
Advanced Search
KEYWORDS
CATEGORY
AUTHOR
MONTH

Please click here to take a brief survey

Negroponte's Hundred Dollar Laptop
Jamais Cascio, 25 Jul 05

HDL.jpgThe design for inexpensive, useful networked information devices for the developing world is a recurring topic here. After all, the evidence is strong -- and getting stronger all the time -- that cheap, functionally-ubiquitous information and communication devices help to accelerate development. The political and cultural impact of personal devices allowing for ongoing access to knowledge and widespread group communication is somewhat harder to measure, but it seems likely that there's a positive result there, too.

We talk quite a bit about the form such devices could take, as function is intimately related to form. Will they be souped-up mobile phones? A handheld computer, like the Simputer (WorldChanging ally and former contributor Taran Rampersad got his Simputer recently, and his recent posts about its use definitely make it sound appealing)? Or perhaps something more conventional, like Nicholas Negroponte's suggestion of a cheap ("hundred dollar") laptop for young students across the developing world.

Except... it turns out that Negroponte's design isn't quite as conventional as I thought.

The Hundred Dollar Laptop (HDL) project now has a website, albeit one with limited information. What's there, however, is suggestive of something other than a cheap copy of a Thinkpad or Powerbook:

The $100 Laptop will be a Linux-based, full-color, full-screen laptop, which initially is achieved either by rear projecting the image on a flat screen or by using electronic ink (developed at the MIT Media Lab). In addition, it will be rugged, use innovative power (including wind-up), be WiFi- and cell phone-enabled, and have USB ports galore. Its current specifications are: 500MHz, 1GB, 1 Megapixel. The cost of materials for each laptop is estimated to be approximately $90, which includes the display, as well as the processor and memory, and allows for $10 for contingency or profit.

An article in MIT's Technology Review gives a few more details:

...the HDL will be loaded with Linux and other open-source software; its display will use either a rear-projection screen or a type of electronic ink invented at the MIT Media Lab; and it will store one gigabyte's worth of files in flash memory.

The HDL has a number of other, intriguing features. Since many villages in the poor world do not have electricity, the machines may be powered by either a crank or "parasitic power"--that is, typing. Once turned on, HDLs will automatically connect to one another using a "mesh network" initially developed at MIT and the Media Lab. In the mesh network each laptop serves as an information-relaying node. Households that have HDLs will be able to communicate with each other by e-mail or voice calls.

Most importantly, Negroponte wants every mesh network to have access to the Internet. The laptops will be loaded with Skype, a communications application that provides free telephone calls. Consider: the most forlorn parts of the globe might become part of the wider world.

Let's leave aside for a moment whether something like this can be built at a cost under $100; ambitious, sure, but the price of the required components is dropping so rapidly that a sub-$100 cost is inevitable. What I find much more interesting are the built-in mesh networking and the features intended to deal with a low- or no-electricity situation. These are the real leapfrog features of the HDL.

Mesh networking means, at minimum, each of the devices will be able to communicate with and along any other within range. It gives any single element of the mesh a range out to the most remote member of the mesh: messages and connections can hop from node to node until they get to the target machine, even if that machine is nowhere nearby. Mesh networking also suggests the possibility of CPU sharing, where some processes can be farmed out to machines not currently in use, something like the SETI@Home/BOINC model. A 500MHz processor may not sound like much these days, but if computationally-intense processes can get an assist from other machines on the mesh, 500MHz is more than enough.

At the same time, the use of electronic ink for the display and flash memory for storage means that the power requirements for the laptop might possibly be less than the power needed for a mobile phone, and certainly less than that required by something like an iPod. Batteries for laptops can hold a lot more juice than the much smaller phone batteries. Low consumption plus large storage would allow relatively low-power inputs -- hand cranks, the "parasitic power" of typing (imagine if every time you hit a key, you added a smidge more power back to the battery), even cases made in part out of cheap plastic solar panels -- to provide sufficient electricity to keep the batteries at least semi-charged.

These two elements -- mesh networks and super-efficient power -- make the HDL a device that's really unlike anything else ever mass-produced (despite a surface similarity to conventional laptops). As long as you're relatively close to another HDL in the community, you can potentially be connected to them all; as long as you have the personal energy to turn a crank, type, or sit in the sun, you can keep working/learning as long as you want to. The stereotyped developed world "nerd" sits alone in a darkened room to work; the HDL makes both of those behaviors counterproductive. These systems don't promote isolation -- in fact, they actively discourage it.

Does this mean I'm now a convert to the Hundred Dollar Laptop? No, although I will give the idea more credit than I have in the past. I still think the form factor is a potential problem, as it's too large to fit into a pocket or purse for easy carrying. A system that enables casual use is one most likely to become integrated into a society. Moreover, while there are clearly good keyboard systems for pictographic languages like Chinese, a handheld device would more readily allow for direct on-screen writing; students who know how to write would be able to pick that up without question, rather than having to learn a potentially difficult keyboarding routine.

Regardless, I'd be intrigued to see what could be done with the low-power components and mesh networking features of the HDL on a smaller, more portable device. What could students do with the HDH (Hundred Dollar Handheld), I wonder...

Bookmark and Share


Comments

Well, I don't have to order a million Simputers... :-)


Posted by: Taran on 25 Jul 05

It's a well meaning initiative and I'm all for inexpensive and ubiquitous computing but I'm not sure why this is better than a more achievable $100 handheld computer based on the Palm, Windows Mobile or Symbian operating systems that already exist and just need evolutionary improvements in areas such as battery life, wireless connectivity, local storage, etc.

Isn't the key here (particularly in rural digital divide projects) the infrastructure, i.e. ensuring the wireless (or more usefully Wi-Max) connectivity and the recharging facilities for fuel-cell batteries etc.

I'm not convinced we need better mouse traps just better bait.


Posted by: Graham Brown-Martin on 26 Jul 05

It will be interesting to see how they deal with the battery drain from Wifi (let alone wifimax). This requires serious battery power.


Posted by: Anthony on 26 Jul 05

This covers the most important bases on the technical side by getting rid of the three power hogs on a laptop: screen, CPU and HDD.

On the marketing front, having it for less than $100 with room for some profit seems like the sweet spot. This is cheaper than the Soekris set-up suggested by CU Wireless which was covered here recently (Ad Hoc Network Leapfrogging).

I like the laptop form-factor for students because it makes it much easier to be productive. However, couldn't we just have separate keyboards and screens at their desks which they connect to their PDAs?

Anthony, do you have any numbers on wireless energy use? The Soekris chips only used 10 watts IIRC.


Posted by: Daniel Haran on 26 Jul 05

I thought the Simputer was a serious failure. Why would this device be different?
And why would it change the plight of 50% of humanity, which still lives in the countryside, where urgent investments are needed in far more basic things like water, electricity, roads, and transport.
I know some people do not like to prioritize things when they're talking about development, but I think it does make sense to do so.
Such a cheap computer may come in handy in urban areas, though (even though we must keep the Simputer's failure in mind).


Posted by: Lorenzo on 26 Jul 05

http://www.synack.net/wireless/consumption.html

Perhaps I'm wrong. Most of my experience is with PocketPc.

Looks like Prism2 uses 100 mW.

http://www.synack.net/wireless/consumption.html


Posted by: Anthony on 26 Jul 05

To make things more concrete: how would a cheap computer prevent 4 million Nigerese from starving to death?
How does it allow 40 million Congolese farmers to get their produce to ports for export?


Posted by: Lorenzo on 26 Jul 05

I think the idea is revolutionary, since it enables the 3rd world to apply PC:s, internet, plus all the standard software, that Palm, Pocket PCs and Symbian mobile phones do not allow.

And more that that, personally I'd prefer it over a standard laptop - no batteries to worry about charging, no hazzles of internet setups while changing the physical location, enough storage space to hold all the important data, large screen and standard keyboard.

From my opinion 500 Mhz runs pretty fast for all the usual day-to-day tasks, and the possibility for computing power sharing throught the mesh network sounds cool.

But I'd be very concerned about spam, supplying advanced mesh-based spam-relay network for the 3rd world where people live less than 1 euros per day is unethical.


Posted by: Pekka on 26 Jul 05

I'm an analyst of emerging display technology, and am well familiar with the E Ink technology. I also know a display engineer/entrepreneur who just joined HDL to help with this portion.

The display was originally slated to take $25 of the $100 budget. The first move my acquaintence made was to push that to $30. And her second move was to make the E Ink implementation the "Gen 3" model, after using a simplified TFT-LCD and using an LCOS rear-projection set-up (it used to be Gen 2 after the LCOS implementation).

I agree fully that the electrophoretic (E Ink) option is a great one, given that the display is bistable, requiring power only to change its image. So when it's being used for reading, as it presumably would be frequently in the case of student, it requires no power. And it also requires no backlight if used in reflective mode.

(Of course, I have also heard that some of these villages have poor light sources, and hence the advent of a laptop PC provides a great light source into the evening hours, if the battery can run for a few hours. A reflective display won't have this perk).

As for the $30 price tag, who knows? At this time, E Ink displays cannot be manufactured in volume in laptop sizes. The closest they've come are in the Sony Librie panels, which are about 6 inches (and are monochrome).

The modules used to be assembled by Philips, which was supplying the backplane (E Ink supplies the ink, and Toppan supplies the packaging for the electrophoretic portion). But.... just a couple months ago Philips sold its E Ink manufacturing unit to Prime View International in Taiwan as part of a company-wide move away from display manufacturing. So things are a bit up in the air-- will Prime View keep making E Ink panels in the same way? Their plans are not yet clear.

My view is that HDL should heed the advice to make a Gen 1 version using a simplified LCD (stripped-down optics and backlight, for instance, or maybe even operate the LCD in reflective mode), put it out there, and see what happens. Then some of the surprising social effects can start to be seen-- such as, how do people actually use these things, and what else would they like them to do? -- and can be responded to in the Gen 2 version.

By the time E Ink panels are technically capable of working in a laptop (say, 4-5 years from now), we might have a good idea of how to use them most effectively.


Posted by: Kim Allen on 26 Jul 05

"But I'd be very concerned about spam, supplying advanced mesh-based spam-relay network for the 3rd world where people live less than 1 euros per day is unethical."

That would be cruel indeed. Dying of malaria (a net costs 2 €uros), starving from hunger (food for a year costs 90 €uros), but having a computer spewing spam!


Posted by: Lorenzo on 26 Jul 05

"To make things more concrete: how would a cheap computer prevent 4 million Nigerese from starving to death?"

It doesn't. You are comparing apples and oranges. Read up on the difference between relief/aid and development.


Posted by: Chris Albon on 26 Jul 05

Lots of good comments here. Some responses:

Graham, if you follow the link in the piece to a post I did back in February ("The $100 Dilemma") you'll see that I'm right there with you in terms of favoring handhelds, and noting that much of what the HDL can do is already possible with cheap (sub-$300, some close to $100) devices.

Anthony, with low-power WiFi chips, solar cases, hand-crank generation and "parasitic power" keyboards, the battery drain from wireless connections shouldn't be too bad.

Lorenzo, where did you get the impression that the Simputer was a "serious failure"? It was delayed, but is now out and selling.

As for the larger development question, cellular networks are expanding to rural areas far faster than you seem to believe, so devices along these lines would have plenty of utility in the countryside. Prioritizing aid is fine, but (as you and I have discussed in the past) that doesn't equate to a "first we deal with water, then food, then power, then education..." type strategy. It is quite possible to apply efforts in different regions that reflect local needs -- in some places, clean water, in others, access to communication tools.

Sending HDLs to starving communities would do little good (as a later commenter noted, there's a difference between relief and development). But -- as we've pointed out on WC probably close to a dozen times by now -- communication and information tools *do* have a strongly positive effect on rural communities in terms of access to markets. Would the HDL per se help farmers get better prices for their goods? Possibly not, but more general info/comm devices certainly would.

Pekka, I linked to an article about spam in developing nation information networks here.

Kim, thanks for your insights on this. Given that the e-ink model would be third generation, are low-end LCD displays low enough in cost to be a feasible alternative?


Posted by: Jamais Cascio on 26 Jul 05

Excellent reply.

I wanted to comment on something you said:

"It is quite possible to apply efforts in different regions that reflect local needs -- in some places, clean water, in others, access to communication tools."

I think this could be applied handheld/mobiles and the HDL.

It is true that in many rural regions, the HDL has the real potential of being useless. Small plot and subsistence farming families typically do not have need for a laptop or a word processor. However, being able to call nearby markets to check prices and even to pay for supplies is a major asset.

However, I believe there are many regions (especially around large cities) where computer/office skills are the "golden key" to a middle class job. The HDL would give children a chance to learn these skills, while a mobile phone or handheld would not.


Posted by: Chris Albon on 26 Jul 05

Just to clarify:

Examples of computer/office skills I was refering to:

Word/Open Office
Windows/Linux
Excel
Access
Even programming: xhtml, css, php, etc...


Posted by: Chris Albon on 26 Jul 05

"To make things more concrete: how would a cheap computer prevent 4 million Nigerese from starving to death?"

It doesn't. You are comparing apples and oranges. Read up on the difference between relief/aid and development.
------------

Sure, I agree on that, as an abstraction. But the truth is that really vast areas of this planet have been permanently and structurally reduced to "relief-zones", so much so that the distinction you make disappears entirely. These regions need a kind of massive, permanent aid-like effort, which comes down to development-as-relief, and cell phones or puters really are very low on the list of priorities to get such an effort going.
As a development tool proper, they might be useful though.


"where did you get the impression that the Simputer was a "serious failure"? It was delayed, but is now out and selling."

I thought I had read this several times (here (reference to an AP story: "Simputer is a Failure") and here (The Age, "Simputer fails to take off", april 2005).

I'm not sure though whether this is because of the failure of the technical development of the puter as such or whether it is because marketing efforts and consumer demand was too low. But then, even the poor are rational consumers, so pushing things because they're good in theory won't work, not even there.

I hope it works better after the relaunch, though.


Posted by: Lorenzo on 26 Jul 05

It looks like the "failure" of the Simputer (and I have spent enough years watching the computer industry to be hesitant to call it that yet) is related directly to price. You'll recall, of course, that the cost issue was something we mentioned pretty much every time we talked about the Simputer (and similar items).

$200 base, $300 for something usable, is simply too much. Even $100 is too much, really; the devices need to be below $50 to make sense for much of the world.


Posted by: Jamais Cascio on 26 Jul 05

"But the truth is that really vast areas of this planet have been permanently and structurally reduced to "relief-zones", so much so that the distinction you make disappears entirely. These regions need a kind of massive, permanent aid-like effort, which comes down to development-as-relief"

I agree but these regions are served with aid (food, health, shelter, security) not with development (IT infrastucutre, education, business development, and infrastruture).

We can both agree with HDL fits in the second category.


Posted by: Chris Albon on 26 Jul 05

$200 base, $300 for something usable, is simply too much. Even $100 is too much, really; the devices need to be below $50 to make sense for much of the world.
=========

Agreed, and consider that 50% of the developing world - the target group - is illiterate. (Take India, one of the better performing countries, certainly not the poorest of all, and already "moderately" developed: 40% or 450 million people there are illiterate; let's not even consider sub-saharan Africa...).

Does a simputer make sense here? Does it have some speech recognition software?

Better invest that money in hiring one teacher per three villages (you can buy one for an entire year for the price of three simputers).

I don't think that even a US$50 computer makes sense for illiterate people. A mobile phone is much more useful. Couple a mobile phone to an internet-relay service, which is operated by live-persons. 2 rupees per bit of information... Or invest in local radio.

Seriously, I would love to see the thorough market research that must have motivated the development of the simputer, and the incredible disparity between the expected sales (50 000) and the realized sales (2000) - this must be the worst market research to have ever been carried out. Anyone know if that's publicly available?


Posted by: Lorenzo on 26 Jul 05

I agree but these regions are served with aid (food, health, shelter, security) not with development (IT infrastucutre, education, business development, and infrastruture).

We can both agree with HDL fits in the second category.
===========

Sure, and it would be artificial to keep these two lines very distinct or put people into groups strictly belonging to one category.

But precisely the experience with the simputer shows that there is a very large proportion of the "second group" that instantly buys ordinary products on the global market, once it is considered "developable" ("business development", "education" and "infrastructure" really are things for the extremely well off poor who've made it to a stage where they're almost middle class, still a very very tiny minority).

So your only option then is to target the group that doesn't need one (the masses). And then you become a pusher, which always ends in failure, certainly when you're targetting people who have the keenest sense of what money's really worth.

Moreover, making a computer for well under US$ 50 at realistically small quantities (say 1000 a year) will be nearly impossible (the first simputer's price setting was based on producing hundreds of thousands per year; it sold 2000 over 5 years). And if you don't get it to well under US$ 50, you don't have a market at all.

It's a sad dead end.


Posted by: Lorenzo on 26 Jul 05

Given that the e-ink model would be third generation, are low-end LCD displays low enough in cost to be a feasible alternative?

Yes, with the provision that they be built for the purpose of the HDL, not retooled afterward. Stripping down an already-existing panel only adds to the cost, of course, and currently a simple TFT-LCD in the 8-inch range costs more than $50.

So it becomes a matter of asking a display maker to set aside factory time for making these simple panels. The negotiations are on. It seems possible that a trial run could take place...


Posted by: Kim on 26 Jul 05

"business development", "education" and "infrastructure" really are things for the extremely well off poor who've made it to a stage where they're almost middle class, still a very very tiny minority

===========================================

I disagree.

I spent some time with Worldvision and have seen their incredible success with business development programs such as micro-loans. The micro-loan programs are not targetted towards the "almost middle class" at all, infact they are targetted at the lowest level of society. The micro-loan programs have pushed hundreds of jobless/illiterate poverty stricken women out of food queues and into micro/small business owners.

Infrastructure includes roads, hospitals, irrigation systems, water pumps etc.. etc.. These are in no way targetted at the "almost middle class" and infact many times it is the poorest who benefit from these projects the most (water pumps and irrigration in particular).


Posted by: Chris Albon on 26 Jul 05

Economics is 90% hoax, but... the simple economic idea of marginal cost/marginal benefit seems relevant.

In less-developed countries, among the poorest of the poor, what would be the most sensible thing to do with the next $100 available? Would the next (the marginal) $100 be best used for a computer? If not - if the next $100 would be better spent on a bore well or pit latrine, for instance - how many $100 outlays would one make until clearly the smartest thing to do with the next $100 is to hand someone a cheap computer?

Some years ago, I did some work in a squatter settlement outside Mexico City. My business partner recently spent a couple of months helping build an orphanage in Bolivia. In both cases, $100 worth of good old duct tape would have done wonders - more than I can imagine from a computer.

The sad thing about "marginal cost/marginal benefit" is that it's just a fancy term for "triage".


Posted by: David Foley on 26 Jul 05

Folks, I hate to break this to you - but the reports of the death of the Simputer have been grossly exaggerated. It's in use. It just hasn't been getting as much press as Negroponte's promises, which is understandable given the Western sway of the press - traditional and otherwise.

A representative of MIT's $100 endeavour at the WSIS-LAC meeting this year, in preparation for Geneva, told one of my colleagues that they had not used the Simputer as a model for the alleged $100 laptop because they couldn't get one.

So, here I am - lowly individual - with apparently more resources than MIT's Media Lab. I feel special. And not only that... the plans for it are freely available at Simputer.org.

Negroponte, Negroponte, Negroponte. The Simputer has been in commercial production since last year, and still the Western biased media talks about Negroponte's promises.


Posted by: Taran on 28 Jul 05

Taran, a few things to note:

Lorenzo's links to media reports of the Simputer's "failure" were to Indian media outlets, not western outlets.

MIT couldn't get a Simputer a few months ago, but you could a few weeks ago -- it's entirely possible that Simputers were essentially impossible to get until very recently.

It really should come as no surprise that the head of the MIT Media Lab (with various books under his byline, appearances at Davos, etc. etc.) gets media attention, while a small Indian company that had delays meeting its release dates doesn't.

I think the Simputer's cool, and have been reading your posts about using one avidly -- hell, I linked to them in this very piece. But because of Negroponte's memetic weight, whatever his group does is going to have an immediately bigger splash, even in failure.


Posted by: Jamais Cascio on 28 Jul 05

I completely agree Jamais.

After reading Taran's overview of the simputer, I went to the Amida website to get one for myself.

I discovered 2 things:

1) There is no prices and no way to order online.

2) Through an email address I asked for pricing information and for shipping information to Miami, it has been 3 days without a response.

So, here I am - lowly individual, so far unable to get my hands on one.


Posted by: Chris Albon on 29 Jul 05



EMAIL THIS ENTRY TO:

YOUR EMAIL ADDRESS:


MESSAGE (optional):


Search Worldchanging

Worldchanging Newsletter Get good news for a change —
Click here to sign up!


Worldchanging2.0


Website Design by Eben Design | Logo Design by Egg Hosting | Hosted by Amazon AWS | Problems with the site? Send email to tech /at/ worldchanging.com
©2012
Architecture for Humanity - all rights reserved except where otherwise indicated.

Find_us_on_facebook_badge.gif twitter-logo.jpg