The recent dispute between OLPC and Intel over Intel’s involvement with the XO laptop has again brought the issues of the OLPC project much attention.From reading around the topic over the last week, there seem to be several points which are worthy of consideration, many of which have not been clarified by either OLPC or Intel.

The author wrote an article about the XO on 29th November 2007, and this new article aims to take stock of the situation a month and a half later, with the events that have recently occurred.

The Classmate PC

Firstly, a look at the Intel and OLPC debacle.

Intel partnered the OLPC project in July 2007, after months of conflict between the two developers. The addition of Intel to the project meant that the company joined the other 11 companies (inc. google and AMD) who were already partners. From a BBC article in July, it appeared that the intention was for Intel to keep selling the Classmate PC, OLPC to keep selling the XO with the AMD chip, but to have the back up servers using Intel technology, and consideration be given by OLPC to using Intel chips in the XO’s. There was also mention that new software developed would be XO and Classmate compatible. OLPC were quoted in the same BBC article as saying ‘I think we will end up with a family of products that run across a wide variety of needs. Intel will be part of that mix.’

At the very beginning of January 2008, Intel withdrew its support from the OLPC project.

The OLPC rationale, as said by Nick Negroponte, the head of the OLPC project, for this decision was:

‘The biggest single reason was that [Intel] were directly selling their Classmate laptop as opposed to having it be a reference design [for the XO].’ (source)

One of the events mentioned most frequently in support of OLPC, is that a salesperson for Intel in Peru apparently made derogatory remarks about the XO, in an attempt to stop the government buying XO’s and encourage them to buy Classmate PC’s instead.

Therefore, OLPC say that one of their other reasons for having a problem with Intel’s actions is that Intel were criticising the XO to promote the Classmate. If that is the case, it does seem underhand on the part of Intel, and likely goes against the agreement made.

‘OLPC head Nicholas Negroponte has lashed out at the company accusing them of being self-serving and undermining the OLPC programme.’ (source)

However, it appears, and has been mentioned on several articles, that this issue of defamation occurred (or at least has been accepted to have occured) in one case, with one salesperson, in one country. As Negroponte himself said in reference to the boss of Intel:

‘He’s got 100,000 people and he can’t control all of them.’ (source)

So perhaps judgement of Intel’s actual intentions to contradict any agreement in this manner should be withheld, as there appears to be no proof of this as yet.

The XO laptop

Intel’s rationale for the split, as explained by an Intel spokesperson, was that “OLPC had asked Intel to end our support for non-OLPC platforms, including the Classmate PC, and to focus on the OLPC platform exclusively. At the end of the day, we decided we couldn’t accommodate that request.” (source)

Having looked at explanations given by both sides, the issue of the Classmate PC, and competition itself, are seemingly the causes of the problem.

There seems to have been a difference in OLPC and Intel’s views as to how Intel were going to deal with the Classmate PC once the partnership had been made. Without knowing exactly what the agreement stated, it is hard to know what was actually agreed, however it is clear that there was at the very least either misinterpretation, or outright breaking of the agreement on at least one side.

From the BBC article from July 2007 quoted earlier, the impression given was that Intel would continue as before with its sales approach to the Classmate PC. There would be benefits to both sides in terms of cross-compatible software, and potentially Intel chips in the XO as well as the back up servers using Intel technology.

However, Negroponte’s statement quoted earlier, that the Classmate should have been used as a reference design rather than Intel ‘directly selling their laptop’, seems to contradict this.

OLPC’s request that Intel end support for it’s own product the Classmate PC, also appears to directly contradict this.

In summary of this point, the author wonders what OLPC actually thought Intel would do. Did they think Intel would forfeit profit from the Classmate PC and not be ‘self-serving’. Negroponte’s quote mentioned earlier that OLPC ‘thought [they] could move towards [the Classmate PC] being a reference design’, seems significant with the word ‘thought’. Does this mean that Intel never agreed to this for definite? The question seems to be, in short, did OLPC expect Intel to not act like a business?

It does not become apparent from researching around this topic, how exactly OLPC and Intel thought that the partnership could work. The mix of business and not-for-profit is unusual in a situation where both are competing. If we consider what would happen in a business to business situation, the partnership would never have happened. If we consider two not-for-profit enterprises, they would probably collaborate, because they could help each other. However, in this case, OLPC seem to have expected help from Intel, and expected to not only give nothing in return, but also that Intel would sacrifice their aims (of profit) for the OLPC cause.

In terms of the competition between the two, this issue again proves extremely contradictory. Firstly, the two clearly are competitors, otherwise this problem would never have arisen in the first place, and the reasons given by both sides for the split, whilst somewhat different both acknowledge that it was because of competition.

Negroponte again contradicted himself after he stated a few months ago:

‘From my point of view, if the world were to have 30 million” laptops made by competitors “in the hands of children at the end of next year, that to me would be a great success. My goal is not selling laptops. OLPC is not in the laptop business. It’s in the education business.’ (source)

This appears to be in contradiction to his recent approach of asking Intel to stop selling the Classmate PC.

As Negroponte said:

‘When I questioned [Intel] about selling laptops directly, Paul Otellini himself would say that one size doesn’t fit all. And we agree with that, of course. That wasn’t the issue. They cannot compete with OLPC and be a partner.’ (source)

So, although there appears that there was no part of the agreement that stated that Intel could not promote or sell the Classmate PC, OLPC knew they could not be both a competitor and a partner. So exactly what was Intel’s approach supposed to be?

In terms of this issue, perhaps the comment from Paul Otellini that one size doesn’t fit all, was misinterpreted by Negroponte when he stated that it meant that they were not competitors. Whilst consumers clearly choose products for different needs or wants, it does not mean that the products are not competitors. Just because the XO has an ideology of constructivist learning, does that mean that consumers will put the two in different categories? Or does it actually mean that consumers will still look at the product and it’s tangible benefits, over the intangible and theoretical education differences.

It seems to the author that perhaps one of the problems with the OLPC is that whilst they are correct that their approach (constructivism) would be beneficial especially in developing world countries (as it emphasises the users particular social situation (culture, language, logic etc) and encourages collaboration), they are not selling an idea in reality, they are selling a product. And they are selling a product that resides in a hugely competitive market (as discussed in an interesting nextbillion.net article), ruled by giant corporations, ruled by profit. What the author means by ‘in reality’ is not that OLPC are not trying to sell an idea, they clearly are, but that markets are ruled by consumers, not sellers, and maybe an idea that is not status quo in terms of education is not going to feature as a priority when considering one laptop versus another in a society where bridging the digital divide is generally considered a priority. Intel clearly supports the bridging the digital divide initiative, with its World Ahead program of Accessibility, Connectivity, Education, and Content, and it’s clear update and support system for their product.

In conclusion, it seems that the partnership was destined to fail, and that there are just too many differences in the way the two developers work. However, the author does sympathise with Intel (apart from the alleged underhand defamation of the XO), and considers that maybe the OLPC’s ideology based approach needs some sort of reality check on its integration of ideology into reality, without which it seemingly will not reach it’s potential.

If OLPC want to help education through their ideology, that is a good thing and the author does believe that the approach of constructivism could work. However, perhaps OLPC need to consider that the product first needs to be able to compete with the existing products in their market, to allow infiltration of their product. This would allow people to actually understand through first hand experience why the approach is beneficial, rather than OLPC relying on people to trust the theory of constructivism, and the XO interpretation of it, neither of which the buyers or users likely have experience or knowledge of.

Related articles of interest:
Wiki on Constructivism theory
New York Times article ‘Intel Quits Effort to Get Computers to Children’
Nextbillion.net article ‘Next Billion Customers? Intel’s Already a World Ahead’
Nextbillion.net article ‘Reality Check for the $100 laptop’

The attitude and priority that a government gives to investing in infrastructure is one that is deemed necessary of consideration prior to any plans for infrastructure development, whether it be funded and developed by external or internal groups of any kind. The term used to describe this is the ‘absorbative capacity’ of a country, which is defined as the capability to use, maintain and develop an infrastructure. The requirements that are needed to allow a sufficient absorbative capacity are defined by retired managing director for global infrastructure investment at the UK CDC with the following quote:

‘There is lots of evidence that unless a country is spending 4-5% of its GDP on infrastructure development, they will not sustain any increase in GDP per capita.’

As an increase in GDP per capita is a key indication of a country’s increasing development, this appears to demonstrate that unless a country is spending the required amount to have an adequate absorbative capacity, there is little point in other external, or private groups developing an infrastructure, as it will not be able to aid development in the country in the long run, as the country will not be able to maintain or develop it further. This point seems particularly relevant to my post about the Connect Africa summit, and its outcomes, which are to develop an infrastructure within Africa. It also seems relevant in light of the East African Submarine Cable system agreed investment over the last few days (talked about in an AfricanLoft article titled ‘East African Submarine Cable System Receives a Boost).

In conclusion, to reiterate the importance of infrastructure investment, I want to show the diagram provided by Henry Kyambalesa in his book ‘Socio-Economic Challenges: The African Context’, of the symbiotic relationship created by infrastructure development:

The Connect Africa Summit was held a couple of weeks ago in Rwanda. Over 1000 people took part, and it seems like the results are potentially quite promising.

The driving force behind this Summit, which is one of a series planned across the world, is the need to speed up the development process to attain the goals defined at the World Summit on the Information Society in 2003.

The key point seemed to be that it has been established that ICT must take the lead in development of Africa as a whole, and that broadband is the next big step in this process. The opinion that this can be successfully implemented by a estimated date of 2012 is drawn from consideration of the growing mobile phone market, which is currently surpassing any other country in terms of growth rate (and has been for several years), and had nearly 200m subscribers in Africa at the end of last year.

Funds and Aims Promised:

An estimated total of US$55,892,750,000 has been promised.

US$50b from mobile operators who have committed to increasing connectivity to interconnect all African cities, and strengthen the connection to rest of world, with 90% coverage of the population by 2012.

Over Euro100m of grants, along with over Euro200m of loans from the European Commission to develop projects that have multi-nation impact, either through cross-border projects or country projects that will develop the continent as a whole (this fund is intended to be added to at the end of 2008). This project mentions specifically the use of ICT.

The World Bank Group has not committed to, but has talked about the likelihood that they will increase their US$1b per year to US$2b per year by 2012, which is a fund aimed at development of the private sector.

The Africa Development Bank has committed nearly US$65m towards infrastructure development (and is intending to spend 60% of its resources on infrastructure in the next few years).

These outcomes seem very positive in terms of infrastructure development, and helping to achieve the 2015 goals. However, it does raise a concern that appears to me to be potentially critical in terms of the overall development of Africa through connectivity, which is - whilst it is very positive that there is clearly to be plenty of money and technical progress in terms of this development, what exactly is the connectivity to be used for?

The one target relating to this is one of the 10 connectivity goals from the 2003 Summit which states:
‘encourage the development of content and to put in place technical conditions in order to facilitate the presence and use of all world languages on the Internet’.

This goal seem quite ambiguous in terms of what content they are referring to. Who is going to develop it? Who is it going to be developed for? What are the aims of this content? How are they going to find out what content people want? The list of questions goes on… I have not yet found (and someone please send me links or information if you have the answers to this) any information as to what software or hardware is to be developed specifically to accompany this newfound connectivity.

My concern lies with the fact that technology is only as good as the applications that it supports. People don’t buy a computer because it has the fastest processor, they buy it because that fast processor can make their software run faster.

It is also concerning that this drive may fall into the same trap as many other drives for African development in the past. A factor that could be linked to the cause of these could be the assumptions made by the people, governments and corporations funding or developing the projects as to what is needed to encourage development, whereas perhaps the reality, as seen by the end users of these projects, is very different.

So maybe in the case of African connectivity, the first question should be – What do we specifically want to be able to do that we can’t do now? (And by this, I don’t mean ‘access the internet’, rather, ‘farmers want to communicate directly with buyers in developed countries’ for example). And then secondly - What will enable us to do that in terms of technology?

Otherwise the risk is that perhaps the huge amount of money and time will not be used to their maximum potential, and the goals set out in 2003 will be achieved in theory, but not in practice.

There does appear to be some evidence that corporations have in some cases recognised this need – Nokia’s head of mobile division was quoted last month saying that they see the key as the content the phones deliver to the potential users.

“We are looking as a company as to how to facilitate and participate in creating content that is actually relevant for the consumers in emerging markets… the biggest barrier in emerging markets to people using the mobile internet was lack of interest.” (Kai Oistamo, Oct 2007)

However, if the Nokia 2626 is anything to go by in terms of what Nokia think is appropriate development of technology, then this company policy may not amount to much in practice…(I will be writing a post overviewing emerging market handsets shortly, where I will go into more detail on the phones themselves).

However, another concern regarding this huge investment in infrastructure comes for a conversation I recently had regarding African infrastructure development with a retired managing director for global infrastructure investment at CDC. With regard to developing infrastructure for Africa, he said:

‘There is lots of evidence that unless a country is spending 4-5% of its GDP on infrastructure development, they will not sustain any increase in GDP per capita.’

And I would question as to whether many of the coutries being provided with this infrastructure do spend this amount? If they don’t, it appears that any development projects will be futile, as in the long-term, how will they be able to achieve the aims of overall development in the country?

Relevant links:
The Nokia 2626:
Tech Digest article
pocketpicks.co.uk article
The Connect Africa Summit:
allafrica.com article
International Telecoms Union (itu.int) article