Homo Naledi and Programming

Not so long ago a man sitting in the sun on a rock cut out 3 rows of numbers on a baboon fibula. 2 of these rows were primes adding to 60 and the other adds to 48. 60 and 48 have 12 as a common factor. What is interesting is that he did this 10 thousand years before primes were discovered.

More interesting was that 40 thousand years before this a woman in the Lebombo mountains made a 29 notch tally. The Lebombo mountains also provide home to the worlds only state sponsored Sangoma school.

Here is a fun claim: the two people cutting notches into the tally sticks mentioned above were the world’s first programmers.

Why do I, a programmer, raise this? We know Charles Babbage made the first computer, that Ada Lovelace was the first programmer and Alan Turing made it work.

It is not my intention to claim Homo Naledi invented programming, although it is a fun idea for 2 micro seconds.

However it is my intention though to question some of the baggage we carry with us from the days of installing MS Dev Studio and running MFC apps.

The weight of our history should not affect our future.

This is my perspective: programming should become more useful and accessible. Don’t worry – I am not going to talk about vendor lock in or open source.

Many of us program for the joy it brings. But programming also provides the key to finding good work and access to markets and information.

The digital economy is not just consuming content but making things. The mortar is data, the bricks are the cloud and mobile. Small business should be using technology to grow, but technology alone will not solve development challenges.

Programming is not supposed to be rocket science. However, programming has become more arcane and technology a source of division. While it is true that if you are programming for an enterprise there are complicated issues. But for the majority of small companies software is a nightmare. It need not be.

We have built our solution, the Tree, to do the plumbing to make coding more useful. We must remember coding is often the easy part. It should not provide a distraction from the important parts: communication, prioritising, planning, learning and  designing systems.  Our solution helps small companies get purpose built software, more quickly and cheaply. Small companies cannot afford ERP solutions, let alone managing a team to produce a bespoke solution. The cloud can reduce costs and this is something we use to make programming better.

This perspective on programming is not new, but it will change everything for small businesses. 10 000 companies turning over $100 mill get all the software. We want to get to the 10 000 000 companies turning over $1 million.

I am sure that one does not have to go back 10 000 years to find someone enjoying programming; someone programming a ZX Spectrum (4kb RAM) had more fun that someone programming the latest mobile phone  (4GB RAM). It can be like this again.

Draft ICT SMME Support Strategy: Unlocking the potential of ICT SMMEs

Response: “Draft ICT SMME Support Strategy: Unlocking the potential of ICT SMMEs” 29 March 2017 Minister of Telecommunications and Postal Services

4 May 2017, Version 1.0, DJ Hislop (pdf -> DTPS_Response)

0 Foreword

The author sets out to respond to the proposed strategy document [0]. We comment within the sections demarcated in the strategy document.

In addition to a point-wise response to the strategy, at the end the author identifies some problems with this strategy and proposes additional interventions.

From the outset it must be noted that if this is a strategy (“a plan of action or policy designed to achieve a major or overall aim”) then it fails to adequately define the plan, actions or aims.

1 Introduction

The report begins with the usual observation of “the need to unlock the potential of SMMEs”[0]. But it has been noted that, despite the lip service to SMME’s, real support goes to brick and mortar companies. “despite the emphasis of policy statements on SMEs, substantial on- and off -budget support continued to be extended to a number of capital- and electricity-intensive … sectors” [1].

This is probably due to the immediacy of interventions regarding jobs and exports. However, the world has changed.  The following figure illustrates the decline of mining and the rise of  financial services (which is a sector heavily using IT and software.)

Furthermore, despite all the resources that have been thrown at SMME’s “the growth in the number of SMMEs from 2008 to 2015 was lower than the economic growth rate” [2]. This is shocking. In fact, the author would contend, the number of SMMEs remains pretty stagnant in this period. See Figure 3.

This stagnation in SMME’s is quite an inditement of the process to date. What has gone wrong? This question needs to be answered and we do it now. As we see it some of the problems are that

  1. interventions have been supply side only (universities, accelerators) and not demand side (consuming innovation)
  2. interventions have been localised in competing government departments
  3. the nature of ICT is cross-sector
  4. the term ICT is unhelpful because it does not recognise the different business paces of telecommunications, software, hardware and broadcasting
  5. government has not seen its role clearly (for example creating standards)
  6. ICT interventions and SMME’s are not in touch with big business.

These points are elaborated at the end.

It should be noted that there is a great deal of interest and activity in (ICT) SMME’s. By some metrics, SMME’s are doing really well. This vibrancy and financial activity cannot be ignored. (There almost seems to be two different conversations.) An interesting question is whether the fate of South African startups is to be acquired.

2 Rationale

In proceeding with a review of the rationale for unlocking the potential of SMME’s one can note that this argument has been won already. What’s lacking is an integrated public action plan. In all the available references, nowhere does one see the economics of the intervention. The lack of focus on the market is its profound failure of the strategy.

It should be observed that there are many reports and position papers, where quotes from experts are cherry-picked. These authors use snapshot-figures and statistics. Quite frankly it is impossible to get a good picture of the state of ICT SMMEs with competing strategies emanating from government departments and different methodologies of consultants, let alone ideologies.

This lack of coherence or leadership is the fundamental problem faced by tech SMMEs. (Herein already lies an important bifurcation – informal SMME’s are probably not “tech” and possibly not central in defining a strategy.)

Highly concentrated ICT market 

The logic of this section starts of being confused. “High capital requirements for deploying infrastructure … limit[s] involvement of small players” – if small players had a lot of capital they would not be small. What can this sentence mean?

One would need to see some kind of justification for the claim “uncompetitive and highly concentrated structure of the ICT sector”. It is too broad. For example: does it mean that the software industry is uncompetitive or the hardware industry (just the import of “big-iron”) or the local manufacturing of PCB boards.

The document clarifies that it is speaking about telecommunications (“spectrum”) in the next paragraph. There are easier ways of giving “access to spectrum” such as using spectrum more efficiently through analogue to digital migration and allowing more content providers onto digital multiplexes. As mentioned in the paper, rolling out a network is capital intensive and a cost that the so-maligned operators have entered into already.

The question of data tariffs is indeed important. (The 3G allocation was a shoo-in for operators, which was a government fail.) Yet it is hard to see where yet another cellco could get the resources to introduce efficient competition.

Inhibiting policy and regulatory environment 

There are quite some problems on this level, such as charging VAT on foreign vendors offering a cloud solution to South African clients. This is clearly an area that needs a substantial amount of work. But who does the work? And who enforces it? It is clear that small innovative SMME’s should not succumb to the same regulatory burden used to snare big foreign companies.

Limited [university]commercialisation of innovation 

It is in the opinion of some, a fundamental misconception that commercialisation happens out of innovation. Steve Jobs has been quoted that Apple does not innovate but builds products that are foremost intended to delight its customers. Universities cannot reasonably claim to understand what real customers would want.

SMME constraints in access to finance 


Limited market access for ICT SMMEs 

Agree: “Current procurement policies favour big businesses to the detriment of small ones” and “to effectively respond to tender bids has been raised as one of the challenges, requiring intervention.”

Skewed deployment of infrastructure 

It makes no sense to try and roll out an infrastructure homogeneously throughout the country due to the inhomogeneous distribution of people and businesses. The strategy should be to encourage the formation of hubs.


There must be many more needs in supporting ICT SMME’s.

Here is what “the  entrepreneurial ecosystem of South Africa: A strategy for Global Leadership 2017 – accelerating technology absorption with a focus on digital technology  SAB-Allan-Gray-Orbis-Foundation-Research-Report-Infographic” observes:

Figure 2

3 Vision of Strategy 

One cannot but agree with “the vision of the ICT SMME Strategy is to create a substantial number of internationally competitive, dynamic, innovative, technologically driven sustainable ICT SMMEs that significantly contribute to the country’s developmental priorities.”

4 Strategic Goal 

The goal “to unlock the business opportunities and create an enabling business and administrative environment for SMMEs in the ICT sector to thrive and advance into successful and sustainable entities” seems suspiciously difficult to quantify and measure.  It could be imagined that useful goals include “improve national growth by 1%”, “increase employment by 5%”, “export more software than we import”, “increase mobile broadband penetration by 5%”, “increase number of SMMEs by 10%”, “increase the average size of SMMEs to 10 people” or “reduce mobile data costs by 10%”.

5 Objectives of the strategy 

In looking at the objectives, they are not particularly aligned to the vision. In particular the “accelerated entry of SMMEs in the ICT sector” has to be nothing but a recipe for failure.

To “increase the levels of uptake and usage of ICTs by SMMEs, across sectors” looks like a good idea. This will not create sustainable and scaleable companies in the long run.


One cannot possibly imagine how the “establishment of a coordinated and integrated planning mechanism” will work (we have had committees of “wise men” before).

6 Definition of SMME’s

This seems like a good idea, but does the definition of an SMME actually make sense: “… a separate and distinct business entity, including co-operative enterprises and Non- governmental organisations, managed by one owner or more which, including its branches or subsidiaries, if any, is predominantly carried on in any sector or sub sector of the economy mentioned in Column I of the Schedule… ”. Maybe one needs to be a lawyer to understand it?

6.1 What is an ICT SMME? 

The paper must actually take the initiative and define what an ICT SMME is. For example, there is no single ICT sector, like Agriculture. Software, by contrast, is a cross sector discipline.

7 Overview of the current position of ICT SMME’s and uptake of ICTs across Sectors in SA.

7.1 Structure of the ICT SMME Sector (Indicators)

Without comment.

7.2 ICT diffusion by SMME’s across economic sectors

Without comment.

7.3 Policy and Regulatory environment

Without comment.

7.3.1 General small business policy and legislation

“… Multinational companies that cannot have local ownership are encouraged to use the Equity Equivalent Investment Programme (EEP) to invest in sustainable enterprise development to increase black ownership of the economy in South Africa” has to be questioned. The programs in which author is familiar were disasters.

Department of Science and Technology and Trade and Industry 

The effectiveness of programs such as SPII have to be questioned. For example in 2013 SPII introduced without warning a unilateral moratorium on funding.

There is no question that government plays a key role in innovation.

However, even when the government wins, it looses. For example the DST won the SKA, a laudable achievement. But the 2016 Innovation index talks about the innovation benefits that go to Australia. We should really see the SKA as an integral part of ICT innovation, entrepreneurship and communications.

8 Analysis of the ICT sector value chain: Identifying challenges and opportunities for SMMe’s.

8.1 ICT sector value chain analysis

Without comment.

8.1.1 Telecommunications

Without comment. Infrastructure

Without comment. Wholesale vs Retail

Without comment.

8.1.2 Information Technology

Without comment. Hardware, including components and devices – the electronics sub-sector

Without comment. Software (Including OS and types of software vendor)

What is critically missing is cloud and how that changes business models.

5 years ago Marc Andreessen wrote to the WSJ on Why Software Is Eating the World – this strategy does not do justice to this point of view. Data Centres, including the internet and the cloud

“Cloud computing or cloud services are just expressions used to describe the ability for a user to access data storage and computing applications located on equipment not on their premises (and often not in their country). Effectively, it is outsourcing some or all of the enterprise or individual computing needs to external resources on a pay-per- use basis.” The cloud changes everything. Business models; legal frameworks and more are completely changed. This statement is incredibly naive.

The understanding of the Internet in this paper is horribly limited. eCommerce seems to have no role. This paper does not frame SA’s ICT needs in the global context of Google and Yahoo! in a way that is at all convincing.

Furthermore, where would the paper locate a bank, like FNB, that provides IT products to SMME’s? The Apps environment, including games and gamification

It is difficult to understand the import of this section. Moreover it is horribly dated and probably wrong (for example the importance of the mobile web is ignored. Interoperability and Cybersecurity

Without comment. (Why is Cybersecurity different from Security?) IT producer Enterprises and IT User Enterprises

Without comment.

8.2 ICT Value Chain: Opportunities and Challenges.

8.2.1 Telecommunications Network Infrastructure

Without comment. Service Provision

Without comment.

8.2.2 Information Technology

Without comment. Hardware Sub-Sector (Including Electronic Components and Devices)

Without comment. Digital Content

Without comment. Software Products – including OS and packages

The statement “there is very little scope for local SMMEs to become developers or distributors of packaged software” is not correct. Games and Apps

This section is sorely lacking. One really hope that the DTPS actually speak to someone who knows a little about this. Data centre and cloud computing (emerging opportunities)

Exactly: “Once the infrastructure has been set-up, there are a wide range of services and products available on top of the core infrastructure. SMMEs have tremendous opportunities to leverage cloud infrastructure to provide a range of specialised and niche offerings to both large and small customers. This is a new and emerging area of business, and SMMEs should develop innovative solutions and business models. As with any new business model, cloud computing is not well understood by potential customers. SMMEs wishing to operate in this area will need to work hard to sell new concepts to sceptical customers.” Interoperability and Cybersecurity

Without comment.

8.2.3 Other Business Opportunities, particularly in the public sector

“During the ICT Policy Review consultation process, stakeholders raised the importance of government being the first client for products and services developed or provided by SMMEs, considering the annual budget it uses to deliver support services to small business. This will instil confidence in other potential clients, especially private sector. As part of its responsibility of ensuring the realisation of an inclusive information society and achieving universal access, government has continued to roll-out infrastructure and ICT services in schools and other public facilities.” Agreed.

“The analysis of the ICT value chain has revealed both areas of opportunities and challenges for small businesses. It is evident from the discussion above that some upstream levels of the market, such as infrastructure deployment, have high barriers to entry, considering the capital intensive nature of some of the activities. However, there are opportunities for small enterprises in the service layer of the value chain, e.g., installation of some components, maintenance of the infrastructure, etc. Although most of the hardware used in South Africa is imported, there are prevailing opportunities in the areas of design, assembly, testing and configuration, requiring high end skills. Applications and gaming development, software development, data centres (set costs are prohibitive, but there are opportunities in service and maintenance) and cloud computing based services seem to have low barriers to entry and as such, strategies to enhance SMMEs participation are required.” Agreed.

9 Proposed ICT SMME Development Support Strategy

9.1 A Model for ICT SMME Development

For the Department to confine itself to a model emanating from a university is extremely concerning. This particular project from which the model emanates has a history of monopolising government channels to the detriment of the industry at large. They do not represent software professionals in general.

The author cannot possibly support the ideas encapsulated in the model. Concerns would be “Sort and Select” criteria are notoriously unreliable (the market should decide), the Lean Startup has been called into question. There are so many things wrong with the model that would require a separate document. (We propose alternative interventions later.)

A top-down or control-and-command approach to tech (ICT) SMME’s is simply wrong.

The strategy mentions hubs and this is indeed a good idea. Amongst other things, the formation of hubs provides a way of supporting the intangible assets through real-estate (capital). Another issue that should be considered is outreach to global ICT firms through the creation of campuses. Local SMME’s should be given offices in these campuses to share knowledge and provide services. It is also a way of capitalising the local ICT industry.

9.2 Strategies and Interventions to support the ICT SMME development model

“The strategic interventions, are presented in alignment with the three objectives highlighted in Section 5 of the Strategy.” What to do? The author has questioned these objectives which leaves commentary on the strategy somewhat pointless.

Strategic Objective 1: Facilitate development and accelerated entry of SMMEs (youth and women entrepreneurs) in the ICT sectors (sic) 

What are the ICT sectors now? Sub-sectors?

Challenge: Highly concentrated ICT market (Telecommunications sub-sector) 

Without comment.

Challenge: Limited participation of SMMEs in the Information Technology sub-sector 

Without comment.

Challenge: Limited access market
Without comment.

Challenge: Lack appropriate business, administrative skills and entrepreneurial capabilities
Without comment.

Challenge: Limited Access to funding instruments and incentives for ICT SMMEs

Without comment.

Challenge: Skewed deployment of appropriate infrastructure
Without comment.

Challenge: Inhibiting enabling policy and regulatory environment for SMMEs to thrive.
Without comment.

Challenge: Limited commercialisation of innovation
Without comment.

Strategic Objective 2: To increase the uptake and usage of ICTs by the general SMME sector (sic). 

Since when is SMME a sector?

Challenge: Low of uptake and usage of ICT by SMMEs
Without comment.

Strategic Objective 3: Coordinated and integrated planning mechanism for development of ICT SMMEs and increasing uptake and usage of ICTs. 

Challenge: Lack of a coordinated and integrated mechanism to drive and monitor the ICT SMME Sector
Without comment.

10 Institutional framework for implementation of the strategy 

“There is a need to strengthen partnerships between government, industry associations, market intermediaries, academia, civil society organisations, grassroots groups, who must all be able to play their roles effectively in the entrepreneurship development ecosystem.” Agreed.

The question is how is this going to be done – there have been previous failed attempts. There are bodies out there  (SAIEE, BITF, IEEE, WASPA, Silicon Cape, Matchi, the Innovation Summit and many other organisations and individuals) that are capable of representing the views of their members. The author has heard claims that there are “too many” to engage and that they are not representative. This is an opinion that stands only because, in the supposed absence of representative bodies, control defaults to the inner circle making this statement.


“Complex, systematic challenges like expanding economic opportunities to SMMEs, present frustrating bottlenecks to unilateral actions, especially from government side. Even the best resourced efforts eventually run into limitations of scale at some point. Therefore, the implementation of this ICT SMME Support Strategy requires concerted efforts from all relevant stakeholders.” Agreed.

“Collaboration allows parties to share knowledge and information, pool scarce or diverse assets and resources, access new sources of innovation and create economies of scale. The public sector, or more exactly the government bodies, will act as partners with the private sector, i.e. with SMMEs, as well as with the organisations of civil society, all for the purpose of contributing to the development of the entrepreneurial sector, and consequently to the South African economy in general.”

This is not true – only academics and government employees have time to attend partnership meetings. Any policy needs to be sufficiently clear and clean that stakeholders can find their role without being told.


Coordination for the implementation of the Strategy 

“It is important to have a single coordinating body responsible to coordinate and facilitate the implementation of the ICT SMME Support Strategy.” Again, this is a recipe for disaster. What is needed is firstly leadership from government (how is this even possible with all the different initiatives?) and then meaningful and respectful dialogue with industry and civic bodies. This is quite possible, but requires the appetite of government to get involved in the messy industry.

“The strategy has to combine and coordinate efforts of institutional structures that make or affect policy, the administration of policy and public support organisations. In this regard, the Industry-Growth Coordinating Mechanism as proposed in the National Integrate ICT Policy White Paper, to be convened by the

  • Ministry of Telecommunications and Postal Services

with other cluster Ministries including the

  • Ministry of Science and Technology,
  • Ministry of Trade and Industry,
  • Ministry of Small Business Development,

will be entrusted with the responsibility for coordination, facilitation and overseeing the implementation of the Strategy.” The author proposes the following heuristic: if the view of government from outside is chaotic and uncoordinated, the plan will fail. Another way of saying this is that governments organisational problems should not be an issue for industry etc. It is.

The following Figure 3 illustrates the abject failure of the efforts to stimulate SMME’s for the past decade.

Figure 3

Establishment of an ICT SMME Programme/Project Implementation Steering Committee  

“An ICT SMME Project/Programme Implementation Steering Committee will be established, co- chaired by officials from the Departments of Telecommunication and Postal Services and Small Business Development.”

There are too many stakeholders and not enough leadership: Department of Small Business Development & Department of Economic Development and DTI and DST. Included in the government stakeholder are SOEs such as the CSIR, …. The plan needs a leader who represents key businesses and institutions. We cannot possibly have the usual structure or institutions.

“The Steering Committee will report to the ICT Industry Growth Coordinating Mechanism, a structure to be created to facilitate synergies across relevant Government departments and ensure bottlenecks are addressed, as provided for in the National Integrated ICT Policy White Paper. The White Paper stipulates that the coordinating body will be structured along the broad sub-sectors of the industry viz.

  • ICT Manufacturing (including the electronics and related hardware sub-sectors),
  • ICT software development (including applications development) and
  • the ICT Services industry (providing maintenance, logistical support, data warehousing, network support etc.).”

Where does web development, software design, cloud services, IT R&D, etc fit in? Where does security, storage, big data, etc fit in? This is an absurd list.

“This distinction is essential to ensure that interventions take cognisance of the uniqueness of each and that targeted support programmes are developed accordingly, also prioritising opportunities for ICT SMME development across the value chain.” Well, yes.

Conclusion to review of Strategy document

There are many outstanding issues in this strategy. The author strongly recommends that the discussion gets broadened.  It is clear that the number of voices in this strategy is limited to the usual likely suspects.

As it stands, industry cannot possibly support this strategy, being a collection of the normal nebulous statements about “interventions” and pot-shots at selected incumbents.

It is quite possible to create an open public innovation system. But government has failed and needs to answer some hard questions.

This strategy does not touch on many important issues:  a discussion on tenders, the role of SITA, where are the big ICT businesses, Public Private Partnership Goals, Internet balkanisation and Fake news and more. Also, as PWC notes, government need to upgrade VAT laws to keep up with the digital economy. “Foreign entities were not subjected to the same tax regime as local companies”

It is possible to create a virtuous cycle but government must change its attitude from being closed and threatened to being open.

General analysis

To be successful, we think the strategy would need

  1. alignment with other departments and initiatives
  2. alignment with industry norms, conventions and requirements
  3. defined “small steps” to achieve the strategy
  4. definition of ICT sector (if one can) and the the separation of concerns such as telecommunications, software, etc into sub-sectors.

In starting with a strategy it would be useful to seek alignment.

DTPS -> Most Important Goal         -> poverty, inequality and unemployment. growth and development. national and international competitivenes  (phew! Which? We pick Growth.)

Technology Industry Goal   -> Improve Productivity ?

Global Marketplace Goal    -> Increase Trade ?

Once this is done, to define the plan, determining actions or aims is possible.  As mentioned, at this stage less emphasis should be put on the “how” and more on the “what”. The “why” question is answered quite simply as “we wish to create a cool, viable and profitable sector”.

While it is important to have a mission and vision, it needs to lead into a scenario where goals are measurable. However, this is not to advocate going through yet another round of measurements and analysis. That is a recipe for analysis paralysis. What is being advocated is the need to define a plan where stakeholders can easily identify their roles and progress can be measured.

As promised the author will now expand upon the problems as outlined above.

  1. interventions have been supply side only (universities, accelerators) and not demand (consuming innovation)

We introduce this point by setting the context through a quote Nimrod Zalk [1]. He reviews “South Africa’s progress with the development and implementation of industrial policy over the post-apartheid era.  This history falls into three broad phases:

  • from the end of the Second World War to democracy in 1994,
  • 1994–2007, and
  • post-2007, with a particular focus on the last period.

“Economic policy, especially between 1994 and 2007, has been overwhelmingly dominated by orthodox laissez-faire economic reforms.  These reforms were meant to achieve a step change in  fixed investment and thereby catalyse higher levels of growth and employment across the economy, including manufacturing. However, they have not delivered significant or sustainable investment, growth or employment gains.”

“In particular, the NIPF [National Industrial Policy Framework, 2007] emphasises that “South Africa cannot rely so heavily on either consumption or commodities as the basis for our growth and development” and sets out four strategic industrialisation objectives

  • To facilitate diversification beyond [the] current reliance on traditional commodities and non-tradable services.  This requires the promotion of increased value-addition per capita characterised particularly by movement into non-traditional tradable goods and services that compete in export markets as well as against imports.
  •  The long-term intensification of South Africa’s industrialisation process and movement towards a knowledge economy.
  •  The promotion of a more labour-absorbing industrialisation path with a particular emphasis on tradable labour-absorbing goods and services and economic linkages that catalyse employment creation.
  • The promotion of a broader-based industrialisation path characterised by greater levels of participation of historically disadvantaged people and marginalised regions in the mainstream of the industrial economy.


Part of GEAR [Growth, Employment and Redistribution, 1996] envisaged a range of grant-based “supply side”, predominantly aimed at assisting small and medium (SMEs) manufacturing firms to adapt to a sharp increase in international competition. In practice, on-budget support for these measures was generally of limited scale and widely dispersed across a range of sectors and multiple policy objectives.

  1. interventions have been localised in multitudinous departments

There have been many attempts to fire up ICT, SMME’s, entrepreneurs and innovation. Most of them fall away – from the AAIICT, to Road mapping, and the Innovation Fund. This strategy cannot going the wrong way and using the same advisors in the same places.

Quite simply government does not know how to speak to business. There is a lot more “industry” bodies in the innovation and entrepreneurship space.

  1. the nature of ICT is cross-sector

A recurring problem is that IT (an aspect of ICT) is cross-sector. The language in this strategy document frequently gets confused talking about “ICTs” and “sectors”.

  1. term ICT is unhelpful because it does not recognise the different business paces of telecommunications, software, hardware and broadcasting

The alignment of thinking in this document to the ITU and rather the needs of local SMME’s is concerning. ITU was an analogue organisation now trying to find relevance in the Digital. “Most everything goes to (digital) data” – there is a need for more data focus.

  1. government has not seen its role clearly (for example creating standards)

Government should be the primary catalyst for innovation.

In one of the references [3] there is a troubling statement that relates to whether SMME’s are the product of a thriving economy or cause the economy to flourish. “While the precise causal relationships between SMEs, growth, and poverty have not been conclusively determined, the numbers do reveal some close interaction between SMEs and overall economic health. It appears that the more advanced a country, the larger the formal SME sector. In developed countries, SMEs contribute 60% of employment and 50% of gross domestic product (GDP). In developing countries, the figures are only about 30% and 17%, respectively.”

Apparently in South Africa – “overall, micro, very small and small businesses accounted for 27-34% of total GDP in 2006. This GDP share remained relatively constant across the period 2001-2006.”

When officials with salaries are talking about SMME’s they are safe with their pensions. This cognitive dissonance needs to be understood. It cannot be right that the people with no skin in the game are suggesting that people leave university to become entrepreneurs.

The discussion around SMME’s and technology is the clash of cultures par excellence. The fears of millennials must not be disregarded. (The author is not a millennial.)

Yet, the author feels government is critical in innovation: Mariana Mazzucato notes that many demand “why doesn’t the government just get out of the way and let the private sector — the ‘real revolutionaries’ — innovate?” But she does not agree [11] stating the importance of government in innovation. We concur.

  1. ICT interventions and SMME’s are not in touch with big business

The distance between tech SMME’s and big  business is quite remarkable. There are very few tech businesses listed on the JSE.

Additional comments

This report uses very old references, then when we to examine where the references are coming from, the source documents use even older references.

The author cannot replicate this result: “The DTI estimated a total of 2.26 million SMMEs in South Africa in 2007, of which more than 75% operated in the informal sector.” They based this estimate on the QLFS published by StatsSA. From where??

In any case, 2 million SMME’s does not sound right. Certainly not ICT and formal. What is the number of these?

Additional Proposals

The author has been extremely critical by the strategy simply by recognising that it is replicating the mistakes of previous such initiatives and the source of advice used in this document. So, as not to be merely critical, the author has been working on or with a number of initiatives for some years.

a. Standards proposal

In the mid 1980’s the European Union incentivised many competing companies in Europe to collaborate around GSM. These companies created the world’s most successful tech industry. The author is working on an Open Data Industry Connections program with the intention of setting up a Standard. The intention of this activity is to create a viable business ecosystem.

b. Silicon [ Cape | Veld | Savannah ] / Innovation Summit

The author did not understand the genius of these initiatives until much later. But creating the right (investor friendly) milieu is absolutely critical. At the moment there is a team in Johannesburg trying to create such a marketing vehicle.

c. Building and Designing is underrated. An award is needed.

Too much emphasis is put on entrepreneurs and innovation. But the simple art of building something useful does not get recognised. There is a need in this country for a recognition of people who build something useful.

Eskom runs an incredible program for science. The Cellcos need to do the same for ICT.

There is too much weight given to experts who act as gatekeepers. We need to encourage a culture of recognising the new and beautiful.

We have a culture of focusing on the people or the process (this is truly a swing from the sublime to the ridiculous). We need to look more at what people make.

d. Many government labs operate to the detriment of ICT innovation

There is a long list of labs that are never engaged or involved with businesses let alone SMMEs.


Fundamentally, the focus on the supply side of innovation (aka ICT SMMEs) is a broken pipeline and this needs to be fixed. For example, graduates leave university and cannot find jobs. Another example of this is policy makers who have an aversion to engage market forces.

The economic imperative to stimulate ICT SMME’s is not to be doubted. But creating a lot of SMMEs and ask them to “compete” will not see them survive and scale. What is needed is a framework where each SMME can find a niche. For example making an “App” is not that useful, unless it is able to ‘speak’ to a corporate or government IT system. To be specific with a simple example, finding a pothole is quite useless unless the guy whose job it is to fill potholes gets a notification and can schedule the work.

There is a considerable amount of exciting technology activity in telecommunications and IT. SMME’s have the agility to engage this innovation profitably.

It quite concerning that ICT SMME’s are regarded as the panacea for unemployment, low growth, poverty, etc. The government needs to think very carefully what its Most Important Goal is, and then how that chains into secondary and other goals in a virtuous cycle.

With the right kind of leadership it is quite possible to create a vibrant ICT SMME environment. But this strategy document is not it.


[0] Draft ICT SMME Support Strategy: Unlocking the potential of ICT SMMEs” 29 March 2017 Minister of Telecommunications and Postal Services

[1] Industrial policy in a harsh climate:The case of South Africa Nimrod Zalk, International Labour Office.

[2] Small Medium and Micro Enterprises in South Africa, Report Commissioned by SEDA, 2016.

[3] Economic Opportunity Series. The Role of Information and Communication Technology Sector in Expanding Economic Opportunity, by William J Kramer, Beth Jenkins and Robert S. Katz. 2007.

[4] United Nations Commission on the Private Sector and Development. 2004. Unleashing Entrepreneurship: Making Business Work for the Poor. Report to the Secretary-General of the United Nations. New York, NY: United Nations Development Programme (UNDP). Page 13.

[5] Survey of Employers & the Self-Employed (SESE) , Presentation to the Portfolio Committee, 3 September 2014, Pali Lehohla Statistician-General

[6] Annual Review of Small Business In South Africa 2005-2007 August 2008.

[7] Department of Science and Technology:    Report on the Research and Development Tax Incentive Programme 2009/10 Annual Report to Parliament

[8] Global Innovation Index 2016 by Cornell University, INSEAD, and the World Intellectual Property Organisation

[9] Georgetown University – DTI Software Development Presentation Draft 2.09.2011 Aleksandra Trpkovska et al

[10] http://www.statssa.gov.za/?p=9720

[11] https://www.ted.com/talks/mariana_mazzucato_government_investor_risk_taker_innovator

About the author

Dr Hislop got his PhD from UCT in 1996. He is a Senior Member of the IEEE and is a Certified Software Development Professional. His professional interests lie in Open Data (he is convenor of the Open Data Industry Connection Program), the dynamics in and off social networks and data science. He is CEO of Korwe Software, a leading and award winning South African ICT startup He has been following the science / entrepreneurship / innovation policy discussion since 2003, organising 4 software engineering colloquia amongst other events and regularly contributes responses to DST, DoC, DTI and other position papers.

Final Comment

The allocated time of 30 days was not sufficient to respond to the document adequately. Furthermore, the author was unable to engage key parties in formulating his reply.

Fishy distributions

I have been looking at a distribution for determining the number of fish in the sea. This is because I am actually interested in calculating the number of bugs in a block of codes, and thought (probably erroneously) that this would be a good start. The size of the fish population was modelled by a Japanese scientist called Niwa in a way that matches the observed shoal sizes quite well.

However, I am not so sure the model has any physical similarities to the number of bugs in a block of code, because although

he modeled this situation with a stochastic differential equation


performed merging-splitting simulations to estimate variance.

I am not so sure this is relevant for bugs. Nevertheless, it is a start.

The fish distribution is called the Niwa distribution (https://sinews.siam.org/Details-Page/modeling-food-systems)

I made a picture of the Niwa distribution with Grapher and compared it to the Poisson (blue), Gaussian (green) and Exponential Function (red).

The Niwa distribution

has exponential-like tails that are much ‘fatter’ than Gaussian ones. Thus, observations of large schools are likely to be much more frequent than Gaussian statistics suggest. This indicates that the use of familiar Gaussian models may easily overestimate the total population.



10 Startup Rah Rah Scams

I get a lot of spam. This is not a boast about the size of my Junk-mail box. In the era of trumped up news, pumped-up stereotypes, musky tropes, frankenstein-monster-like memes and recycled group-think, I thought I would tout a diagnostic 10-of-the-best innovation bullshittery.

Most pre-approved personal loans, must-see housing estates, certifications and Agile training land up in Trash. (To think that some poor shmo spent hours crafting the mail shot that has a 1-in-a-gazillion chance of being read is just sad. And then, to think they spent money on a sad-ass mailing list with email addresses like junk@korwe.com.) Ah well.

However, what still gets through is the highly personalised startup / innovation / entrepreneur sludge. Admittedly I am sensitive to this world. But I am standing up and saying enough to the sharks and duds and politicians that feed off the dreams of our  dreamers.

#1 First up, politicians always say yes.  This is a warning to the people who may think that going through politicians is a way forward. I have not yet seen government using their spending power in anything other than short term spending. Really they should be doing more, but quite frankly they are so wrapped up in twars to understand really understand your tech, or anyone else’s for that matter. (As for tenders, just don’t do it.)

#2 Making stone soup used to be a funny aphorism to me, until I realised that many events get organised by people who get paid to sit around.  The organisations they represent do not have anything real to offer. They are not incentivised by doing a deal. I got invited to a big national conference, with speakers and agendas and a Plan. We was shocked, when the roadmap turned out to be “we have got you all in a room, now talk”. Big deal.

#3 Beware of the usual suspects. Look at the organisations sponsoring the pitching sessions / conference / etc. There are some organisations who are really not worthy (they may have a big brand, but in terms of delivery they take more than they give).

#4 Trust your friends. They are better than strangers bearing gifts.

#5 Marketing experts are best at marketing themselves. (In any field, an expert is a rare bird.)

#6 Hustlers will hustle you. Give me someone who actually makes something any day.

#7 Beware of invitations to submit to an award. Especially awards you have apparently already won (no kidding). These are often just information gathering. I never unsubscribe from a mailing list, especially if I never joined. If an unsubscribe URL (e.g. http://www.<name>.com/unsubscribe?<david>) is to the organisation in question (http://www.<name.com) that requires no more information than already in-play I may unsubscribe.

#8 Beware of permutations of the basic 419 scams. They go like this: we want to publish a position piece on your company that will be seen by Bill Gates, Elon Musk and Angelina Jolie. All you have to do is buy an advert in our illustrious mag.

#9 Never ever ever give up your day job to be an entrepreneur. Especially when the guy telling you to do it gets a pension and uses logic like “a real entrepreneur perseveres despite …” Rubbish – a few exceptions might have succeeded despite … But a real entrepreneur picks their fights, runs away scared and never ever takes chances.

#10 Real innovators don”t read about real innovators. They are busy making stuff that matters.

Unfortunately the list goes on. I will conclude with an opening statement at a recent conference: “Despite investment what is it [that entrepreneurs] really need …” Here are three things: economic growth, an addressable market and, oh, investment.

Bates Distribution

I have been playing with MacOS’s grapher application. It is free and works really well, to understand the Bates distribution (compared to Normal / Gaussian).

As an intermediate step I plotted the following which I find quite cute:

Here is the normal distribution plotted against the Bates distribution.

Here are the two equations and the plot params (we used n= 4

The plot goes wild for x>1 or x<1 – it almost looks chaotic.

Here it is again for n=7


Open Data Plan

Hi all,

I have been under the weather recently, but now I am back!

After two initial meetings I bewimpel’ed a plan here:

Open Data

I am now looking to take this plan forward. Clearly, the above 6-point plan can be extended.

I would like to meet at 2pm until 3pm on Thursday 30 March 2017 to grow Open Data as a shared resource and a business opportunity.

The Agenda is:

(1) what the IEEE brings

(2) what would business really like

(3) why should we collaborate

(4) how can we get the plan in place

(5) setup an ad-hoc management committee

(6) links to Big Data & Smart Cities

For that we need a board room with internet connectivity.

Please contact me should you wish to attend @davidjhislop on Twitter or email me at david dot hislop at ieee dot org.

LGPL question

Help, I am confused about LGPL.

We have a integration platform called the Core. It is licensed under LGPL. A former client has built their multi-million rand service on it.

We do not get a dime from this client.

They provide an online service using the the Core and probably resell their service as a product using   the Core.

(I suspect they have modified our source now, but I cannot prove this. So I will leave it for the moment.)

The question is how do we get income from the former client for our product?



My friend Duncan wanted to teach Quantum Mechanics in the township. To me this was the equivalent of the story of the Mexican boy throwing starfish back into the sea. I thought I would rather be in industry, make things and create value. Little did I know this would become a trip into innovation and entrepreneurship, which has become a lurching Frankenstein monster.

I am frequently shocked by the divide between C-level executives and their project teams. In South African companies it is the CEO who has has the dream. She must take risks (innovate) to keep her business competitive. It is the techies who need to execute the plan and this is a comedy of errors of miscommunication. While techies are self righteous and avoid responsibility, executive-speak is a language that only C-level executives can speak, and plans that are not executable are good as useless. However, the gap between government technology managers and the real world is even larger than that in companies.

We are in this country blessed with a DST and a credible Minister. But consider for a moment the decision making process of a SET manager in government: in contrast to the DTI, they need to produce a return on investment on long-term nebulous criteria. To help themselves, they can speak to academics and consultancies, but there are problems with these conversations. On the other hand they can speak to successful innovators, but the problem with successful people is that they are the exception and not the rule.

And so they hit upon the idea of hiding behind innovation and entrepreneurship, as have the thinkers in many other countries. The fundamental problem about this programme is that it is supply side only, along with punting universities, pumping up accelerators and pandering to metrics.

The theory around innovation is mathematically justified. Schumpeter started this program about 90 years ago and it has received many iterations.

But Schumpeter also argued the importance of large companies with capital to invest in research and development of new products and services. Modern theorists such as Mariana Mazzucato agree but are slanted towards government. In fact the GSM family of technologies comes from the European Union (government) getting companies to cooperate. Even that famed device of private enterprise, the iPhone, came largely from government labs.

It is from this structure and management that innovation comes. The company that was founded in a one-car garage in Palo Alto by Bill Hewlett and Dave Packard is an anti-pattern.

I am running perilously close to agreeing with a recent headline “adequate support and access to funding will ensure SMMEs’ success”. It is not the funding that makes companies successful (although that is useful), it is the access to markets. It is paying clients that drive innovation.

What we need to do is create a culture of consuming our own innovation. Instead in South Africa we get SMME’s competing against each other. SMME’s are experts at filling in forms.  The much vaunted new tender system is just terrible.

Supply and Demand. Cause and Effect. This is what is lacking in the Innovation discussion.


Open Data

Open Data Industry Connection Activity Report 1st Year – Feb 2017

As part of the IEEE we are required to submit an annual report. Here is a summary of the report.


We started with a 6 point plan. This we communicated at 2 meetings with representatives of business and government stakeholders at 2 inaugural meetings at the beginning of 2016. At the meeting we shared what we regarded as normative references Appendix A.

We received initial support from local and regional government and various businesses. .

We briefly review the 6 pillars of the plan and successes to date.

Project #1- Local and Regional Government

  • Help municipalities get data into cloud specifically Provincial Government of the Western Cape
  • Look at User Journey for use of the City of Cape Town’s Open Data Portal
  • Look at City of Cape Town’s Open Data business model (See Appendix B for a top level view)
  • Build a community at the City of Cape Town’s Open Data portal
  • Speak about standards, specifically the ETSI / GSM model

In  this regard we had two successes:

We were nominated as the City of Cape Town Open Data public representative and we developed an App using the CoCT Open Data (see Appendix D for screen shots).

Project #2 – Governance and Management

  • Proposed  Structure
    • Scalable
    • Professional
    • Lead up into full standard
    • Include statutory bodies South African Bureau of Standards (SABS), Statistics South Africa (StatsZA) and South African Institute of Electrical Engineers (SAIEE).

We made a proposal around a structure (see Appendix C) and engaged various honourary professional societies who all indicated their support.

Project #3 – Business Engagement

Here we proposed to our initial stakeholders a tiered membership.

Businesses initially proposed some shared projects.

  • FICA Exchange
  • Free Hosting

Project #4 – Tools

Here we wish to investigate various tools on behalf of our stakeholders, such as visualisation, privacy, security, clustering, building solutions, …

Project #5 – Digital Divide

“A digital divide is an economic and social inequality with regard to access to, use of, or impact of information and communication technologies (ICT)”. The question is what are we actually doing about it.

Project #6 – Collaboration National International

We wish to host a Conference on Open Data. We participated in the UN Open Data forum (http://undataforum.statssa.gov.za) held in Cape Town.


A – Normative References

B – CoCT Business Drivers

3.1 ‘Well-run City’: The transparent government programme.

3.2  ‘Opportunity City’: Create an enabling environment to attract investment that generates economic growth and job creation.

3.3 The policy supports the Economic Growth Strategy programme on prioritising competitiveness in City business improvement initiatives, specifically where this relates to governance and oversight and improving the competitiveness of the broader economy of Cape Town.

C – Proposed Open Data Management structure

D – App screenshots – City of Cape Town Open Data Pollution

Welcome to the new Korwe blog

We have started our blog again. We were extremely diligent bloggers, then we decided to get rid of Drupal. The old blog is here.

We will be keeping you updated about developments.

We are creating some showcase for our platform the Tree. We recently created opendata.tree4.mobi. We will do more in the coming weeks. If you have a challenge for us, let us know.

These are showcases of things you can do on the Tree, not the Tree itself. The Tree is the self service provisioning backend PaaS.