The user is at the core of their strategy and they have a passion to provide some value to that user in a measurable way. The product engineer also has an affinity to understand the domain and delve into the conceptual design space of opportunities for the product in that domain. The design space here does not merely refer to a visual design space but a more foundational functional or operational design space of possibilities on how to solve a problem. They know how to build a product that fits the users needs today but can evolve fast in response to the opportunities in the design space triggered by direct feedback from users or indirect insights derived from evaluating the product in use.

A good product engineer participates actively in the distinct phases of a product delivery process. During discovery, they are able to ask questions about the domain that exposes the design space enough to understand how to map real world user needs to the right abstract concepts in technology. They understand the value of prototyping and know when to prototype at different levels of detail. They are able to architect and make technology choices that ensures both functional and non functional product requirements are met while iteratively creating a product that meets the user's needs.

Since the product engineer is keen to measure value delivered to user, they continuously think about methods for evaluating an intervention during delivery and show keen interest in methods for researching the problem space during discovery. They are familiar with the concept of experimentation; they are excited to understand outputs from user research; and they relish the opportunity to look at production usage data in order to get objective insights on real user behaviour and explore creative ways to continue to bring value to the user.

There is a lot more to building a great product team than having engineers (or indeed product managers, experience designers etc) who share the mindset described above. For instance the detail of research and experimentation is a significant part of a great product strategy and requires expertise to ensure things are done rigorously. The mindset discussed here goes alongside the technical rigour of building applications that are secure, resilient and scalable.

Modern product teams are responsible for the entire software delivery life cycle: discovery, building, shipping, and keeping the application running for end users. Building and running a product instils a sense of ownership and responsibility on team members. It also drives the quality of the solutions and provides context for contribution when discovering new ideas for improvements on future versions of the product. Building high quality software products and getting it in front of users is one thing. Supporting the product and knowing what issues users are facing is another. This article expresses some thoughts on principles for an effective support system for product teams.

Shared Understanding of System Architecture

Your team must have a clear understanding of where your services lie in the overall system context. What categories of users depend on your service? What user journeys does your service help fulfil? What other services does your service depend on? What are the failure modes of those services, and what impact do they have on the overall system?

Diagram formats that might help communicate the system here are context diagrams (from C4). In addition, any diagrams showing how requests travel from your clients to the compute running your services would help. An example is a sequence diagram. When used, make sure things like load balancers, firewalls, caches, or proxies are visible. Most of the time these work as expected, but when they go wrong, it is crucial to have a way to determine if they are the source of the issues.

Most observability or app performance management (APM) tools (e.g., newrelic, dynatrace, datadog)  have the ability to automatically generate a service map or a backtrace of how your service has been called over time. Use this where possible, as it provides near real-time (and real-life) data and is unlikely to go stale as the service changes. Be wary of sampling bias when deriving an understanding of the overall system from non-production environments, as this might only be executing and highlighting the easy paths through your systems that humans usually find when running test scenarios.

Clear understanding of responsibility

Once your team has a reasonable understanding of the services they own and their dependencies, they also need to know the team or party responsible for those dependencies. Different organizations refer to this with various terms, including owners, stewards, caretakers or custodians. These are all terms that might be used to describe those tasked with the responsibility for ensuring a service is running as expected in different environments in an organization. Have a clear way to reach out to those responsible for  different services and have a shared place to look up who is on-call for any given service. Tools like pagerduty or opsgenie help to manage such rotas.

Tools and access

In addition to clearly knowing who is responsible, ensure your team have the right access to the tooling that provides visibility into the services you support. This primarily includes access to logs from different services or infrastructure health metrics e.g., cpu, memory, threads etc. Your team should also have access to pipelines that enable the deployment of changes across environments. The ability to accurately and easily determine what version of the application is running in a given environment allows engineers to reproduce and address any issue that might occur. The right access to the right tools would make it easier to find the root cause of the problem. Is it an error on the server-side? Is it a client-side browser version issue? Could it be a misconfigured firewall? Or is it a network issue? If your service is run using containers, ensure the team is able to access the container images  and run them locally when reproducing issues.

Build dashboards that provide easily digestible information about the momentary state of your services, with an option to look at historical trends. Leverage built-in configurations for anomaly detection and alerting, and ensure this is refined frequently enough to be a useful signal and not noise. Fine tune alerts to remove false positives, and ensure your dashboards show at a glance if something is not working as expected.

Know the limits of your tools

Finally, ensure you understand the limits of log retention for your service. How far back in time can you see your service logs? Longer retentions are better but typically more expensive. Work with your team to find a good balance. Your log retention period should ideally span multiple deployments of your service. The reason for this is to allow comparison of multiple application versions when investigating an issue.

Invest in growing knowledge in the team

Set aside time in the team to ensure everyone knows how to navigate the tools required to support the system. Encourage updates and contributions to fine-tuning your alerts  and the dashboards. Ensure every engineer in the team has the opportunity to be part of the rotation to learn from incidents when they occur.

Conclusion

In a world where software products are increasingly complex and integral to operations, building an effective software engineering support team is critical. Such a team not only understands the intricacies of the system architecture and has a clear comprehension of their responsibilities, but is also equipped with the right tools and knowledge to troubleshoot and resolve issues efficiently.

A good product team cultivates a culture of continuous learning and improvement. Each incident is an opportunity to learn and fortify the system against future issues.  Doing so ensures the smooth running of operations and earns the trust of users, which is an invaluable asset in today's digital era.

Acknowledgements

This post was inspired by questions from Siva Subramanian - an insightful, inquisitive and perceptive technologist.

1. remove the *.firebaseapp.com domain from the sign in page

2. add links to your site's privacy policy and terms and conditions to the sign in page

3. add your sites logo to the sign in page (this would require verification)

Step 1: Verify your custom domain on GCP

Ensure your firebase project is selected in GCP. This should be the same project created in firebase console and the same project your site is publishing google metrics and analytics to. Failure to do this might result in a misconfigured project. You could be modifying a completely different project on your account.

Head over to GCP and in the navigation menu select APIs & Services > Domain verification. Follow through the steps for adding your custom domain to Google Cloud Platform.

Step 2: Create and host a subdomain in firebase hosting

Head over to Firebase console and select your project. In the Build section select Hosting. Click through the steps to install firebase cli, initialise your project and deploy to firebase hosting. You don't actually have to run any of the suggested commands for what we are trying to do. Finally, select Continue to console.

Once in the Hosting console, select Add custom domain (e.g., authorization.yourdomain.com) and follow the wizard. You will need to verify ownership of the domain.

Step 3: Customize the authorized redirect uris

Head over to GCP and select APIs & Services > Credentials and in the OAuth 2.0 Client IDs section, edit your OAuth Client.

Ensure your custom auth domain (from step 2) is added as an authorized redirect URI. https://authorization.yourdomain.com/__auth/handler.

Ensure also that your naked domain (e.g., https://yourdomain.com and https://www.yourdomain.com) is an authorized javascript origin. This depends on where your website is served.

Save the OAuth Client edit form.

Step 4: Update your OAuth Consent Screen

In the GCP under APIs & Services > OAuth consent screen, select EDIT APP

Update the application name if required and in the App domain section, add links to your website's home page, terms and conditions and privacy policy.

Ensure your domain is listed as an Authorized domain in the Authorized domains section.

If you wish to add a logo, then upload a logo. Note that this would require verification and might take a few weeks.

Save and Continue the form - while accepting defaults. If you need additional restricted scopes you will need to justify that to google and go through the verification process.

Step 5: Update your application's firebase authDomain config

Head to you application code configuration. Look for the authDomain firebase config property and update it to authorization.yourdomain.com

That's it! Congrats. You have completed the process of customising the firebase google sign in page.

The Nigerian Bureau of Statistics (NBS) publish some data on selected food prices across different states in the country starting from January 2017. You can find the data in the elibrary under Selected Food Prices Watch ([Month, Year]). Anecdotal evidence from social media posts contain emotional messages that signal that the people's purchasing power has significantly declined. Some are reporting that the price of staple foods like beans has tripled in the span of a month.

This article uses heatmap visualisations to explore the trends in the price of selected food items tracked by the Nigerian Bureau of Statistics between January 2017 and August 2021 across Nigeria.

Notes about the data

• It is unclear exactly how the data is captured and for each state, how many samples are taken to derive the value that is openly shared by the NBS.

• The data published (and specifically the dataset used in this article) is published for the preceding month. Hence there is a delay in what people are momentarily feeling and expressing across social media and the story the dataset might tell.

• Understandably, it is not possible to capture the price of all food items. I trust that the NBS has good reason for selecting the foods they publish data on.

• Some of the entries in the dataset contained outliers which where immediately evident when visualising the data. Specifically the entries for Kwara state, January 2018 for Catfish (dried) was ₦13,465, Imo state, October 2018 for Catfish (dried) was ₦24,780.5, Kano state March 2019 for Frozen chicken was ₦11,400. All three entries were ten time larger than entries for the preceding and succeeding months. For the visualisations in this article, these entries were rendered as a tenth of their reported values.

• The precise quantity or weight of some of the food items is unclear. For example a tuber of yam come in different sizes as does beef with bone in.

• For the purposes of this article, the January 2000 entry in Benue state dataset was relabelled as December 2019 based on it's position in the sheet.

Regional Differences in trends

When there is a lot of data, it is generally a good idea to look for trends. I'm a fan of Shneiderman's visualisation mantra: Overview first, zoom and filter, then details on demand. This article presents the 'overview' part of the exploration. Filtering and details are left for the curious reader to explore in the interactive notebook that accompanies the article.

South-South and South-East zones have seen the largest changes in the price of onions. According to the data, the spike started around November 2020. Conversely, Northern zones (except Abuja) generally saw some stability in the price of onions. The cost of onions reduced in 2021 in many North-Central and North-Eastern states.

In general, the North-Eastern and North-Central zones (except Abuja) have seen a drop in the price of fresh Tilapia and Yam since 2017. There is a similar trend for the price of catfish which you can explore in more detail in the accompanying observable notebook.

The data also shows that in many states, in 2017 and 2018, yam is more expensive in the middle of the year than it is towards the end of the year. The reported cost of yam is also less in North-Eastern and North-Central Nigeria than it is in other parts.

Things got better then worse

A close look at foods like white and yellow gari show that there were similar levels of reported prices for these food groups between 2017 and 2018. Things got cheaper around 2019 with some variations across states. There are some exceptions for states like Kebbi, Kogi and Bayelsa, which saw more expensive prices when compared to states in the same regions. The prices then got higher again in 2020.

The dip in the cost of foods like gari in 2019 and the consequent increase in cost might contribute to how people are feeling about the affordability of basic food items. The data shows that some parts of Nigeria, experienced higher prices in 2017 than in 2021.

Further exploration

According to the data recorded by the NBS, the South-South and South-East have seen the sharpest increase in price for most of the selected foods. Due to the delayed nature of the data reported by the NBS and the unknown nature of how the data is actually collected, these pictures are merely a visualisation of the data available and do not attempt to say anything about reality. Indeed based on a few checks with personal contacts in Lagos, the prices reported in the real world were higher than what the NBS data reported. Perhaps there is opportunity to crowdsource this information (in addition to  the work the NBS is doing open sourcing access to the dataset).

If you enjoyed the visualisations in this article, there is an accompanying observable notebook where you can view the heatmap for all the selected foods across all the different states. There is also a histogram showing the price distribution for the selected food. Check it out here.

Acknowledgements

Data used in the visualisations shown in this article is provided by the Nigerian Bureau of Statistics and was downloaded on October 9, 2021. The preview image used in this article is by Dennis Irorere on Unsplash.

Appendix

Selected foods

Here is the list of selected foods tracked by the NBS:

• Agric eggs medium size

• Agric eggs(medium size price of one)

• Beans brown,sold loose

• Beans:white black eye. sold loose

• Beef Bone in

• Beef,boneless

• Bread sliced 500g

• Bread unsliced 500g

• Broken Rice (Ofada)

• Catfish (obokun) fresh

• Catfish :dried

• Catfish Smoked

• Chicken Feet

• Chicken Wings

• Dried Fish Sardine

• Evaporated tinned milk carnation 170g

• Evaporated tinned milk(peak), 170g

• Frozen chicken

• Gari white,sold loose

• Gari yellow,sold loose

• Groundnut oil: 1 bottle, specify bottle

• Iced Sardine

• Irish potato

• Mackerel : frozen

• Maize grain white sold loose

• Maize grain yellow sold loose

• Mudfish (aro) fresh

• Mudfish : dried

• Onion bulb

• Palm oil: 1 bottle,specify bottle

• Plantain(ripe)

• Plantain(unripe)

• Rice agric sold loose

• Rice local sold loose

• Rice Medium Grained

• Rice,imported high quality sold loose

• Sweet potato

• Tilapia fish (epiya) fresh

• Titus:frozen

• Tomato

• Vegetable oil:1 bottle,specify bottle

• Wheat flour: prepacked (golden penny 2kg)

• Yam tuber

Inspiring people to have the sense of agency to drive change starts first with providing the opportunities for change to happen and providing enough feedback and transparency that links their actions to the resulting changes. A necessary change I envision in a significant part of the world today is access to data. Nigeria in 2021 has an estimated median age of 18, with about 55% of the population between ages 15 - 65. A household survey by the World Bank showed that about 74% and 20% of people have access to telephones and internet respectively (with regional variations). The state of electronic information currently available is fragmented in most sectors. Information and data exist and are provided by different sources often aggregated and summarised for general consumption. Deriving any source of insights from the data remains challenging.

There is opportunity to educate and pick the narrative we use to create factual information that are culturally relevant and that people can relate to. The first step in achieving this is curating and collecting foundational data and ensuring people have access to consuming the data as well as contribute to enriching and modifying the data. Some are already doing great work in this space including Tracka, BudgIT, shineyoureye, and the National Bureau of Statistics. These platform have the Nigerian populace as a direct intended consumer. I believe it is time to empower the young technophiles with access to structured data upon which they can build tools, derive insights and tell their own stories.

Citizen Science Nigeria is a technology-first platform with initial focus on curating and publishing records on tenures of government officials in Nigeria's Fourth Republic. Over the last decade, I've heard statements such as:

"We don't have any young politicians..."  and "We have been governed by the same people for the past 5 decades..."

I understand these sentiments given the state of the country and its potential. They however lack the actionable insights I'd like to inspire in a growing young population. Such statements are often paired with heavy negative emotions due to constant frustrations felt in the country and are rarely backed up with any objective data because even when it exists, data is not reliably available. Moreover Nigeria is a large country with a diverse population. General statements like these undermine any efforts and progress achieved by sub-communities across the country. So how does one nudge those with access to technology to seek out objective data? Assuming a world where coherent and cohesive data exists, for instance on public office holders in the Nigerian government, would it change the way people feel about politics? If the platform hosting the data encouraged engagement and objective discourse about insights that are being derived as a result of the curation and aggregation of the data, can we nudge people towards making more data-driven decisions?

There is a very long road ahead and I do not claim to have answers to the questions I pose. I have ideas. Lots of ideas. This is one very small step in providing access to information that I believe will have a multiplier effect on decision making for future generations.

Acknowledgements

The thumbnail image used in this post was by Rohan Makhecha on Unsplash. Jeremiah Iyamabo provided valuable comments on draft versions of this document.

In March of 2020, Nigeria declared a state of emergency and went into national lockdown for 4 weeks. The Nigerian Bureau of Statistics in collaboration with the World Bank conducted phone surveys starting April 2020 to get a sense of the impact of COVID-19 on households in Nigeria. This article walks through a series of heatmaps to get an overview of typical home schooling strategies that households from different parts of Nigeria are utilising.

The data used for the visualisations were collected during the first 5 rounds of phone surveys conducted in 2020. The households interviewed were from those in the living standards measurement survey. The table below summarises the periods when the data for each round of survey was collected.

Start End Cycle 20-04-2020 11-05-2020 Round 1 (Baseline) 02-06-2020 16-06-2020 Round 2 06-07-2020 20-07-2020 Round 3 09-08-2020 24-08-2020 Round 4 07-09-2020 21-09-2020 Round 5

Except for in round 1 (the baseline round), the households that  participated in the survey were roughly evenly distributed across geopolitical zones and approximately 60% of the households interviewed were from rural areas.

Amongst several other key indicators, the interviews explored the impact of COVID-19 on education. Specifically, it explored the strategies adopted by different household to cope with the school closures. These strategies included: the use of assignments, the use of educational mobile applications, the use of educational television, the use of educational radio, studying, tutoring by parents or other household members and tutoring by a teacher.

Each household could respond affirmatively to using a combination of these strategies. The heatmaps presented below serve at least two purposes:

1. Are there home schooling strategies that are particularly popular across all states?

2. Are there any clear changes in these strategies over time , i.e., across the different rounds?

Interpreting the heatmap

Each cell in the heatmap is the number of households in the given state that reported using that strategy as a coping strategy for managing education while schools were closed. The heatmap is rendered for each of Round 1 through 5 and the for each state.

The benefit of a heatmap is the quick perception of hotspots in the data. The value encoded in the coloring (i.e., the heat) in this case is the number of households in a given state who responded affirmatively to using the corresponding strategy. Hence we can see at a glance, the relative difference between the adoption of different education strategies across states and over time (i.e., the rounds). Each column represents the households that reported using the strategy for homeschooling. Each cell in the column represents a state and the color of the cell represent the proportion of households who use that strategy and are from that state. A red cell would simply mean more people reported using the strategy than a yellow cell. Vertical streaks show the strategies that are popular across states. The chart shows that nationally, and perhaps across all rounds, tutoring by parents/other household members, as well as studying were popular strategies. This visualisation merges responses from both rural and urban sectors. The cluster of strategies we see are by chance and we are still not clear on how the popularity of the strategies change over the rounds.

The charts below show how the popularity of the different strategy change over the rounds. Both axes are now sorted. On the vertical axis, the state with the highest response rate is at the top. On the horizontal axis, the strategy with the highest response rate is on the left. This is why we see the darker red cells on the top left corner. While studying and parent tutoring are consistently popular across rounds, engagement with tutors were not as popular earlier in the year. Now let's look at the data split by sector.

Effect of sector on home schooling strategies

The heatmap below shows households from urban areas. The reader might notice more holes in the visualisation. These holes represent states who did not report using the corresponding strategy. Where the state is missing entirely, then there were no households from that state who reported continued engagement with education for that round.

If we plotted the same for households in the rural parts of the country, we see states with sparse representations in the rural sector. For this visualisation these were states such as Borno, FCT, Lagos or Ekiti.

We could also just focus on a single strategy to see how different states compare over different rounds for that strategy. That would drive discussions such as "Which states are more likely to engage in learning by mobile apps?" The image below shows that Lagos is where people are most likely to report having access to mobile apps for education. Curiously, Benue, Kwara and Enugu states started off stronger with households reporting use of mobile apps for education purposes. At later rounds, we see households reporting less use of this strategy. Adamawa on the other hand has seen a steady increase in households reported to have used this strategy to cope with the closure of schools.

We could take the visualisation a step further and render the heatmap just for a single state over the 5 round of interviews. This would allow us to see how different strategies compare across different rounds for a single state. The possibilities are endless.

These visualisations were generated using vega-lite in observablehq. There is an accompanying interactive notebook with the dataset used to generate the pictures and all the code. Do check it out.

Acknowledgements

This article was written based on data collected by National Bureau of Statistics. Nigeria COVID-19 National Longitudinal Phone Survey (COVID-19 NLPS) 2020. Dataset downloaded from https://microdata.worldbank.org/index.php/catalog/3712/get-microdata on 31 December 2020.

Jeremiah Iyamabo provided valuable comments on draft versions of this document.

This is the second part in a two part series that explores access to digital technology in Nigeria. The first part explores the effect of geographical location and sector (rural vs urban) on access to technology reported by nearly 5000 households. This article will explore the effect of age and gender on access to digital technology in Nigeria.

About the data

The survey covered 17,713 (9,040 female and 8,673 male) individuals across 4,971 (1,593 urban and 3,378 rural) households. All individuals were at least 10 years old and the households were sampled from all 36 states in the country, including the nation's capital.

Access to Telephones

Those who reported not having access to phones were mainly young people under 16 years old. Although 50% of those without access to phones and younger than 25 years old were female, a larger proportion (73%) of those over 25 years old were female.

It has been well reported and documented that there is a gender divide when it comes to access to technology. This is particularly true for developing countries. We took a closer look to see the effect of gender on reported access to a phone. Although the data shows that nationally, 49% of those with access to phones were female, there is more of a gender divide when looking at access by state. The chart below shows that in states like Kebbi, Zamfara, Taraba, Kaduna and Katsina, females disproportionately have less access to a phone compared to males.

Access to Internet

Participants with access to internet ranged between ages 10 - 90 years old (n = 4658) with 75% of participants being 39 years old and younger. Of those with internet access, 51% came from urban areas across the country with 49% from rural areas.

In a similar way that access to phones were biased towards males, access to the internet was even more so. Even though  51% of those who participated in the survey were female, only 42% of those who had access to the internet were female. Simply put, there were only two states in Nigeria where internet access for females reached or exceeded the level of female representation. The chart below shows a state level breakdown of internet access by gender.

Combining the state data into groups of Nigeria's geo-political zones, one can see that the Northern regions typically have a lower proportion of females with internet access as well as access to phones. The lowest proportions were recorded in the North West region. The South East region has the highest proportion of females with access to phones (54%) and internet (47%).

Closing thoughts

The regional differences in female representation evident across the country from this study is worth some reflection. One would expect that the level of education of individuals  affects their access to jobs. Employment affects standard of living, which in turn affects likelihood of affording the cost of technology. If education and employment are the strongest barriers to technology adoption, then in comparison to other regions in Nigeria, those from South-East Nigeria appear to be doing a better job of bridging the gender divide.

So, to what extent are you willing to risk excluding market segments from that tech idea or innovation? Have you carefully considered lowering the entry barriers to include existing conditions that the target market currently lives by? If you would like a closer look at the data and visualisations from this article, see this observable notebook.

Acknowledgements

This article is based on data collected by the Nigeria National Bureau of Statistics. General Household Survey, Panel (GHS-Panel) 2018-2019. Dataset downloaded from www.microdata.worldbank.org on June 9 2020.

What is the true market size of that tech idea you have? What demographic does your product exclude? Read on to find out the effect of location on access to digital technology in Nigeria.

Digital technology has no doubt had a significant impact on the manner and scale in which things are done across the globe. From profound differences in ways information is processed to various utilities in entertainment, technology has affected all aspects of our livelihoods in one way or another. It is easy to overestimate the efficacy of digital technology at solving problems especially when creating solutions for emerging markets. Technology grows so fast, we often forget that for emerging markets, its dissemination and adoption is nowhere near as fast as seen in other parts of the world.  

The World Bank, in collaboration with the Nigerian Bureau of Statistics, has been conducting longitudinal surveys to get a better understanding of the living standards of Nigerian households. This article reports on typical households in different parts of Nigeria and their access to technology. Specifically, we use access to phone and access to internet as proxies for access to technology and we report on the regional distributions of technology access across different parts of the country.

About the data

The survey covered 17,713 (9,040 female and 8,673 male) individuals across 4,971 (1,593 urban and 3,378 rural) households. All individuals were at least 10 years old and the households were sampled from all 36 states in the country, including the nation's capital.

Access to Telephones

Since the telecommunications boom that started in the early 2000s in Nigeria, access to telephones has got better over time. Of all individuals included in the survey, 79% had access to at least one phone. When looking from the point of view of households, 94% of households had access to a phone. The majority (81%) of those with access to phones obtained access using personal devices and these were mostly evenly split between rural and urban areas.

The proportion (79%) of participants that have access to phone is respectably high. In the U.S, access to mobile phones is 98% as at 2019 according to Pew Research Center. Access to smartphones is much lower at 81%. A closer look at states shows some differences. Participants from Osun state had the highest likelihood (98%) of having access to phones while participants from Zamfara were least likely (49%) to have access to phones.  In addition, most states consistently showed more likelihood of having participants with access in urban than in rural areas. There are a few exceptions to this as shown in the chart below.

Since its introduction in 2002, Nigerians have increasingly gained access to GSM mobile telephones. Although, high transaction cost limited access at the early stages, competition reduced entry barriers with such initiatives as, unbundled offers, per second billing, and free SIM cards. Recently, the industry regulator, Nigerian Communications Commission (NCC), reported that telecoms contribution to Nigeria's GDP increased to 10.6%  as at Q4 2019. In July 2020, FBN Quest Capital attributed telecoms growth to instant payment transactions conducted through mobile devices. Clearly, disproportionate access to mobile phones would have carryover effects to other key functionalities such as, finance and access to information.

Access to Internet

Another important aspect of technology is the internet. It enables digital communication across devices at different geographical locations. Nationally, only 26% of individuals (n = 4,658) indicated they had access to the internet. This varies significantly by state as shown in the charts below. States like Lagos and Abuja had the highest proportion of individuals with internet access. Further analysis also shows that individuals from urban areas are more likely to have access to internet than those in rural areas.

Internet access modes

The most common reported source of access to internet was from personal devices (69%) followed by devices owned by other members within the same household (17%). About 7% accessed the internet through cyber cafes while 6% had internet access through a relative or friend (i.e., in a different household).

Nationally, 55% of the individuals with internet access on their personal devices came from urban areas while 45% came from rural areas. Lagos and Akwa Ibom were the states with the most people who had internet access on personal devices while Yobe and Sokoto had the least. Individuals from Akwa Ibom, Gombe and Niger were most likely to access internet from another household device. Over a third of individuals who reported accessing the internet by cyber cafes were from Kaduna, Gombe and Abuja, with 15% coming from Kaduna alone.

Implications

The data shows that access to phones is far more widespread than access to internet. As a result, access to phones should not be used to imply access to internet. Instead, services and solutions should be built with regional sensitivity to what is available to the demographic. Moreover there are several factors that we have not explicitly considered in this article. For instance, of those with internet access, what sort of bandwidth can they afford? What sort of bandwidth do the Internet Service Provider's infrastructure provide access to? What utility does the internet  mostly serve e.g., communication, social media, e-commerce, education, entertainment? We talk about the speed of internet access in Nigeria in a separate post.

Increasing tech adoption in Nigeria typically implies significant investment. Yet, nailing down how Nigerians would respond to such innovations has been evasive. To cite just one example, results from large scale interventions to increase women's access to mobile banking in Nigeria, showed no difference between treatment versus control conditions. Such outcomes should cause stakeholders to pay attention to the basics; that is, what does the data we already have tell us about access to technology in Nigeria, particularly for underrepresented and marginalised groups?

If you would like a closer look at the data and visualisations from this article, see this observable notebook. A future post will explore the effect of gender and age on access to digital technology.

Acknowledgements

This article is based on data collected by the Nigeria National Bureau of Statistics. General Household Survey, Panel (GHS-Panel) 2018-2019. Dataset downloaded from www.microdata.worldbank.org on June 9 2020.

Revision History

11 July 2021 - updated 74% to 79% in second paragraph of the Access to Telephones section.

It is easy to get carried away by the potential of the internet and lose sight of the numbers behind the utilities we are creating. Today's technology companies are filled with rich user experiences that require a minimum amount of internet bandwidth for a frustration free experience. As a result of Covid-19, 2020 has proven to be the year where distributed computing is accelerated, and people all over the world have been forced to operate remotely where possible.

The world has come a long way since the days of dial-up internet through landlines. The majority of people in Nigeria today access the internet through their mobile phones. What does the current infrastructure in the country allow with respect to speed of the internet? This article looks at typical download speeds that one can expect to get across different locations in Nigeria.

While doing research for this article, the majority of sources about internet speed pointed to speedtest by ookla. According to Ookla, the mobile internet speed for Nigeria for the month of July 2020 was 15.76Mbps (download) and 7.18 Mbps (upload) while the fixed broadband speed was 11.87Mbps (download) and 10.59 Mbps (upload). You can find a typical country level report here.

The speedtest.net websites says:

"Results are updated mid-month for the previous month. January 1, 2019 onward countries must have at least 300 unique user results for mobile or fixed broadband to be ranked in either category. Prior to January 1, 2019 we required 670 unique user results for mobile and 3333 for fixed broadband."

Ookla's detailed report is not easily accessible. I had several questions about the chart above. What's the actual sample size? From what locations were the samples taken? I am usually very sceptical about overly summarised datasets. For a metric that bears so much significance in today's world,  transparency should be the norm.  So I made an inquiry to ookla.

Open technology for measuring the internet

I didn't hold my breath for a response from ookla. I continued my research and came across measurement lab (M-Lab). According to M-Lab:

"M-Lab provides the largest collection of open Internet performance data on the planet."

After obtaining access to the GCP MLab BigQuery project, I was able to get some data on internet speeds in Nigeria.

The data for Nigeria from M-Lab for 2020  (between January  and August 30 2020) contains 347,899 samples. The samples cover 30 states with Lagos responsible for 78% of the samples. The data shows that for the year 2020, 75% of the sample had download speeds of 3.8Mbps or less with a mean speed of 3.6Mbps and the median of 1.34Mbps. This is significantly less than what is reported by speedtest for July 2020 and indeed for most of the past few years as evident in several reports. The speed (15.76Mbps) reported for July 2020 by ookla  is only available for the top 5% of the samples that contributed to the data from M-Lab.

Internet speed across states

What connection speeds can one expect to get from the different states across the country? The sample size and range of connection speeds used to derive the mean values varied significantly across states, with Lagos contributing the most. Although the top speeds were recorded by samples from Lagos, Niger and Abuja, other states recorded higher mean download speeds.

A deep dive into the data available for Lagos might be useful to see the locations with typically good connections. One might also explore the effects of population density or living standards on the recorded download speeds. Note that although the MLab data contain information about upload speeds, this article only reports on download speeds.

So what?

How closely internet speed estimates reflect the target population has significance in several domains. It determines how accurately you can predict the dissemination of technology solutions for that target population and the successful adoption of such solutions. A badly predicted network speed would have a significantly negative effect on any product that depends on a baseline throughput for an optimal user experience. For instance what's the true market size for pain free streaming video media in Nigeria? How can you rethink digital media for a market with poor internet connection speeds? The earlier your product team knows about this, the earlier they can design the right product for the right market. For policy makers or those in a position to make decisions on the global growth an accessibility of the internet, it is essential that the feedback loop used to measure success is representative of the population that is being served.

Acknowledgements

The visualisations in this article were created in ObservableHQ using M-Lab NDT Data Set 2020-01-01–2020-08-30. The data were specifically obtained from the Bigquery table measurement-lab.ndt.unified_downloads

The type of foods consumed and the quality all have effects on health and well being. This article explores the cost of living across different states in Nigeria, with a focus on how much people can expect to spend on food.

This is part of a series to generate more visual representations of data captured by the National Bureau of Statistics during the General Household Survey, Panel 2018-2019, Wave 4. This article specifically looks at the post-harvest questionnaire with a focus on typical expenditure on food over a period of 7 days.

The data used for the visualisations can be found here. More specifically, the charts in this article were generated from the following datasets:

• sect1_harvestw4 - the number of individuals in each household at the time of the survey was derived from this data file.

• sect10b_harvestw4 - the reported food purchases by each household (see here) was summed to get a total amount spent on food per household. In order to get a normalised view, the total amount was divided by the size of the household. The charts below report the amount spent per person over a period of 7 days.

A primary goal of this article is to provide insights to the reader about how much they spend on food and how they compare with the sample population that took part in the survey. More generally, given a target demographic, the reader should be able to derive estimates of their food expenditure.

Typical food expenditure over a period of 7 days

The distributions below show the amount spent per head over a period of 7 days for rural and urban areas at the national level. Not surprisingly, we see typically higher spends in urban areas for the different quantile ranges highlighted (in red vertical lines). For instance to get into the top 1% of spenders in the rural regions, you should spend about ₦9,000 per head over a course of 7 days. For the urban region you would need to be spending around ₦11,000 over the same period. The mean spend per head over a 7 day period for rural and urban areas are ₦1473 and ₦2292 respectively.

Mean food expenditure by state

Visualising the food expenditure by state shows that the southern part of the country spends more on food per person. Enugu, Bayelsa and Abuja had the highest mean food expenditure. This could be for a few reasons. It could be that food is more expensive in these regions. It could also be that people are buying more food or that people are buying food of a certain type which happens to be more expensive. In the following sections we explore other factors that might affect the amount spent on food.

What affects amount spent per head?

A scatter plot of the amount spent per head by the number of people in the household shows there might be a negative relationship between these two variables. The more people are in a household, the less that household appears to spend per head. The charts below show data for all rural and urban areas across the country. Here, we see the mean spend per head for each household size. The size of the discs in the chart represents the number of households that contributed to the mean score.

Both of these charts show that the more people there are in a household,  the less that household spends on an individual member. This is not surprising. Most households have just two adults earning a living to support the livelihoods of the other members. An increase in household size given the same disposable income implies there is less money to allocate to each member. It is also possible that larger households are more common in geographical areas where people tend to have less disposable income. The level of poverty in an area might also drive various factors that encourage larger households.

Although the data shows we have significantly lower number of households who have more than ten members (see the sizes of the discs for reference), it is reasonable to expect that given enough data, this relationship stabilises on an amount that designates the baseline cost of food per person, in a given region. Hypothetically then, a bare minimum amount you can expect to spend on food over a period of 7 days might be ₦500 and ₦1200 for rural and urban areas nationally. That equates to a daily amount of ₦71 and ₦171 for rural and urban areas.

If you would like to explore the data in a bit more detail, have a look at the observablehq notebook where the visualisations that appear on this page were generated.

Acknowledgements

These visualisations are driven by data collected by the Nigeria National Bureau of Statistics. General Household Survey, Panel (GHS-Panel) 2018-2019. Dataset downloaded from www.microdata.worldbank.org on June 9 2020.

In June 2020, I noticed the Living Standards Measurement Study (LSMS)  organised by the World Bank. These studies are carried out in different parts of the world to better understand households and provide policy makers with information that would help to positively affect society. I got particularly interested in the Nigerian Living Standards Survey. There are official reports published for the datasets. I however decided to explore the data and create my own visual interpretations. This article is a summary of the tools I found useful and the processes I go through when exploring the datasets.

Data Files

The datasets are provided as CSV, Stata or SPSS files. I opted for the CSV as it did not tie me into having any special licences. My intent for the dataset was to first get a visual overview of the dataset to understand if there are any identifiable trends before deep-diving into more rigorous statistical analysis. Each  of the LSMS survey is accompanied by a a series of data files that fit a relational database model.

AWS Athena

Amazon's Athena is part of the data ops tools for creating a platform that allows data analysis in the cloud. You provide it datasources such as logs, CSV or JSON files. These files are typically uploaded to AWS S3.  You then create a crawler that transform the input data into a relational format using AWS Glue. Glue is able to run a scheduled job that automatically updates the generated tables. Finally, Athena provides a dashboard for querying the tables using SQL.

One gotcha for the CSV dataset I was processing was that the AWS Glue crawler expects that each datafile exists in its own folder before it created a table for each file. I remedied that by creating a shell script that looped through the files in a given directory, created a folder with the same name and moves the file into the folder. See below for an example.

#!/bin/sh
for filename in  --REPLACE-WITH-PATH--
do
  directoryName=`echo $filename | cut -d'.' -f1`
  mkdir $directoryName
  mv $filename $directoryName
  echo "moved $filename to $directoryName"
done

echo "done"

I was able to complete the data import of 94 tables into AWS in under 15 minutes. I am also exploring ways to build some automation around provisioning users who might be interested in querying or analysing the data using SQL on AWS.

Note that Athena costs some money to run. The costs are proportional to the amount of data scanned while processing a query. Familiarise yourself with the pricing before using it.

Observables HQ

Once I get a good view of the data which might involve filters, grouping and joining multiple tables, I export new views of the data to CSV and visualise the new datasets using d3 or vega-lite on observablehq. Observablehq provides an interactive platform for quickly generating visualisations.

Why not use MySQL workbench?

MySQL workbench community edition is a free tool and provides the ability to import CSV files as tables into a MySQL database. It attempts to determine the column type based on the values in the first few rows of the data file. I was pretty optimistic. My optimism dwindled once I saw how error prone the process was. For example, determining the column type from the first few rows means that if the type has been determined to be an INTEGER and the import engine later finds an empty value, it throws an error or fails to import the entire record instead of  writing a NULL value to the cell. The wizard creates a VARCHAR type for columns whose fields are empty in the first few rows. This quickly became a tedious process. Some of the tables had over 20 columns with nullable INTEGER values. I found myself changing most of the columns to VARCHAR since it was the only way to be sure that all the records in the data files were imported into the database tables.

Aside from manually updating the column types, the import process was slow. For instance, it took nearly an hour to import a data file with half a million records. Attempting to run multiple imports at the same time showed lots of errors in the logs and did not fill me with confidence about the records that were successfully imported into the tables.

I’m publishing this as part of 100 Days To Offload. You can join in yourself by visiting https://100daystooffload.com.

The National Bureau of Statistics in Nigeria in collaboration with the World Bank have been running a series of surveys to better understand the Nigerian household and to provide much needed data when making policies. The following charts are an attempt to provide a visual digest of demographic data from the most recent General Household Survey - Panel that took place between 2018-2019.

There is an official report of the survey published here. It's a table heavy report and does not afford an easy digest of the underlying data. The following are some of the charts I generated from the data. These were built on observablehq. The goals of these charts are as follows:

1. Get an overview of the age distribution of individuals in the households that were surveyed with any insights on differences in urban or rural sectors.

2. Get an overview of gender distribution for age-groups of the individuals and see if there are any differences for different sectors, geographical regions or states.

Age Distribution

The distribution of ages shows that 75% of the population are under 35 years old. There is also an interesting artefact in the data showing there are more frequencies of ages divisible by 5, i.e., 20, 25, 30, 35 etc. The national median age is 18 and mean is 24. There seems to be a slight difference between rural and urban areas, with a  median age of 17 and 19 respectively and a mean of 24  and 25. This is on par with the Wikipedia entry for Nigeria on the list of countries by median age.

Age distribution by sector and gender

The survey sampled households from all states in Nigeria from both rural and urban sectors. With the exception of population size, the chart below shows very similar trends for both sectors - a population skewed to the right. Some notable things are that there are more men under 20 years old in rural areas and more women than men in their thirties and forties.

The shape of the distribution shows a very similar trend for the rural and urban sectors. In general most of the individuals in the study were young and less than 20 years old.

The ratio of male to female across states

The distribution above shows differences in genders across age groups and sector. In order to see if there are any notable differences between different states I decided to create a different chart that visualises the ratio between male and female for the different age groups for every state.  By using a ratio derived from a variable with two possible values, it is possible to visually represent the proportion of male to female in each age group over a continuum across different ages. So instead of using two separate lines (one for each gender), I draw a single line whose y value varies between 0 - 100. When the line goes below the 50 mark it designates having more women for that age  and when it goes above the 50 mark it designates having more men. In order to get a sense of how the state compares to the country as a whole, the black dashed curve shows the ratio at the national level. For a bit of context, I plotted a histogram above the chart to give the reader information about the number of individuals that are in each age group. Clearly the number of samples contributing to the ratios decline significantly for the older age groups.

Below is the trend for each zone. Aggregating by zone allows a higher level picture to see typical variations from the national trends. Here we see the North West zone of Nigeria has more older men (between fifty and early eighties). The South East zone typically has more women across age groups above 20.

It is not surprising to see difference across states for gender ratio in different ages. Nigeria is a very large and diverse country with very different cultural practices and norms across different geographic regions. If the data were representative of the region, the gender ratio charts should signal that the states that trend significantly above or below the national trend should be explored in more detail. Looking through these charts should inspire you to ask more questions about how we support people of different gender and ages across different parts in the country.

In a future post, I'll be looking at changes in households over the years that the survey was carried out. If you enjoyed these visualisations and you are interested in drilling down into the data a bit more, there is a live interactive notebook here. Do reach out to me if you have any questions about this data or the visualisations.

Acknowledgements

These visualisations are driven by data collected by the Nigeria National Bureau of Statistics. General Household Survey, Panel (GHS-Panel) 2018-2019. Dataset downloaded from www.microdata.worldbank.org on June 9 2020.

Special thanks to Jeremiah Iyamabo and Michael Braimah for helping to refine this article.

I consider myself a pretty chilled person but supermarkets often get me stressed. A few years ago, an environmentally conscious friend of mine said to me: "I really like that shop XYZ now allows you the option of not printing out a receipt. I think they should remove the need for receipts all together." I remember thinking: "Nah. You can't take away receipts. Society is not ready for that."

People who find themselves in a minority community change their day-to-day behaviours to avoid reinforcing society's biases and stereotypes. A lot of these behavioural changes are hidden from people outside the communities. This is in no way an attempt to imply that the  experiences I share in this article are common to many. I have no idea. I am a black man in the UK. I acknowledge that people are different and there are different challenges around the world.

These days it is common when at the till of a supermarket to be asked : "Would you like a receipt?" My response to this question depends on a few things. Firstly, is this question being asked by an attendant? Or is it being asked by a self-checkout bot. If the question is posed by an attendant, are we sufficiently close to the exit? This proximity is important to allow the attendant easily clarify any potential misunderstandings that might ensue if a security tag in my purchase sets off the alarm and a security guard decides to ask for my receipt. You would think I set off the alarms a lot. I actually don't. However if I walk out of a store at the same time as someone who is not black and the alarm goes off, I am more likely to be stopped and searched. I dread these situations.

If I am at a self-checkout till,  the number of items I'm purchasing also influence whether or not I opt for a receipt. When buying more than a couple of items, I'd be opting for a receipt.

I started thinking a bit more about this after I went to the local WHSmith supermarket to buy a pack of sweets. I paid the attendant with my contactless card.  Since I was close to the exit, I bought a single item and thanks to modern day banking, I had a transaction notification on my phone that matched my purchase, I opted out of getting the receipt. Shortly after stepping out of the shop, my wife called  to ask if I could pick up something else from the shop. I don't have a receipt for the bag of sweets I'm wielding and no, I'm not confident enough to walk back into the shop with a bag of sweets I've just paid for. Even if I'm lucky enough to meet the same attendant at the till, it would be unfair to expect them to remember the transaction. The most reasonable course of action was to go home, drop my bag of sweets and return to the supermarket to purchase the other item. Lucky for me, home was less than 10 minutes walk away.

On a separate trip to the local supermarket, I went through the self-checkout and got my receipt when asked if I wanted one. I was grocery shopping for the week. As I wheeled the shopping cart out of the store the alarms went off. Oops! I waited for a few seconds because I was not sure what to do. Since no one approached me I carried on walking towards the main exit where I was confronted by a security guard. He asked if I set off the alarm, I said: "I think so." He then asked if I had any steak in my trolley to which I answered yes. I presented my receipt on request and he let me go. He appeared to be checking that I had paid for the steaks. Good thing I opted to print the receipt.

I spent sometime to reflect on the alternate scenario where I opted not to print the receipt and the outcome is not very pleasant. I could show him the notifications from my mobile banking app that I did pay some amount of money to the supermarket just now. Sadly that does not show that I am entitled to the steaks I bought. The statements are not itemised. Perhaps he'd have to take me back to the tills to scan all the items and check it against the amount on my app? Maybe customer service can help re-print the receipts? But what about other shoppers observing my interaction with the security guard? What stories are they telling themselves about the scenario before their eyes? Would they realise they actually don't know anything about what is going on? Would they choose to make up an unpleasant explanation about what they saw? Is the security guard properly trained to respond fairly to this scenario?

I do not feel this way when I visit a store that provides a bag at the till. I think it is because I feel the bag is a token that signals onlookers that I have been through a process that clears me from any potential misunderstandings that might arise from failures in other parts of the system. I am not a security guard so I don't know if any of these have an effect on how they react to alarms.

I don't have very much faith in a system that portrays a race with a higher amount of suspicion than others. As a result I change my behaviour. There are countless other seemingly innocuous changes people make to their lives when performing mundane tasks like going to the supermarket. Most of this happens internally to each individual and it inevitably has a compounding effect on quality of life and interactions with society. I am constantly thinking about how society perceives my actions especially when those actions have a chance of an outcome that reinforces a negative stereotype.

Thank you for reading. Remember that you probably don't have all the context when you interact with people outside your immediate households. Be slow to making judgements and no matter what your individual experiences are, try not to project it on other people.

I’m publishing this as part of 100 Days To Offload. You can join in yourself by visiting https://100daystooffload.com.

In the past 18 months I've been privileged to work as Technical Lead of some very talented software engineers in a complex organisation with lots of moving parts. In addition to having fun building products that are technically interesting, I have found a lot of joy in working with people and helping my engineering team progress along their professional journeys in the little way I can. I've worn the Technical Lead hat for nearly 5 years now in a variety of settings and companies.

Here are a few things I will keep doing as I get along in my career and as I continue learning to inspire people.

Safe space

It is important to create a safe space where people feel comfortable expressing themselves in a manner that works for them. Understand that people are different and might require different approaches. Some people are shy, some are more energetic and some might not be empathetic enough to understand the perspective of others.

Ownership

Encourage ownership of processes and solutions especially in line with stretch goals for the different members of the team. It is important to be ready to provide support through the process of achieving those stretch goals whilst allowing for room to make mistakes.  Allowing engineers to anchor activities or lead the process for finding technical solutions helps with the sense of ownership and growth.

Learning in public

I'm a strong advocate of well crafted questions and I encourage it in my teams. I also encourage posing questions in spaces that are accessible to a wide group so that the solutions are shared across teams and others can learn from it. It is a lot easier to get younger folks to align with this mentality and it might be tougher to get more senior folks on board. My strategy for this is mainly leading by example. I ask questions publicly (even trivial sounding questions). When in the context of digital spaces like microsoft teams, slack or google chat I'd tag a team member who might initially have reservations about asking questions in public. I find people don't mind being mentioned or cc'ed in a question even if they mind being the ones asking the question. It is a huge opportunity for growth - particularly in communicating clearly and succinctly.

Listen and empathise

Always listen, empathise and try to see the world from the other person's perspective. People are very different and might be going through things that are not obvious. I found it useful to be able to observe body language or reactions which I could indirectly or directly check on later with the team member.

Admitting not knowing

I made it a habit to be clear when I don't have an answer to something. I think this is an important lesson for team members to understand and accept that no single person should be burdened with knowing all things. I also strongly encouraged brainstorming solutions, finding out who would be able to help us and going ahead to share the solutions in spaces with the highest impact. Depending on context, this could be across teams or the entire organisation.

Share context

It is important for teams to share context to maintain the notion of a coherent unit. This ensures there isn't much of a knowledge gap between engineers in the team and minimises any knowledge silos. I created a culture where context is regularly shared by always pair programming (swapping pairs twice a week) and having 1 hour a week where everyone comes together to talk about some interesting technical feature they worked on or would like to hear about.

Regular 1:1 check-ins

This is regular time set aside to listen and be listened to if needed. My approach to this definitely evolved over time. I set up recurring check-ins with members of my team the first week I joined and about 3 weeks in, I noticed the check-ins weren't actually happening as regularly as my invite stated. There were a couple of reasons for that. The first reason was that I assumed that everyone had calendar notifications set up and they would message me if they wanted to check in. This was not the case. I later found out in our regular feedback session that: "it would be nice to make the feedback sessions happen." The second reason was that some people preferred impromptu sessions - i.e., chat during lunch or whenever was convenient for them. I was pretty flexible and I accommodated this. The check-ins got easier and more organic. I also adjusted the frequency to match each individual's needs.

Feedback culture

In addition to 1:1s, it is important to find some time as a team to give feedback to each other. One way to achieve this is by running speedback sessions where all possible pairs in the team sit together for a short amount of time (say 6 minutes) to give feedback to each other. Our speedback sessions evolved pretty nicely. We started in a physical location with chairs in different rooms and later ran them consistently remotely. It was also helpful to openly (within the team) track the things we were trying to grow in or receive feedback on.

Feel free to reach out to me if you have any questions about how to implement this in your team.

I’m publishing this as part of 100 Days To Offload. You can join in yourself by visiting https://100daystooffload.com.

Time flies and I worry that you worry about things out of your control.

I think you know change is inevitable but the changes you are seeing are far flung from those you  imagined were possible. When you were young there were gramophones,  radios and black and white televisions. The TVs eventually got a bit of colour. You interacted with these technologies at will - in your own time. You tuned to a radio station to hear the news. You swapped out the discs in the gramophone player depending on your mood. You had a handful of them. You had local control over what you engaged with. You read magazines you bought monthly from the market or books from the bookstore whenever we made a trip to Ikeja.

The things you needed tomorrow you carefully planned today. Indeed a bit of uncertainty was baked into our existence - you got by fine. "Dad is running late from work? It must be the Pen Cinema traffic." You made trips to confirm regular plans. Telephones existed but you didn't have one. So, you made trips to the telephone booths to make plans with those who had telephones. You relied on the local community for information about the bigger community. Of course, these were often embellished by the information bearer. That was not a problem. The information was local enough for you to verify it with a few other people you trusted. If you were lucky, you would get close enough to the source of information to get first hand confirmation. Most importantly, you could look into the eyes of the information bearer and tell without explicitly trying, how much trust you would attribute to what you are hearing.

Then came mobile phones. Expensive gadgets that made communication a lot easier! Gone are the days of proactively planning for the next day, week or month. "Why plan now while I am with you when I can call you later to finalise the details?" "Dad is running late from work? Why don't we call him?" Sometimes, he picks up the phone. Sometimes, he doesn't. "Why does he not answer the phone? Did the phone ring three times and then cut off? Why did the phone go straight to voicemail?" Why this? Why that?

The phone is now in your pocket or your handbag for the day you are wearing Iro and buba. It's also in your hands a whole lot of times and it opens the door to a wild world where trust is earned and most people pretend to be who they are not. You receive messages from people without being in a real time conversation with them. You've lost the visual and the audio cues that come with human communication. You only have your imagination to fill in the blanks for the right emotional cues the communicator intended. The scheduled hourly news you previously tuned to are now constantly buzzing beside your bed. In a world where anybody can create and share news with almost anyone else in the world, please slow down. Please pause, please think, please ask questions - lots of questions.

Maybe you still try to tune into the daily news on the TV. I guess this is mostly pointless, there is no electricity. There is still some charge on your phone and you still have a bit of credit. You do the next best thing. You call your friend to make plans for tomorrow and while you're at it, you exchange news with minor embellishments.

The world is changing mum. It's changing faster than most of us can imagine. Technology can operate without active human interaction. You were happy to tune into TV or radio. You were happy that video or audio records could be captured on a tape and replayed later. You are realising that technology is a double-edged sword. It heals. It burns.

There are people who work very hard to understand the world. These scientists are constantly running experiments on a variety of things - the cure for various ailments, better material for building, more efficient sources of energy. Like most things however, there can be two sides to discoveries and knowledge can be used to benefit or to take advantage.

The herbs and plants you've learned to use to cure ailments were curated and researched by our scientist ancestors.

There is a lot we do not know about the world. A lot of people who realise this are humbled and excited by the opportunity to learn why things are the way they are. I am one of those people. Some people (most people) instead of admitting that they do not know,  fill in the gaps based on what they do know. This is very dangerous. Ask them questions.

I want to finish by telling you something I learned last week. I learned this listening to radiolab on an episode titled Octomom. Mother octopuses do not eat while brooding their eggs. In order to make sure as many of their babies successfully hatch, they sit on their eggs until they hatch without moving much or eating. They also die in the process. The longest known brooding period for an octopus is about 53 months. Let that sink in. That's 53 months without eating. It's not magic nor is it witchcraft. Don't try to explain how this happens because we currently don't have an explanation for how this is possible. Be prepared to admit that you don't know and be wary of those who claim to know everything or have an explanation for everything.

I’m publishing this as part of 100 Days To Offload. You can join in yourself by visiting https://100daystooffload.com.

Special thanks to Jeremiah Iyamabo for proofreading this article.

Bawo is a board game played in many parts of sub-saharan Africa. It is a counting game in the mancala family played on a board with 32 holes. It is a two player game. The board is divided equally between the players - each having 16 holes grouped in rows of 8. I first came across Bao in Mzuzu, Malawi. Bao is ubiquitous in Malawi.  Shortly after learning the rules of what I've come to know as the simplest version of the game, I was obsessed with understanding its complexity. For instance, is there a winning strategy? If so is the first or second player at an advantage?

I did some research to find out what others had done in analysing Bao. I came across  Alexander de Voogt who has dedicated some time in studying board games. I also found  Programming Bao, a paper by Jeroen Donkers and Jos Uiterwijk. At the time, I could not find an interactive web version of Bao. So I created one.

It's a simple version of the game where each hole is started with 2 seeds each.  Players take turns in picking, sowing and capturing the opponent's seeds until the opponent can no longer make a move. You can play it here against an alpha beta algorithm with varying levels of foresight into the game tree.