Category Archives: ITU

Collaboration and competition in Covid-19 response

A week ago, I wrote a post about the potential of crowdsourcing and the use of hashtags for gathering enhanced data on infection rates for Covid-19.  Things have moved rapidly since then as companies, civil society organisations, international organisations, academics and donors have all developed countless initiatives to try to respond.  Many of these initiatives seem to be more about the profile and profits of the organisations/entities involved than they do about making a real impact on the lives of those who will suffer most from Covid-19.  Yesterday, I wrote another post on my fears that donors and governments will waste huge amounts of money, time and effort on Covid-19 to little avail, since they have not yet learnt the lessons of past failures.

I still believe that crowdsourcing could have the potential, along with many other ways of gathering data, to enhance decision making at this critical time. However the dramatic increase in the number of such initiatives gives rise to huge concern.  Let us learn from past experience in the use of digital technologies in development, and work together in the interests of those who are likely to suffer the most.  Eight issues are paramount when designing a digital tech intervention to help reduce the impact of Covid-19, especially through crowdsourcing type initiatives:

  • Don’t duplicate what others are already doing
  • Treat privacy and security very carefully
  • Don’t detract from official and (hopefully) accurate information
  • Keep it simple
  • Ask questions that will be helpful to those trying to respond to the pandemic
  • Ensure that there are at least some questions that are the same in all surveys if there are multiple initiatives being done by different organisations
  • Work with a globally agreed set of terminology and hashtags (#)
  • Collaborate and share

Don’t duplicate what others are already doing

As the very partial list of recent initiatives at the end of this post indicates, many crowdsourcing projects have been created across the world to gather data from people about infections and behaviours relating to Covid-19.  Most of these are well-intentioned, although there will also be those that are using such means unscrupulously also to gather data for other purposes.  Many of these initiatives ask very similar questions.  Not only is it a waste of resources to design and build several competing platforms in a country (or globally), but individual citizens will also soon get bored of responding to multiple different platforms and surveys.  The value of each initiative will therefore go down, especially if there is no means of aggregating the data.  Competition between companies may well be an essential element of the global capitalist system enabling the fittest  to accrue huge profits, but it is inappropriate in the present circumstances where there are insufficient resources available to tackle the very immediate responses needed across the world.

Treat privacy and security very carefully

Most digital platforms claim to treat the security of their users very seriously.  Yet the reality is that many fail to protect the privacy of much personal information sufficiently, especially when software is developed rapidly by people who may not prioritise this issue and cut corners in their desire to get to market as quickly as possible.  Personal information about health status and location is especially sensitive.  It can therefore be hugely risky for people to provide information about whether they are infected with a virus that is as easily transmitted as Covid-19, while also providing their location so that this can then be mapped and others can see it.  Great care should be taken over the sort of information that is asked and the scale at which responses are expected.  It is not really necessary to know the postcode/zipcode of someone, if just the county or province will do.

Don’t detract from official and (hopefully) accurate information

Use of the Internet and digital technologies have led to a plethora of false information being propagated about Covid-19.  Not only is this confusing, but it can also be extremely dangerous.  Please don’t – even by accident – distract people from gaining the most important and reliable information that could help save their lives.  In some countries most people do not trust their governments; in others, governments may not have sufficient resources to provide the best information.  In these instances, it might be possible to work with the governments to ehance their capacity to deliver wise advice.  Whatever you do, try to point to the most reliable globally accepted infomation in the most appropriate languages (see below for some suggestions).

Keep it simple

Many of the crowdsourcing initiatives currently available or being planned seem to invite respondents to complete a fairly complex and detailed list of questions.  Even when people are healthy it could be tough for them to do so, and this could especially be the case for the elderly or digitally inexperienced who are often the most vulnerable.  Imagine what it would be like for someone who has a high fever or difficulty in breathing trying to fill it in.

Ask questions that will be helpful to those trying to respond to the pandemic

It is very difficult to ask clear and unambiguous questions.  It is even more difficult to ask questions about a field that you may not know much about.  Always work with people who might want to use the data that your initiative aims to generate.  If you are hoping, for example, to produce data that could be helpful in modelling the pandemic, then it is essential to learn from epidemiologists and those who have much experience in modelling infectious diseases.  It is also essential to ensure that the data are in a format that they can actually use.  It’s all very well producing beautful maps, but if they use different co-ordinate systems or boundaries from those used by government planners they won’t be much use to policy makers.

Ensure that there are at least some questions that are the same in all surveys if there are multiple initiatives being done by different organisations

When there are many competing surveys being undertaken by different organisations about Covid-19, it is important that they have some identical questions so that these can then be aggregated or compared with the results of other initiatives.   It is pointless having multiple initiatives the results of which cannot be combined or compared.

Work with a globally agreed set of terminology and hashtags (#)

The field of data analytics is becoming ever more sophisticated, but if those tackling Covid-19 are to be able readily to use social media data, it would be very helpful if there was some consistency in the use of terminology and hashtags.  There remains an important user-generated element to the creation of hashtags (despite the control imposed by those who create and own social media platforms), but it would be very helpful to those working in the field if some consistency could be encouraged or even recommended by global bodies and UN agencies such as the WHO and the ITU.

Collaborate and share

Above all, in these unprecendented times, it is essential for those wishing to make a difference to do so collaboratively rather than competitively.  Good practices should be shared rather than used to generate individual profit.  The scale of the potential impact, especially in the weakest contexts is immense.  As a recent report from the Imperial College MRC Centre for Global Infectious Disease Analysis notes, without interventions Covid-19 “would have resulted in 7.0 billion infections and 40 million deaths globally this year. Mitigation strategies focussing on shielding the elderly (60% reduction in social contacts) and slowing but not interrupting transmission (40% reduction in social contacts for wider population) could reduce this burden by half, saving 20 million lives, but we predict that even in this scenario, health systems in all countries will be quickly overwhelmed. This effect is likely to be most severe in lower income settings where capacity is lowest: our mitigated scenarios lead to peak demand for critical care beds in a typical low-income setting outstripping supply by a factor of 25, in contrast to a typical high-income setting where this factor is 7. As a result, we anticipate that the true burden in low income settings pursuing mitigation strategies could be substantially higher than reflected in these estimates”.

 

Resources

This concluding section provides quick links to generally agreed reliable and simple recommendations relating to Covid-19 that could be included in any crowdsourcing platform (in the appropriate language), and a listing of just a few of the crowdsourcing initiatives that have recently been developed.

Recommended reliable information on Covid-19

Remember the key WHO advice adopted in various forms by different governments:

  • Wash your hands frequently
  • Maintain social distancing
  • Avoid touching eyes, nose and mouth
  • If you have fever, cough and difficulty breathing, seek medical care early

A sample of crowdsourcing initiatives

Some of the many initiatives using crowdsourcing and similar methods to generate data relating to Covid-19 (many of which have very little usage):

Lists by others of relevant initiatives:

 

Global Covid-19 mapping and recording initiatives

The following are currently three of the best sourcs for global information about Covid-19 – although I do wish that they clarified that “infections” are only “recorded infections”, and that data around deaths should be shown as “deaths per 1000 people” (or similar density measures) and depicted on choropleth maps.

 

2 Comments

Filed under Africa, AI, Asia, capitalism, cybersecurity, Development, digital technologies, Education, Empowerment, ICT4D, ICTs, inclusion, India, ITU, Latin America, mobile phones, Pakistan, Sustainability, technology

The gendering of AI – and why it matters

Digital technologies are all too often seen as being neutral and value free, and with a power of their own to transform the world.  However, even a brief reflection indicates that this taken-for-granted assumption is fundamentally flawed.  Technologies are created by people, who have very specific interests, and they construct or craft them for particular purposes, more often than not to generate profit.  These technologies therefore carry within them the biases and prejudices of the people who create them.

This is as true of Artificial Intelligence (AI) as it is of other digital technologies, such as mobile devices and robots.  Gender, with all of its diversity, is one of the most important categories through which most people seek to understand the world, and we frequently assign gender categories to non-human objects such as technologies.  This is evident even in the languages that we use, especially in the context of technology.  It should not therefore be surprising that AI is gendered.  Yet, until recently few people appreciated the implication of this.

The AI and machine learning underlying an increasing number of decision-making processes, from recruitment to medical diagnostics, from surveillance technologies to e-commerce, is indeed gendered, and will therefore reproduce existing gender biases in society unless specific actions are taken to counter it.  Three issues seem to be of particular importance here:

  • AI is generally used to manipulate very large data sets.  If these data sets themselves are a manifestation of gender bias, then the conclusions reached through the algorithms will also be biased.
  • Most professionals working in the AI field are male; the World Economic Forum’s 2018 Global Gender Gap Report thus reports that only 22% of AI professionals globally are women. The algorithms themselves are therefore being shaped primarily from a male perspective, and ignore the potential contributions that women can make to their design.
  • AI, rather than being neutral, is serving to reproduce, and indeed accelerate, existing gender biases and stereotypes.  This is typified in the use of female voices in digital assistants such as Alexa and Siri, which often suggest negative or subservient associations with women.  A recent report by UNESCO for EQUALS, for example, emphasises the point that those in the field therefore need to work together to “prevent digital assistant technologies from perpetuating existing gender biases and creating new forms of gender inequality”.

These issues highlight the growing importance of binary biases in AI.  However, it must also be recognised that they have ramifications for its intersection with the nuanced and diverse definitions of gender associated with those who identify as LGBTIQ.  In 2017, for example, HRC and Glaad thus criticised a study claiming to show that deep neural networks could correctly differentiate between gay and straight men 81% of the time, and women 74% of the time, on the grounds that it could put gay people at risk and made overly broad assumptions about gender and sexuality.

The panel session on Diversity by Design: mitigating gender bias in AI at this year’s ITU Telecom World in Budapest (11 September, 14.00-15.15) is designed specifically to address these complex issues.  As moderator, I will be encouraging the distinguished panel of speakers, drawn from industry, academia and civil society, not only to tease out these challenging issues in more depth, but also to suggest how we can design AI with diversity in mind.  This is of critical importance if we are collectively to prevent AI from increasing inequalities at all scales, and to ensure that in the future it more broadly represents the rich diversity of humanity.


THIS WAS FIRST POSTED ON THE ITU’S TELECOM WORLD SITE ON 17TH JUNE 2019.  It is reproduced here with their kind permission.

 

Leave a comment

Filed under AI, Conferences, Gender, ICT4D, ITU

ITU and UNESCO Chair in ICT4D session at WSIS Forum 2018: International decision-making in ICT – where are the women?

The ITU is strongly committed to achieving gender equality across its organisational structures, and has been one of the driving forces for achieving gender equality in and through ICTs across the world, not least through its involvement in creating the EQUALS initiative.

One of the key international gatherings convened by the ITU has been the series of World Radiocommunication Conferences held periodically to reach international agreements on Radio Regulations, with new and revised Resolutions and Recommendations.  Traditionally, these have been very male dominated, and the ITU has therefore taken steps to encourage greater involvement of women at all levels in its decision-making processes.  One aspect of this has been the creation of the Network of Women for WRC-19 (NOW4WRC19), led by Dr. Hanane Naciri, which aims to encourage increased participation of women in the conference being held in 2019.  Its main objectives are to have a better gender balance among delegates, to prepare women for key roles in WRC-19, and to grow the women’s community capacity and contribution.

As part of this process, the ITU and the UNESCO Chair in ICT4D convened Session 113 at the WSIS Forum 2018.  This began with a lively panel discussion, opened by Dr Hanane Naciri (Radiocommunication and Software Engineer, Radiocommunication Bureau, ITU), with Sahiba Hasanova (Vice-Chairman, ITU-R Study Group 4 / Leading Adviser, Ministry of Transport, Communications and High Technologies, the Republic of Azerbaijan), Caitlin Kraft-Buchman (CEO/Founder Women@theTable, Geneva, Switzerland) and Brigitte Mantilleri (Director of the Equal opportunities office of the University of Geneva).  The speakers shared some of their experiences of leadership in the field of ICT, commented on the challenges facing women who wish to participate in such events, and suggesting what needs to be done to involve more women at all levels in such processes (summary).

workshop

Building on these inspirational introductions, participants then shared their experiences, insights and suggestions for what still needs to be done to ensure that women contribute fully and appropriately to international ICT decision making, and especially to WRC-19.  Twelve themes were identified, and these were captured in a mind map which is available on the ITU and UNESCO Chair for ICT4D sites:

  • Top leadership and champions: it is essential that top leadership supports the increased participation of women, and that champions are identified who can promote such participation;
  • Ensuring that women are in powerful positions: women need to be supported throughout their lives, and particularly encouraged to take leadership roles;
  • Building and promoting networks: it is essential that we work together in intergenerational networks that can support and advise women participating in such decision-making activities;
  • Involving men: we must have male feminists as well as female ones who are willing to help change attitudes and cultures of oppression;
  • Training: more effective training programmes are necessary, particularly ones that help men to understand the relevant issues;
  • Organisational structures: addressing elements of organizational culture is key, and it is important to equip women to survive and flourish in the environments where they work;
  • Awareness and communication: the need to provide much more information about how women can contribute to such decision-making gatherings, and to confront people who have negative behaviours;
  • Changing norms: the need to address and revisit many underlying assumptions;
  • Incentivisation: the need to provide incentives to organisations and individual women to participate in such events;
  • The role of recruitment: recruitment agents can play a key role in ensuring balanced interview panels and processes, and in supporting a charter code of practice on gender;
  • Remember that inclusion is not the same as diversity: diversity is not enough and we need to be inclusive to ensure that women feel comfortable in whatever environment they find themselves; and finally
  • Recognising it may not happen overnight: given how slow change has been so far, we need to recognize it may not happen swiftly, but we must develop the momentum so that it will happen as quickly as possible.

Participants were committed to supporting EQUALS and working with the ITU to ensure that there is much greater involvement of women at all levels in WRC-19.

Leave a comment

Filed under Conferences, Gender, ICT4D, ITU

ICTs, sustainability and development case studies: the Restart Project

Earlier this year, I was privileged to work on a co-authored book project for the ITU.  This was published by the ITU as ICT-Centric Economic Growth, Innovation and Job Creation, and was launched at the World Telecommunication Development Conference in Argentina in October.  The chapter that I led was entitled ICTs, sustainability and development: critical elements, and provided a challenging account of ICTs and sustainability.

Each chapter was accompanied by a single case study – although I had argued strongly that there should be more than one case study for each chapter, so that a range of different examples and perspectives could be included.  I had worked with several colleagues to produce great examples that would exemplify some of the key arguments of the chapter, but sadly these were not published.

Hence, as a supplement to the book, I am including these now as blog posts.  This is the first, and was written with the help of the amazing Ugo Vallauri, co-founder of the Restart Project:

The Restart Project: local, community driven initiatives moving beyond the throw-away economy

One effective way of reducing the environmental impact of ICTs is simply to use them for longer.  The Restart Project, a London-based social enterprise that encourages and empowers people to use their electronics longer in order to reduce waste, is an excellent and innovative example of such initiatives.  Launched in 2012 with its first “Restart Party” pop-up community repair event in the UK, it has inspired groups in 10 other countries to replicate similar initiatives in  Europe, North Africa and North America.

Restart

Most energy used and most emissions generated during the life of mobile phones occur during its production process.  Hence, if people use their mobile phones for longer, and repair them when they are faulty, their overall energy impact can be dramatically reduced.  The figures are striking: the average mobile phone made in 2015 produced 36 kg of carbon emissions in manufacture, equivalent to 16 weeks of laundry in affluent countries; the total carbon footprint of the 1.9 billion mobiles sold in 2015 was roughly equivalent to Austria’s total carbon emissions; if every mobile phone were used for one-third longer than the typical 3 years, there would be an emissions saving equivalent to Singapore’s total annual emissions.

The Restart Project is both about changing people’s attitudes and also helping them to make a practical difference.  It works with communities, schools and companies to value and use ICTs longer, and to document the barriers to so doing.  This is done through convening hands-on learning events, known as Restart Parties, where volunteers  help people fix their own small electrical and electronics, and also through helping others to do the same globally, not least through developing educational resources to inspire younger people and sharing tips  for repairing different kinds of equipment.  Acting together, they draw on the skills that everyone has, and collect and publish data on the products they fix. Just over 50% of all products taken to Restart events get fixed by volunteers. By collecting data on common failures and barriers to repairability, Restart hopes to inspire designers, manufacturers and policy makers to fix some of the problems that cannot be solved: early software obsolescence, ease of disassembly and availability of spare parts are all common problems. The combined impact of the over 200 Restart Parties held  by April  2017 prevented 4,011 kg of waste, and 88,687 kg of CO2 emissions, which is equal to driving a car 739,000 km or the emissions caused in the manufacture of 15 cars.

Their guidance for hosting Restart Parties is clear and simple:

  1. Offer free entry to the public (although you can suggest a donation);
  2. Promote a collaborative learning process;
  3. Fix other stuff like bikes if you want, but you’ll need at least three-to-four electronics repairers;
  4. Tell the Restart Project about your party beforehand, and share the results with them; and
  5. Be insured! The Restart Project is not liable for events we do not organise. If uninsured, please work in partnership with a group that is.

Such efforts, though, require funding, at least of the central team running and administering the parties and undertaking the research.  Not everything can be done by volunteers.  The Restart Project has to date been funded by the Esmée Fairbairn Foundation and the Shuttleworth Foundation. Some of their activities are supported by running special events for local authorities, cultural institutions and companies. They are actively looking to  generate additional income from consultancy built on their insights on participants’ frustrations and recurrent faults and direct donations from the general public.

Many more initiatives such as the Restart Project can readily be created by local community groups across the world; as the Restart Project claims, “We’re fixing our relationship with electronic – putting people and planet first”.  Such initiatives are truly focused on finding ways through which ICTs can indeed deliver a more sustainable world, and thus help to make progress in achieving the SDGs.  If everyone kept their mobile phones, tablets or laptops longer, manufacturers would have to prioritise provision of better repair services, spare parts and refurbishing of devices, and the environmental impact would be significant.  It would be one way through which everyone in the world who owns a digital devices could contribute to achieving the SDGs.

1 Comment

Filed under Development, ICT4D, ITU, SDGs, Sustainability

EQUALS Research Group Meeting in Macau

EQUALS 5is a global initiative committed to achieving gender equality in the digital age.  Its founding partners are the ITU, UN Women, UNU Computing and Society (UNU-CS) institute, the International Trade Centre, and the GSMA, and it has been a real privilege to work with colleagues from these organisations and other partners over the last 18 months to try to help forge this partnership to reduce the inequalities between men and women in the digital age.   There are three partner Coalitions within EQUALS: for Skills (led by GIZ and UNESCO); Access (led by the GSMA); and Leadership (led by the ITC).  These are supported by a Research Group, led by the UNU-CS. The picture above shows the first Principals meeting held in September 2017 at the edges of the UN General Assembly in New York.

Despite all of the efforts to achieve increasing female participation in STEM subjects, in employment and leadership positions in the ICT sector, and in the use of ICTs to help towards women’s empowerment, most of the indicators show that gender digital inequality is increasing.  At the broadest level, this means that most of the initiatives undertaken to date to reduce these inequalities have failed.  Business as usual is therefore not an option, and the EQUALS partnership is intended to encourage committed partners to work together in new ways, and on new initiatives, to help deliver Sustainable Development Goal 5,  to “achieve gender equality and empower all women and girls”. 

The first face-to-face physical (rather than virtual) meeting of the Research Group was convened by the UNU-CS in Macau from 5th-6th December (official press release), and it was great that both Liz Quaglia and I were able to represent the UNESCO Chair in ICT4D (at Royal Holloway, University of London) at this meeting, which was attended by researchers and policymakers from 21 universities and organizations around the world. This meeting established the group’s research agenda, drafted its work plan for 2018, and finalized the content and schedule of its inaugural report due to be published in mid-2018.  In particular, it provided a good opportunity for researchers to help shape the three Coalitions’ thinking around gender and equality in the  areas of skills, access and leadership, and also to identify ways through which they could contribute new research to enable the coalitions to be evidence-led in their activities.

This slideshow requires JavaScript.

Huge thanks are due to Araba Sey, who convened the meeting with amazing enthusiasm, insight and professionalism, and all of the other staff at UNU-CS who contributed so much to the meeting.  It was a great occasion when some of the world’s leading researchers in gender and ICTs could meet together, not only to discuss EQUALS, but also to explore other areas of related research, and to build the trust and openness necessary to increase gender equality both in the field of ICTs, and also through the ways that ICTs influence every aspect of people’s lives.  The BBQ and dancing on the last night ensured that memories of this event will last for a long time in everyone’s minds!

Leave a comment

Filed under Gender, ICT4D, ICTs, Inequality, ITU, United Nations, Universities