Content
Welcome to the Club of Amsterdam Journal.
Watch the new edition of The Future Now Show with John Nosta, Digital Health Philosopher and Katie Aquino about Digital Health!
Felix F Bopp, Founder & Chairman
Reinventing Leadership Development
George Pór is an evolutionary thinker, Teal mentor, and advisor to culture change and system transition in organisations.
Gallup research shows that, worldwide, only 13% of employees are engaged in their jobs. The statistic is shocking and means more than three-quarters of employees are wasting their time, energy, and their organisation’s resources. How can we mobilize the creativity and collective intelligence in our organisations to turn this around?
For a challenge this massive, we need more than yesterday’s mindsets and leadership development approaches.
The crisis of employee engagement, like any of our global crises, cannot be solved by today’s dominant modes of thinking. “The next cycle is the one of an integral, holistic consciousness that enables the integration of the inner and outer technologies and sciences, deep intuition and systems thinking, spirituality and precision of inquiry.”[1] We need to develop competences in each. The sooner the better.
Think of it as scaling a mountain. As we ascend, we leave useless, old stuff beside the path, stuff that burdens and blinkers us. The climb is worth it, though. When we get to the top, we gain an eagle-eye perspective. We now enjoy a broader view. We see realities that were previously hidden from us. And we see how everything is interconnected. As our insight deepens, so does our compassion and inner coherence.
As we scale the mountain, we renew ourselves personally and professionally, and step more fully into our potential. Looking back, we wonder how we ever believed in the myth of controlling people, predicting the future, or escaping the consequences of an unhealthy life. And we know that we couldn’t have made this journey alone. Not this quickly and not this smoothly.
Our sherpa guides or mentors have been here before. They know how to guide us through the shortcuts and rocky terrain. That’s because they have honed the capacity to sense, think, and relate from a more expansive sense of self. They also help us unlearn unproductive habits and drop them beside the path.
As a mentor who guides leaders in next-stage organisations, I can share with you a few things I have learned so far. Some of it comes from the three breakthroughs that Frederic Laloux discovered while researching and writing his book. Their implications are far-reaching, practical, and powerful. Not only can we reinvent our organisations, but also ourselves, and the way we develop as leaders. Let’s look into how.
Evolutionary Purpose
You may be tempted to start by wondering where your organisation could evolve to next or how it could reach its creative potential. However, that is step two. Step one is discovering that breakthrough in your own self.
This is not about deciding what you’ll be when you grow up. And it’s not about pushing forward with your head or will. To get clarity about your evolutionary purpose, it is best to bypass the analytical mind for a little while. Instead, ask the question “who is the world inviting you to be, in service of creating a future that we all want?” and listen to the answer that comes from your heart. Or ask simply where are your deepest talents and highest aspirations meeting a need in your world, one that you feel passionately called to address.
Take time to re-read either of those questions that speak to you. Sit with it for a moment. Even better, move away from the computer, take your question to a quiet corner or take it for a walk. Turn your attention to your breath. Know that there is a creative impulse within you, and it wants to manifest through you. But it can only reveal its secret when you become silent or still enough to hear its whisper.
A more psychological way to access your inner wisdom is to access your intuition and ask: what kind of work gives me the greatest joy? What is the need/tension in my world that is crying loudest for my help? Where the answers to those questions dovetail, that’s where you will find your evolutionary purpose. At least, for now.
I ask myself these questions at least once a year, on my birthday, keeping them fresh, vibrant and dynamic. I use them as a North Star on my journey.
Self-Management
Leading by example is par for the course in self-managed organisations. As Laloux says, “an organisation cannot evolve beyond its leadership’s stage of development.” This makes your personal development more than personal. You need to embody the qualities you want to inspire in others. And you need to operate from a deeper knowledge of both yourself and the work environment than what is required in a traditional workplace.
This involves gaining a high level of knowledge about your organisation’s operating conditions and the various roles and accountabilities that dovetail with yours. Only then can you use your creative potential to the fullest and contribute to the whole.
It gets personal. As a leader, your self-management includes the capacity to recognize your needs, values, and moods, and the skills to manage the latter. That calls for moment-to-moment awareness of what is arising from within, as well as the tensions and opportunities for individual action arising from the workplace. Only then can you, as an individual along with groups of individuals, develop skills and practices that allow everyone to work together harmoniously under any conditions, even in the most challenging situations.
Developing quiet-mind muscles, in addition to the active-mind muscles you already have, is crucial. This involves stilling yourself enough to gather information through mindful attention, sensing, and feeling. This is about being present, receptive, and simply being.
You can try this right now. As you read the rest of this blog, simultaneously keep some of your attention on your breath. It is as if you are attending the words in one hand, and your breath in the other, and you are aware of both. This is not an easy practice because the mind tends to jump to either the sentence or the breath instead of remaining alert to both. With practice, you can become better at it.
Try this the next time you talk with someone. Pay attention to your breath or heartbeat while fully absorbing what you hear from the other person. Notice how this changes the quality of your presence.
Quiet-mind skills include self-reflection, sense-making, and perspective-taking. They help you nurture the unfolding leadership qualities in yourself and in each member of the organisation. I sense another blog about them in the making…
Wholeness
Imagine dropping your professional mask and bringing the whole of who you are to work. This is what happens in next-stage organisations. You are free to show up in an authentic way without hiding your vulnerability. You are in touch with your emotions and able to express even the so-called “negative” ones, without harming others. Like you, they value genuine relationships and have no need to to play “games” or to “play nice.”
Another aspect of wholeness is about developing a portfolio of the roles you energize, both personally and professionally. Which of your talents come out to play in each role? How do your roles strengthen each other? Which of your roles can generate the larges ripple effects on others’ roles, and then impact the whole. In this way you learn to optimize your contribution and impact on the whole organisation.
This also applies to the totality of your lifework roles, which I illustrated here. So, reinventing leadership development has to also address the ways in which you, as a leader, can optimise the investment of your attention/energy in multiple contexts.
These are some of the wholeness competencies you may want to develop: conflict resolution; intuition; imagination; generative listening; celebrating accomplishments; and managing the distribution of your energy across a portfolio of your roles. Most important is caring for the wellbeing of all aspects of your diversity, and aligning these parts functionally into a working whole.
If you want to go deeper in the individual dimension of wholeness, read the blog of my colleague Celine McKeown “What does Wholeness mean in the context of Teal Organising?”.
Wholeness, at the focus of reinventing leadership development, provides an invitation to pay conscious attention to making every decision from the biggest context that can put your arms around. I.e., one that is personally meaningful to you. For some leaders, it will be the well-being of the organisation’s members and other stakeholders. For others it may be the evolutionary purpose of the enterprise. Yet for others it will be evolution itself. It’s all good, really. But attention training will be needed till attending to the largest whole becomes your second nature.
If you are an intentional learner, and wish to accelerate your evolutionary journey, here are some tips:
- 1. Read the Reinventing Organizations book, watch the video, and browse the wiki.
2. Join a meet-up group or a community of practice or an online network, focused on next-stage, or self-managing, or Teal organisations, under any other name, and explore with them how to discover next “next stage” in one’s own life.
3. Subscribe to and get involved with Enlivening Edge, the online magazine of next-stage organisations.
This approach to reinventing leadership development is not for everyone, but it can be introduced on top of more traditional leadership development effort, especially for those who want to move beyond what it offers.
Your highest-leverage action to build capacity for organisational reinvention is to become a “next-stage” mentor. Evolutionary mentoring, as introduced here, is foundational to reinventing leadership development, and it is always cross-mentoring that transforms both the mentor and the mentee. That’s why it’s one of my greatest joys.
Along with my colleague Jackie Thoms, I will be offering a workshop during the Integral European Conference May 4-8, on how to mentor leaders of “next-stage,” or what Laloux calls, “Teal” organisations.
This is work in progress, evolving through the generative conversations with our clients and colleagues. If you want to join the conversation, please comment or ask your questions below.
We, at Future Considerations, are working with our clients to help accelerate their journey to the next level of potential. Our Teal mentoring service is an emergent process that starts with a free, in-depth, generative interview. The communication channels we use include: face-to-face meetings, video calls, email, and a dedicated private collaboration spaces. For more details, please get in contact
George Pór is an evolutionary thinker, Teal mentor, and advisor to culture change and system transition in organisations. He is a Fellow of Future Considerations and founder of “Enlivening Edge: News from Next-Stage Organizations” and the Teal Practice Group (London). George is the publisher of the Blog of Collective Intelligence, and an independent scholar. His former academic posts included INSEAD, London School of Economics and UC Berkeley.
The Future Now Show with Chris Dancy and Katie Aquino
Every month we roam through current events, discoveries, and challenges – sparking discussion about the connection between today and the futures we’re making – and what we need, from strategy to vision – to make the best ones.
May 2016
featuring
John Nosta, Digital Health Philosopher, USA
Katie Aquino, aka “Miss Metaverse”, Futurista™, USA
Paul Holister, Editor, Summary Text
In an age where biometric devices have become a popular consumer item, where you can go online and get your genome analysed (23andMe), where gene therapy is promising to treat everything from genetic disorders to cancer and where we are mastering the ability to precisely target specific genes for control or editing (CRISPR being the latest buzz technology), how are these myriad developments going to change the way our health is managed? Or the extent to which we can manage it? Add AI into the mix and the possibilities can be dizzying, if not sometimes a little scary.
Wisdom of the Crowd or Wisdom of a Few?
Ricardo Baeza-Yates. Yahoo! Labs. Barcelona, Spain
and
Diego Saez-Trumper. Universitat Pompeu Fabra. Barcelona, Spain
An Analysis of Users’ Content Generation
ABSTRACT
In this paper we analyze how user generated content (UGC) is created, challenging the well known wisdom of crowds concept. Although it is known that user activity in most settings follow a power law, that is, few people do a lot, while most do nothing, there are few studies that characterize well this activity. In our analysis of datasets from two different social networks, Facebook and Twitter, we find that a small percentage of active users and much less of all users represent 50% of the UGC. We also analyze the dynamic behavior of the generation of this content to find that the set of most active users is quite stable in time. Moreover, we study the social graph, finding that those active users are highly connected among them. This implies that most of the wisdom comes from a few users, challenging the independence assumption needed to have a wisdom of crowds. We also address the content that is never seen by any people, which we call digital desert, that challenges the assumption that the content of every person should be taken in account in a collective decision. We also compare our results with Wikipedia data and we address the quality of UGC content using an Amazon dataset. At the end our results are not surprising, as the Web is a reflection of our own society, where economical or political power also is in the hands of minorities.
Categories and Subject Descriptors
H.2.8 [Database Management]: Database applicationsData mining;; J.4 [Computer Applications]: Social and Behavioral Sciences
General Terms
Human factors, measurement.
Keywords
Social networks; user generated content; wisdom of crowds. Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than the author(s) must be h onored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from Permissions@acm.org.
HT’15, September 1–4, 2015, Guzelyurt, TRNC, Cyprus.
Copyright is held by the owner/author(s). Publication rights licensed to ACM. ACM 978-1-4503-3395-5/15/09 …$15.00.
http://dx.doi.org/10.1145/2700171.2791056.
1. INTRODUCTION
The wisdom of crowds is a well known concept of how “large groups of people are smarter than an elite few, no matter how brilliant, they are better at solving problems, fostering innovation, coming to wise decisions, and even predicting the future” [20]. On the other hand, although all people that use Internet can contribute to web content (or any type of activity), most people do not. In fact, in any social network, the set of people that just looks at the activity of others (passive users or digital voyeurs) is much larger than the people that is active. Similarly, among the active users most of them do little, while a few do a lot (digital exhibitionists). We are interested in the characterization and interplay of these groups of people regarding the generation of content.
Let us take a specific case, say the world of blogs in the Web. Most people do not have a blog and few people have good blogs. Conversely, most blogs are not read and few blogs are well read. Indeed, people contribute to content in a social network or in the Web because they have the (possibly wrong) perception that someone will look at and read their contribution. This perception that they are speaking to the whole world, when the truth is that most of the time they are speaking alone, creates a very long tail of content that nobody sees, a huge digital desert where people write to an empty audience, metaphorically speaking.
Although we believe that there is a high correlation between the quality of content and the activity of users interacting with that content, in this paper we explore this process: how people contributes to content and what is the impact of the content generation process in the so called wisdom of crowds. As we cannot study this in the context of the whole Web, as most usage data is private, we use two different datasets: a small sample of New Orleans Facebook users and a large one coming from a micro-blogging platform, Twitter. Both are good case studies for the problem being tackled. In fact, today Facebook and Twitter are the two largest social networks in term of users. In one of these cases we estimate a weak lower bound of how much of the UGC produced is never seen. We also compare the content generation process in these social networks to the content generation of Wikipedia as well as the estimations of unique users per month that visit that website.
Moreover, using another UGC dataset from Amazon’s movies reviews, where the quality of content it is explicitly rated, we study the relation between quantity and quality of content produced by people, finding also that the majority of high-quality content is generated by a small set of users
Our main results are:
• The percentage of users that generate more than 50% of the content is small, less than 7% in our two examples;
• These top users are quite stable in time, more than 70% of the initial people in our two examples stay on that group during all the time observed;
• The quality of content it is not strongly correlated with amount of users’ activity, but;
• Given that quality of content it is (almost) equality distributed among users, more active users produces in absolute numbers – more high quality content than less active users.
• The number of users that do not contribute to the generation of content is the majority of them, some because of inaction while thers because their content is not taken in account;
• There is a significant volume of content that nobody sees, and hence is not taken in consideration; and
• The bias seems to be even worse in non social contexts such as content creation in Wikipedia, where there are also is higher amount of content that is never visited.
The reminder of this paper is organized as follows. Sections 2 and 3 give the background. Sections 4 to 6 present the experimental results and discuss them
2. RELATED WORK
The concept of the wisdom of crowds was introduced by Francis Galton in 1907 [6], and used by James Surowiecki in his seminal book “The Wisdom of Crowds” [20], where he posits – among other things – that the aggregated knowledge of a group would be bigger than the knowledge of any of its single components. Although wisdom is difficult to measure, on the Web this concept has been translated – and widely applied – as using the data provided directly (e.g., content) or indirectly (e.g., clicks) by users to discover knowledge in a crowd sourcing approach [14, 9, 5, 8]. A good example of how this wisdom can be used, is exploiting the clicks that users do after issuing a query in a web search engine. This allows to extract semantic relations between queries in an automatic manner [1, 2, 3]. Therefore, in this example and others, more user generated content implies more knowledge that can be potentially discovered.
In Online Social Networks, wisdom can be related to the amount of content produced by users. Previous studies suggests that the amount of user’s activity (e.g., number of tweets) it is related with her/his number of followers [19], and also with the monetary value that they produce [18]. Similarly, in social graphs – where node in-degree has a power-law distribution [10, 13, 11, 7] – most of the content produced (i.e., activity) is generated by a small subset of users, while the majority of users act as passive information consumers [16]. Moreover, previous studies have shown that the around 50% of URLs consumed in Twitter are produced by a tiny portion of users (less than 1%) [22]. However, while previous work shows that to have a lot of followers cannot be considered as synonym of influence [4], nowadays we do not know enough about most active users. In this paper we try to understand the importance and characteristics of most active users regarding the generation of content.
3. EXPERIMENTAL FRAMEWORK
3.1 Assumptions and Definitions
We consider that each unit of activity (tweets or posts) is one unit of content and that the overall activity is proportional to the wisdom of the crowd. A possible variation is to consider the length of the text of the tweet or the post. Nevertheless, as these texts are small (e.g. tweet length is capped by 140 characters), the results should be similar. Later on this paper, we discuss about the content’s quality, and how it relates with the concept of wisdom.
To distinguish top contributors (wise users or digital exhibitionists) from the rest of the active users, we use the following arbitrary definition for a given time period: wise users are the set of most-active users such that they contribute with 50% of the content (or half of the wisdom). Other definitions are possible, for example based in a larger percentage, but we consider that 50% is already a majority of the content. Nevertheless, the results would be similar as all the distributions involved resemble power laws. We call the rest of the users, in fact, the majority of them, others.
3.2 Datasets
We use two different datasets from two different kind of social networks: Twitter, a micro-blogging social network; and Facebook, a pure social network. For all the experiments we consider only the active users, meaning users that have shown some posting activity in the time period considered.
Facebook: This dataset corresponds to the New Orleans Facebook’s Regional Network (Regional Networks were deprecated by Facebook in August of 2009). [21]. We have two lists: the first one contains the social graph (friendships) and a second list with user-to-user wall posts (where u,v means user v posting in u’s wall), and the timestamp. All these data has
been anonymized.
The social graph has 1,545,686 edges between 63,731 users. The information about wall posts has 876,993 actions, with 39,986 users doing at least one post (active users). Hence, we can estimate that at least 37% of users are passive or inactive. Notice that these are users that have a public profile, so although the set is partial, is what can be compared to other datasets based on public data as the next one.
This dataset has wall posts from 14th of September 2004 to 22th January of 2009 according to Table 1. We use the last three years as the two first are too small. The Pearson correlation of the number of posts with respect to the number of friends, using a logarithmic transformation to linearize the distributions, is 0.64. That is, the distributions are partially correlated. The distribution of posts versus users follows a power law of parameter -1.58.
Twitter: Our dataset contains almost all the tweets done in Twitter between March 1st until May 31st of 2009. We also have the complete social graph of Twitter for that period. This information is a subset of the dataset obtained in [4]. Specifically, tweets are represented as a list of pairs (user id,timestamp), and the social graph is an adjacency list. The social graph has 1,963 of edges between 42 million of users (688 million edges between 12 millions of active users). The activity considers 440 million of tweets produced by the active users. In fact, the number of nonactive users (71%) is more than 2.4 times larger than the number of active users (29%). Hence, our analysis would be more striking if we take percentages over the whole user population.
Our Twitter data has two limitations: (i) we have “only” the last 3,200 tweets from each user, but we have found only 167 of users with more tweets than this threshold in around 50 millions users; 2( In any case, this implies that our results are a good lower bound because we are trimming the most active users ) and (ii) from the social graph, we cannot establish when each edge was created, therefore we are working with the final snapshot of that graph.
In order to study the UCG with different time granularity, in our experiments we use this dataset in two different ways: first, the full dataset split in months and next, a smaller sample where we split in weeks the first three weeks of May. Tables 2 and 3 gives the details of them. The distribution of tweets versus users can be approximated by a power law of parameter -2.1. On the other hand, the Pearson correlation of the number of tweets with respect to the number of followers, using again the logarithmic transformation, is 0.68. That is, the distributions are again partially correlated.
4. EXPERIMENTAL RESULTS
4.1 Wise and Others
We start by finding the proportional sizes of the user groups defined in the previous section. Figure 1 shows that the distribution of user activity is very skewed. For Twitter – Figure 1 (right) – where we have more data points, we show that the distribution also depends on the time window considered, as a longer time window implies that a smaller group of users produced most of the content. For three years, we found that in the Facebook dataset, the wise users were just 7.0% of the total. On the other hand, in a period of three months, we found that in the Twitter dataset the wise users accounted for just 2.4% of them.
In the case of the Twitter dataset, looking at the social graph we found that even though wise users are less than 3%, they concentrate more than the 35% of the incoming edges, therefore they have also a higher in-degree (see Table 4). On the other hand, in the Facebook dataset, 7% of the users concentrate 21% of the links. This is not surprising as the Facebook social graph is more sparse as friendship is bidirectional and both users have to accept the relation.
To measure the connectivity of each group, we use the Gamma index, that is the ratio between the links observed over all possible links in the complete graph of active users [15]. A larger Gamma Index means higher connectivity into the graph. In Twitter we see that wise are the most cohesive group. In Facebook the differences are even bigger, as wise users are five orders of magnitude more cohesive than the rest. Differences between Twitter and Facebook might be due to the different nature of the link creation process in each platform (in Facebook both parts needs to agree to create a link, while in Twitter each user can decide alone) and the type of graph. However, in both cases wise users are more cohesive than the rest, suggesting that they are a highly connected elite.
We can partially compare these results to the content generation process ofWikipedia. Indeed, according to data published by Wikipedia itself, the top 10,000 editors produce 33% of the content editions. Considering that there arealmost 20.8 million registered editors, the top editors represent just 0.04% of them. As the number of passive users, that is, people that use Wikipedia but do not contribute with content, is more than one billion, the percentage of active editors with respect to the total number of users is negligible. Something similar happens with the creation of the almost 4.5 million Wikipedia articles in English, where the 2,005 most prolific authors account for the creation of 50% of the articles [17]. This is less than 0.01% of registered editors, and this number would be even smaller if non-registered users could be taken into account.
4.2 Evolution Along Time
Now we find the percentage of wise users for different periods of time for both datasets. Results are detailed in Tables 5 and 6. This shows that even though the percentage of wise users decreases with larger time windows, the absolute number is pretty stable. However, are those wise users always the same? Table 7 shows the percentage of users that were in the wise group in the first year and stay there in the next two years for the Facebook dataset. Table 8 shows the percentage of users that were in the wise group in the first week and stay there during the next two weeks for the small Twitter dataset. As can be seen the wise users are very stable, as more than 70% or 80% remains after three years or weeks, respectively.
In Figure 2 we show the dynamics of the wise and others groups, during three years or months for both datasets, showing the percentage of people that come from the groups in the previous month as well as the percentage of new users. The numbers displayed in the edges, represents the percentage of users going to a given group in previous/next time slot. Outgoing edges pointing to previous time slot (e.g. from month 2 to month 1) shows where the users come from, while edges pointing to the next time slot shows the destiny of those users. For example, in Figure 2 (b), in month 2, 66% of wise users come from the wise group in month 1, 27% from others and, 7% from new users (users that were not active in month 1). Next, 92% of those wise users, stay in the wise group in month 3, and 8% went to others. Another way to understand this, would be look at symmetric edges. For example, in month 3, 1% of the new users went to wise group, while 4% percent of the total of wise users come from new users.
Overall, we can see that groups are quite stable if we do not consider new users. In fact, at the end of three periods, most of the wise users have always been in this group, for example in Figure 2 (a), 74% of wise users stay in that group from year 1 to year 2, and then 98% stay in that group from year 2 to year 3, confirming the stability of this group.
4.3 The Digital Desert
In this section we analyze the phenomena of the content that is uploaded for some users, but is never seen by anyone else. We refer to this content as the digital desert. We can estimate a lower bound for the content that is never seen for the case of the Twitter dataset. (For the Facebook Dataset we cannot estimate the size of the digital desert because was obtained through a snowball sampling) In fact, a lower bound for the digital desert can be computed as the percentage of content generated by people that has no followers. (Potentially, content uploaded by users without followers can be reached through the search page of Twitter or a generic search engine. However, tweets posted by users without followers are unlikely to be top-ranked in any search results.) This percentage of people is only 0.06% for the wise group but 20.58% for the others group. This accounts for 0.03% and 1.08% of the whole content, respectively. Hence, the digital dessert in this dataset is at least 1.11%. Although small, this implies that the opinion of some people is not really considered and hence is not part of the collective wisdom.
The size of the digital desert increases if we look at another kind of UGC platform such as Wikipedia. Comparing the logs of requested pages in the English Wikipedia during a month (June 2014) (https://dumps.wikimedia.org/other/pagecounts-raw/) with the new content added in the previous month (May 2014) we see that from the 1,350,554 articles edited/added during that month, 31% of them were not visited (These visits include humans and bots.) in June. This is an upper bound for the digital desert in this dataset and time period.
5. QUALITY OF THE WISDOM
In previous sections we have used an arbitrary definition of wisdom that is directly related with the amount of content produced by users. However, one can argue that quantity of content produced (i.e., activity) does not imply equal contribution to the global wisdom. To address this problem we need to measure the quality of content. Unfortunately, it is not simply to measure content quality in a social network, because it would be difficult to define what is a “good tweet” or a “good post” in Facebook. One option is to relate quality with popularity (e.g., retweets or likes), but such metric would be clearly biased towards popular users. Therefore,it is preferable to use a dataset where the quality of users’ contributions it is clearly ranked by the readers. A good example of such kind of content are Amazon’s products reviews, where readers can evaluate the helpfulness of a review by answering yes or no to the following question: “Was this review helpful to you?”
Specifically, we use a public Amazon’s movie reviews dataset released by [12] in 2013. This dataset contains almost 8 million reviews, from 889,176 users, of around 250K different movies, in a p erio d of 15 years (from 1997 to 2012). From each review we have – among other things – the (anonymized) author, the content, and also a field called “helpfulness”, that contains the numb er of readers that have rated the review as helpful or not.
In order to make this data comparable with the previous experiments, first we divided users in wise and others, following the definition given in Section 3.1 that based in the amount of activity (previously numb er of p ost or tweets, now numb er of reviews). In this case we found that 4% of users pro duced 50% of all reviews. This is similar to Facebook (7%) and Twitter (2%), suggesting that the pro cess of content generation is comparable with the previous cases. For future comparison we denote this group of users as activitybased-wise.
Next, we want to redefine wisdom by adding the dimension of content’s quality. To do that, we say that users are contributing to the wisdom only if each review has been rated as helpful by at least one reader. The intuition behind this definition is that if a review help ed at least one user, the review is a contribution to the total wisdom. Obviously, stronger requirements can b e imp osed (e.g., that at least 50% of the users rating a review found it useful). However, our definition will establish a lower bound for the content’s value. Hence, now the total wisdom will be the sum of all helpful reviews. Surprisingly, we found that 64% of the reviews were helpful for at least one reader, and 66% of users have produced at least one helpful review, showing that a wide group of users contribute to the total good content and that almost two thirds of the whole wisdom generated is valuable. However, breaking down the results we found that – again – just 2.5% of the users produced 50% of the total helpful reviews. We denote these users as quality-basedwise. Moreover, we found that quality-based-wise users is a proper subset of the activity-based-wise users in this dataset.
We also compute the (review) entropy for each user. To that aim, we grouped the reviews of each single user, and then computed the Shannon entropy in that text. Interestingly, we found that the Spearman correlation between activity and entropy is low (0.32), while the correlation between entropy and helpfulness is slightly higher (0.43). Figure 3 shows that from a certain level of users’ entropy the reviews tend to be more useful, but that also there is a saturation point where more entropy does not imply more helpfulness. This relation between entropy and value (helpfulness) is useful to generalize these results because we expect that users that introduce more information per word (i.e., higher text entropy) are – at the same time – adding more wisdom.
6. CONCLUSIONS AND FUTURE WORK
Our results, added to social influences, undermines the independence principle that is needed to have a real wisdom of crowds [20], as the percentage of people that produces most of the content is really small. Moreover, if we consider that very active people is highly connected among them, compared with the rest of users, creating a cohesive elite. The diversity principle is also challenged, as many users do not contribute to the wisdom, either because they do not exercise this option or because their opinion is not taken in account (the digital desert).
The distribution of how people contribute to wisdom becomes more skewed when a filter of quality is introduced. Although many people show the capability of producing helpful content, the majority of such content is produced by just a
subset of the elite.
The datasets used are already a bit old, but on the other hand one of them is complete and less noisy than a more current dataset, as at that time Twitter had less spam than nowadays. For sure the results in other datasets would be
different, but we believe the issues addressed in this paper will remain valid for the majority of UCG.
Finally, although we would like to believe that the Web is a more democratic environment as all the people has the same opportunities, at the end the Web mimics our society. Indeed, the economic or political power in most countries
belongs to a minority of the people. Even when explicit decisions must be taken through elections or referendums, many people choose not to exercise their right to vote.
REFERENCES
[1] R. Baeza-Yates and A. Tiberi. Extracting semantic relations from query logs. In Proceedings of the 13th ACM SIGKDD international conference on Knowledge discovery and data mining, pages 76–85. ACM, 2007.
[2] P. Boldi, F. Bonchi, C. Castillo, D. Donato, A. Gionis, and S. Vigna. The query-flow graph: Model and applications. In Proceedings of the 17th ACM Conference on Information and Knowledge Management, CIKM ’08, pages 609–618, New York, NY, USA, 2008. ACM.
[3] H. Cao, D. Jiang, J. Pei, Q. He, Z. Liao, E. Chen, and H. Li. Context-aware query suggestion by mining click-through and session data. In Proceedings of the 14th ACM SIGKDD international conference on Knowledge discovery and data mining, pages 875–883. ACM, 2008.
[4] M. Cha, H. Haddadi, F. Benevenuto, and K. Gummadi. Measuring User Influence in Twitter: The Million Follower Fallacy. In Proc. of the 4th Int’l AAAI conf. on Weblogs and Social Media (ICWSM), Washington DC, USA, 2010.
[5] A. Fuxman, P. Tsaparas, K. Achan, and R. Agrawal. Using the wisdom of the crowds for keyword generation. In Proceedings of the 17th international conference on World Wide Web, pages 61–70. ACM, 2008.
[6] F. Galton. Vox populi (the wisdom of crowds). Nature, 75, 1907.
[7] R. Gonzalez, R. Cuevas, R. Motamedi, R. Rejaie, and A. Cuevas. Google+ or google-?: dissecting the evolution of the new osn in its first year. In Proceedings of the 22nd international conference on World Wide Web, pages 483–494. International World Wide Web Conferences Steering Committee, 2013.
[8] C. Hsieh, C. Moghbel, J. Fang, and J. Cho. Experts vs the crowd: Examining popular news prediction perfomance on twitter. In Proceedings of ACM KDD conference, Chicago, USA, 2013.
[9] A. Kittur and R. E. Kraut. Harnessing the wisdom of crowds in wikipedia: quality through coordination. In Proceedings of the 2008 ACM conference on Computer supported cooperative work, pages 37–46. ACM, 2008.
[10] H. Kwak, C. Lee, H. Park, and S. Moon. What is twitter, a social network or a news media? In Proc.of Int’l conf. on World wide web, WWW’10, pages 591–600, NY, USA, 2010. ACM.
[11] G. Magno, G. Comarela, D. Saez-Trumper, M. Cha, and V. Almeida. New kid on the block: Exploring the google+ social graph. In Proceedings of the 2012 ACM conference on Internet measurement conference, pages 159–170. ACM, 2012.
[12] J. J. McAuley and J. Leskovec. From amateurs to connoisseurs: modeling the evolution of user expertise through online reviews. In Proceedings of the 22nd international conference on World Wide Web, pages 897–908. International World Wide Web Conferences Steering Committee, 2013.
[13] A. Mislove, M. Marcon, K. P. Gummadi, P. Druschel, and B. Bhattacharjee. Measurement and Analysis of Online Social Networks. In Proceedings of ACM SIGCOMM Internet Measurement Conference, 2007.
[14] M. Pasca. Organizing and searching the world wide web of facts–step two: harnessing the wisdom of the crowds. In Proceedings of the 16th international conference on World Wide Web, pages 101–110. ACM, 2007.
[15] C. Ricotta, A. Stanisci, G. Avena, and C. Blasi. Quantifying the network connectivity of landscape mosaics: a graph-theoretical approach. Community Ecology, 1(1):89–94, 2000.
[16] D. M. Romero, W. Galuba, S. Asur, and B. A. Huberman. Influence and passivity in social media. In Machine learning and knowledge discovery in databases, pages 18–33. Springer, 2011.
[17] K. Rutherford. The few who write Wikipedia. https://en.wikipedia.org/wiki/Wikipedia: Wikipedia_Signpost/2014- 01- 22/Special_report, 2014. [Online; accessed 19-Jun-2014].
[18] D. Saez-Trumper, Y. Liu, R. Baeza-Yates, B. Krishnamurthy, and A. Mislove. Beyond cpm and cpc: determining the value of users on osns. In Proceedings of the second edition of the ACM conference on Online social networks, pages 161–168. ACM, 2014.
[19] D. Saez-Trumper, D. Nettleton, and R. Baeza-Yates. High correlation between incoming and outgoing activity: a distinctive property of OSNs? In Fifth International AAAI Conference on Weblogs and Social Media,, Barcelona, Spain, 2011.
[20] J. Surowiecki. The Wisdom of Crowds: Why the Many Are Smarter Than the Few and How Collective Wisdom Shapes Business, Economies, Societies and Nations. Random House, 2004.
[21] B. Viswanath, A. Mislove, M. Cha, and K. P. Gummadi. On the Evolution of User Interaction in Facebook. In Proceedings of the 2nd ACM SIGCOMM Workshop on Social Networks (WOSN’09), Barcelona, Spain, August 2009.
[22] S. Wu, J. M. Hofman, W. A. Mason, and D. J. Watts. Who says what to whom on twitter. In Proceedings of the 20th international conference on World wide web, pages 705–714. ACM, 2011.
A Harvard Mad Scientist Invented Ice Cream That Has Skin
In a sleek new restaurant in Cambridge, Massachusetts, David Edwards is trying to change the way we take in nutrition. The professor, writer, entrepreneur and inventor has created a range of food innovations, from a carafe that will turn your scotch into an inhalable cloud to a device to print smells sent from your iPhone. His best known creation is wikicells, an edible skin meant to replace traditional food packaging. Edwards’ biggest problem isn’t creating these alternatives, it’s selling the public on them. (Video by Drew Beebe, David Yim and Alyssa Zahler)
News about the Future
Artificial muscle
Stanford researchers create super stretchy, self-healing material that could lead to artificial muscle
Researchers show how jolting this material with an electrical field causes it to twitch or pulse in a muscle-like fashion. This polymer can also stretch to 100 times its original length, and even repair itself if punctured.
This new material, in addition to being extraordinarily stretchy, has remarkable self-healing characteristics. Damaged polymers typically require a solvent or heat treatment to restore their properties, but the new material showed a remarkable ability to heal itself at room temperature, even if the damaged pieces are aged for days. Indeed, researchers found that it could self-repair at temperatures as low as negative 4 degrees Fahrenheit (-20 C), or about as cold as a commercial walk-in freezer.
Freight Farms
Freight Farms is addressing the needs of the world’s changing food landscape by providing physical and digital solutions for creating local produce ecosystems on a global scale. Freight Farms customers are located across North America and range from entrepreneurs and small businesses, to hotels and restaurants, to corporations and educational institutions. By decentralizing the food supply chain and bringing production closer to consumers, Freight Farms is drastically reducing the environmental impact of traditional agriculture and empowering any individual, community or organization to sustainably grow fresh produce year-round, no matter their location, background or climate.
Built entirely inside a 40’ x 8’ x 9.5’ shipping container, freight farms are outfitted with all the tools needed for high-volume, consistent harvests. With innovative climate technology and growing equipment, the perfect environment is achievable 365 days a year, regardless of geographic location.
Mathematics and sex
Mathematics and sex are deeply intertwined. From using mathematics to reveal patterns in our sex lives, to using sex to prime our brain for certain types of problems, to understanding them both in terms of the evolutionary roots of our brain, Dr Clio Cresswell shares her insight into it all.
Dr Clio Cresswell is a Senior Lecturer in Mathematics at The University of Sydney researching the evolution of mathematical thought and the role of mathematics in society. Born in England, she spent part of her childhood on a Greek island, and was then schooled in the south of France where she studied Visual Art. At eighteen she simultaneously discovered the joys of Australia and mathematics, following on to win the University Medal and complete a PhD in mathematics at The University of New South Wales. Communicating mathematics is her field and passion. Clio has appeared on panel shows commenting, debating and interviewing; authored book reviews and opinion pieces; joined breakfast radio teams and current affair programs; always there highlighting the mathematical element to our lives. She is author of Mathematics and Sex.
Recommended Book: Daemon
Daemon
by Daniel Suarez
Technology controls almost everything in our modern-day world, from remote entry on our cars to access to our homes, from the flight controls of our airplanes to the movements of the entire world economy. Thousands of autonomous computer programs, or daemons, make our networked world possible, running constantly in the background of our lives, trafficking e-mail, transferring money, and monitoring power grids. For the most part, daemons are benign, but the same can’t always be said for the people who design them.
Matthew Sobol was a legendary computer game designer — the architect behind half-a-dozen popular online games. His premature death depressed both gamers and his company’s stock price. But Sobol’s fans aren’t the only ones to note his passing. When his obituary is posted online, a previously dormant daemon activates, initiating a chain of events intended to unravel the fabric of our hyper-efficient, interconnected world. With Sobol’s secrets buried along with him, and as new layers of his daemon are unleashed at every turn, it’s up to an unlikely alliance to decipher his intricate plans and wrest the world from the grasp of a nameless, faceless enemy — or learn to live in a society in which we are no longer in control. .
Futurist Portrait: Thornton May
Thornton May is a futurist, educator and author. His extensive experience researching and consulting on the role and behaviors of “C” level executives in creating value with information technology has won him an unquestioned place on the short list of serious thinkers on this topic. Thornton combines a scholar’s patience for empirical research, a stand-up comic’s capacity for pattern recognition and a second-to-none gift for storytelling to address the information technology management problems facing executives. The editors at eWeek honored Thornton, including him on their list of Top 100 Most Influential People in IT. The editors at Fast Company labeled him ‘one of the top 50 brains in business.’
Thornton has established a reputation for innovation in time-compressed, collaborative problem solving. Thornton designs the curriculum that enables the mental models that allow organizations to outperform competitors, delight customers and extract maximum value from tools and suppliers. He specializes in creating action-based learning spaces for high performance organizations. He ran the multi-client research program at the Nolan Norton Institute, led the Management Lab at Cambridge Technology Partners, co-founded the Olin Innovation Lab, and founded the CIO Institutes at UCLA and UC-Berkeley. He co-manages the CIO Solutions Gallery at THE Ohio State University, co-directs the CIO Practicum program at the University of Kentucky and was the Executive Director and Dean of the IT Leadership Academy at Florida State College at Jacksonville.
Thornton serves as Futurist – External Technology Advisory Board at the Franklin W. Olin College of Engineering and is on the Advisory Board of Mobiquity, Inc.
Thornton’s research has been acknowledged in such seminal business books as Seth Godin’s Permission Marketing; Michael Schrage’s Serious Play: How the World’s Best Companies Simulate to Innovate; Moshe Rubenstein’s The Minding Organization; Bill Jensen’s Simplicity; and Jeff Williams’ Renewable Advantage: Crafting Strategy Through Economic Time.
Thornton’s book, The New Know: Innovation Powered by Analytics, analyzes what organizations know, how they come to know and how they act upon what they know/don’t know.
Thornton obtained his bachelor’s degree in Asian Studies from Dartmouth College; his master’s degree in Industrial Administration from Carnegie-Mellon University, and developed his Japanese language competence at the Center for Japanese Studies at the University of Michigan and Keio University in Japan.
Thornton May – Computerworld: “For the past seven years, I have traveled around the world asking organizations what they know, what they don’t know, what they need to know and how they come to know. Answers in hand, I have set about examining the data in the context of business outcomes and mission accomplishment (in the case of not-for-profit enterprises). I have come to some broad conclusions about the general state of knowing in the world today.
A paradox has emerged. Generally, our capacity to know (that is, what is knowable) is expanding exponentially, thanks to technology improvements (for example, affordable sensors and improved and accelerated analytics) and the apparently never-ending emergence of new sharing platforms (Facebook, YouTube, etc.).
What we actually know, on the other hand, appears to be advancing linearly — when it advances at all. Thus, I respond to Nick Carr’s question “Is Google Making Us Stupid?” in the negative. I maintain that Google isn’t making us stupid.”
Thornton May: Technology Futurist, Author, Educator and Keynote Speaker
Agenda
| 15 Nov 2016 morning | Amsterdam How Fashion Meets Impact A Blue Ocean Investment Strategy for the Global Textile & Garment Industry in cooperation with Impact Economy Location: Denim City | Amsterdam, Hannie Dankbaarpassage 47, 1052 RT Amsterdam |
Leave a Reply