Control-Generativity Paradox – Visiting student Michael Blaschke is working with me for the next year.

For the next year Michael Blaschke, is visiting me at the LSE. He is a final year PhD student from University of St Gallen (HSG) and the SAP Innovation Center St.Gallen. His research mainly focuses on digital platforms and value co-creation.

The following  paper-summary written by Michael gives an idea of his research. 

Abstract

The platform economy represents the most profound global macroeconomic change since the industrial revolution. Digital platforms afford organizations to synergistically co-create value in digital third-party ecosystems. Considering these ecosystems’ specificities, digital platforms require a delicate balance of two conflicting ends: control and generativity. While pure control makes adaptation difficult, pure generativity suffers the costs of experimentation without gaining associated benefits. In turn, embracing the complementary benefits of simultaneous control and generativity is challenging given its inherent contradictions. Beyond summarizing the control-generativity paradox of digital platforms, this blog post makes four alternative modes of balancing control and generativity available to platform managers. The publication can be found here.

The Control-Generativity Paradox

Key Takeaways

What? Digital platforms—digital core technologies upon which third parties add peripheral derivatives—afford organizations to co-create value in networked business ecosystems.

So What? While platform owners aim for stabilization to exploit their third-party ecosystem (control), third parties aim for autonomy to explore unanticipated avenues of innovation (generativity).

Now What? Platform owners draw upon at least four modes of balancing control and generativity in digital platforms—contextual, structural, temporal, and domainal balance.

Managerial and scholarly interest in digital platforms is mounting. Some of the most valued companies—including Alibaba, Amazon, and Alphabet—embrace the platform logic with surprisingly short histories. At the same time, many long-lived companies are considering how they can adopt the platform logic to improve performance. Prominent digital platform exemplars are social media platforms (e.g., Facebook and LinkedIn), mobile operating system platforms (e.g., Android and iOS), payment platforms (e.g., PayPal and Apple Pay), and peer-to-peer platforms (e.g., Uber and Airbnb).

Digital platforms are characterized by synergistic value co-creation in digital third-party ecosystems. These ecosystem make digital platforms subject to a delicate tension between (1) maintaining control and, at the same time, (2) stimulating—not directly managing—generativity through dynamically recombining third-party resources (Blaschke and Brosius 2018). While control captures mechanisms that encourage desirable outputs or behaviors by third parties (Tiwana et al. 2010), generativity describes a technology’s overall capacity to produce unprompted change driven by large, varied and uncoordinated audiences (Zittrain 2006).

Notably, control and generativity are not incompatible or mutually exclusive goals. Successful digital platforms meet both ends as pure control makes adaptation difficult and pure generativity suffers the costs of experimentation without gaining associated benefits. As balancing has in fact become the innate mindset of digital platform management, we ask: How do digital platforms balance simultaneous control and generativity?

Balancing Control and Generativity

Based on our research, we extracted a set of four modes of balancing control and generativity in digital platforms, namely contextual, organizational, temporal, and domanial balance. This set of modes is drawn based on the premise that digital platforms seek both (1) for stabilization to exploit the given ecosystem of third-party actors (through control) and (2) for dynamism to explore new avenues of resource integration in adapting to third-party actors’ external stimuli (through generativity). Next, we summarize these modes of balancing control and generativity in digital platforms.

 

 

Contextual balance denotes a situation-dependent combination of concurrent control and generativity. It is a form of contextual buffering, whereby the platform owner maintains control and generativity activities (1) situation-dependent for each platform partner individually and (2) simultaneously at any given organizational level. For instance, Microsoft (Windows) employs contextually configures control and generativity within the contexts of either exchanging, adding, or synergistically integrating third-party resources.

Structural balance refers to different types of partners that are subject to either control or generativity. It is a form of spatial buffering, whereby the platform owner maintains control and generativity (1) simultaneously on the platform ecosystem level, but (2) are situated within distinct organizational units for distinct partner types (e.g., new and existing partners), respectively. For instance, SAP (SAP Cloud Platform) runs one unit to negotiate and onboard new partners (control), while a different unit explores novel software with already existing partners (generativity).

Temporal balance denotes sequential shifts over time from control to generativity, and vice versa. It is a form of temporal buffering, whereby control and generativity (1) coexist for the same given platform partner but (2) at different points in time, so that the platform owner switches sequentially between control and generativity for each platform partner. To illustrate, Alibaba Group (Alibaba.com) predominantly maintained generativity to become a two-sided platform (1994-2004), relied on control to mitigate the threat of platform envelopment (2005-2006), and fostered generativity again to pursue a digital ecosystem strategy (2007-present).

Domainal balance denotes control in one domain with simultaneous generativity in another domain. It is a form of domanial buffering, whereby any given platform partner is subject to both control and generativity (1) organizational domains while (2) the platform owner balances these domain-dependent control and generativity activities globally across domains. For instance, Microsoft (LinkedIn) differentiates a platform’s core, interfaces, and complements as key architectural domains, each of which require different control-generativity configurations.

Recommendations

  1. Thriving platforms simultaneously seek (1) for stabilization to exploit the given digital ecosystem (through control) and (2) for dynamism to explore innovation in adapting to third parties’ external stimuli (through generativity).
  2. Thriving platforms balance control and generativity through a platform-specific adoption and adaption of the four proposed balancing modes.
  3. Effective platform mangers identify novel modes and mechanisms to achieve the targeted control-generativity balance.

About the paper

References

Blaschke, M., and Brosius, M. 2018. “Digital Platforms: Balancing Control and Generativity,” in: 39th International Conference on Information Systems (ICIS2018). San Francisco, US.

Tiwana, A., Konsynski, B., and Bush, A. A. 2010. “Platform Evolution: Coevolution of Platform Architecture, Governance, and Environmental Dynamics,” Information Systems Research (21:4), pp. 675-687.

Zittrain, J. L. 2006. “The Generative Internet,” Harvard Law Review (119:7), pp. 1974-2040.

 

I’m recruiting! Post-doc position at UCL/LSE: Interface Reasoning for Interacting Systems Research Fellow

I am recruiting for a joint UCL/LSE position examining the modelling of interfacing. The role will be split between UCL and LSE (managed by Prof. David Pym and myself) and will focus on modelling complex distributed systems. It would strongly suit an information systems PhD with a technical/computing/IT consulting type background. 

Post-doc position UCL/LSE: Interface Reasoning for Interacting Systems Research Fellow in Programming Principles, Logic, and Verification

Interface Reasoning for interacting Systems (IRIS) — a project funded by the UK’s EPSRC.

https://interfacereasoning.com

Today’s large enterprises are harnessing a complex mix of cloud computing services, APIs, legacy applications and service-oriented architectures to build complex information systems. You will work with an interdisciplinary team consisting of computer scientists, Information Systems researchers, logicians and modellers to explore the modelling of such complex distributed digital ecosystems. This fellowship will involve working with industry partners to analyse and model their ecosystems. Ideally you will have a technical/engineering background with experience in programming, formal methods, business modelling and business analysis, and an understanding of qualitative and quantitative research techniques. An understanding of information systems and management would be highly desirable, as would experience of action research or design science. Good communication skills are essential.

The role will be jointly managed by David Pym at UCL and Will Venters at LSE.

While based at UCL, the role will involve working at the LSE for around two days per week where you will have a desk.

Applicants must hold, or be about to receive, a PhD with relevant expertise and research interests; for example, in systems modelling, software engineering, formal methods, business analysis, and/or information systems. Advanced programming skills and knowledge of, or some interest in, distributed systems and/or information and systems security are highly desirable.

Appointment at Grade 7 (£35,328 – £42,701 per annum) is dependent upon having been awarded a PhD; if this is not the case, initial appointment will be at research assistant Grade 6B (salary £30,922 – £32,607 per annum) with payment at Grade 7 being backdated to the date of final submission of the PhD thesis.

Appointment is subject to UCL’s terms and conditions.

Th post is funded for 12 months in the first instance with a possible extension up to 36 months.

Closing date 23 June 2019.

Informal enquires to David Pym (d.pym@ucl.ac.uk; http://www0.cs.ucl.ac.uk/staff/D.Pym/)

or Will Venters (w.venters@lse.ac.uk; https://www.willventers.com).

 

For full details and to apply, please see UCL’s recruitment page for this position:

https://atsv7.wcn.co.uk/search_engine/jobs.cgi?SID=amNvZGU9MTgwNzA2NSZ2dF90ZW1wbGF0ZT05NjUmb3duZXI9NTA0MTE3OCZvd25lcnR5cGU9ZmFpciZicmFuZF9pZD0wJnZhY194dHJhNTA0MTE3OC41MF81MDQxMTc4PTkyNzg2JnZhY3R5cGU9MTI3NiZwb3N0aW5nX2NvZGU9MjI0

The 5-Es of AI potential: What do executives and investors need to think about when evaluating Artificial Intelligence?

I spent last week in Berlin as part of a small international delegation of AI experts convened by the Konrad-Adenauer Foundation[1]. In meetings with politicians, civil servants and entrepreneurs, over dinners, conferences and a meeting in the Chancellery[2], we discussed in detail the challenges faced in developing AI businesses within Germany.

A strong theme was the difference between AI as a “thing” and as “component”. Within most commercial sales-pitches AI is a “thing” developed by specialist AI businesses to be evaluated for adoption. Attention is focused on what I will term efficacy. Such efficacy aligns with pharmacology definitions as “the performance of an intervention under ideal and controlled circumstances” and is contrasted with effectiveness which “refers to its performance under ‘real-world’ conditions.[3]. AI efficacy is demonstrated through sales-pitch presentations based on specially tagged data or upon human-selected data-sets honed for the purpose.

As a “component” however, AI only becomes when it is incorporated into real-world consequential and ever evolving business processes. To be effective not just efficacious, AI must bring together real-data sources in real-world physical technology (usually involving cloud services, complex networking and physical devices) for consequential action. AI “components” then become part of a complex digital ecosystem within the Niagara-like flow of real businesses rather than a “thing” isolated from it. Since business processes, data-standards, sensors and devices, evolve and change so the AI must evolve as well while continuing to meet the needs of this flow.

To be effective (not efficacious) AI components must also be:

  • efficient in terms of energy and time (providing answers sufficiently quickly as to be useful.
  • economic in terms of cost-benefit for the company (particularly as the cost of human tagging of training data can be extreme),
  • ethical by making correct moral judgements in the face of bias in data and output, ensuring effective oversight of the resultant process, and transparency of the way the algorithm works. For example resent research shows image classifier algorithms may work by unexpected means (for example identifying horse pictures from copyright tags or train types by the rails). This can prove a significant problem when new images are introduced.
  • established in that it will continue to run long-term without disruption for real world business processes and data-sets.

The final two of these Es are particularly important: Ethics because business data is usually far from pure, clean and will likely include many biases. And Established because any process delay or failure can cause a pile-ups and overflow to other processes, and thus cause disaster. In my opinion, only if all these 5Es are achieved should a business move AI into core business processes.

For business leaders seeking to address these Es the challenges will not be in acquiring PhDs in AI-algorithms but instead in (1) hiring skilled business analysts with knowledge of AI’s opportunity but also knowledge of real-world IT challenges, (2) hiring skilled Cloud-AI implementors who can ensure these Es are met in a production environment and (3) appointing AI ethics people to focus on ensuring bias, data-protection laws, and poor data quality do not lead to poor ineffective AI. Given the significant competition for AI skills, digital transformation skills and for cloud-skills [4] this will not be easy.

So while it is fun to see interesting wizz-bang demos of AI products at industry AI conferences like those in Berlin this week, in my mind executives should remain mindful that really harnessing the potential of AI represents a much deeper form of digital transformation. Hopefully my 5Es will aid those navigating such transformation.

(C) 2019 W. Venters

[1] https://www.kas.de/ also https://www.kas.de/veranstaltungen/detail/-/content/international-perspectives-on-artificial-intelligence

[2] https://en.wikipedia.org/wiki/Federal_Chancellery_(Berlin)

[3] Efficacy : https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3912314/  Singal AG, Higgins PD, Waljee AK. A primer on effectiveness and efficacy trials. Clin Transl Gastroenterol. 2014;5(1):e45. Published 2014 Jan 2. doi:10.1038/ctg.2013.13. I acknowledge drawing on Peter Checkland’s SSM for 3Es (Efficacy ,Efficiency and Effectiveness) in systems thinking.

[4] Venters, D. W., Sorensen, C., and Rackspace. 2017. “The Cost of Cloud Expertise,” Rackspace and Intel.

[5] https://en.wikipedia.org/wiki/Industry_4.0

Wardley Mapping and building situational awareness in the age of service ecosystems.

How do executives make sense of their complex digital ecosystem of cloud services? How do they gain situational awareness? One method gaining increasing popularity in a large number of organisations is Simon Wardley’s “Wardley Mapping” technique. With Simon, and with Roser Pujadas and Mark Thompson, we have been developing and researching of how and why this technique is used. The following paper, to be presented in June at ECIS Stockholm[1], outlines the basics of the technique and our early findings.

Pujadas, R, Thompson, M., Venters, W., Wardley, S. (2019) Building situational awareness in the age of service ecosystems. 27th European Conference on Information Systems, Stockholm & Uppsala, June 2019. 

Paper Abstract:

We discuss the little-explored construct of situational awareness, which will arguably become increasingly important for strategic decision-making in the age of distributed service ecosystems, digital infrastructures, and microservices. Guided by a design science approach, we introduce a mapping artefact with the ability to enhance situational awareness within, and across, horizontal value chains, and evaluate its application in the field amongst both IS practitioners and IS researchers. We make suggestions for further research into both construct and artefact, and provide insights on their use in practice.

Keywords: Situational awareness, Distributed systems, Design Science, Strategy, Digital Ecosystems, Digital Infrastructure, modularity, servitization.

[1] ECIS, the European Conference on Information Systems, is the meeting platform for European and international researchers in the field of Information Systems. This 27th edition will take place in Sweden. We will present our paper in the “Rethinking IS Strategy and Governance in the Digital Age” research track.

For more on Simon’s Wardley Mapping see: https://en.wikipedia.org/wiki/Wardley_map or https://www.wardleymaps.com/ 

Fully funded PhD studentships in Interfaces / Boundary Resources / Digital Ecosystems / Cloud Computing.

I am part of my large research grant with UCL, QMUL and Imperial (https://binaryblurring.com/2017/12/04/win-of-6-million-to-research-digital-interfacing/  )  titled “Interfaces Reasoning for Interacting Systems.

As part of this we have four fully funded PhD studentships available –  http://www.cs.ucl.ac.uk/prospective_students/phd_programme/funded_scholarships/

While the advertisement focuses on computer science issues, the final bullet point “Tools for modelling and reasoning about organizational architecture” directly relates to Information Systems and Digital Innovation areas. Essentially if you would like to undertake a funded PhD focused on the managerial, social and organisational impact of Digital Interfacing (e.g. Digital ecosystems, Cloud Computing, Platforms, Boundary Resources, APIs,) please apply! While the advert insists on computer science or mathematics degrees – for those seeking to work with me a strong Information Systems Masters, or First-Class Degree in a related discipline would be sufficient.

[Note the PhD may need to be based at UCL while supervised by myself with Prof. David Pym as joint supervisor. The starting stipend will be approximately £17,000, with an approximate annual uplift of 3%. ].

It is possible that the studentship would be supported by AWS (Amazon) and involve working with AWS (Amazon), BT or Facebook among others .

Please apply ASAP as we are looking to recruit very very soon! Applications need only be a couple of pages long. 

Best wishes,

Will.

 

Five days of trials, tech and teamwork: welcome to Sprint Week

Innovation can transform the world. So how can it be encouraged and nurtured? Sofia Klapp, studying my course in “Innovating Organizational Information Technology” for her MSc Management of Information Systems and Digital Innovation (MISDI), reveals how our Sprint Week concept challenged her and her fellow classmates to generate, develop and pitch genuinely groundbreaking ideas. 

It was Monday morning, and 18 multidisciplinary teams were assembled at their desks. It was the beginning of the Sprint Week. We all had our materials ready (post-its, tape, markers, cardboards, and one big whiteboard) and plenty of healthy snacks to keep our energy levels high. Visa, one of the world’s leading payment brands, were explaining their global innovation challenges. From this moment until Friday afternoon, we would have to work in an “agile manner” to create an  innovative digital solution to win this innovation competition.

The Sprint Week: A learning-by-doing process framed as an innovation race

Will Venters and Carsten Sorensen, scholars on the “Innovating Organizational Information Technology” course, came up with a better use for the reading week for the MISDI Programme at LSE. Instead of just teaching about digital innovation and agile theory, why not use this week of no classes to immerse the students in a hands-on learning experience? They called it “The Sprint Week”, and this is the second year they´ve run this 5-day bootcamp.

As if making Sprint Week 50 % of our course assessment wasn’t enough, to add some extra adrenaline the teachers framed it as an innovation competition. Two key partners (Visa and Roland Berger) were invited to make things even more exiting. Both would be judges and choose the best projects for the grand final on Friday. Visa shared its strategic digital challenges to inspire our innovation ideas. Trending topics like mobility, digital identity, and a cashless society, served as fuel to ignite our imaginations. At the same time, Roland Berger, a strategic consulting firm and design sprint expert, was there to support our hands-on learning process.

The Sprint Week Methodology: The MISDI approach to developing digital innovations

But how did it all work? Sprint Week addresses digital innovation development by combining the best of two approaches: Design Sprint Methodology (a five-day work process for answering business questions through design, prototyping, and testing digital ideas with customers created by Google Ventures) and Soft Systems Methodology (a socio-technical approach broadly used to understand, design and intervene in information systems and digital innovation). While the first approach encouraged us to work in an agile manner as a multidisciplinary team, the second allowed us to understand the digital challenges from a systemic perspective considering their social and human implications.

The Sprint Week Experience Challenges: It’s not about intellectual capacity, but about the right mind-set and team-work skills.

Initially, these methodologies seemed simple. But as we moved forward we realised that putting them into practice wouldn’t be easy. For me, the biggest challenges we faced weren’t intellectual, but mostly related to how we managed uncertainty and how we interacted and communicated as a team. Whether we felt lost or on track depended on how well we managed our teamwork, triggering a roller coaster of emotions in our team throughout the week.

Managing the uncertainty that every innovation process entails can be very hard. We humans seem to have a control seeking mind-set that also looks for right answers. Yet working in an agile manner is not a linear step-by-step process. The agile mind-set is about learning and discovering the answers as you go, navigating in a disciplined way the messiness of the innovation process. If you are a control freak, you will suffer a lot. A good strategy was to keep trusting the methodology, accepting uncertainty as a normal feeling during the process while being open to be suppressive by the outcomes of applying it.

All the teams were highly diverse in their backgrounds and personalities. My team mates were from Indonesia, China, and the UK, whereas I´m from Chile; and their backgrounds ranged from IT-engineering, linguistic, international business, innovation and psychology. The methodology encouraged us to interact and discuss in an active and collaborative way. But it also meant dealing with disagreements among team members. We all speak English but our cultural differences and accents meant we had to focus extra hard. Getting to know each other before the Sprint Week and negotiating working styles was very important. We also ran open-heart sessions after each day gave feedback about what we liked and what we could improve for the next day.

What did I get out of all of this? From connecting theory and practice to being inspired by my classmates

As a MISDI student with previous work experience in innovation and agile development, I did not expect to learn as much as I did. The Sprint Week has definitely been the highlight of the MISDI programme so far.

Getting the opportunity to work on a real-life case challenge for a global company, with the input from industry experts, helped to link the theory I´d learned on the course with real world challenges.  And the ideas and discussions it generated between team members from different backgrounds, life-experiences and nations were amazing. Honestly, I feel that in one week it made me a better team player!

Moreover, seeing the teams´ project presentations on Friday was inspiring (all of them, not just the finalists). All the initiatives were so diverse and creative.  They greatly exceeded my expectations: from a data monetization platform that allows individuals to gain control of and get value from their digital data, to a futuristic payment chip inserted in consumer’s hand linked to an integrated app. Even some of the social projects surprised me; there was a donations platform that streamlines the funding of NGOs for increased transparency and another that provides digital sovereign identity and financial inclusion to the unbanked population.

This hands-on experience helped us gain a practical understanding of breakthrough methodologies while developing the multidisciplinary team skills needed to craft digital innovations. But most importantly, this week reminded me that at LSE your classmates are one of the main sources of learning and inspiration.

ABOUT THE AUTHOR

Sofia Klapp is from Chile and holds a BA in Organisational Psychology, plus diplomas in Business Management and Innovation & Entrepreneurship. Her experience in leading customer experience evaluations in technology projects in a global IT consultancy enabled her to understand the strategic complexities that digital transformation brings, encouraging her to pursue her MSc Management of Information Systems and Digital Innovation (MISDI) at LSE.

Lecturing for Cambridge Executive Education

It was great to be back in Cambridge last week lecturing with Dr Mark Thompson on Digital Innovation and Transformation @ Judge Business School Executive Education Programmes.

20181115_095722My contribution was a deep dive into the digital infrastructures which are transforming our digital economy. I talked about Cloud Computing as transformational in enabling Data and Algorithms to have agency in changing business environments.

Central to this transformation is Artificial Intelligence (AI) which I argued to be a means of industrialising data-analytics at scale. Through AI and cloud computing, organisations can share data across their organisational boundaries in order to derive new business benefit.

For example an FMCG company might harness AI to automatically integrate data from wholesalers, distributes and retailers with complex production data, external statistics on consumer behaviour, logistics movement or meteorology. Through this integration an AI algorithm may better forecast demand fluctuations and thus reduce costs than a closed data-process.

Achieving this though requires effective, secure and agreed interfacing between companies for large data-sets and complex pooled data processes. This digital interfacing is the focus of my current research efforts: https://interfacereasoning.com/  

(Banner image (CC) from Rept0n1x : Used with thanks).