Higher education platforms and cloud infrastructures in the ‘dataist state’

Ben Williamson

Higher education is being reimagined by companies and agencies as being integrated with digital platforms and cloud infrastructures. Photo by Bahador on Unsplash

Prepared for the Centre for Global Higher Education research seminar Student learning in the midst of the pandemic: the case of the UK, 10 November 2020

Privately-owned digital platforms have become integral to societies and public life, and are advancing into the provision of higher education too. As Jose van Dijck, Thomas Poell and Martijn de Waal argue in The Platform Society, two types of platforms are especially significant for HE. First, infrastructural platforms, such as those of Microsoft, Google and Amazon, undergird other platforms used in universities. They enable and manage global data flow, storage and analysis. And second, sectoral platforms have become integrated into public services and the public sector, significantly so in HE, while depending on infrastructural platforms for their functioning.

The expansion of privately-owned sectoral platforms and infrastructural platforms into HE has accelerated as COVID-19 and the declaration of states of emergency worldwide enabled the global technology sector to seize new market opportunities and carve out advantageous public-private partnerships across both the health and education sectors. Educational technologies have become key sites of private sector and commercial influence, initially by offering short term ‘relief’ from the crisis in the shape of online platforms for remote teaching and learning, and then by promoting them as a model for long-term ‘reconstruction’ involving private platform providers as fully-embedded partners in HE delivery. What does this increase in the participation of private platform actors signify for teaching and learning in universities, for the public role of higher education, and for the lives of students?

Intelligent networks

During the pandemic, higher education in the UK and internationally became a focus of intense ‘reimagining‘ by sprawling webs of think tanks, consultancies, sector agencies, edu-businesses, financial organisations and technology companies, as part of both a longer history of multisector HE reform efforts and recent projections of the ‘digital transformation‘ of HE. The Jisc Learning and Teaching Reimagined Initiative launched in October 2020 is a good example. It is a multisector programme of HE reconstruction involving the organisations Jisc (UK’s digital learning agency for HE), Universities UK (the sector’s representative body), Emerge Education (an edtch startup investment company), and Salesforce (a Silicon Valley cloud-based software company).

In the joint report Digital at the Core, a strategy for ‘digital transformation’ of UK HE, this multisector group projects a shared imaginary of digital, data-intensive HE:

The technology now exists to connect the variety of applications used within the university, where the IT landscape tends to be more fragmented than in the enterprise. Replacing these siloed ‘information systems’ with intelligent information networks will enable highly personalised engagement with students and staff, individualised experiences, and actionable strategic intelligence. … Now, advanced analytics with augmented intelligence have the capacity to predict what will happen and prescribe actions which can be taken to cause an outcome.

The report supports the idea that aspiring ‘data-empowered universities’ should emulate the data-driven approach of Silicon Valley platform companies:

Consumer market leaders such as Netflix, Apple, or Uber apply data-driven decisions and provide dynamic experiences based on an individual consumer’s information. Applying these same design principles to higher education can transform the way that our stakeholders experience learning, teaching, research, and professional services.

The Jisc digital strategy anticipates HE being reconstructed to match the logics of the platform society. It involves the datafication of all activities and data-driven decision making; the commodification of education as ‘unbundled’ components provided by market suppliers; and the selection of ‘personalised’ educational content based on individualised data analytics. Being ‘digital at the core’ also means universities being connected to interoperable cloud and data systems provided by giant infrastructure partners in new public-private partnership configurations.

The partnership with Salesforce on the strategy is suggestive of how such a vision of HE could appear in practice. It has pushed its Education Cloud for HE as a key solution for universities during the pandemic. On the launch of the Jisc strategy, the general manager and senior vice-president of Salesforce.org Education Cloud said:

Higher education is at a turning point globally. And there is a tremendous opportunity to reimagine how the sector in the UK should evolve using technology, a pivotal step required to thrive in the next normal. Salesforce.org is proud to work alongside industry peers to develop a framework for digital transformation that will inevitably shape and contribute to future learner and institution success.

Salesforce enacts this vision through its Education Data Architecture offer to institutions. The EDA is designed ‘to configure Salesforce for education’ as a ‘360-degree’ infrastructure for viewing and using student data. It combines student information, learning management, and other institution systems, as well as third-party plug-in apps available through the Salesforce AppExchange, into one interoperable system.

The salesforce Education Cloud is also ‘infused with Einstein,’ the company’s machine learning and predictive ‘artificial intelligence.’ As another Salesforce executive put it:

Now, campuses of all sizes are expected to operate more like nimble software organizations—innovating quickly, scaling up virtual service centers, and putting infrastructure in place to support the always-on digital engagement needs of students. … Picture using the underlying technologies in Amazon one-click, Spotify recommendations, or the Apple Watch’s health tracking for higher education.

The promise of the Education Cloud is to compile all student data into one interoperable intelligent network for 360-degree observation, analysis and prediction, while offloading or outsourcing the technical demands to the Salesforce AI, cloud and data architecture.

Cloudsourced education

Salesforce is not the only infrastructure provider for student data cloud hosting and analytics, as other companies including Google, Microsoft and Amazon Web Services have entered into a global competition for infrastructural dominance over universities’ information and learning management systems. They offer both sectoral platforms (e.g. Google G Suite, MS Education tools, AWS Educate programs) and infrastructure services to enable a host of other platform providers. These infrastructures are important aspects of online learning platforms that have proliferated across HE during the pandemic.

The for-profit MOOC provider Coursera, for example, is a sectoral platform that relies on third-party cloud infrastructure provided by Amazon Web Services. AWS has become a major infrastructure provider in HE, providing back-end services to Coursera, as well as leading management systems providers Blackboard and Canvas too. Powered by AWS, Coursera offered Coursera for Campus free degree content during the pandemic. Institutions could sign up for ‘bundles’ of ‘job-relevant’ degree content in the absence of existing institutional online learning arrangements. Coursera also announced a new business opportunity, with plans to sell the courses it has developed as ‘courseware’ to institutions. This enables the outsourcing of key teaching functions to ready-made content and the AI software for ‘personalised learning’ that Coursera has also built into the platform.

In this sense, as a sectoral platform serving universities, Coursera anticipates the future cloud-sourcing of higher education through public-private partnerships on both the demand and supply sides. Subscribing institutions purchase the unbundled components of a degree and AI support from Coursera, which acts as a platform hosting content provided by its partners network, all of it enabled by back-end infrastructure provided by Amazon Web Services. According to a senior Coursera software engineer, ‘Quite simply Coursera could not exist without AWS. We rely on AWS every single day for the speed, service, agility and scale required to serve our students worldwide’ and to ‘provide deeper insights into student learning that would not be possible without AWS.’ AWS also runs its own popular public-facing training courses on cloud computing and machine learning on Coursera, as it both provides infrastructure for the platform and upskills new experts to work in its cloud services at the same time.

To a significant extent, then, the ‘reimagined’ HE of Jisc’s UK digital strategy would be increasingly integrated into the commercial ecosystem of the global cloud and data infrastructure providers. Jose van Dijck and colleagues argue that MOOC platforms such as Coursera are entangled in a ‘political agenda where formerly defined public and government functions are administered towards yielding private profits,’ and public funding is increasingly lured toward platforms that capitalise on ‘data-based, technology-intensive forms of teaching and learning, at the expense of investments in human-based, labor-intensive pedagogical and didactic skills.’

The dataist state

These shifts to embed HE in global computing infrastructures and public-private partnerships might be understood as a manifestation of a new form of ‘digital statecraft’ as Marion Fourcade and Jeff Gordon have conceived it. ‘By “statecraft,”’, they argue, ‘we mean the state’s mode of learning about society and intervening in it.’ In a dataist state, the state learns from tracing ‘discrete slices of people and things’ as data, and then seeks to intervene to change, ‘nudge’ or optimise their behaviours, for example through machine learning-based personalised learning technologies.

But the dataist state lacks the in-house technical capacity to undertake this analysis and intervention at huge scale. So it must outsource data collection and analysis to private platforms and the infrastructure providers on which they depend, particularly in public sectors such as healthcare and education. In the dataist state, machines take over significant parts of the state’s operations through asymmetrical public-private partnerships that raise private challenges to the state provision of services. And this, Fourcade and Gordon argue, anticipates ‘the private appropriation of public data, the downgrading of the state as the legitimate producer of informational truth, and the takeover of traditional state functions by a small corporate elite.’ The dataist state is not just digitalised but hugely privatised too, reshaping the infrastructures of the public sectors and the welfare state.

Their tentative solution is to seek to adapt big data technologies to an alternative, citizen-led political rationality. ‘Given our view that the fundamental technologies of the digital age (distributed systems, machine learning) are not going anywhere,’ Fourcade and Gordon conclude, ‘the question is whether we can put them to more solidaristic and less extractive uses.’ They suggest a mode of statecraft that identifies social problems ‘from the perspective of those affected,’ such as reorganising certain digital infrastructures as public utilities and decommodifying a number of essential internet-based services. Education could be a key sector for such experiments in alternative statecraft.

These provocations on the privatised dataist state and how to reorganise it raise questions about the kinds of values and purposes of HE promoted as parts of post-pandemic recovery and reconstruction. Public values of education as a common good and as a bedrock of democracy are now in tension with private values of individually-targeted, hyperpersonalized, job-relevant education. These are challenges to raise with sectoral bodies such as Jisc and the Office for Students, which lean towards the public-private partnership paradigm of the dataist state, as signified by political appetite for flexible and remote learning focused on economic and skills needs. They are ceding the authority of higher education institutions and the state to ‘learn’ about students, institutions and the sector to global private platform and infrastructure providers that currently only see them in economic and job-relevant terms. We need alternative ways of deploying platforms and infrastructures that are led from the perspectives of those affected within education itself.

This entry was posted in Uncategorized and tagged , , , . Bookmark the permalink.

1 Response to Higher education platforms and cloud infrastructures in the ‘dataist state’

  1. paulmartin42 says:

    I can see why Salesforce might see the fragmented IT infra-structure as an opportunity, but its all about perspective. A HeadTeacher has a handful of different IT platforms to manage a school all different with different cultures & worse a requirement by Local Authorities if there is need for further staff in a school to learn the system there is “train the trainer” delivery.

    From the teacher & student perspective it seems a lot simpler. There have been some crocks of VLEs but nowadays the main players dominate the student facing front. It’s hardware (eg computer trolleys, and old IT suites) inc peripherals that restrict application, novel or otherwise.

    BYOD was problematic I suspect that recent financial strictures will make progress in this field difficult

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s