Yesterday I went to a workshop at LSE on ‘Digital Platforms’.
This considered the theory behind social platforms such as Twiter and also the Apple apps store. Dr Kevin Boudreau spoke about the economics of app development and hosting on the Apple platform. His argument was that this platform provides a very low entry cost ($99 per year plus 30% of any income through the app), so that the normal selection and retention processes don’t apply: anybody can put something up, and they do. (1103 ‘bubble shooter’ games, and so on.) Apps take little time to develop, and are developed for non-financial reasons: as hobby projects, for self-esteem, or to support another interest, with little or no expectation of market returns. This leads to a collapse of quality: 97% of the apps on the platform do not make any money, although Dr Boudreau argued that it also led to some very high quality apps as well.
There was then a discussion on what platforms actually are. Is almost anything a platform (eg a business itself)?
Three or four definitions were suggested:
– a distinct functional cluster with its own set of resources (but if so, these have existed for a long time, eg any company.)
– a market (but if so, why not just call it a market?)
– both of the above plus a ‘social space’
– a coordination space or mechanism between supply and demand (eg Uber)
The best answer, I thought, was Dr Boudreau’s own comment that platforms are actors in the economy who regulate beyond their own boundaries and shape behaviours beyond their own boundaries.
Dr Anne Helmond then spoke about the ecosystem between advertising platforms (such as Facebook) and third party partners. Sadly her talk is not yet available.
She showed some fascinating diagrams of linkages between Facebook, Twitter, etc, and some of the many partner sites who help you to develop campaigns using social media. These partners offer:
– Audience Onboarding: Find your current customers on Facebook via Custom Audiences.
– Audience Data Providers: Find new customers on Facebook using 3rd-party data.
– Audience Outcomes: Understand and optimize how you connect with people by measuring reach, frequency, targeting and cross-device performance
and so on.
They may pay for access to more FB or TW data than is usually available, and combine this with other paid-for 3rd party data in order to build target lists.
All of this struck a chord after my experiences with Google the day before (see previous post). Similarities:
– two or three major competing platforms, all existing on a massive scale (Facebook/ Google Cloud)
– large numbers of users
– large ecosystem of consultants and ‘partners’ available to help new users, sometimes developing their own related software to run on or with the original platform.
– large amounts of data being generated and used. The difference here is that FB etc own your data, and sell it or use it. Google claim they don’t own or use your business data.
– emphasis on new and easier business models (‘understand and optimise how you connect with people’) – all of which require no further effort to collect or to analyse the data, because the data is already there and you can rent the analytical engines. You can have it now, more or less.
– the idea that all these major platforms shape behaviours beyond their own boundaries in a way that perhaps only governments have done up to now.
Then today the ‘Economist’ has an essay about ‘The Data Economy’.
This makes several good points:
1. the quality of data is changing: “They are no longer mainly stocks of digital information—databases of names and other well-defined personal data, such as age, sex and income. The new economy is more about analysing rapid real-time flows of often unstructured data: the streams of photos and videos generated by users of social networks, the reams of information produced by commuters on their way to work, the flood of data from hundreds of sensors in a jet engine.”
2. Peple are finding new uses for data: “Facebook and Google initially used the data they collected from users to target advertising better. But in recent years they have discovered that data can be turned into any number of artificial-intelligence (AI) or “cognitive” services, some of which will generate new sources of revenue. These services include translation, visual recognition and assessing someone’s personality by sifting through their writings”
3. but data are not homogenous: “flows of data are not a commodity: each stream of information is different, in terms of timeliness, for example, or how complete it may be. This lack of “fungibility”, in economic lingo, makes it difficult for buyers to find a specific set of data and to put a price on it: the value of each sort is hard to compare with other data.”
4. “Trading data is tedious”…. As a result, data deals are often bilateral and ad hoc. They are not for the fainthearted: data contracts often run over dozens of pages of dense legalese, with language specifying allowed uses and how data are to be protected…In the case of personal data, things are even more tricky.”
5. “Data, argues Hal Varian, Google’s chief economist, exhibit “decreasing returns to scale”, meaning that each additional piece of data is somewhat less valuable and at some point collecting more does not add anything. What matters more, he says, is the quality of the algorithms that crunch the data and the talent a firm has hired to develop them.”
The essay then discusses EU moves to regulate data collection and use, and concludes that individuals need to be aware of the value of the data they produce. It also points out that most of the big data companies are in the USA, and “Past skirmishes between America and Europe over privacy give a taste of things to come. In China draft regulations require firms to store all “critical data” they collect on servers based in the country. Conflicts over control of oil have scarred the world for decades. No one yet worries that wars will be fought over data. But the data economy has the same potential for confrontation.”
Danah Boyd’s essay ‘Six Provocations for Big Data’ refers to ‘apophenia’ which she defines as “seeing patterns where none actually exist, simply because massive quantities of data can offer connections that radiate in all directions”. An essay by Hal Varian points out various ‘confounding variables’ that make some inferences dubious.
As an article by Burns et al points out, Google has now had three sets of ‘cloud’ management software: Borg, Omega and Kubernetes. Each new one was developed to extend or improve the capabilities of the previous one. The Google cloud offering is quite complex and offers several routes to the cloud: one session on Wednesday called ‘where should I run my code’ explained that “Google Cloud Platform offers solid support for the full spectrum of compute models. We’ll help you navigate the tradeoffs and decide which models are the best fit for your systems as well as how the models map to Google Cloud services – whether Compute Engine, Container Engine, App Engine and/or Cloud Functions.”
If future uses for data or for corporate ICT are discovered, will we all be set in a rut that makes it more difficult to open up to them?
I still find myself wondering how the flood not only of data, but also of busy people developing new uses for it, will affect our world. Varian’s decreasing returns to scale may impact fairly soon. The focus on quality, however, is not easy: even those who create algorithms often have no idea how well they work. But as the infrastructure and technologies develop and become more standard, we may trap ourselves into a way of doing things that creates a pervasive mindset.