The Recent State of the Web
Hi! My name is Yulia. I'm a curriculum engineer at Inrupt.
Web 1.0’s purpose was to help connect individuals and organizations all around the world. It was built for ease of use and compatibility as more and more devices and people were looking to connect and be a part of a burgeoning information distribution system. Web 1.0 gave us the blogosphere where static HTML documents were linked together with URLs, and all of those documents were accessible through the HTTP.
Let's look at an example of how the traditional Web 2.0 way of building applications that are strongly coupled with data results in scattered personal data for users and their experiences on the web. We'll start with a person who is subscribed to a number of streaming services. They can be Showflix, Prime TV, Pear Plus, and let's say Superhero Media. Each of these services require a subscription, and for that they require the user to create an account. There might be a way around creating yet another account by using a “single sign on”, but not all services offer that, and not all people have a single sign on type of account. An example of single sign on solutions are Google, who acts as an identity provider. You have likely seen a “Login with Google, Facebook, Apple, or Okta” when using various services on the web. We'll talk more about the role of identity providers in later videos. For now, let's get back to this person with their streaming services.
Say they signed up for the video streaming services and proved that it is them using their email, Facebook, or Apple account.
Now each service starts to collect data about them. What they like, what they browse for, and what they don't like. Each service wants to keep them as a customer, so each builds a profile of the usage of their services, in order to cater to the person's preferences. Yay! Convenience! Each of these services ends up having a different accumulated portrait of the user though. Showflix found that they liked horror more than drama. Pear Plus is under the impression that this same person is a comedy buff, meanwhile, Superhero Media learned that they're more of a funny-horror kind of viewer. The viewing patterns of all of these versions of the same person are this user's personal data, captured by each of the independent online services they use. However, this person cannot reuse this information for other streaming services, they can't update this information when it is no longer relevant, nor can they easily know what their personal data reveals about them, and finally they don’t know where all that personal data is stored. Note: the streaming services are also trapped in their own collection mechanism. They don’t know any more about the person than they can collect on their own. If our user starts to enjoy Korean dramas and signs up for a Korean streaming service, both the user and the streaming service will have to start new - one sharing their personal data while the other learning and guessing about preferences and needs of their customer.
The user still has little to no control in making sure that it is relevant to the rest of their tastes, or that the viewing recommendations are based on content consumption from other streaming services creating a more complete view of them for the service providers. A view that is an amalgamation of all of their viewing preferences, history, and up-to-date information about their viewing patterns, which is all part of their personal data.
While the user is limited in their ability to control, update, or view their data, the streaming services are also limited in what they can offer to their users, due to limited access to data about them. Each service likely has a different application architecture, recommendation algorithms, and data models. They are locked in their proprietary silos, where even if they wanted to share the data about their common users, to provide a better user experience, it would take a lot of time and effort for them to do so.
If the streaming services had access to the most complete view of a users' viewing data, they could offer better services and compete for the user's attention by offering a more personalized, better tailored content that is more relevant to their user at any given time.
Using this streaming services scenario we ran into a few key challenges that hold true for Web 2.0 as a whole.
Web 2.0 has created siloed environments that are increasingly complex and dynamic in nature, where apps are inseparable from the data users generate through them and users have little control over the data that's held about them.
The increased complexity and dynamic nature of tooling has created siloed environments where apps are inseparable from the data that they gather.
This, in turn, permeates into a few key areas of improvement for the web, namely user agency, interoperability, and data governance.
In the case of user agency, user data is scattered across applications and web services, as a result people lack control and knowledge of their personal data, and generally lose out on the value embedded in it. This lack of user agency on the web is what the next phase of the web solves.
Then there is the lack of interoperability. Data being tightly coupled with the applications that gather it inhibits interoperability, which contradicts the initial vision for the web. HTML, HTTP, URL, and other open standards broke down proprietary software and hardware barriers, but new tooling and application barriers were raised in this era, and we're back to siloed environments. The next step in the web evolution, which is referred to as Web 3.0 solves this problem as well.
Another area that is tackled by Web 3.0 is the data governance imbalance and complexity. At the moment, data has become a corporate currency. The rationale is that the more data an application can collect, the more useful services it can provide to its users because of insights from that data, this in turn gives a competitive advantage to the organization that created the app. Often this translates to innovation focused primarily on the volume of data gathered to deliver value from aggregated insights rather than nurturing relationships with their customers.
These limitations of agency, interoperability, and data governance apply to digital services in all verticals, including, but not limited to medical, government, and social media. They all provide a service, gather personal data, and as a result each has just a snapshot of data about their users. Some digital services might even have entirely opposing ideas about the same user due to outdated or inaccurate information that they collected at some point, and due to lack of data interoperability.
In this video we talked about the current state of the web and some of its challenges. We looked at how personal information gets scattered and trapped in silos and remains inaccessible to the people that this data pertains to. We explained how this results in a lack of user agency on the web. We also noted that because data is tightly coupled with applications, it tends to be proprietary and difficult to extend and evolve, making it difficult to share with anything other than the application that created it or that it serves.
Last, but not least, we talked about how data governance is often correlated with market share and commercial success, which relies on volume data collection to produce insights from the aggregate in order to establish and nurture customer relationships.
In the next video we'll discuss interoperability and ways to tackle some Web 2.0 hurdles using existing web standards in innovative new ways.