I keep on having this recurring idea that would allow people to host their own app data in their own personal data silo, maintaining control over it and importantly privacy by potentially encrypting it.

Technically, this is doable. But what are the advantages of being able to do this and would things actually be better than with centralised so-called “Walled Garden” apps?

Messaging is a good example of a service that could be decentralised in this way. However, if I decided to build my data silo today, no apps would be compatible with it - all of them are proprietary and already have data centres dedicated to storing their users’ data. So I’d have to build my own messaging app too. But what about everyone else? They’d have to either use my app or build their own. The other option is to agree on a standard protocol over which these messaging apps could communicate with my data store.

This is a lot of effort to go through in order to send a message to me, and let’s not forget that the people messaging me will still also be using Facebook messenger, WhatsApp, WeChat, SMS, etc. There isn’t much incentive to switch because of the very nature of these apps that take advantage of the network effect.

WhatsApp was successful in gaining traction and once a threshold was reached, it went viral. The fact that you only needed a phone number to use it was one of the main reasons for this. It solved the problem of identity by delegating to an existing protocol, eliminating the need to create yet another online account.

By reducing the requirements to be a participating member of the network, WhatsApp boot-strapped itself to become one of the most popular messaging apps today.

But how could a decentralised competitor possibly rival such a tech giant as WhatsApp owners, Facebook?

Looking back in history at why other decentralised networks became so popular might be useful in order to consider this question.

The introduction of the Web in the 90s caused a huge tech boom that changed how we shop forever. With the commercialisation of the Web, commerce shifted from being mostly on the high-street and in catalogs to what it has become today - a mix of online retail and in-shop retail. This was in part due to the relative ease of going online and clicking “buy” in comparison to either walking to the shop or getting on the phone (or possibly sending a paper order form in the post!) But it also allowed new companies to come to market with fewer costs and it opened up avenues that weren’t possible before.

It also boosted the growth of home and personal computers which are ubiquitous today.

The Web was catapulted into existence in part due to its potential uses. The potential use cases of a website are unbounded and this is clear today.

So what could decentralisation offer that the Web can’t?

Perhaps a new model of data ownership? The Web changed how we request services and goods in the home, but it didn’t change how we own our data. Put another way, the Web didn’t need to change how we own our data to become what it is today.

And it still doesn’t - despite events such as the Cambridge Analytica scandal, most people continue to put trust in Facebook and companies like it. The convenience offered outweighs the lack of control with respect to how a user’s data is used, which namely means more targeted advertisement and more influence on users’ opinions.

Does that mean decentralisation of these systems could offer a less influential, more controlled alternative?

Some would argue that it does, but in truth if a decentralised network enjoyed as much success as a centralised one, there would be no reason why any societal problem that exists in a centralised system wouldn’t also exist in a decentralised world.

For example, in the case of influential advertisements, if there isn’t a dedicated mechanism for delivering ads to users, there is at the very least a mechanism to send (unsolicited) adverts to users - i.e. spam.

Other issues that would still plague users would include bullying, hate speech, marginalisation and harassment - it’s true that technology is a reflection of society; a society that is comprised of the same people, online or off.

But is it a good idea despite these issues? Would a change point us in the right direction as a society?

I personally don’t think so - these issues are societal, not technological.

However, I really think we can move the needle a little bit towards having more private lives online. I don’t believe we can allow for fully encrypted, anonymous communication without bad actors making use of the same technology, but I do believe the average web user should feel safe and confident that they can use the web privately.

Circling back to “Data Ownership as a Service” - how do we get there? And should we try?

Well, we discussed that decentralisation doesn’t solve any of the web’s main societal issues but it would change the way people control their data, potentially to the benefit of individual privacy.

This topic can be broken up into a few different aspects that I’d like to discuss separately:

1. Societal issues - how decentralisation could amplify the worst parts of the web and make moderation and content policing near impossible.
2. Technical issues - the technical complexities that would inconvenience even the most tech-savvy user.
3. Communal issues - the combinatorial friction that comes with any mainstream network-effect app.

Be sure to check back soon for further posts on the subject.

Thanks for reading!