--- published: true layout: post title: 'An Observable Regulatory Provider Or Industry API Management Gateway' image: https://s3.amazonaws.com/kinlane-productions2/algorotoscope-master/35201856153_61bc075e4b-nazi-invasion.jpg ---

I wrote a separate piece on an API gateway specification standard recently, merging several areas of my research and riffing on a recent tweet from Sukanya Santhanam (@SukanyaSanthan1). I had all these building blocks laying around as part of my research on API gateways, but also from the other areas of the API lifecycle that I track on. Her tweet happened to coincide with other thoughts I had simmering, so I wanted to jump on the opportunity to publish some posts, and see if I could slow jam a conversation in this area. Now, after I defined what I’d consider to be a core API gateway set of building blocks, I wanted to take another crack at refining my vision for how we make it more observable and something that could be used as a tech sector regulatory framework.

Copying and pasting from my API gateway core specification, here is what my v1 draft vision for an API gateway might be:

That is a pretty long list. I know there are other building blocks missing, but this represents my first pass through my API research. It reflects the building blocks I want available when I put an API gateway to work in any cloud, on-premise, or on-device use case. However, in this post I wanted to go beyond what I’d consider the core API gateway, and talk about how we make it more observable, and something that could be applied to regulate an industry. Not something that happens behind the scenes, but something that happens out in the open, bringing in some sunlight, and pulling back the curtain on the black boxes some companies enjoy operating. While this won’t solve all our problems, I think it will provide a pretty good first step towards bringing some much needed observability to the tech sector, using common solutions that already exist. Here are a handful of building blocks I think could contribute to this conversation.

This model isn’t without precedent. Last year I spent a couple months studying the approach by UK regulators to bring more observability into the banking sector, and the formation of Open Banking organization (https://www.openbanking.org.uk/). Learning more about what open banking was in the UK (http://apievangelist.com/2018/02/21/what-is-open-banking-in-the-uk/), how it provides a common API definition (http://apievangelist.com/2018/02/21/open-banking-in-the-uk-openapi-template/), who the common stakeholders were (http://apievangelist.com/2018/02/26/the-banking-api-actors-in-the-uk/), and exploring how I could emulate this approach as an open source API industry template (http://open.banking.blueprint.apievangelist.com/). In 2019, I want to go even further with this open source API Blueprint, understand how it can be used to define a common open source API gateway specification, while adding the necessary building blocks to ensure there is observability at the industry level.

I am proposing that this model be defined, standardized, and applied at the single provider, or industry level. If required, an independent entity can be setup to operate the API gateway as an independent platform, funded by regulators, and the monetization layer that leverages a mix of providers, consumers, researchers, and industry analyst access to the platform. While there will be private layers of the platform, the goal is to provide as much as possible out in the open, operating as many public API platforms have been operating for the last 20 years. Ironically, some of the worst behaved technology platforms have operated using this model in the past, but sadly have been actively tightening down access levels. I’m proposing we take their formula, open source it, and operate it independently, out in the open, and make them pay for it.

As I said before, this will not solve all problems. However it will define a layer that ALL desktop, web, mobile, and device applications can be required to go through. Requiring that provides only develop applications on top of APIs deployed within regulated API gateways. Requiring that all internal, partner, and 3rd party public consumers access API via the single, or regionally distributed gateways. If auditors ever find that an API provider is leverage APIs not listed in the API gateway directory, or their partners and the public have access to data that is not defined in schema within the API gateway—then there is a problem, and enforcement can follow. Of course, all of this isn’t as simple as proposed in this post, but it provides the scaffolding and blueprint for how it can be applied. There is no reason this can’t be applied to regulate technology companies, and applied to existing industries that are already regulated, helping bring more observability into already existing regulatory practices.

Nothing I’m proposing here is rocket science, or theoretical. It is based upon proven practices of tech companies. All I’m saying now, is rather than just advising companies, organizations, institutions, and government follow these best practices, I’m saying that we should begin working to establish a standard, and craft policy that requires everyone to participate. I know many tech folks don’t like the idea of regulation, but for certain industries, and certain platforms, it might be a positive thing to inject some regulation into the equation, and begin doing things out in the open, rather than behind the curtain. If you have any questions or comments on this blueprint, feel free to email me at [email protected], or submit an issue on the GitHub repository for this blueprint.