Detailed Plan

Context:

There’s currently no fully open source API Layer on top of Cardano. Meaning your options are to build it yourself or look at fee-paying vendor solutions which are not open-source.

This is source of huge friction for those who want to create new data-driven value streams.

We want to remedy this by implementing a first class end-to-end API solution and open it to everyone. You’ll be able to consume our API infrastructure or run your own if you prefer as our whole stack will be open-source.

In addition to raw on-chain data API, we will also make available new business intelligence analytic-driven API to bring added value to SPOs, investors and businesses alike in order to enable new data-driven services on Cardano.

Specifically for small SPOs, for whom it is so difficult to differentiate themselves from larger pools, we want to provide the kind of data-driven insights that will allow them to better attract potential investors and to keep existing ones engaged.

Lastly, we will create reusable front-end & backend components, which by consuming our APIs, will offer service providers the ability to improve the user-experience of their customers or to create new customer journeys altogether.

Plan:

There are 5 EPICS in this project:

1. API definition: in this phase we look at defining the API contracts and express them in the relevant standard (OpenAPI for REST API, AsyncAPI for STREAMING API)

2. API implementation: based on onion-architecture principles and micro-services

3. API management: from Identity, Authentication, Subscription and Monitoring all key components which allow managing the API lifecycle in a secure and industrialized fashion.

4. API Catalogue, Data-Insight visualization: https://cardanobi.io will be our portal for API subscription, API catalogue, API documentation as well as key data-insight dashboard for each of our pre-define customer profiles (SPO & Investors initially).

5. Reusable components: these will be standalone frontend widgets or backend CLI, built on top of our APIs to enable new user experiences/interactions (e.g. components for SPOs to leverage important business events like a new block being forged, a new address delegating to their pool, a new ATH for the max number of ADA processed by their pool, etc…).

Roadmap:

From the above 5 EPICS, we will break them down into features and look at optimizing the sequencing such that we can deliver value incrementally and get client feedback quickly.

EPIC #1: API definition

• Feature 1: REST API following db-sync data model
  Prepare swagger files.
• Feature 2: New value added REST API
  Prepare swagger files.
• Feature 3: New value added ASYNC API
  Prepare avro schemas.

EPIC #2: API implementation

• Feature 1: Infrastructure
  Provisioning of necessary cloud based resources using Infrastructure As Code (IaC).
  Leveraging load-balancing where necessary (db-sync, app servers)
• Feature 2: db-sync
  Install db-sync (https://github.com/input-output-hk/cardano-db-sync) in a load-balanced pattern.
• Feature 3: Micro-services
  Build services following Service Oriented Architecture (SOA) principles, services topology yet to be determined but certainly will follow entity domains as per db-sync data model enriched with new value added entities.
• Feature 4: Resilience + monitoring + NFRs
  Implement resource monitoring through cloud service or adhoc solution (TBD)
  Make sure above components are fault tolerant through load-balancing and can resume from failure.

EPIC #3: API management

• Feature 1: API gateway
  Leverage cloud provider API gateway.
  Configuration, link to authentication server, setup Oauth2, API Keys to manage incoming traffic (north/south)
• Feature 2: Microgateways
  Leverage open source microgateways (nginx).
  Manage traffic among microservices.

EPIC #4: API catalogue, data-insight portal

• Feature 1: Catalogue
  Auto-generate API documentation from source code using relevant tool (eg. redoc)
• Feature 2: Portal front-end
  React native app consisting of landing page, API key generation page, API documentation pages, User subscription page (to access specific user profile data insights), data-insights page to present both key on-chain data analytics and new added-value data analytics (user will have to create an account to access profile specific data analytics (e.g. SPO, investors))
• Feature 3: Portal back-end
  Nodejs backend to fulfil certain interaction end-point and backend function (e.g. API key generation and safe keeping in dedicated vault)

EPIC #5: Reusable components

• Feature 1: Front-end components
  React components + javascript code to retrieve relevant data points from APIs and present them in a modern widget.
  Scope to cover: epoch, slot, tip, time to epoch boundary, new block, transaction count, volume exchanged etc…
• Feature 2: Back-end components for SPOs
  CLI component allowing realtime pool data to be sent to Fron-end components.
  CLI component allowing customer channel interaction (twitter, telegram) when business events are triggered (new block, new delegator etc…)

Timeline / Success criteria:

Here we provide estimated completion time horizons of each feature identified with E(epic number)F(feature number) (e.g. E1F1 for FEATURE #1 of EPIC #1)

• 1 month:
  E1F1, E2F1, E2F2
  Outcome: the core infra is set up and core REST APIs are defined
• 3 months:
  E2F3, E3F1, E3F2
  Outcome: microservices framework and first instances deployed, gateway and microgateway in place. (this can be considered a first MVP where the API consumption can be tested end to end).
• 6 months:
  E1F3, E2F4, E4F1, E4F2, E4F3, E5F1, E5F2
  Outcome: new value added API have been defined, the services are resilient and follow relevant non-function requirements (NFRs), monitoring is in place, the portal is in place with its API catalogue. (this can be considered as the industrialized MVP enriched with the customer front-end)
• 12 months:
  Outcome: we’ve done the necessary social media marketing, IOG is helping as they see the great product that has been built, we have 10,000+ API queries a day.

Expected start date: February 2022

Expected public launch date: Summer 2022

Team (currently being staffed):

1. Lead, architect, backend/frontend dev: Thomas Langlois, https://www.linkedin.com/in/langloisthomas/
2. Data architect, backend/frontend, event-driven: Ahmed Abouelnour, https://www.linkedin.com/in/ahmed-abouelnour-91828b8/

3. Front end developer: TBD

4. UX developer: TBD

Cost breakdown:

• Cost per resource type:

• Lead/backend: $50/hour

• Data architect / cloud developer: $50/hour

• Front-end developer: $25/hour

• UX developer: $25/hour

• Estimated hourly contribution per resource type:

• Lead/backend: 336 hours (6 months x 14 hours per week)

• Data architect / cloud developer: 336 hours

• Front-end developer: 168 hours (3 months x 14 hours per week)

• UX developer: 84 hours (6 weeks x 14 hours per week)

• Total cost for staff: $39,900

• Total cost for servers for 1st year : 16 servers x $60 per month x 12 = $960

• Total $ request: $40860

Intellectual Property:

All software will be open-source and distributed under a to be defined OSI-approved license.