Low Code Versus Developer Freedom for Data Visualization

• Read original article here

Properly democratizing the management and automation of company data is crucial to core operations as well as the digital projects or experiences they deliver. But powering data pipelines is only half the equation — you need to prepare for what’s next, and that’s where data insights come into play. Integrated insights allow for your in-situ data to be used for decision-making, user and market insights, as well as process and operational analytics — all via a single source of truth (point of access), even when data is spread across disparate systems. Streaming this capability in real-time (e.g. WebSockets) makes the company more adaptive, fostering market-driven innovations, predictive maintenance, event-driven alerting, and faster time-to-customer value.

The surge of no-code/low-code tools has helped make it easier for any business user (technical or other otherwise) to build data visualization dashboards. That said, many of those tools are proprietary, closed-source, and/or rigid — with limited extensibility options. This buckets them into one of two categories:

Stop-Gap: Simple “band-aid” tools that get the job done and serve as an easy way to gather general insights… but don’t scale and have low-hanging feature ceilings that limit their lifespan and utility.

Complex: Feature-rich and highly configurable, but at the expense of supporting less technical business users. These are better long-term solutions… but do not properly democratize data visualization.

The key to getting the best of both worlds is a platform with a simple and intuitive UX/UI that can be leveraged by all users, but that also embraces extensibility by design, ideally, one that is also open source (the ultimate escape hatch).

Without an API, your platform is an island unto itself. Beyond simple event-based webhooks, APIs are the glue that binds all a business’s various systems and services. Whether you’re powering a SaaS, digital experiences, applications, internal tools, or even a fleet of IoT devices… the API is what enables connecting, distributing, importing, and ingesting the underlying data. It’s a crucial part of any software that exists in a modern company’s data fabric.

Beyond simply having an API, it’s important to understand that there are different API specifications that have implications for your company’s tech stack. GraphQL is a new and robust option, but REST is still powering the majority of the market. Instead of betting on one or the other, it is important to build flexibility into your foundational layer by supporting both (as Directus does, natively).

That I/O is also relevant to the two questions above. To centralize data and insights, you need to properly connect, ingest, and sync across those disparate data systems using an API. And for complete extensibility, you need to have underlying API tools that enable programmatically broadcasting aggregate and grouped data to external tools or custom experiences.

Directus can automatically wrap any #SQL database with a GraphQL and REST API for developers. How has Directus made this easier? Sorry, but I can’t use any marketing-like copy here, I’m afraid. Please just explain in a straightforward way.

Directus has full database abstraction that allows it to layer on top of any SQL vendor and introspection that allows it to work with any data architecture. This unopinionated approach allows it to instantly conform to your project requirements or company’s tech stack, and deliver auto-documented APIs you can start using in seconds.

The alternative is to build a custom API for your needs (which can take months or years) or adopt a more opinionated data platform that may not support your current or future API specification or database vendor requirements.

InfluxData writes that its tools can “expedite device-to-cloud data transfers so developers can get centralized insights from their time series data in real time. The capabilities introduce the fastest way for developers to get time series data from third-party brokers into InfluxDB Cloud without additional software or new code.”

In this context, how might this work with a platform such as Directus for data visualization? And with a Grafana panel? (If this does not apply to you, feel free to ignore this question.)

There are many potential integrations and a good Venn diagram of capability overlap, but I’ll summarize all this by saying that real-time data is just one of the many data types that Directus supports. When your project needs to go beyond temporal data to include geospatial data, raw data, content authoring, or even file assets… Directus enables a suite of tools to fill in the gaps of these more domain-specific tools.