Cloud is backend. For the lion’s share of what it does, cloud computing and the web exists as a backend service layer delivered by a cloud services provider hyperscaler from a datacenter, down a connectivity pipe… and onward to our desktops and mobile devices. Ultimately we interact with cloud-based web applications at the frontend, through a graphical user interface and increasingly also via voice, or perhaps even through touch or some combination of all of the above.
Monitoring these services to make sure they perform is the work of observability platform specialists, of which there are several.
Working at both ends of the observability spectrum but now with a renewed focus on frontend (i.e. user level) observability issue is Honeycomb. The company’s extremely logically named Honeycomb for Frontend Observability product offers debugging capabilities aligned to the frontend of applications. It is designed to enable software engineering teams to gain insights into web application performance and user experiences.
According to Honeycomb CEO Christine Yen, as websites [and cloud services] become more interactive, the technologies behind them are becoming just as complex as what’s on the backend.
RUM Punch, Too Weak
“Traditional frontend monitoring tools, like real user monitoring [often known as RUM], are not designed to make the rich context around every user interaction available in real-time and haven’t been able to keep up with that rapidly expanding complexity. Honeycomb for Frontend Observability removes the limitations of legacy RUM products through OpenTelemetry-based instrumentation and support for hundreds of custom attributes at no cost. These attributes are immediately accessible in powerful visualizations to surface new insights into customer and performance issues,” said Yen.
She insists that the frontend element of the application stack is critical. Why? Because it’s where users spend their time and where revenue is generated. The Honeycomb team claim that there’s a lot of outdated frontend observability tools out there that only offer aggregated metrics and limited insights.
What Honeycomb is trying to do is to provide a more holistic view of an entire application – from the cloud datacenter server, through the networking and interconnectivity layer (not necessarily a core competency for Honeycomb, but we can infer some relevance here when the company uses the term holistic) and then onward to the browser and the user themselves.
Fender, No Bender
“E-commerce is critical to Fender,” said Michael J Garski, director of software engineering at Fender Instruments. “Honeycomb for Frontend Observability removes the guesswork from diagnosing our site’s performance issues by tracking precise page speeds, filtering sessions and identifying the cause of speed spikes, enabling targeted site optimization for maximizing conversions and delivering better customer experiences.”
Honeycomb says its new software tools integrate with its existing observability platform to provide a unified interface for debugging across the entire application stack. The Honeycomb’s OpenTelemetry wrapper is meant to simplify frontend instrumentation, collecting performance data and allowing integration of custom attributes (elements of software engineering that describe the standard properties of data assets, such as name and description) so that software teams can use end-to-end tracing for efficient cross-team issue resolution.
Core Web Vitals
There’s also comprehensive user interaction tracking and context capture to allow software teams to query high-cardinality data, enabling faster problem-solving and better cross-team collaboration. By capturing attribution data for core web vitals, developers gain precise insights into the causes of poor performance scores and clear guidance on how to improve them.
As we move forward with the still-evolving world of web, cloud and all manner of virtualized technologies delivered through the ether, gaining a hold on the frontend user experience is of course fundamental to our management of backend architecture.
While we might need to ask some organizations why they chase their company name, Honeycomb is obviously named to convey an impression of complex interconnected elements that ultimately deliver strength and rigidity. With all that branding, this lot hardly even needs any marketing buzzwords.
