Thread by @copyconstruct

These days when I hear about yet-another-monitoring-company planning on integrating tracing/high-cardinality events into their product, part of me is glad the balkanization of o11y signals/viz is ending, but another part of me can’t help think this is a form of cargo-culting.

I’ve been - and continue to remain - sceptical of the dream of a “single pane of glass”.

That said I also think iterative drill-down and exploration across various axes, dimensions and systems is going to be critical to successfully troubleshoot issues.

Put differently, it will take an enormous amount of careful thought to design a system that will layer these different signals in a manner optimized for easy exploration and deduction.

That’s what the end user cares about the most and not so much about exposition formats, push vs pull, read time vs write time aggregation etc.

Sure, these factors influence everything the user *does* care about - but only tangentially.

There needs to be thought put into how the system’s topology at the time of debugging can be can be abstracted in a way that helps forming and validating hypothesis, given one’s mental model of the system.

The challenge isn’t so much having a single system for logs, metrics and traces; it’s about these higher level abstractions and visualizations.

That’s where there’s potential for real innovation - since that’s the true value prop to a customer.

Companies that are failing to succeed on this front when they’re dealing with a *single* signal (say, metrics) can’t possibly think they can succeed if only they also supported tracing as well. 🙄

This is a hard problem that can’t be solved by “unifying” the “three pillars”.

Don’t get me wrong - there’s *tremendous* value in being able to correlate along different dimensions.

What I’m calling for is more thought and innovation to happen on this front - there’s a huge opportunity to push the envelope forward.

But that’s not what I’m seeing happen.