Charity Majors+ Your Authors @mipsytipsy CTO @honeycombio; co-wrote Database Reliability Engineering; loves whiskey, rainbows. I test in production and so do you. 🌈🖤Black Lives Matter🖤 Oct. 09, 2019 2 min read + Your Authors

Thread of the month. Go read it. I'll wait.

In 2016 when we started working on honeycomb, observability was, to the software industry, just an obscure synonym for telemetry.

If you had heard it before, you probably knew someone on the internal tools team at Twitter.

Over the past 3.5 years, we have done a *ton* of work to flesh out that definition, to propose a technical spec that is useful, generalizable, and effective for complex systems observability, and to explain why earlier solutions had stopped working.

This has resonated, especially with people who are working on newer architectures, where everything is a series of high cardinality, high dimensionality problems.

Which is why all the vendors in monitoring, logging and APM categories are glomming on to our messaging.

They know that YOU know you need observability. So they plaster it all over their website and marketing collateral.

But this is o11ywashing, plain and simple. And it makes me angry.

Defining observability as "metrics, logs and traces" is saying that observability is still nothing but a generic synonym for telemetry. It's saying the last 3.5 years didn't happen.

They're selling you metrics, logs and traces because those are the shitty products they happen to have. They haven't changed a goddamn thing.

They are selling you day old bread after they popped it in the microwave and slapped a BAKERY label on it. Be insulted.

Ben calls them spans, we call them events, but we are saying the same thing)

I would argue that we are technically more accurate because spans are narrowly defined within a tracing context, whereas I see traces as just one way of visualizing events; but this is only a linguistic distinction.

Honeycomb is inches away from lightstep, philosophically speaking. You might think we would be bitter rivals, but nah.

We're actually allies against the bad old ways of doing things, of trying to understand our systems with metrics, dashboards and hope.

Besides the billions of dollars in DataRelicSplunkFXDog capital and armies of marketers churning out crappy content, we face two main obstacles: inertia and cynicism.

Lots of people are so beat down, they don't actually believe their lives can be any better. 💔

Competition is good for users. But that's not what they are doing. Using their position to bait and switch users will set the entire industry back at a time we can ill afford it.

Just fix your fucking products, guys. I even told you how.  https://www.honeycomb.io/blog/so-you-want-to-build-an-observability-tool/ 


You can follow @mipsytipsy.



Bookmark

____
Tip: mention @threader_app on a Twitter thread with the keyword “compile” to get a link to it.

Enjoy Threader? Sign up.

Since you’re here...

... we’re asking visitors like you to make a contribution to support this independent project. In these uncertain times, access to information is vital. Threader gets 1,000,000+ visits a month and our iOS Twitter client was featured as an App of the Day by Apple. Your financial support will help two developers to keep working on this app. Everyone’s contribution, big or small, is so valuable. Support Threader by becoming premium or by donating on PayPal. Thank you.


Follow Threader