HN user

odyslam

39 karma
Posts6
Comments19
View on HN

I used extensively the theta mention in a forecasting course and it's amazing how good it is, even with default parameters.

ML techniques have negligible improvement, considering the computational overhead that they require.

Hey OP,

I have made a similar setup. One question that I have not yet answered and thus I am changing setup is how secure is to "give" your IP to everyone on the internet.

My assumption is that your home-router has security vuln that we don't know about and could be bypassed by hackers who are willing to mess with you. Even though you may have taken precautions, such as opening only very specific ports, etc.

I hear you, we know that our audience is sensitive to their privacy. We are all are.

Here are a couple of thoughts that have guided us. Thank you for engaging in this conversation and for caring enough.

1) This data is crucial for us. We need as much as we can get and it's highly specialized to Netdata. (e.g a sudden increase in crashes will prompt our team to see recent changes) 2) The more friction we add (opt-in), the less people will do it (because people choose the easier route, always) and thus we will have less data to work with. 3) People who care enough, as you said, about their privacy, can *very* easily disable the anonymous statistics, by both adding a flag to the install script or doing a small config change afterwards. I feel that we are communicating in many different places that we take anonymous data, so most of our users should be informed. 4) It's a fairly standard industry tactic and I don't believe that other solutions are not doing. Of course this is not an excuse for anything, just noting that we are not an outlier.

Thanks again for engaging. Feedback is great for us, it makes us both happy (because someone cares enough) and better.

That's a bit unfair. In the docs we are being very upfront that you can opt-out of anonymous telemetry: https://learn.netdata.cloud/docs/get

we use the data we gather in order to make smarter product decisions. We want to invest resources where it matters, so we need to know how our users use the product.

We are also very detailed on what we gather: https://learn.netdata.cloud/docs/agent/netdata-security

Lastly, we just changed our analytics engine, from google-analytics to a self-hosted posthog, which is an open-source product analytics platform

We do love ebpf. Guilty as charged ¯\_(ツ)_/¯

We have a whole bunch of metrics that we keep track and we are currently implementing a load more.

Soonish, we will greatly increase the number of metrics that we gather with ebpf. That coupled with our per-second granularity, should give you a very detailed view of the system.

Docs: https://learn.netdata.cloud/docs/agent/collectors/ebpf.plugi... Community Forums discussion: https://community.netdata.cloud/t/linux-kernel-insights-with...

Actually, we give a lot of thought in defining sane default alarms for most of the data sources that we have.

We want our users to get 80% of the value with 20% of the effort.

It's an opinionated approach that liberates a lot of users from having to setup and maintain everything.