Looking for books & other references mentioned in this video? Check out the video description for all the links! Want early access to videos & exclusive perks? Join our channel membership today: ruclips.net/channel/UCs_tLP3AiwYKwdUHpltJPuAjoin Question for you: What’s your biggest takeaway from this video? Let us know in the comments! ⬇
Martin is an excellent presenter and this talk is probably the best OpenTelemetry 101 that I've seen so far. It was a no-brainer for us to feature this talk the latest 💥Tech Talks Weekly newsletter issue 🎉
This was one of the most fun talk I have listened to. Very informational with just the right bit of humor here and there. I laughed throughout the session and learned a lot. Started following Martin on LinkedIn
My servers are failing to sync time right now .... due to firewall rule limits .... should really spin up a local server for them lol Thanks for reminding me !
@@vnzinki I'm reviewing frontend observability products like umami, posthog also OTel based products as well but which one will be the right fit for RUM
Browser based telemetry is still in its infancy with Open Source/standards based libraries. OpenTelemetry is still trying to work out what it needs to be for the frontend. SPAs need tracing and they need CWV. Just remember that you don't need Metrics (time-series aggergates) to build graphs, and you can (and should) use the collector which can convert the raw data into Metrics if you know which graph you want to build.
Shows "Better Documentation" as coming soon. Then asks what can we do to make it easier to adopt.. Quit making Open Source so easily Silo-able into rockstars, make it easier to implement by having a documentation-first approach.
Better documentation is part of it, not the whole thing. We talk to varying types of people from the Otel side, and as a vendor we hear that struggle from prospects and customers. The idea that its plug and play is the problem in my opinion. We had that as part of the APM world and we found ourselves wanting for more. Its work to achieve good observability. Could it be easier than it is right now? Sure... should it be completely automated behind an agent, no. Documentation just isn't all of it, usage, availability, integration, those are all major things that contribute to the wider adoption issue, which is what we need data on.
I’m surprised that he claims that most people don’t sync the clocks on their servers. Is that really true? I have never seen a server where clocks weren’t synced.
@kkklllaaa1234 to be clear I'm talking about whether people check to ensure that they're in sync. In cloud services it should happen automatically, and that's the assumption, but rarely checked and proven to be true. I've hit numerous situations where they've drifted over time in EC2 instances because someone blocked ports, etc.
Whimsy is cultural, and specific. We should do this less. We don't name our servers or software after obscure Dr who characters anymore. We exclude the majority of the planet with this behaviour. You want standards? You want the "young uns" to listen to you? Use plain English.
Take that as the joke it was meant to be, a dig at a past era where tools were named in a fun way. We can keep our sense of fun without excluding, it is harder though. everything doesn't need to be corporate named though.
Looking for books & other references mentioned in this video?
Check out the video description for all the links!
Want early access to videos & exclusive perks?
Join our channel membership today: ruclips.net/channel/UCs_tLP3AiwYKwdUHpltJPuAjoin
Question for you: What’s your biggest takeaway from this video? Let us know in the comments! ⬇
Martin is an excellent presenter and this talk is probably the best OpenTelemetry 101 that I've seen so far.
It was a no-brainer for us to feature this talk the latest 💥Tech Talks Weekly newsletter issue 🎉
look at this presentantion to undesrtand how much text should be on the slides and how much to show at one time, dude is genious
This was one of the most fun talk I have listened to. Very informational with just the right bit of humor here and there. I laughed throughout the session and learned a lot. Started following Martin on LinkedIn
Thanks
Thank you very much @juliojordan. This is much appreciated! ⭐️
Great presenter
I am more a viewmodel kind of guy
@@avwie132 I'm not a fan, I prefer ports and adapters
Excellent presentation.
Does anyone perhaps know where I could find Martin's talk on the OpenTelemetry collector that he mentioned at 38:00? thank you
I've had a look and I can't actually find a recorded version of that talk.
I do have numerous blogs on our website though.
@@DotNetMartin Ah alright. Sounds good 👍
My servers are failing to sync time right now .... due to firewall rule limits .... should really spin up a local server for them lol
Thanks for reminding me !
Thanks.
it good
28:38 I wish there was auto instrumentation of METRICS for angular. It seems all the auto instrumentation for JavaScript is geared towards traces
you can try Grafana Faro for SPA and otel SDK for SSR
@@vnzinki I'm reviewing frontend observability products like umami, posthog also OTel based products as well but which one will be the right fit for RUM
Browser based telemetry is still in its infancy with Open Source/standards based libraries.
OpenTelemetry is still trying to work out what it needs to be for the frontend. SPAs need tracing and they need CWV.
Just remember that you don't need Metrics (time-series aggergates) to build graphs, and you can (and should) use the collector which can convert the raw data into Metrics if you know which graph you want to build.
@@martinthwaites1378 so where are the examples to do that? I am new to JavaScript. The backend written in Java has been a much better experience
Will there be otel libraries for clojure?
Prerequisite to talk about server: You must have a LONG BIG ROBUST STABLE RESILIENT beard.
Judging by the grey coming in, I'm worried about the "robust" side
nice to see Ramzan Kadyrov getting off of tiktok
Shows "Better Documentation" as coming soon. Then asks what can we do to make it easier to adopt.. Quit making Open Source so easily Silo-able into rockstars, make it easier to implement by having a documentation-first approach.
Better documentation is part of it, not the whole thing. We talk to varying types of people from the Otel side, and as a vendor we hear that struggle from prospects and customers.
The idea that its plug and play is the problem in my opinion. We had that as part of the APM world and we found ourselves wanting for more. Its work to achieve good observability. Could it be easier than it is right now? Sure... should it be completely automated behind an agent, no.
Documentation just isn't all of it, usage, availability, integration, those are all major things that contribute to the wider adoption issue, which is what we need data on.
I’m surprised that he claims that most people don’t sync the clocks on their servers. Is that really true? I have never seen a server where clocks weren’t synced.
@kkklllaaa1234 to be clear I'm talking about whether people check to ensure that they're in sync. In cloud services it should happen automatically, and that's the assumption, but rarely checked and proven to be true.
I've hit numerous situations where they've drifted over time in EC2 instances because someone blocked ports, etc.
Whimsy is cultural, and specific. We should do this less. We don't name our servers or software after obscure Dr who characters anymore. We exclude the majority of the planet with this behaviour. You want standards? You want the "young uns" to listen to you? Use plain English.
Take that as the joke it was meant to be, a dig at a past era where tools were named in a fun way.
We can keep our sense of fun without excluding, it is harder though.
everything doesn't need to be corporate named though.
OTEL is a complicated piece of bad taste.