Daan Gerits

New Blog Series: Rethink Data | A new mindset for using data in your organization. Check out the first in a new series from @daangerits "We just can’t wait a day, or even an hour for that matter, for the insights we need to steer our organizations." synadia.com/blog/introduci…

Recording from today's @NATS_io office hours now available. Love these! Great open discussion with lots of interactive Q&A, awesomely hosted by @TheCodeGangsta from @Synadia. Agenda: > @DaanGerits: Next, NATS, Benthos and beyond! > NATS HTTP Gateway released for Synadia Cloud > Let's play with http gateway! > Blog post: Delegate with Trust > Introduction to OtterJet Watch: youtube.com/live/qOE7gIXjg… #iot #iiot #ai #ml #edge #edgecomputing #connectedthings #connectedcars #cncf #opensource #distributedcomputing #llm #devops #platformengineering #kafka #redis #rabbitmq #wasim #webassembly #multicloud

You are cordially invited to join @jbfletch_ and @matthiasjsax for the inaugural episode of The Duchess & The Doctor Show, where our distinguished hosts mix equal parts whimsy and know-how to concoct answers to your most urgent event streaming questions! youtu.be/aSeTAkYjeRM

Finally, we interviewed @Jeffail and @MihaiTodor about Benthos: an OSS project with the tagline "fancy stream processing made operationally mundane". /4

Had a great chat with @krisajenkins on Developer Voices youtube.com/watch?v=labzg-…, we basically solved every social issue and there is a bit about stream processing as well

NASA astronaut demonstrating the Dzhanibekov effect. An astronaut noticed this physical behavior of a handle, that turned out to be the proof of a theorem: the tennis racket theorem (also dubbed the Dzhanibekov effect) The Tennis Racket Theorem states that an object with three unique moments of inertia, rotation about the axis of intermediate moment of inertia is unstable, while rotation about the other two axis is stable. The effect is named after the Russian cosmonaut Vladimir Dzhanibekov who discovered this phenomenon while in space in 1985.

Ok, laten we met een GPT partij beginnen. We laten alle beslissingen nemen door chatGPT. Politiek gezien klopt het, kunt perfect met de vinger wijzen en zeggen dat “het door de computer kwam”. These people … seriously …

@fonskeVL Puerto Escondido -> Mexico City -> Amsterdam -> Brussel. Elke keer een paar uur in de luchthaven.

@daangerits goede trip gewenst ... maar er móet een kortere reisweg zijn 🤔

Preparing for a 36h trip back home. So much looking forward to being reunited with my family again after close to a month away. New friendships and ideas to get through the upcoming year!

@krisajenkins Interesting. Will have a look! On a similar note, what’s your take on warpstream.com ?

@daangerits Ooh, thanks for the tip - I'll put it on my research list. In a similar vein, have you heard of typestream.io?

I have no idea which language to tackle #AdventOfCode in this year. Haskell's always an option. Gleam, Unison and Zig are interesting new contenders. And Rust would probably do me good. Decisions, decisions... 🤷‍♂️

Artworks sync via browser's localStorage, super clever. Love it!

@peeterskris @apachekafka Don’t get me started. Turks & Caicos

@daangerits @apachekafka Where are you ? Btw, that seems more like a batch of water than a stream. Just sayin'

Honestly, there are worse locations to perform your data engineering job from. I love doing on-site stuff at places nobody expects to even know about streaming tech. #streaminglife @apachekafka

@MatthiasJSax @TheDanicaFine @kafkastreams You bring the sigars, I’ll bring the whiskey what could possibly go wrong #famouslastwords

@daangerits @TheDanicaFine @kafkastreams I would love to say: let’s do it! How hard can it be? — But well, I happen to know how much effort it is… maybe a good retirement project? But need to learn go first… #lifeGoals

To celebrate these recordings being live, starting tomorrow, I'll be counting down the top 10 sessions from #kafkaSummit as rated by the conference attendees! If you're looking for the best of the best content from this year, you'll want to bookmark this thread.

@MatthiasJSax @TheDanicaFine @kafkastreams If only it was written in go ;)

@TheDanicaFine @daangerits Not necessary. @kafkastreams is just the best! 😄

Happy Friday! The countdown continues... In 4th place of the top 10 #kafkaSummit sessions is @daangerits with his talk "The Possibilities And Pitfalls of Writing Your Own State Stores." Have a watch to level up your #kafkaStreams apps ⤵️ confluent.io/events/kafka-s…

Techies around the world, I need your help! What are some good public documentation tools, not just a read the docs, but a slick experience you are attracted to? #devrel

The recordings for #kafkaSummit London are live! Now you can re-live your favorite sessions or experience the great #apacheKafka content you may have missed. Check it out! 👀 confluent.io/events/kafka-s…

Powered by The Log © "Genius is making complex ideas simple, not making simple ideas complex" The log is a fundamental data structure in many distributed systems. You can't fully understand databases, NoSQL/KV stores, replication, consensus algorithms, version control, Kafka, or almost any other software system without understanding logs. 🙅‍♂️ Why is that? Its quick disk persistence makes it ideal for durably capturing the state prior to actually doing it. This ensures consistency and avoids “ghost reads” in failure scenarios. It’s used widely in databases but is even more critical for distributed systems because it's a good tool to solve a hard distributed systems problem - ordering. Logs simply record what happened and when. ✍️ Such consistency can be a problem in distributed systems. But logs are very simple to keep consistent, as they are immutable! You only have to agree on what order things were appended to it. You can reduce the problem of: 😥 - making multiple machines all do the same thing to the problem of 👍 - feeding their processes the same distributed consistent log. It’s no coincidence that a log is at the heart of many distributed consensus algorithms like Zab (ZooKeeper) and Raft (Kafka). It is a great way to represent a series of events. ⭐️ Any state in any system is the result of the series of events that happened to it in its precise order. 💰 e.g. double entry accounting - give me an ordered set of debit/credit transactions (a log) and I can compute your current bank balance (state). Given its immutability, if you give two machines the same log, they should deterministically reach the same state. State here is nothing more than a particular offset in the log! With an active-passive model, it becomes trivial to keep this log consistent - whatever the leader wrote is the truth. Since the log is so universally useful, the moonshot idea then becomes to package this incredibly elegant data structure into a universal integration mechanism for systems within companies. 🤯 A company’s tech stack is nothing but an even larger distributed system - all the systems want to be consistent with each other. To make that happen, you need to package the log in a fast, cheap, and scalable solution to make it practical to use at scale… and that’s what Apache Kafka is :)