First release of SatPulse was focused on a fairly narrow use case: transferring time from GNSS to a PTP Hardware Clock, which required some pretty niche hardware. It's now moved far beyond that, and is turning into a fairly general purpose tool for doing precision timing and positioning with GNSS. I just created a new doc satpulse.net/recent-changes.… that has an list of all the new features implemented since the first release. It's a lot! The website has fallen behind, but the man pages section is up to date.
It would have been completely impossible to implement this amount of stuff in this time without AI help. SatPulse started off a several years ago in the pre-AI era. As coding agents have got better, I have been making increasing use of them. But SatPulse is definitely not AI slop. I care deeply about code quality. Coding agents can help not just improve velocity but also improve quality. But it takes sustained effort, because the default is slop. So that's another thing I am going to talk about.
It's very hard to tell the difference between slop AI and legitimate AI. For an example, one can create a SKILL file by just prompting, while another can create it by iteratively growing it to be minimal yet effective. Both will look the same.
Future desktop software should have a built-in MCP interface. UI should reflect the actions taken via MCP. Eg: select-address tool will select a UI element in the address book app.
I'm super hyped about joining @lambda_conf in a couple of weeks! Can't wait to talk about @ballerinalang with all the programming language enthusiasts there.
'Introducing Ballerina: A Pragmatic Programming Language Designed for Cloud-Native Apps' by @sameerajayasoma at LambdaConf. Explore how Ballerina simplifies cloud-native app development with its unique features.
Sign up ➡️ buff.ly/3wbaG7U#CloudNative#BallerinaLang
The latest update to the Ballerina (@ballerinalang) VSCode extension now allows for XML files to be seamlessly converted into records.🎉🎉 #ballerinalang#vscode
The @ballerinalang programming language works hard to ensure that error management isn't something you do as an afterthought, when production outages force you to--it's something you do as part of everyday software engineering. 👇
stackoverflow.com/collective…
The #BallerinaLang#compiler involves many #algorithm, including inferring numeric literal types. Check out my demo built with the help of OpenAI & Copilot to see how it works 👉 bal.tips/docs/expressions/li…. Let me know if you implement this in Ballerina. I'd love to see your code!
ALT A value written in a numeric literal always represents a specific type, which is determined by the literal itself. The type of a literal can be one of the basic types, such as int, float, or decimal.
When determining the basic type of a literal, this 3 steps algorithm is used.
Here, I am visualizing the application design just from the code in #VSCode, and this image never goes out of sync with the code. How cool is that? 🙂
This is the Google cloud microservices demo.
My internship project at WSO2 is to implement a WebAssembly backend for the @ballerinalang native compiler nBallerina. Here is a blog post with details on initial implementation
poorna.info/posts/wasm-backe…
Ever wished that creating and combining network services were a little more ... effortless?
The leap into Ballerina is COMING SOON! 💫
#sliit#sliitfoss#foss#WSO2#ballerina@wso2
I will be presenting some practical aspects of adopting SRE for your organization & how Kubernetes can help in that journey, based on my experience in building the SRE practice at WSO2. The WSO2 SRE team runs the @ChoreoDev & @asgardeo clouds. Join me @ 2nd Feb 2022, 5:30PM IST.
Ever had the question when SREs and Kubernetes are worth it and when they aren’t?
Mr. @afkham_azeez, Vice President & Deputy CTO-Cloud at WSO2 will join us to host a webinar on SREs and Kubernetes. Get ready to grasp knowledge
Stand a chance to win amazing swags from @wso2