Random interesting stuff
Bray Falls posted his beautiful image of the 2017 solar eclipse.
Martin Casado shared his list of most usable metrics for companies. The whole thread is worth reading.
In-depth article about reverse engineering the Tesla Firmware Update Process.
In-depth case study of rewriting their “Read State” service from Go to Rust. The issues with Go implementation were primarily coming from memory management and garbage collection (GC). This is not “rewrite everything to Rust” type of article, it is more “use more appropriate tool for the problem”.
The main goal behind
jlrsis to provide a simple and safe interface to the Julia C API. Using this crate you can call arbitrary Julia code from Rust, including your own, and share data between the two languages.
Overview of techniques / libraries for writing async Rust on embedded systems.
An alternative service to crates.io written in Rust.
After working in Go for more than half a decade, I’m starting to think that it is probably a better idea to impact developer velocity and force them to write software that is more correct. Go works if you are careful about how you handle it. It however amounts to a giant list of rules that you just have to know (like maps not being threadsafe) and a lot of those rules come from battle rather than from the development process.
Jimmy Angelakos at FOSDEM 2020 helping to navigate the rich but confusing field of (Full) Text Search in PostgreSQL.
4 hints to improve you Terraform scripts:
- Use Terraform Modules to abstract specific pieces of infrastructure into logical groupings.
- Use Terraform Data calls to provide information.
- Be Smart About Where Interpolation and Concatenation Happens.
- Implement State Locking for Ease of Deployments.
Overview of some required changes and non-obvious bugs you’ll see when using Helm3.
withdefinitely is fantastic, short article shows how one can leverage
withto get rid of entangled
casestatements and improve readability of the code.
SimCLR shows 7-10% improvement on ImageNet for self-supervised and semi-supervised learning.
Authors claims a significant improvement over the previous work:
When applied to the MNIST optical character recognition tasks, our approach achieved 99.25% accuracy which significantly outperforms state-of-the-art solutions and is close to the accuracy of the best non-private version.
Authors suggest use of Jacobi iteration to speed up standard feedforward computation through allowing for parallel iterations. They have seen between 1.2 and 33 speedup factors.