OTC CatchUp #170 Summary

He told us about an Okta access token issue that he solved by implementing a cache for tokens and a quarantined area for revoked tokens.

He said that he made the mistake of starting to code as soon as he got the requirements of his task, without properly understanding the context of the job and without consolidating all the requirements into a cohesive plan. He advised everyone to collect all requirements and plan properly, to avoid repetitive and unnecessary work.

He also told us about their Quality Assurance (QA) process, which involves Smoke Testing, Unit Testing and more. They also use a platform called SonarQube for linting and Static Code Analysis.

Log4j got a mention as well and we’ve talked about the major security incident that affected loads of projects using the library in OTC CatchUp #70 and #58.

To combat this, Shubham suggested that legitimate Open Source contributions really help, because a lot of experience is gained by setting up complicated project infrastructure and adding a fix into a codebase. This experience can then also be shown in interviews, as a qualifying metric.

A New Attempt At Async Device Shutdown Support For Linux

Introducing Sudo for Windows!

Breaking Bitlocker - Bypassing the Windows Disk Encryption

Serverside Dart

Serverside Dart #2

I hate twitter

How to be Happy *\*Tutorial\*\*

Happiness

Interview with Web3 Founders in 2023

He was asked about what happens when one types 'google.com' in the browser.

Dheeraj’s college projects helped him a lot, as he was able to talk about technical experience and a variety of things like HTTP Live Streaming (HLS) that he was able to explore because of those projects.

In this age when AI is able to generate data and art because of being trained on publicly available data, there is a question of copyright and license infringement that is very fair to ask.

There was discussion that we don’t have to pay to listen to music that might inspire us to create our own music, so even AI shouldn’t have to pay for things it uses to train itself to be able to produce art or other data. Well, the counter to that is that even if we don’t pay for the music we stream, the music service that we’re using counts every stream and pays the artist for those streams. So the original owner is getting paid. With AI, we don’t know the method in which data is collected. Also, a lot of general licenses prohibit mass usage of data or commercial usage of data, so special attention needs to be paid to copyright and license rules, so that the correct people and artists get paid.

Dheeraj Lalwani suggested watching a TV series The Playlist, which explains all of this from each perspective (the artist, the law, the industry, the coders, etc.).