The Weekly Dev - 202223
The weight of Experience
Sometimes it looks like Testing has already become a new religion, but still apparently, the industry does not learn enough from the experiences.
Out of the systems that we build, it should be possible to see clearly what works, what could be done better, and what is clearly unsustainable.
A lot of technologies are borderline: they do make sense at the scale of well-known $privacymonster and $metadistopy, but in some context the same technology might not apply, might not solve any real issue, or simply be not sustainable.
We cannot simply buy a model for our needs: we need to always make a model ourselves; the technology can be bought or built, but only when the need is clear, never because it's the current hype.
You can buy a T-shirt because it's cool, not a cluster or a database license, this way you will end up impacting other people lives, and possibly not for the better, sorry.
Of course, it feels terribly uncool to type in an XTerm, but it could end up being much better than using a text editor that makes you join a botnet...
Enterprise Computing
An extensive article about the Java memory model. Interesting from various perspectives, and not only for Java nuts
Close Encounters of The Java Memory Model Kind
Source: shipilev.net
Data-Intensive Applications
The following is an article that generated a lot of productive discussions and interesting counter-arguments. So, it highly depends on what you're up to, but still, I think it has some strong point.
Why you should never use sticky sessions
Source: dev.to/gkoniaris
Boring Programming
Some drop of wisdom about code review findings.
Learnings from 5 years of tech startup code audits
Source: kenkantzer.com
A seemingly contrarian article about strong typing, with actually a lot of insight into the Rust way, why it is good, and why it is a pain.
Source: fasterthanli.me
[privacy] [programming] [java]