There's more of us it seems! I've been in HFT for the past 15 years and after reading Flash Boys I had to do a full 180 on Lewis. Every single book from him I now treat as mainly fiction sprinkled with some historical facts to maybe make it more relatable.
Yup, I'm really surprised by people defending Lewis. Don't get me wrong, I own most of his books but his 60 minutes interview about SBF was a cringe fest.
Arch is weird. I consider myself a Linux power user, including kernel hacking but no way I'm going to manually set up partitions and type gazillions of commands to set my system up.
This is off putting, I have no idea what's wrong with an installer. WTF Arch?
I watched the 60 minutes with Lewis who used to be one of my favourite writers but I had to do a 180 after I saw him defending SBF for the whole time. It was hard to watch and made me question the accuracy of every single book of his that I had read.
This is exactly the problematic attitude. I'm an OpenJDK author but not linux kernel contributor for mainly this reason. There are many people like me.
> majority of contributors to Linux, an extremely successful free software project, prefer the email-based workflow
This reminds me the many absurd conversations I had from my time in Goldman Sachs, few years ago. People with 15+ year tenures claiming Slang is the best answer to any- and everything. They just didn't know any better and stubbornly stuck to tooling and mindset straight out of 1995.
This is like saying that pyramids were constructed without modern cranes and bulldozers so obviously it's the right way to build stuff today. It's clearly not true but there are many power-wielding individuals in these projects who prefer to browse the web with lynx or develop in emacs and will impose this on anyone that would like to participate.
Maybe so, but my interpretation is that there's personal opposition to change and improvement from people like Linus or TGLX or other greybeards. At least in the linux kernel world. The contribution by email is an antiquated approach and needs to go.
Now, knowing the HN crowd this comment will become grey in 3... 2... 1...
I wish this was mentioned more often. Air fryers are tiny compared to ovens. In my case where I typically cook a meal for two days for a family of four an air fryer is utterly useless.
It's Google's own fault that anything new they launch these days has a meme status. I don't see anything wrong in pointing this out.
Similarly, many people avoid watching Netflix shows until they survive to season 3 at least because it's not worth investing your time in something that's likely to get cancelled after a season or two.
It's their own doing and mocking it is fully deserved
Yeah, GP has no idea what they are talking about. They replied something similar on another thread without ever providing evidence for their claims. They even claimed their RC implementation in Haskell is faster than hand-crafted C++.
> RC may turn reads into writes, but of course, GC ends up having to go through literally every piece of memory ever from bottom to top once in a while.
> RC limits itself to modifying only relevant objects, whereas GC reads all objects in a super cache-unfriendly way. Yes, an atomic read-modify-write is worse than a read, but it's not worse than a linked-list traversal of all of memory all the time.
This is patently untrue. Contemporary GCs have had card marking/scanning for 10+ years now.