vault backup: 2024-06-18 17:57:37

content/education/omscs/cs-6200.md

@@ -8,7 +8,7 @@ My first class I took at Georgia Tech's OMSCS program was cs-6200, or Graduate I
 
 I loved the class. It's probably my favorite class I've taken in anything, ever. 
 
-In GIOS, the cool thing is that the projects are open-ended. All you have to do is pass a test suite they provided. For Project 1, I did it strictly by the book. For Project 3, I did something very weird and zany to the point where the Autograder actually got confused by my project, leading to a manual grading required. Project 4 was my least favorite: it uses gRPC and C++, and I ended up disliking both by the end of it.
+In GIOS, the cool thing is that the projects are open-ended. All you have to do is pass a test suite they provided. For Project 1, I did it strictly by the book. For Project 3, I did something very weird and zany to the point where the Autograder actually got confused by my project. Project 4 was my least favorite: it uses gRPC and C++, and I ended up disliking both by the end of it (mostly the C++ API of gRPC, I've used the protocol in other languages successfully).
 
 ### Notes I wrote because of this class
 

content/sports/nba/hoopers-i'll-remember/derrick-white.md

@@ -0,0 +1,7 @@
+---
+title: "Derrick White: Optimal Basketball player"
+tags:
+  - thoughts
+---
+Derrick White is literally me if I were a basketball player. We even have similar hairlines.
+

content/sports/nba/hoopers-i'll-remember/index.md

@@ -0,0 +1 @@
+There are only a limited amount of accolades available in basketball. Years from now, there will be kids (probably mine) that will be browsing basketball-reference and clicking on random players. There are players whose stats won't tell their game. I will make sure they are remembered.

content/sports/nba/hoopers-i'll-remember/jamal-murray.md

@@ -0,0 +1,20 @@
+---
+title: "Jamal Murray: The Big Games Merchant"
+tags:
+  - thoughts
+---
+Is Jamal Murray the greatest 0x All-Star in league history? I would say so. The man has rightfully earned the nickname of a "big games merchant". I have NEVER seen a man rise like this in the playoffs. I don't know if playoff risers are truly a real phenomenon, or some players just don't even bother to try in the regular season. Jamal Murray might be the exception.
+
+ In January of 2024, the Nuggets came to Boston where I witnessed Jamal hitting turnaround 20 foot jumpers with Jrue Holiday and Derrick White, the best guard defenders in the league. Like what?
+
+![[good-teams-man.png]]
+
+I witnessed this man [[jamal-murray-game-winner-2024|hit a game winner on the Lakers]] after going 3-15, and then he proceeded to hit another game winner on the Lakers 3 games later.
+
+Nikola Jokic is arguably one of the top 3 greatest offensive players of all-time and his excellence carries into the playoffs, but Jamal completes this team. The Nuggets went nowhere without his ridiculously tough shotmaking in the seasons following his ACL tear. The Jokic-Murray PnR is the stuff of nightmares.
+
+If the Nuggets win it all this year (and defeat my beloved Celtics, got forbid), he's gonna be in the Hall. His personal career accolades will read:
+
+> NBA All-Rookie Second Team (2017)
+
+I guarantee you kids will wonder why the Oldheads speak about him elusively, citing the lack of accolades. But we will never forget you, Jamal Murray.

content/sports/nba/hoopers-i'll-remember/james-harden.md

@@ -0,0 +1,16 @@
+---
+title: "James Harden: a half-hearted defense"
+tags:
+  - thoughts
+---
+Let me be clear: James Harden will be remembered. But I really don't think he's going to be remembered as he should. 
+
+James Harden, for three straight months in 2019, averaged **40** points a game. He did this before the offensive explosion of the 2020s. Does anyone realize how bonkers that is? I don't think we talk about it enough. For three seasons, he was pretty much the greatest offensive (regular season) player in NBA history. 
+
+I understand that blah blah blah playoffs, but I don't think he's a "playoff dropper". It's more apt to say he couldn't maintain his ridiculous efficiency from the regular season, but he still in the playoffs managed to put up elite numbers, especially relative to his peers. His biggest detriment was being in the same conference as the KD-era Golden State Warriors, which basically broke the sport for three seasons. Had the Rockets not missed 27 straight 3s — mind you, assuming each 3 is an independent event with a 35% of making them, has a 0.000888506% probability, roughly 6x the probability of getting hit with lightning — we'd be speaking differently about him.
+
+In the 2021 playoffs, he played on an injured hamstring to play point for the 2021 Superteam Nets, missing Kyrie. He nearly brought them across the finish line. The injury practically destroyed his iron man career, and subsequent seasons were fraught with injury concerns. I want people to remember what this man did, how he sacrificed his body.
+
+Let me be clear. I don't really like James Harden that much. He's a flopping whiner, he's quit on multiple teams, and like, look at his stupid-ass beard.
+
+But I will not let anyone deny that he was one of the greatest hoopers this game has ever seen.

content/sports/nba/hoopers-i'll-remember/kate-martin.md

@@ -0,0 +1,10 @@
+---
+title: "Kate Martin: Dirk Herwitzki"
+tags:
+  - thoughts
+---
+NCAAW Final 4 2024. UConn-Iowa. Caitlin Clark is hard carrying her team, but she needs someone to step up and hit some timely shots.
+
+Enter Kate Martin.
+
+This girl was hitting fadeaway contested jumpers off one leg like Dirk Nowitzki. UConn was fighting back and she was taking ridiculously tough shots. And MAKING them.

content/sports/nba/hoopers-i'll-remember/michael-porter-jr.md

@@ -0,0 +1,10 @@
+---
+title: "Michael Porter Jr: the sane one?"
+tags:
+  - thoughts
+---
+This is an ode to Michael Porter Jr, aka MPJ. He was selected with the 14th overall pick in the 2018 NBA Draft, but he honestly may have gone first overall if it weren't for that pesky back of his. 
+
+As a sidenote, I want to salute modern science, it's incredible how we've managed to make things that would literally kill people a hundred years ago into living a fruitful life.
+
+There's literally nothing going on in his head. I understand this is probably mean of me to say but it's literally like a DVD logo bouncing around. But he hoops.

content/sports/nba/moments-i'll-remember/jayson-tatum-game-winner-2022.md

@@ -0,0 +1,6 @@
+---
+title: Jayson Tatum wins the game at the buzzer
+tags:
+  - thoughts
+---
+The 2021-22 basketball season was an incredible one for me. It didn't end the way I wanted it to, and in fact it ended so poorly I basically suffered a minor mental breakdown. Whatever. 

content/sports/why-do-we-care-about-sports?.md

@@ -0,0 +1,5 @@
+---
+title: Why do we care so much about sports?
+tags:
+  - thoughts
+---

content/technology/hardware/nvidia/installing-cuda-nvidia-drivers-on-lxc.md

@@ -0,0 +1,19 @@
+---
+title: Installing CUDA/Nvidia Drivers on a host to be shared with Linux Container(s)
+tags:
+  - observations
+---
+I have a Proxmox server with an RTX 4090. I want to share the GPU across multiple lightweight Linux Containers (LXCs), of which they might share them with Docker containers, which could theoretically share... whatever, you get it. I want access to `nvcc` on LXCs for development. I also want access to the drivers so applications that use hardware acceleration (such as [Jellyfin](https://github.com/jellyfin/jellyfin)) can... accelerate.
+
+This is a little bit more of a convoluted process than it may seem like. It involves messing with the kernel, building kernel modules, passing through drivers, and downloading multiple packages, but in the end, you get a GPU that can be used in multiple contexts avoiding the overhead of virtualization, with all the upside of containerization and isolation.
+
+> [!info]
+> If you can avoid it, don't install CUDA on your host or LXCs. You should use devcontainer ecosystem and develop inside a docker container, pulling [`nvidia/cuda-devel`](https://hub.docker.com/r/nvidia/cuda/) for `nvcc` access, and you *should* do this to pin your dependencies. However, you *will* need to have the corresponding drivers installed on the host and any additional layers of isolation. This guide includes the installation of drivers, you can skip the CUDA stuff if it's not relevant.
+
+I'm doing this on a fresh installation of Proxmox. My server is using a 7950x3D + a PNY RTX 4090 (my beloved). I will be using debian (btw) as my LXC guest since it's the gold standard, but all of this can be adapted to your relevant distro. 
+
+# Getting the relevant driver
+
+Here's a fun fact: NVIDIA releases CUDA releases separately from their gaming/workload drivers, but the latter drivers may sometimes be ahead of their CUDA-based drivers. This can lead to some *fun* (author's note: horrible) scenarios. NVIDIA does not directly publish what version of the driver is associated with a CUDA release, you need to figure it out yourself. *woohoo!* 
+
+God, NVIDIA, I might be your only defender left, but can you please make it easier for the rest of us?

content/technology/programming/concepts/rbac.md

@@ -0,0 +1,8 @@
+---
+tags:
+  - notes
+title: Role-based access control
+---
+- Sources:
+	- https://tailscale.com/blog/rbac-like-it-was-meant-to-be
+	- https://axellarsson.com/blog/rbac-and-nodejs/

content/technology/programming/work-stealing.md → content/technology/programming/concepts/work-stealing.md

@@ -1,6 +1,7 @@
 ---
 tags:
   - notes
+title: Work-stealing executors in a multithreaded environment
 ---
 - Work-stealing is an algorithm that enables threads in an application to "steal" work from other threads to balance tasks across threads
 	- Context: In a multithreaded environment, a thread may be working on a set of tasks, but tasks may take a variable amount of time to finish, leaving the workloads unbalanced. a work-stealing algorithm dynamically balances the workload across a set of threads to balance workloads. 

content/technology/programming/languages/javascript/why-so-many-config-formats?.md

@@ -0,0 +1,14 @@
+---
+title: Why does Javascript have so many config formats?
+tags:
+  - thoughts
+---
+I've been learning Typescript recently. I kinda love it in all honesty! I was turned off from Javascript because of the lack of a type system, but Typescript's is *very* strong and ergonomic. 
+
+But. I'm astonished at how many freaking configuration files I need.
+
+Vite needs its own config file. Eslint needs its own configuration file. The configuration file formats aren't unified. Some of them are actual Javascript files, some of them are JSON files, some of them are YAML files.
+
+Python may have its issues, but at least the community has agreed to let `pyproject.toml` be the source of truth. flake8 [tried to play games](https://github.com/PyCQA/flake8/issues/234) and has been taken over by `ruff`. Why the hell are we allowing the execution of code in a configuration file that should be invariant to the actual application, does anyone realize how much of a security nightmare that is?
+
+I'm loving the devex of Typescript, but I really can't get over this. C'mon, we need a dictator, not even benevolent, to support an unified packing format like `pyproject.toml`. Please, for the love of god, what are we doing here?

content/technology/programming/operating-systems/index.md

@@ -0,0 +1 @@
+Notes related to programming functionality related to operating systems in specific.