2092391605882028391

2026-08-25 22:57:02 -04:00
1 parent 3b8e3e3488
commit 26779fdb8a
@@ -0,0 +1,37 @@
+---
+title: "@nicbarkeragain I think it's actually *easier* to reason about your code when you pre-allocate and just \"fill\" buffers that never move."
+type: archive
+source: twitter
+source_url: "https://x.com/big_tony299/status/2092396351392796893"
+author: "Antoine Gagnon"
+handle: big_tony299
+post_id: "2092396351392796893"
+date: 2026-08-25
+archived: 2026-08-25
+draft: false
+tags:
+  - archive
+  - twitter
+  - nicbarkeragain
+description: "@nicbarkeragain I think it's actually *easier* to reason about your code when you pre-allocate and just \"fill\" buffers that never move."
+in_reply_to: ""
+parent_post_id: "2092392060280307994"
+---
+
+## Source
+
+- URL: https://x.com/big_tony299/status/2092396351392796893
+- Author: Antoine Gagnon (@big_tony299)
+- Posted: 2026-08-25 23:38:59
+
+## Branch
+
+**1/** **@big_tony299** ^2092396351392796893
+
+**@nicbarkeragain**
+
+I think it's actually *easier* to reason about your code when you pre-allocate and just "fill" buffers that never move. So many opportunities for DMA-style optimizations!
+
+## Related
+
+- Spine: [[archive/threads/nicbarkeragain/2026-08-25-i-think-one-of-the-biggest-level-ups-i-had-in]]
@@ -0,0 +1,37 @@
+---
+title: "@nicbarkeragain I still find myself sketching data flow diagrams when designing data processing oriented systems, but I agree that it often doesn't translate well to concrete interfaces in code, it misses any problems you might have with push/pull impedance mismatches, who \"holds\" memory etc."
+type: archive
+source: twitter
+source_url: "https://x.com/smolus/status/2092393961771970898"
+author: "jakub smolaga"
+handle: smolus
+post_id: "2092393961771970898"
+date: 2026-08-25
+archived: 2026-08-25
+draft: false
+tags:
+  - archive
+  - twitter
+  - nicbarkeragain
+description: "@nicbarkeragain I still find myself sketching data flow diagrams when designing data processing oriented systems, but I agree that it often doesn't translate well to concrete interfaces in code, it misses any problems you might have with push/pull impedance mismatches, who \"holds\" memory etc."
+in_reply_to: ""
+parent_post_id: "2092391605882028391"
+---
+
+## Source
+
+- URL: https://x.com/smolus/status/2092393961771970898
+- Author: jakub smolaga (@smolus)
+- Posted: 2026-08-25 23:29:29
+
+## Branch
+
+**1/** **@smolus** ^2092393961771970898
+
+**@nicbarkeragain**
+
+I still find myself sketching data flow diagrams when designing data processing oriented systems, but I agree that it often doesn't translate well to concrete interfaces in code, it misses any problems you might have with push/pull impedance mismatches, who "holds" memory etc.
+
+## Related
+
+- Spine: [[archive/threads/nicbarkeragain/2026-08-25-i-think-one-of-the-biggest-level-ups-i-had-in]]
@@ -0,0 +1,37 @@
+---
+title: "@nicbarkeragain Something that CS suffers from is not being clear in what is true in the ideal model vs the actual implementation of a “computer.” For example, in relation to EE, ideal op-amps vs the actual transistor level circuitry of an op-amp allow for different approximations to be made."
+type: archive
+source: twitter
+source_url: "https://x.com/interlarper/status/2092403755966713992"
+author: "loper"
+handle: interlarper
+post_id: "2092403755966713992"
+date: 2026-08-26
+archived: 2026-08-25
+draft: false
+tags:
+  - archive
+  - twitter
+  - nicbarkeragain
+description: "@nicbarkeragain Something that CS suffers from is not being clear in what is true in the ideal model vs the actual implementation of a “computer.” For example, in relation to EE, ideal op-amps vs the actual transistor level circuitry of an op-amp allow for different approximations to be made."
+in_reply_to: ""
+parent_post_id: "2092391605882028391"
+---
+
+## Source
+
+- URL: https://x.com/interlarper/status/2092403755966713992
+- Author: loper (@interlarper)
+- Posted: 2026-08-26 00:08:25
+
+## Branch
+
+**1/** **@interlarper** ^2092403755966713992
+
+**@nicbarkeragain**
+
+Something that CS suffers from is not being clear in what is true in the ideal model vs the actual implementation of a “computer.” For example, in relation to EE, ideal op-amps vs the actual transistor level circuitry of an op-amp allow for different approximations to be made.
+
+## Related
+
+- Spine: [[archive/threads/nicbarkeragain/2026-08-25-i-think-one-of-the-biggest-level-ups-i-had-in]]
@@ -0,0 +1,39 @@
+---
+title: "some distinction between allocating new memory, copying src->dst, and then freeing src."
+type: archive
+source: twitter
+source_url: "https://x.com/mirbacx/status/2092402560745570814"
+author: "mirbacx"
+handle: mirbacx
+post_id: "2092402560745570814"
+date: 2026-08-26
+archived: 2026-08-25
+draft: false
+tags:
+  - archive
+  - twitter
+  - nicbarkeragain
+description: "some distinction between allocating new memory, copying src->dst, and then freeing src."
+in_reply_to: ""
+parent_post_id: "2092391605882028391"
+---
+
+## Source
+
+- URL: https://x.com/mirbacx/status/2092402560745570814
+- Author: mirbacx (@mirbacx)
+- Posted: 2026-08-26 00:03:40
+
+## Branch
+
+**1/** **@mirbacx** ^2092402560745570814
+
+some distinction between allocating new memory, copying src->dst, and then freeing src. copying/moving is expensive.
+
+updating pointers (to point to elsewhere in already allocated memory) is different.
+
+there's also the matter of data locality. the cpu prefetches data (from RAM) onto caches. only so much can fit closest to the CPU (L1), so you don't want your algos accessing memory all over the place (eg, linked lists). "cache misses"
+
+## Related
+
+- Spine: [[archive/threads/nicbarkeragain/2026-08-25-i-think-one-of-the-biggest-level-ups-i-had-in]]
@@ -0,0 +1,42 @@
+---
+title: "I think one of the biggest level ups I had in programming was shaking off the metaphor associated with data \"moving\"."
+type: archive
+source: twitter
+source_url: "https://x.com/nicbarkeragain/status/2092391605882028391"
+author: "Nic Barker"
+handle: nicbarkeragain
+post_id: "2092391605882028391"
+date: 2026-08-25
+archived: 2026-08-25
+draft: false
+tags:
+  - archive
+  - twitter
+  - nicbarkeragain
+description: "I think one of the biggest level ups I had in programming was shaking off the metaphor associated with data \"moving\"."
+in_reply_to: ""
+---
+
+## Source
+
+- URL: https://x.com/nicbarkeragain/status/2092391605882028391
+- Author: Nic Barker (@nicbarkeragain)
+- Posted: 2026-08-25 23:20:08
+
+## Thread
+
+**1/** **@nicbarkeragain** ^2092391605882028391
+
+I think one of the biggest level ups I had in programming was shaking off the metaphor associated with data "moving". There's no such thing as moving bits, you can only copy them from one place to another. You can't "remove" array / list elements, you can only overwrite them.
+
+Branches: [[archive/threads/nicbarkeragain/2026-08-25-i-think-one-of-the-biggest-level-ups-i-had-in/2026-08-25-smolus-i-still-find-myself-sketching-data-flow-diagrams]], [[archive/threads/nicbarkeragain/2026-08-25-i-think-one-of-the-biggest-level-ups-i-had-in/2026-08-26-mirbacx-some-distinction-between-allocating-new-memory]], [[archive/threads/nicbarkeragain/2026-08-25-i-think-one-of-the-biggest-level-ups-i-had-in/2026-08-26-interlarper-something-that-cs-suffers-from-is-not-being-clear]]
+
+**2/** **@nicbarkeragain** ^2092391608239190239
+
+Computers don't work at all like the physical world. We tend to mentally conceptualize a computer like a factory floor with things moving from one place to another, but a running program is much more like a painting. To "move" something you need to cover and repaint it elsewhere.
+
+**3/** **@nicbarkeragain** ^2092392060280307994
+
+It's so easy as a student / junior to think of a list type structure as being "empty". When in reality you're using the length / count to represent emptiness, the element memory is still there regardless (maybe zeroed, maybe not)
+
+Branches: [[archive/threads/nicbarkeragain/2026-08-25-i-think-one-of-the-biggest-level-ups-i-had-in/2026-08-25-big_tony299-i-think-its-actually-easier-to-reason-about-your]]
@@ -13,3 +13,4 @@ tags:
 - [[archive/threads/nicbarkeragain/2025-07-22-branchless-programming-is-a-term-used-to-describe]]
 - [[archive/threads/nicbarkeragain/2026-08-03-imo-basically-every-object-creation-api-you-have]]
 - [[archive/threads/nicbarkeragain/2026-08-05-an-example-of-this-a-classic-way-to-construct-a]]
+- [[archive/threads/nicbarkeragain/2026-08-25-i-think-one-of-the-biggest-level-ups-i-had-in]]