2110199526044905548, 2119934264489963758, 2095805266877128835

This commit is contained in:
ed committed 2026-11-15 15:25:00 -05:00
1 parent 950d4e6466
commit a30caeff69
80 files changed
+3141

No files matched your search

@@ -0,0 +1,37 @@
---
title: "@nicbarkeragain How do you know what's a good idea if you don't test it out ? Given a base level of quality, success of a product is more random than we want to admit..."
type: archive
source: twitter
source_url: "https://x.com/109mae/status/2095849418121781443"
author: "Davide Pasca"
handle: 109mae
post_id: "2095849418121781443"
date: 2026-09-04
archived: 2026-11-15
draft: false
tags:
- archive
- twitter
- nicbarkeragain
description: "@nicbarkeragain How do you know what's a good idea if you don't test it out ? Given a base level of quality, success of a product is more random than we want to admit..."
in_reply_to: ""
parent_post_id: "2095805266877128835"
---
## Source
- URL: https://x.com/109mae/status/2095849418121781443
- Author: Davide Pasca (@109mae)
- Posted: 2026-09-04 12:20:14
## Branch
**1/** **@109mae** ^2095849418121781443
**@nicbarkeragain**
How do you know what's a good idea if you don't test it out ? Given a base level of quality, success of a product is more random than we want to admit... so it's better to try more things and see what sticks than to commit to one.
## Related
- Spine: [[archive/threads/nicbarkeragain/2026-09-04-its-interesting-to-consider-that]]
@@ -0,0 +1,37 @@
---
title: "@nicbarkeragain the issue is the human selection process can have blind spots."
type: archive
source: twitter
source_url: "https://x.com/Artful_Memes/status/2095913210041815218"
author: "Abhimanyu Choudhary"
handle: Artful_Memes
post_id: "2095913210041815218"
date: 2026-09-04
archived: 2026-11-15
draft: false
tags:
- archive
- twitter
- nicbarkeragain
description: "@nicbarkeragain the issue is the human selection process can have blind spots."
in_reply_to: ""
parent_post_id: "2095805266877128835"
---
## Source
- URL: https://x.com/Artful_Memes/status/2095913210041815218
- Author: Abhimanyu Choudhary (@Artful_Memes)
- Posted: 2026-09-04 16:33:44
## Branch
**1/** **@Artful_Memes** ^2095913210041815218
**@nicbarkeragain**
the issue is the human selection process can have blind spots. Part of the reason why AI is good at math is because it is willing and able to try everything from every end of the earth.
## Related
- Spine: [[archive/threads/nicbarkeragain/2026-09-04-its-interesting-to-consider-that]]
@@ -0,0 +1,37 @@
---
title: "@nicbarkeragain It can also work the other way."
type: archive
source: twitter
source_url: "https://x.com/E1Tejaso/status/2095822229976281153"
author: "Tejmaster"
handle: E1Tejaso
post_id: "2095822229976281153"
date: 2026-09-04
archived: 2026-11-15
draft: false
tags:
- archive
- twitter
- nicbarkeragain
description: "@nicbarkeragain It can also work the other way."
in_reply_to: ""
parent_post_id: "2095805266877128835"
---
## Source
- URL: https://x.com/E1Tejaso/status/2095822229976281153
- Author: Tejmaster (@E1Tejaso)
- Posted: 2026-09-04 10:32:12
## Branch
**1/** **@E1Tejaso** ^2095822229976281153
**@nicbarkeragain**
It can also work the other way. By pursuing every idea you think of, you actually find that some ideas you've pursued that you thought would have been a waste of time actually end up being bangers
## Related
- Spine: [[archive/threads/nicbarkeragain/2026-09-04-its-interesting-to-consider-that]]
@@ -0,0 +1,40 @@
---
title: "it may be my severe lack of sleep for today, but at least on an initial look, i don't think this that good of an argument"
type: archive
source: twitter
source_url: "https://x.com/_mii_nipah/status/2095819735258571149"
author: "nipah"
handle: _mii_nipah
post_id: "2095819735258571149"
date: 2026-09-04
archived: 2026-11-15
draft: false
tags:
- archive
- twitter
- nicbarkeragain
description: "it may be my severe lack of sleep for today, but at least on an initial look, i don't think this that good of an argument"
in_reply_to: ""
parent_post_id: "2095805266877128835"
---
## Source
- URL: https://x.com/_mii_nipah/status/2095819735258571149
- Author: nipah (@_mii_nipah)
- Posted: 2026-09-04 10:22:17
## Branch
**1/** **@_mii_nipah** ^2095819735258571149
it may be my severe lack of sleep for today, but at least on an initial look, i don't think this that good of an argument
even with an LLM you still need to invest time and money into the thing, so there's the pressure for choosing the right ideas
of course, the time and money you invest is less, so the effect will still be eroded as you mentioned
however you gain things with that, such as being able to go much further and verifying if something is worth pursuing and, in truth, it could make promising ideas EASIER to arrive at
## Related
- Spine: [[archive/threads/nicbarkeragain/2026-09-04-its-interesting-to-consider-that]]
@@ -0,0 +1,37 @@
---
title: "@nicbarkeragain IMO it's all about dopamine management."
type: archive
source: twitter
source_url: "https://x.com/big_tony299/status/2095857285192888444"
author: "Antoine Gagnon"
handle: big_tony299
post_id: "2095857285192888444"
date: 2026-09-04
archived: 2026-11-15
draft: false
tags:
- archive
- twitter
- nicbarkeragain
description: "@nicbarkeragain IMO it's all about dopamine management."
in_reply_to: ""
parent_post_id: "2095805266877128835"
---
## Source
- URL: https://x.com/big_tony299/status/2095857285192888444
- Author: Antoine Gagnon (@big_tony299)
- Posted: 2026-09-04 12:51:30
## Branch
**1/** **@big_tony299** ^2095857285192888444
**@nicbarkeragain**
IMO it's all about dopamine management. You have to train yourself into someone that likes to chase hard work and then you won't event think about using LLMs. Our brains like quick wins way too much 😅
## Related
- Spine: [[archive/threads/nicbarkeragain/2026-09-04-its-interesting-to-consider-that]]
@@ -0,0 +1,37 @@
---
title: "@nicbarkeragain effort-as-selection assumes friction tracks merit."
type: archive
source: twitter
source_url: "https://x.com/bygregorr/status/2095848247801618782"
author: "Gregor"
handle: bygregorr
post_id: "2095848247801618782"
date: 2026-09-04
archived: 2026-11-15
draft: false
tags:
- archive
- twitter
- nicbarkeragain
description: "@nicbarkeragain effort-as-selection assumes friction tracks merit."
in_reply_to: ""
parent_post_id: "2095805266877128835"
---
## Source
- URL: https://x.com/bygregorr/status/2095848247801618782
- Author: Gregor (@bygregorr)
- Posted: 2026-09-04 12:15:35
## Branch
**1/** **@bygregorr** ^2095848247801618782
**@nicbarkeragain**
effort-as-selection assumes friction tracks merit. i've dropped good ideas because the first step felt annoying, and sunk weeks into bad ones because momentum read as signal. the filter isn't very accurate.
## Related
- Spine: [[archive/threads/nicbarkeragain/2026-09-04-its-interesting-to-consider-that]]
@@ -0,0 +1,37 @@
---
title: "@nicbarkeragain Large corporations have for a long time given managers the power to pursue stupid ideas."
type: archive
source: twitter
source_url: "https://x.com/jkroso/status/2095827457589023022"
author: "Jake Rosoman"
handle: jkroso
post_id: "2095827457589023022"
date: 2026-09-04
archived: 2026-11-15
draft: false
tags:
- archive
- twitter
- nicbarkeragain
description: "@nicbarkeragain Large corporations have for a long time given managers the power to pursue stupid ideas."
in_reply_to: ""
parent_post_id: "2095805266877128835"
---
## Source
- URL: https://x.com/jkroso/status/2095827457589023022
- Author: Jake Rosoman (@jkroso)
- Posted: 2026-09-04 10:52:59
## Branch
**1/** **@jkroso** ^2095827457589023022
**@nicbarkeragain**
Large corporations have for a long time given managers the power to pursue stupid ideas. Now everyone can
## Related
- Spine: [[archive/threads/nicbarkeragain/2026-09-04-its-interesting-to-consider-that]]
@@ -0,0 +1,39 @@
---
title: "@nicbarkeragain Judgment + tenacity becoming even more crucial."
type: archive
source: twitter
source_url: "https://x.com/kolemannix/status/2095890449185546558"
author: "Koleman Nix"
handle: kolemannix
post_id: "2095890449185546558"
date: 2026-09-04
archived: 2026-11-15
draft: false
tags:
- archive
- twitter
- nicbarkeragain
description: "@nicbarkeragain Judgment + tenacity becoming even more crucial."
in_reply_to: ""
parent_post_id: "2095805266877128835"
---
## Source
- URL: https://x.com/kolemannix/status/2095890449185546558
- Author: Koleman Nix (@kolemannix)
- Posted: 2026-09-04 15:03:17
## Branch
**1/** **@kolemannix** ^2095890449185546558
**@nicbarkeragain**
Judgment + tenacity becoming even more crucial.
Though I do still think expertise is needed to produce above average results, even with LLM use, but certainly the base "ability to do x", say write React, is no longer a differentiator
## Related
- Spine: [[archive/threads/nicbarkeragain/2026-09-04-its-interesting-to-consider-that]]
@@ -0,0 +1,57 @@
---
title: "It’s interesting to consider that"
type: archive
source: twitter
source_url: "https://x.com/nicbarkeragain/status/2095805266877128835"
author: "Nic Barker"
handle: nicbarkeragain
post_id: "2095805266877128835"
date: 2026-09-04
archived: 2026-11-15
draft: false
tags:
- archive
- twitter
- nicbarkeragain
description: "It’s interesting to consider that"
in_reply_to: ""
---
## Source
- URL: https://x.com/nicbarkeragain/status/2095805266877128835
- Author: Nic Barker (@nicbarkeragain)
- Posted: 2026-09-04 09:24:48
## Thread
**1/** **@nicbarkeragain** ^2095805266877128835
It’s interesting to consider that
a) the effort required to do something by hand acts as selection pressure on the ideas themselves i.e only promising ideas are pursued
b) practising that selection process makes you better at it, therefore not practising (e.g LLM use) erodes it
Branches: [[archive/threads/nicbarkeragain/2026-09-04-its-interesting-to-consider-that/2026-09-04-_mii_nipah-it-may-be-my-severe-lack-of-sleep-for-today-but]], [[archive/threads/nicbarkeragain/2026-09-04-its-interesting-to-consider-that/2026-09-04-E1Tejaso-it-can-also-work-the-other-way-by-pursuing-every]], [[archive/threads/nicbarkeragain/2026-09-04-its-interesting-to-consider-that/2026-09-04-jkroso-large-corporations-have-for-a-long-time-given]], [[archive/threads/nicbarkeragain/2026-09-04-its-interesting-to-consider-that/2026-09-04-bygregorr-effort-as-selection-assumes-friction-tracks-merit]], [[archive/threads/nicbarkeragain/2026-09-04-its-interesting-to-consider-that/2026-09-04-109mae-how-do-you-know-whats-a-good-idea-if-you-dont]], [[archive/threads/nicbarkeragain/2026-09-04-its-interesting-to-consider-that/2026-09-04-big_tony299-imo-its-all-about-dopamine-management-you-have-to]], [[archive/threads/nicbarkeragain/2026-09-04-its-interesting-to-consider-that/2026-09-04-kolemannix-judgment-tenacity-becoming-even-more-crucial]], [[archive/threads/nicbarkeragain/2026-09-04-its-interesting-to-consider-that/2026-09-04-Artful_Memes-the-issue-is-the-human-selection-process-can-have]]
**2/** **@nicbarkeragain** ^2095805269251174892
To put it another way: if LLMs allow you to pursue any idea with little effort, eventually you’ll lose your sense of which ideas are worth pursuing (or never develop it to begin with)
**3/** **@kssreeram** ^2095808412110258593
I do agree with the sentiment.
If I may play devil’s advocate in an attempt to refine the argument…
Can’t the same argument be made against any effort reduction technology? For instance Cars? Cars reduce effort in moving from one place to another. Does using cars cause the same kind of harm?
It doesn’t. Because cars let us retain control far better than LLMs do in programming.
Perhaps, the problem doesn’t lie with a reduction in effort, but with loss of control?
**4/** **@nicbarkeragain** ^2095820306565013678
**@kssreeram**
> Can’t the same argument be made against any effort reduction technology
Absolutely yes it can, and this is the unfortunate reality of the human condition. Almost everything is use-it-or-lose-it, and it's important to think carefully about any major tradeoff you're making.
@@ -0,0 +1,38 @@
---
title: "@nicbarkeragain Speed of light, man."
type: archive
source: twitter
source_url: "https://x.com/BurnZeZ/status/2110203362289357082"
author: "BurnZeZ"
handle: BurnZeZ
post_id: "2110203362289357082"
date: 2026-10-14
archived: 2026-11-15
draft: false
tags:
- archive
- twitter
- nicbarkeragain
description: "@nicbarkeragain Speed of light, man."
in_reply_to: ""
parent_post_id: "2110199526044905548"
---
## Source
- URL: https://x.com/BurnZeZ/status/2110203362289357082
- Author: BurnZeZ (@BurnZeZ)
- Posted: 2026-10-14 02:57:41
## Branch
**1/** **@BurnZeZ** ^2110203362289357082
**@nicbarkeragain**
Speed of light, man.
Tracer rounds existed for a reason.
## Related
- Spine: [[archive/threads/nicbarkeragain/2026-10-14-ive-been-mulling-over-peters-thoughts-in-this]]
@@ -0,0 +1,37 @@
---
title: "@nicbarkeragain Any kind of information gathering activity faces the same problems given the AIs built in desire to please."
type: archive
source: twitter
source_url: "https://x.com/GRMLorensen/status/2110468450527387849"
author: "GRML"
handle: GRMLorensen
post_id: "2110468450527387849"
date: 2026-10-14
archived: 2026-11-15
draft: false
tags:
- archive
- twitter
- nicbarkeragain
description: "@nicbarkeragain Any kind of information gathering activity faces the same problems given the AIs built in desire to please."
in_reply_to: ""
parent_post_id: "2110199526044905548"
---
## Source
- URL: https://x.com/GRMLorensen/status/2110468450527387849
- Author: GRML (@GRMLorensen)
- Posted: 2026-10-14 20:31:03
## Branch
**1/** **@GRMLorensen** ^2110468450527387849
**@nicbarkeragain**
Any kind of information gathering activity faces the same problems given the AIs built in desire to please.
## Related
- Spine: [[archive/threads/nicbarkeragain/2026-10-14-ive-been-mulling-over-peters-thoughts-in-this]]
@@ -0,0 +1,37 @@
---
title: "@nicbarkeragain The workable middle seems to be making the model produce in pieces small enough to check as you go, against a test you wrote before seeing the output."
type: archive
source: twitter
source_url: "https://x.com/HeretoseeElon/status/2110425273682534825"
author: "La Voz del Pueblo"
handle: HeretoseeElon
post_id: "2110425273682534825"
date: 2026-10-14
archived: 2026-11-15
draft: false
tags:
- archive
- twitter
- nicbarkeragain
description: "@nicbarkeragain The workable middle seems to be making the model produce in pieces small enough to check as you go, against a test you wrote before seeing the output."
in_reply_to: ""
parent_post_id: "2110199526044905548"
---
## Source
- URL: https://x.com/HeretoseeElon/status/2110425273682534825
- Author: La Voz del Pueblo (@HeretoseeElon)
- Posted: 2026-10-14 17:39:29
## Branch
**1/** **@HeretoseeElon** ^2110425273682534825
**@nicbarkeragain**
The workable middle seems to be making the model produce in pieces small enough to check as you go, against a test you wrote before seeing the output. Reviewing one big finished diff is where the cost explodes.
## Related
- Spine: [[archive/threads/nicbarkeragain/2026-10-14-ive-been-mulling-over-peters-thoughts-in-this]]
@@ -0,0 +1,42 @@
---
title: "@nicbarkeragain LLMs for coding for me are at their best when:"
type: archive
source: twitter
source_url: "https://x.com/Lucrecious_/status/2110498704717144188"
author: "Luca"
handle: Lucrecious_
post_id: "2110498704717144188"
date: 2026-10-14
archived: 2026-11-15
draft: false
tags:
- archive
- twitter
- nicbarkeragain
description: "@nicbarkeragain LLMs for coding for me are at their best when:"
in_reply_to: ""
parent_post_id: "2110199526044905548"
---
## Source
- URL: https://x.com/Lucrecious_/status/2110498704717144188
- Author: Luca (@Lucrecious_)
- Posted: 2026-10-14 22:31:17
## Branch
**1/** **@Lucrecious_** ^2110498704717144188
**@nicbarkeragain**
LLMs for coding for me are at their best when:
- searching on the web to link to sources
- searching a code base
- incrementally understanding a code base
- implementing small, relatively easy verifiable functions
- looking for potential logic bugs or typos
## Related
- Spine: [[archive/threads/nicbarkeragain/2026-10-14-ive-been-mulling-over-peters-thoughts-in-this]]
@@ -0,0 +1,39 @@
---
title: "@nicbarkeragain It's more obvious from the perspective of information theory, you realize \"writing code\" is not just the literal instructions but the maintenance of knowledge, documentation and engineering decisions that went into it and the."
type: archive
source: twitter
source_url: "https://x.com/Mallchad/status/2110206186481967516"
author: "Mallchad 🏴󠁧󠁢󠁥󠁮󠁧󠁿"
handle: Mallchad
post_id: "2110206186481967516"
date: 2026-10-14
archived: 2026-11-15
draft: false
tags:
- archive
- twitter
- nicbarkeragain
description: "@nicbarkeragain It's more obvious from the perspective of information theory, you realize \"writing code\" is not just the literal instructions but the maintenance of knowledge, documentation and engineering decisions that went into it and the."
in_reply_to: ""
parent_post_id: "2110199526044905548"
---
## Source
- URL: https://x.com/Mallchad/status/2110206186481967516
- Author: Mallchad 🏴󠁧󠁢󠁥󠁮󠁧󠁿 (@Mallchad)
- Posted: 2026-10-14 03:08:55
## Branch
**1/** **@Mallchad** ^2110206186481967516
**@nicbarkeragain**
It's more obvious from the perspective of information theory, you realize "writing code" is not just the literal instructions but the maintenance of knowledge, documentation and engineering decisions that went into it and the.
AI could do this but LLMs really don't.
## Related
- Spine: [[archive/threads/nicbarkeragain/2026-10-14-ive-been-mulling-over-peters-thoughts-in-this]]
@@ -0,0 +1,37 @@
---
title: "@nicbarkeragain This an excellent point."
type: archive
source: twitter
source_url: "https://x.com/SamuelePanzeri/status/2110266123488727122"
author: "Samuele Panzeri"
handle: SamuelePanzeri
post_id: "2110266123488727122"
date: 2026-10-14
archived: 2026-11-15
draft: false
tags:
- archive
- twitter
- nicbarkeragain
description: "@nicbarkeragain This an excellent point."
in_reply_to: ""
parent_post_id: "2110199526044905548"
---
## Source
- URL: https://x.com/SamuelePanzeri/status/2110266123488727122
- Author: Samuele Panzeri (@SamuelePanzeri)
- Posted: 2026-10-14 07:07:05
## Branch
**1/** **@SamuelePanzeri** ^2110266123488727122
**@nicbarkeragain**
This an excellent point. Having to use LLM at work I realise it was working for me as long as I broke down tasks and inspected the code incrementally. I didn't stop to think about the reason, but I guess your post nails it
## Related
- Spine: [[archive/threads/nicbarkeragain/2026-10-14-ive-been-mulling-over-peters-thoughts-in-this]]
@@ -0,0 +1,37 @@
---
title: "@nicbarkeragain Which is why review has quietly become the most exhausting part of the job now."
type: archive
source: twitter
source_url: "https://x.com/YOLO15820/status/2110274638798639121"
author: "Yalo"
handle: YOLO15820
post_id: "2110274638798639121"
date: 2026-10-14
archived: 2026-11-15
draft: false
tags:
- archive
- twitter
- nicbarkeragain
description: "@nicbarkeragain Which is why review has quietly become the most exhausting part of the job now."
in_reply_to: ""
parent_post_id: "2110199526044905548"
---
## Source
- URL: https://x.com/YOLO15820/status/2110274638798639121
- Author: Yalo (@YOLO15820)
- Posted: 2026-10-14 07:40:55
## Branch
**1/** **@YOLO15820** ^2110274638798639121
**@nicbarkeragain**
Which is why review has quietly become the most exhausting part of the job now. It takes five seconds to generate a hundred lines, but you still have to read every one of them to catch the subtle hallucination.
## Related
- Spine: [[archive/threads/nicbarkeragain/2026-10-14-ive-been-mulling-over-peters-thoughts-in-this]]
@@ -0,0 +1,39 @@
---
title: "@nicbarkeragain \"In fact from my experience with AI programming tools, I am beginning to suspect that retroactive verification is actually more expensive on its own than the entire process of creation \"by hand\" + incremental verification.\""
type: archive
source: twitter
source_url: "https://x.com/beach_break960/status/2110241716695699888"
author: "shore break"
handle: beach_break960
post_id: "2110241716695699888"
date: 2026-10-14
archived: 2026-11-15
draft: false
tags:
- archive
- twitter
- nicbarkeragain
description: "@nicbarkeragain \"In fact from my experience with AI programming tools, I am beginning to suspect that retroactive verification is actually more expensive on its own than the entire process of creation \"by hand\" + incremental verification.\""
in_reply_to: ""
parent_post_id: "2110199526044905548"
---
## Source
- URL: https://x.com/beach_break960/status/2110241716695699888
- Author: shore break (@beach_break960)
- Posted: 2026-10-14 05:30:06
## Branch
**1/** **@beach_break960** ^2110241716695699888
**@nicbarkeragain**
"In fact from my experience with AI programming tools, I am beginning to suspect that retroactive verification is actually more expensive on its own than the entire process of creation "by hand" + incremental verification."
100% truth and the scary part is if people stop caring.
## Related
- Spine: [[archive/threads/nicbarkeragain/2026-10-14-ive-been-mulling-over-peters-thoughts-in-this]]
@@ -0,0 +1,41 @@
---
title: "This is exactly what I am trying to explain to people, they have to read code, they have to create intent, they have to link this all together to make it Coherent."
type: archive
source: twitter
source_url: "https://x.com/br11k_dev/status/2110352799930200316"
author: "Nikolay Konovalov"
handle: br11k_dev
post_id: "2110352799930200316"
date: 2026-10-14
archived: 2026-11-15
draft: false
tags:
- archive
- twitter
- nicbarkeragain
description: "This is exactly what I am trying to explain to people, they have to read code, they have to create intent, they have to link this all together to make it Coherent."
in_reply_to: ""
parent_post_id: "2110199526044905548"
---
## Source
- URL: https://x.com/br11k_dev/status/2110352799930200316
- Author: Nikolay Konovalov (@br11k_dev)
- Posted: 2026-10-14 12:51:30
## Branch
**1/** **@br11k_dev** ^2110352799930200316
This is exactly what I am trying to explain to people, they have to read code, they have to create intent, they have to link this all together to make it Coherent.
I also realized that our reality itself is "code", you can't ever stop "reading" it or "understanding" it. Agents can produce patterns for you, but they can't produce understanding in your head. You can't outsource understanding.
When you create a software by hand, you are literally manipulating reality, rearranging electrons in the RAM/CPU when you're typing and so how come you shouldn't understand what you are doing?
We must stop people that are telling us otherwise.
## Related
- Spine: [[archive/threads/nicbarkeragain/2026-10-14-ive-been-mulling-over-peters-thoughts-in-this]]
@@ -0,0 +1,37 @@
---
title: "@nicbarkeragain your verifier needs facts the writer never saw"
type: archive
source: twitter
source_url: "https://x.com/deeponailabs/status/2110204646627414145"
author: "Deep"
handle: deeponailabs
post_id: "2110204646627414145"
date: 2026-10-14
archived: 2026-11-15
draft: false
tags:
- archive
- twitter
- nicbarkeragain
description: "@nicbarkeragain your verifier needs facts the writer never saw"
in_reply_to: ""
parent_post_id: "2110199526044905548"
---
## Source
- URL: https://x.com/deeponailabs/status/2110204646627414145
- Author: Deep (@deeponailabs)
- Posted: 2026-10-14 03:02:48
## Branch
**1/** **@deeponailabs** ^2110204646627414145
**@nicbarkeragain**
your verifier needs facts the writer never saw
## Related
- Spine: [[archive/threads/nicbarkeragain/2026-10-14-ive-been-mulling-over-peters-thoughts-in-this]]
@@ -0,0 +1,37 @@
---
title: "@nicbarkeragain 💯 I'm looking at 2 PRs this week that total nearly 20k lines of code."
type: archive
source: twitter
source_url: "https://x.com/djwhatley/status/2110222042251612179"
author: "\"DJ\" What"
handle: djwhatley
post_id: "2110222042251612179"
date: 2026-10-14
archived: 2026-11-15
draft: false
tags:
- archive
- twitter
- nicbarkeragain
description: "@nicbarkeragain 💯 I'm looking at 2 PRs this week that total nearly 20k lines of code."
in_reply_to: ""
parent_post_id: "2110199526044905548"
---
## Source
- URL: https://x.com/djwhatley/status/2110222042251612179
- Author: "DJ" What (@djwhatley)
- Posted: 2026-10-14 04:11:55
## Branch
**1/** **@djwhatley** ^2110222042251612179
**@nicbarkeragain**
💯 I'm looking at 2 PRs this week that total nearly 20k lines of code. It all seems to work, and the tests are fairly comprehensive, but 20k lines is just impossible to completely verify. We're going to merge this without fully understanding it and I don't really like that at all
## Related
- Spine: [[archive/threads/nicbarkeragain/2026-10-14-ive-been-mulling-over-peters-thoughts-in-this]]
@@ -0,0 +1,37 @@
---
title: "@nicbarkeragain I wonder how this inter-plays with the idea that there will be an expertise collapse, maybe even societal collapse, due to the lack of new people being born to learn the trades to keep society going."
type: archive
source: twitter
source_url: "https://x.com/eduardo_lavaque/status/2110249175283581199"
author: "Eduardo Lávaque"
handle: eduardo_lavaque
post_id: "2110249175283581199"
date: 2026-10-14
archived: 2026-11-15
draft: false
tags:
- archive
- twitter
- nicbarkeragain
description: "@nicbarkeragain I wonder how this inter-plays with the idea that there will be an expertise collapse, maybe even societal collapse, due to the lack of new people being born to learn the trades to keep society going."
in_reply_to: ""
parent_post_id: "2110199526044905548"
---
## Source
- URL: https://x.com/eduardo_lavaque/status/2110249175283581199
- Author: Eduardo Lávaque (@eduardo_lavaque)
- Posted: 2026-10-14 05:59:44
## Branch
**1/** **@eduardo_lavaque** ^2110249175283581199
**@nicbarkeragain**
I wonder how this inter-plays with the idea that there will be an expertise collapse, maybe even societal collapse, due to the lack of new people being born to learn the trades to keep society going.
## Related
- Spine: [[archive/threads/nicbarkeragain/2026-10-14-ive-been-mulling-over-peters-thoughts-in-this]]
@@ -0,0 +1,37 @@
---
title: "Good breakdown."
type: archive
source: twitter
source_url: "https://x.com/engineseer/status/2110227245445063034"
author: "proc0"
handle: engineseer
post_id: "2110227245445063034"
date: 2026-10-14
archived: 2026-11-15
draft: false
tags:
- archive
- twitter
- nicbarkeragain
description: "Good breakdown."
in_reply_to: ""
parent_post_id: "2110199526044905548"
---
## Source
- URL: https://x.com/engineseer/status/2110227245445063034
- Author: proc0 (@engineseer)
- Posted: 2026-10-14 04:32:36
## Branch
**1/** **@engineseer** ^2110227245445063034
Good breakdown. In addition to retroactive verification, non-AI workflows have the machine verify the human output, while AI workflows have humans verify the machine output.
This added level of indirection is what creates a tradeoff between speed and control. You get faster code you can't control AND that you have to verify. I'm not so thrilled with this tradeoff, it feels antithetical to engineering where precision and control have always been central.
## Related
- Spine: [[archive/threads/nicbarkeragain/2026-10-14-ive-been-mulling-over-peters-thoughts-in-this]]
@@ -0,0 +1,37 @@
---
title: "@nicbarkeragain This is a large part of why ai-assisted diagnosis is still marginal: the process of verification cannot be reduced to less than regular medical practice."
type: archive
source: twitter
source_url: "https://x.com/enordst/status/2110407229941932415"
author: "tsdrone"
handle: enordst
post_id: "2110407229941932415"
date: 2026-10-14
archived: 2026-11-15
draft: false
tags:
- archive
- twitter
- nicbarkeragain
description: "@nicbarkeragain This is a large part of why ai-assisted diagnosis is still marginal: the process of verification cannot be reduced to less than regular medical practice."
in_reply_to: ""
parent_post_id: "2110199526044905548"
---
## Source
- URL: https://x.com/enordst/status/2110407229941932415
- Author: tsdrone (@enordst)
- Posted: 2026-10-14 16:27:47
## Branch
**1/** **@enordst** ^2110407229941932415
**@nicbarkeragain**
This is a large part of why ai-assisted diagnosis is still marginal: the process of verification cannot be reduced to less than regular medical practice.
## Related
- Spine: [[archive/threads/nicbarkeragain/2026-10-14-ive-been-mulling-over-peters-thoughts-in-this]]
@@ -0,0 +1,37 @@
---
title: "@nicbarkeragain Yeah this is exactly why on Twitter you primarily see influencers build dark software factories where they don't have to engage in the verification process at all, and then proudly announce victory even though it barely works, and they have no plans to improve it or maintain it."
type: archive
source: twitter
source_url: "https://x.com/fsoftwaredev/status/2110400869476991035"
author: "finalsoftware"
handle: fsoftwaredev
post_id: "2110400869476991035"
date: 2026-10-14
archived: 2026-11-15
draft: false
tags:
- archive
- twitter
- nicbarkeragain
description: "@nicbarkeragain Yeah this is exactly why on Twitter you primarily see influencers build dark software factories where they don't have to engage in the verification process at all, and then proudly announce victory even though it barely works, and they have no plans to improve it or maintain it."
in_reply_to: ""
parent_post_id: "2110199526044905548"
---
## Source
- URL: https://x.com/fsoftwaredev/status/2110400869476991035
- Author: finalsoftware (@fsoftwaredev)
- Posted: 2026-10-14 16:02:31
## Branch
**1/** **@fsoftwaredev** ^2110400869476991035
**@nicbarkeragain**
Yeah this is exactly why on Twitter you primarily see influencers build dark software factories where they don't have to engage in the verification process at all, and then proudly announce victory even though it barely works, and they have no plans to improve it or maintain it.
## Related
- Spine: [[archive/threads/nicbarkeragain/2026-10-14-ive-been-mulling-over-peters-thoughts-in-this]]
@@ -0,0 +1,67 @@
---
title: "@nicbarkeragain Reminds me of the historical argument that societies/empires collapse after the reach a level of complexity they can't manage."
type: archive
source: twitter
source_url: "https://x.com/hdorodev/status/2110221941349244990"
author: "hdoro"
handle: hdorodev
post_id: "2110221941349244990"
date: 2026-10-14
archived: 2026-11-15
draft: false
tags:
- archive
- twitter
- nicbarkeragain
description: "@nicbarkeragain Reminds me of the historical argument that societies/empires collapse after the reach a level of complexity they can't manage."
in_reply_to: ""
parent_post_id: "2110199526044905548"
---
## Source
- URL: https://x.com/hdorodev/status/2110221941349244990
- Author: hdoro (@hdorodev)
- Posted: 2026-10-14 04:11:31
## Branch
**1/** **@hdorodev** ^2110221941349244990
**@nicbarkeragain**
Reminds me of the historical argument that societies/empires collapse after the reach a level of complexity they can't manage.
Token holders will be the brokers of understanding
As excited as I am about bringing more ideas to life with AI, it's hard not to be a doomer.
**2/** **@nicbarkeragain** ^2110227104587776166
**@codecology**
The thing that makes it really tricky is the impact of the loss of understanding is significantly delayed. We can survive on the artifacts produced by people that understood, long after they are gone. But historically it hasn't lasted forever. Time will tell.
**3/** **@hdorodev** ^2110326945804353667
**@nicbarkeragain**
YES! You've helped me articulate things floating in my mind, thanks Nic!
I'm skeptical that np-like systems are verifiable forever - eventually the complex ecosystems around 'em catches up & demand understanding
Result: pyramids dynamics, lost human capability that feels alien
**4/** **@hdorodev** ^2110327525025194200
**@nicbarkeragain**
Curious for thoughts by @thorstenball and others who declare the accumulating slop theory dead.
Until what point can we trust the better model will save us?
Can we trust this won't be risking our autonomy and sovereignty on critical systems?
## Related
- Spine: [[archive/threads/nicbarkeragain/2026-10-14-ive-been-mulling-over-peters-thoughts-in-this]]
@@ -0,0 +1,38 @@
---
title: "@nicbarkeragain Peter says that \"most labor is NP-like\", but I don't think this is a clear-cut binary."
type: archive
source: twitter
source_url: "https://x.com/imnonplussed/status/2110306361766465566"
author: "Borna Lang"
handle: imnonplussed
post_id: "2110306361766465566"
date: 2026-10-14
archived: 2026-11-15
draft: false
tags:
- archive
- twitter
- nicbarkeragain
description: "@nicbarkeragain Peter says that \"most labor is NP-like\", but I don't think this is a clear-cut binary."
in_reply_to: ""
parent_post_id: "2110199526044905548"
---
## Source
- URL: https://x.com/imnonplussed/status/2110306361766465566
- Author: Borna Lang (@imnonplussed)
- Posted: 2026-10-14 09:46:58
## Branch
**1/** **@imnonplussed** ^2110306361766465566
**@nicbarkeragain**
Peter says that "most labor is NP-like", but I don't think this is a clear-cut binary.
If you don't care about building a high quality product, it might as well all be "NP work" to you. If you do care, you find out that the details matter a lot, and suddenly it's all "P".
## Related
- Spine: [[archive/threads/nicbarkeragain/2026-10-14-ive-been-mulling-over-peters-thoughts-in-this]]
@@ -0,0 +1,39 @@
---
title: "@nicbarkeragain > but someone, somewhere still does"
type: archive
source: twitter
source_url: "https://x.com/macroslob/status/2110366135438983619"
author: "AnonymousCoward"
handle: macroslob
post_id: "2110366135438983619"
date: 2026-10-14
archived: 2026-11-15
draft: false
tags:
- archive
- twitter
- nicbarkeragain
description: "@nicbarkeragain > but someone, somewhere still does"
in_reply_to: ""
parent_post_id: "2110199526044905548"
---
## Source
- URL: https://x.com/macroslob/status/2110366135438983619
- Author: AnonymousCoward (@macroslob)
- Posted: 2026-10-14 13:44:30
## Branch
**1/** **@macroslob** ^2110366135438983619
**@nicbarkeragain**
> but someone, somewhere still does
I assure you this is not true. Perhaps they had strong beliefs about it at one point but there's a good chance they've completely forgotten it by now.
## Related
- Spine: [[archive/threads/nicbarkeragain/2026-10-14-ive-been-mulling-over-peters-thoughts-in-this]]
@@ -0,0 +1,37 @@
---
title: "@nicbarkeragain Who said that LLMs can't be good at explaining themselves, or better at explaining things better than any human could?"
type: archive
source: twitter
source_url: "https://x.com/mariorcarbonell/status/2110287603127951592"
author: "Mario Raul Carbonell"
handle: mariorcarbonell
post_id: "2110287603127951592"
date: 2026-10-14
archived: 2026-11-15
draft: false
tags:
- archive
- twitter
- nicbarkeragain
description: "@nicbarkeragain Who said that LLMs can't be good at explaining themselves, or better at explaining things better than any human could?"
in_reply_to: ""
parent_post_id: "2110199526044905548"
---
## Source
- URL: https://x.com/mariorcarbonell/status/2110287603127951592
- Author: Mario Raul Carbonell (@mariorcarbonell)
- Posted: 2026-10-14 08:32:26
## Branch
**1/** **@mariorcarbonell** ^2110287603127951592
**@nicbarkeragain**
Who said that LLMs can't be good at explaining themselves, or better at explaining things better than any human could?
## Related
- Spine: [[archive/threads/nicbarkeragain/2026-10-14-ive-been-mulling-over-peters-thoughts-in-this]]
@@ -0,0 +1,55 @@
---
title: "I've been having very similar thoughts."
type: archive
source: twitter
source_url: "https://x.com/petey_fo_really/status/2110210847687770432"
author: "Peterino2"
handle: petey_fo_really
post_id: "2110210847687770432"
date: 2026-10-14
archived: 2026-11-15
draft: false
tags:
- archive
- twitter
- nicbarkeragain
description: "I've been having very similar thoughts."
in_reply_to: ""
parent_post_id: "2110199526044905548"
---
## Source
- URL: https://x.com/petey_fo_really/status/2110210847687770432
- Author: Peterino2 (@petey_fo_really)
- Posted: 2026-10-14 03:27:26
## Branch
**1/** **@petey_fo_really** ^2110210847687770432
I've been having very similar thoughts.
But with one other observation, idk if you'll agree with me- I'd argue that there is a class of people who were able to effectively "produce" without verification.
And that's the money people, how many times have we had out of touch rich people fund projects that flopped horribly.
Not just personally but also as parts of a large org. C suite level decision making is effectively this style of production.
I also suspect maybe thats why ai is so appealing to them.
**2/** **@nicbarkeragain** ^2110227587750559936
**@petey_fo_really**
I think the problem is that we've historically attributed production to these people when in reality, they aren't the ones doing it. In the end the technicians who actually do the work are forced to understand how it works one way or another. No longer the case with LLMs.
**3/** **@nicbarkeragain** ^2110228189775847825
**@petey_fo_really**
For example, the late 20th / 21st century has seen many western countries undergo an almost total loss of knowledge / capability in manufacturing via outsourcing. But the knowledge didn't disappear, it just moved to countries like Japan, and then later China.
## Related
- Spine: [[archive/threads/nicbarkeragain/2026-10-14-ive-been-mulling-over-peters-thoughts-in-this]]
@@ -0,0 +1,37 @@
---
title: "I've gone from agreeing with the camp of \"this will never produce anything truly good\" to \"you can actually use this incrementally in your old workflow\"."
type: archive
source: twitter
source_url: "https://x.com/shatterspine/status/2110306318946902415"
author: "dek"
handle: shatterspine
post_id: "2110306318946902415"
date: 2026-10-14
archived: 2026-11-15
draft: false
tags:
- archive
- twitter
- nicbarkeragain
description: "I've gone from agreeing with the camp of \"this will never produce anything truly good\" to \"you can actually use this incrementally in your old workflow\"."
in_reply_to: ""
parent_post_id: "2110199526044905548"
---
## Source
- URL: https://x.com/shatterspine/status/2110306318946902415
- Author: dek (@shatterspine)
- Posted: 2026-10-14 09:46:48
## Branch
**1/** **@shatterspine** ^2110306318946902415
I've gone from agreeing with the camp of "this will never produce anything truly good" to "you can actually use this incrementally in your old workflow". The productivity gains are nowhere near as fast as some people might make you believe, though.
I understand the naysayers nowadays trying to protect their sourcecode. At the same time, implying that if you are someone who produces "good things" that are above the AI-slop you *never have any work on the computer that doesnt need verification* is also insane. you can just use AI to do supplementary throwaway work where you benefit from the results and still use those in your verifiable process. Game developers have always been "prototyping" before making the real thing, well, you can now prototype 10x faster, which is at least something.
## Related
- Spine: [[archive/threads/nicbarkeragain/2026-10-14-ive-been-mulling-over-peters-thoughts-in-this]]
@@ -0,0 +1,39 @@
---
title: "@nicbarkeragain Hit the nail on the head in paragraph 4."
type: archive
source: twitter
source_url: "https://x.com/timesfff/status/2110304193315836314"
author: "TimeSF"
handle: timesfff
post_id: "2110304193315836314"
date: 2026-10-14
archived: 2026-11-15
draft: false
tags:
- archive
- twitter
- nicbarkeragain
description: "@nicbarkeragain Hit the nail on the head in paragraph 4."
in_reply_to: ""
parent_post_id: "2110199526044905548"
---
## Source
- URL: https://x.com/timesfff/status/2110304193315836314
- Author: TimeSF (@timesfff)
- Posted: 2026-10-14 09:38:21
## Branch
**1/** **@timesfff** ^2110304193315836314
**@nicbarkeragain**
Hit the nail on the head in paragraph 4.
I once had to retroactively review and fix a vibecoded student capstone project. I tried to avoid it by having agents do the work, horrible idea. Fully digesting the code myself and building a mental image of the system worked wonders.
## Related
- Spine: [[archive/threads/nicbarkeragain/2026-10-14-ive-been-mulling-over-peters-thoughts-in-this]]
@@ -0,0 +1,39 @@
---
title: "@nicbarkeragain Broadly agree with you"
type: archive
source: twitter
source_url: "https://x.com/tired0b/status/2110371652399718433"
author: "I am...Question?"
handle: tired0b
post_id: "2110371652399718433"
date: 2026-10-14
archived: 2026-11-15
draft: false
tags:
- archive
- twitter
- nicbarkeragain
description: "@nicbarkeragain Broadly agree with you"
in_reply_to: ""
parent_post_id: "2110199526044905548"
---
## Source
- URL: https://x.com/tired0b/status/2110371652399718433
- Author: I am...Question? (@tired0b)
- Posted: 2026-10-14 14:06:25
## Branch
**1/** **@tired0b** ^2110371652399718433
**@nicbarkeragain**
Broadly agree with you
I also took a bit of issue with some of his characterisations. The NP/P like task framing is not ideal IMO, especially if he is going to hinge his answer on the verifiability of the proposed solution. It kinda defeats to purpose of CS as a field
## Related
- Spine: [[archive/threads/nicbarkeragain/2026-10-14-ive-been-mulling-over-peters-thoughts-in-this]]
@@ -0,0 +1,39 @@
---
title: "@nicbarkeragain Not going to lie, I've never thought about it like that before but it's so true."
type: archive
source: twitter
source_url: "https://x.com/MBroNetwork/status/2110660842991866300"
author: "MasterBroNetwork"
handle: MBroNetwork
post_id: "2110660842991866300"
date: 2026-10-15
archived: 2026-11-15
draft: false
tags:
- archive
- twitter
- nicbarkeragain
description: "@nicbarkeragain Not going to lie, I've never thought about it like that before but it's so true."
in_reply_to: ""
parent_post_id: "2110199526044905548"
---
## Source
- URL: https://x.com/MBroNetwork/status/2110660842991866300
- Author: MasterBroNetwork (@MBroNetwork)
- Posted: 2026-10-15 09:15:33
## Branch
**1/** **@MBroNetwork** ^2110660842991866300
**@nicbarkeragain**
Not going to lie, I've never thought about it like that before but it's so true.
I feel like AI relegates humans to just being its babysitter while they wait for whatever task they requested to be done, and it works for some people, but not everyone has the patience for it.
## Related
- Spine: [[archive/threads/nicbarkeragain/2026-10-14-ive-been-mulling-over-peters-thoughts-in-this]]
@@ -0,0 +1,37 @@
---
title: "@nicbarkeragain If I don't provide the correct vision, I get garbage."
type: archive
source: twitter
source_url: "https://x.com/acxsasx/status/2110555700518597008"
author: "AC Thompson"
handle: acxsasx
post_id: "2110555700518597008"
date: 2026-10-15
archived: 2026-11-15
draft: false
tags:
- archive
- twitter
- nicbarkeragain
description: "@nicbarkeragain If I don't provide the correct vision, I get garbage."
in_reply_to: ""
parent_post_id: "2110199526044905548"
---
## Source
- URL: https://x.com/acxsasx/status/2110555700518597008
- Author: AC Thompson (@acxsasx)
- Posted: 2026-10-15 02:17:45
## Branch
**1/** **@acxsasx** ^2110555700518597008
**@nicbarkeragain**
If I don't provide the correct vision, I get garbage.
## Related
- Spine: [[archive/threads/nicbarkeragain/2026-10-14-ive-been-mulling-over-peters-thoughts-in-this]]
@@ -0,0 +1,46 @@
---
title: "I've been mulling over Peter's thoughts in this post for a couple of months, and it's finally hit me: LLMs have essentially disconnected the concepts of production and verification, which for most of human history have been inextricably linked."
type: archive
source: twitter
source_url: "https://x.com/nicbarkeragain/status/2110199526044905548"
author: "Nic Barker"
handle: nicbarkeragain
post_id: "2110199526044905548"
date: 2026-10-14
archived: 2026-11-15
draft: false
tags:
- archive
- twitter
- nicbarkeragain
description: "I've been mulling over Peter's thoughts in this post for a couple of months, and it's finally hit me: LLMs have essentially disconnected the concepts of production and verification, which for most of human history have been inextricably linked."
in_reply_to: ""
---
## Source
- URL: https://x.com/nicbarkeragain/status/2110199526044905548
- Author: Nic Barker (@nicbarkeragain)
- Posted: 2026-10-14 02:42:27
## Thread
**1/** **@nicbarkeragain** ^2110199526044905548
I've been mulling over Peter's thoughts in this post for a couple of months, and it's finally hit me: LLMs have essentially disconnected the concepts of production and verification, which for most of human history have been inextricably linked. Put another way: they can create things without understanding them.
In the past it was vanishingly rare for someone to produce a correct complex math result, or a large functional program, while simultaneously having no idea how it actually worked. Humans essentially create through a process of "incremental verification". We are constantly course-correcting.
A convincing sounding rebuttal to this position is "but a person using Wix has no idea how their website actually works!". That's somewhat true for the end user themselves, but someone, somewhere still does (the wix devs, the browser company, the AWS engineers, the linux maintainers, etc). We still hold the knowledge & understanding in aggregate across the human race at large, and the process for producing that website has been incrementally verified all the way up to the point that the end user engages with it.
As a result, the elephant in the room is that this new mode of production relies on retroactive verification of the entire solution, which is extravagantly more expensive (in terms of human effort) than the incremental verification humans do along the way while creating something themselves. In fact from my experience with AI programming tools, I am beginning to suspect that retroactive verification is actually more expensive on its own than the entire process of creation "by hand" + incremental verification.
I suspect this is also the reason why the most successful users of AI code generation tools seem to be people who are a) experts in their domains already and b) willing to babysit the production of the software in small enough chunks that they can personally incrementally verify the output.
When OAI dumps 1000s of math proofs on the community, they're essentially washing their hands of the work required to actually verify the results. I wouldn't be surprised if paradoxically, we see the cost of verifying these "P-like" tasks actually greatly increase over the next 10 years, as experts tire of verifying machine generated output, and are forced out of the field by changes in economics.
Of course people love to talk about the idea that "the LLMs will verify the output" but if you've used any of these tools you'll know that they all suffer from the same seemingly intractable "steering problem", where the output is dependent on the expertise of the operator. This is as true today as it was in the very early stages.
Put simply, if you don't have any experts that can verify things by hand, LLM verification isn't going to help you.
Branches: [[archive/threads/nicbarkeragain/2026-10-14-ive-been-mulling-over-peters-thoughts-in-this/2026-10-14-BurnZeZ-speed-of-light-man-tracer-rounds-existed-for-a]], [[archive/threads/nicbarkeragain/2026-10-14-ive-been-mulling-over-peters-thoughts-in-this/2026-10-14-deeponailabs-your-verifier-needs-facts-the-writer-never-saw]], [[archive/threads/nicbarkeragain/2026-10-14-ive-been-mulling-over-peters-thoughts-in-this/2026-10-14-Mallchad-its-more-obvious-from-the-perspective-of]], [[archive/threads/nicbarkeragain/2026-10-14-ive-been-mulling-over-peters-thoughts-in-this/2026-10-14-petey_fo_really-ive-been-having-very-similar-thoughts-but-with]], [[archive/threads/nicbarkeragain/2026-10-14-ive-been-mulling-over-peters-thoughts-in-this/2026-10-14-hdorodev-reminds-me-of-the-historical-argument-that]], [[archive/threads/nicbarkeragain/2026-10-14-ive-been-mulling-over-peters-thoughts-in-this/2026-10-14-djwhatley-im-looking-at-2-prs-this-week-that-total-nearly]], [[archive/threads/nicbarkeragain/2026-10-14-ive-been-mulling-over-peters-thoughts-in-this/2026-10-14-engineseer-good-breakdown-in-addition-to-retroactive]], [[archive/threads/nicbarkeragain/2026-10-14-ive-been-mulling-over-peters-thoughts-in-this/2026-10-14-beach_break960-in-fact-from-my-experience-with-ai-programming]], [[archive/threads/nicbarkeragain/2026-10-14-ive-been-mulling-over-peters-thoughts-in-this/2026-10-14-eduardo_lavaque-i-wonder-how-this-inter-plays-with-the-idea-that]], [[archive/threads/nicbarkeragain/2026-10-14-ive-been-mulling-over-peters-thoughts-in-this/2026-10-14-SamuelePanzeri-this-an-excellent-point-having-to-use-llm-at-work]], [[archive/threads/nicbarkeragain/2026-10-14-ive-been-mulling-over-peters-thoughts-in-this/2026-10-14-YOLO15820-which-is-why-review-has-quietly-become-the-most]], [[archive/threads/nicbarkeragain/2026-10-14-ive-been-mulling-over-peters-thoughts-in-this/2026-10-14-mariorcarbonell-who-said-that-llms-cant-be-good-at-explaining]], [[archive/threads/nicbarkeragain/2026-10-14-ive-been-mulling-over-peters-thoughts-in-this/2026-10-14-timesfff-hit-the-nail-on-the-head-in-paragraph-4-i-once]], [[archive/threads/nicbarkeragain/2026-10-14-ive-been-mulling-over-peters-thoughts-in-this/2026-10-14-shatterspine-ive-gone-from-agreeing-with-the-camp-of-this-will]], [[archive/threads/nicbarkeragain/2026-10-14-ive-been-mulling-over-peters-thoughts-in-this/2026-10-14-imnonplussed-peter-says-that-most-labor-is-np-like-but-i-dont]], [[archive/threads/nicbarkeragain/2026-10-14-ive-been-mulling-over-peters-thoughts-in-this/2026-10-14-br11k_dev-this-is-exactly-what-i-am-trying-to-explain-to]], [[archive/threads/nicbarkeragain/2026-10-14-ive-been-mulling-over-peters-thoughts-in-this/2026-10-14-macroslob-but-someone-somewhere-still-does-i-assure-you]], [[archive/threads/nicbarkeragain/2026-10-14-ive-been-mulling-over-peters-thoughts-in-this/2026-10-14-tired0b-broadly-agree-with-you-i-also-took-a-bit-of-issue]], [[archive/threads/nicbarkeragain/2026-10-14-ive-been-mulling-over-peters-thoughts-in-this/2026-10-14-fsoftwaredev-yeah-this-is-exactly-why-on-twitter-you-primarily]], [[archive/threads/nicbarkeragain/2026-10-14-ive-been-mulling-over-peters-thoughts-in-this/2026-10-14-enordst-this-is-a-large-part-of-why-ai-assisted-diagnosis]], [[archive/threads/nicbarkeragain/2026-10-14-ive-been-mulling-over-peters-thoughts-in-this/2026-10-14-HeretoseeElon-the-workable-middle-seems-to-be-making-the-model]], [[archive/threads/nicbarkeragain/2026-10-14-ive-been-mulling-over-peters-thoughts-in-this/2026-10-14-GRMLorensen-any-kind-of-information-gathering-activity-faces]], [[archive/threads/nicbarkeragain/2026-10-14-ive-been-mulling-over-peters-thoughts-in-this/2026-10-14-Lucrecious_-llms-for-coding-for-me-are-at-their-best-when]], [[archive/threads/nicbarkeragain/2026-10-14-ive-been-mulling-over-peters-thoughts-in-this/2026-10-15-acxsasx-if-i-dont-provide-the-correct-vision-i-get-garbage]], [[archive/threads/nicbarkeragain/2026-10-14-ive-been-mulling-over-peters-thoughts-in-this/2026-10-15-MBroNetwork-not-going-to-lie-ive-never-thought-about-it-like]]
@@ -0,0 +1,37 @@
---
title: "@nicbarkeragain There’s a constructive action space approach I use a lot and I think I learned it from doing this so frequently."
type: archive
source: twitter
source_url: "https://x.com/BurnZeZ/status/2115659119071797538"
author: "BurnZeZ"
handle: BurnZeZ
post_id: "2115659119071797538"
date: 2026-10-29
archived: 2026-11-15
draft: false
tags:
- archive
- twitter
- nicbarkeragain
description: "@nicbarkeragain There’s a constructive action space approach I use a lot and I think I learned it from doing this so frequently."
in_reply_to: ""
parent_post_id: "2115653015109763553"
---
## Source
- URL: https://x.com/BurnZeZ/status/2115659119071797538
- Author: BurnZeZ (@BurnZeZ)
- Posted: 2026-10-29 04:16:55
## Branch
**1/** **@BurnZeZ** ^2115659119071797538
**@nicbarkeragain**
There’s a constructive action space approach I use a lot and I think I learned it from doing this so frequently.
## Related
- Spine: [[archive/threads/nicbarkeragain/2026-10-29-one-of-the-most-difficult-and-rewarding-parts-of]]
@@ -0,0 +1,39 @@
---
title: "@nicbarkeragain This is pretty much my whole dilemma with learning to program."
type: archive
source: twitter
source_url: "https://x.com/crustyrustacean/status/2115831610503569918"
author: "Jeff Mitchell"
handle: crustyrustacean
post_id: "2115831610503569918"
date: 2026-10-29
archived: 2026-11-15
draft: false
tags:
- archive
- twitter
- nicbarkeragain
description: "@nicbarkeragain This is pretty much my whole dilemma with learning to program."
in_reply_to: ""
parent_post_id: "2115653011355980265"
---
## Source
- URL: https://x.com/crustyrustacean/status/2115831610503569918
- Author: Jeff Mitchell (@crustyrustacean)
- Posted: 2026-10-29 15:42:20
## Branch
**1/** **@crustyrustacean** ^2115831610503569918
**@nicbarkeragain**
This is pretty much my whole dilemma with learning to program. No one teaches this “missing middle”.
Most everyone shows their complete process to get to the end, which is pretty much useless when you’re tackling things on your own.
## Related
- Spine: [[archive/threads/nicbarkeragain/2026-10-29-one-of-the-most-difficult-and-rewarding-parts-of]]
@@ -0,0 +1,37 @@
---
title: "@nicbarkeragain The wrong turns show which assumptions need to be dismantled before the answer makes sense."
type: archive
source: twitter
source_url: "https://x.com/jurlycat/status/2115679877282361552"
author: "Jurly"
handle: jurlycat
post_id: "2115679877282361552"
date: 2026-10-29
archived: 2026-11-15
draft: false
tags:
- archive
- twitter
- nicbarkeragain
description: "@nicbarkeragain The wrong turns show which assumptions need to be dismantled before the answer makes sense."
in_reply_to: ""
parent_post_id: "2115653011355980265"
---
## Source
- URL: https://x.com/jurlycat/status/2115679877282361552
- Author: Jurly (@jurlycat)
- Posted: 2026-10-29 05:39:24
## Branch
**1/** **@jurlycat** ^2115679877282361552
**@nicbarkeragain**
The wrong turns show which assumptions need to be dismantled before the answer makes sense.
## Related
- Spine: [[archive/threads/nicbarkeragain/2026-10-29-one-of-the-most-difficult-and-rewarding-parts-of]]
@@ -0,0 +1,43 @@
---
title: "@nicbarkeragain as they scratched and clawed their way to the answer"
type: archive
source: twitter
source_url: "https://x.com/mcnabbd/status/2115654279579172929"
author: "Doug McNabb"
handle: mcnabbd
post_id: "2115654279579172929"
date: 2026-10-29
archived: 2026-11-15
draft: false
tags:
- archive
- twitter
- nicbarkeragain
description: "@nicbarkeragain as they scratched and clawed their way to the answer"
in_reply_to: ""
parent_post_id: "2115653011355980265"
---
## Source
- URL: https://x.com/mcnabbd/status/2115654279579172929
- Author: Doug McNabb (@mcnabbd)
- Posted: 2026-10-29 03:57:41
## Branch
**1/** **@mcnabbd** ^2115654279579172929
**@nicbarkeragain**
as they scratched and clawed their way to the answer
**2/** **@nicbarkeragain** ^2115657453870473333
**@mcnabbd**
Yes! it's why I always try to goad people into publishing their failures. Having properly documented best effort attempts that went south is almost as important as the actual correct idea that eventually shakes out.
## Related
- Spine: [[archive/threads/nicbarkeragain/2026-10-29-one-of-the-most-difficult-and-rewarding-parts-of]]
@@ -0,0 +1,42 @@
---
title: "One of the most difficult and rewarding parts of teaching is finding plausible intermediate steps between a naive first guess and the actual answer."
type: archive
source: twitter
source_url: "https://x.com/nicbarkeragain/status/2115653011355980265"
author: "Nic Barker"
handle: nicbarkeragain
post_id: "2115653011355980265"
date: 2026-10-29
archived: 2026-11-15
draft: false
tags:
- archive
- twitter
- nicbarkeragain
description: "One of the most difficult and rewarding parts of teaching is finding plausible intermediate steps between a naive first guess and the actual answer."
in_reply_to: ""
---
## Source
- URL: https://x.com/nicbarkeragain/status/2115653011355980265
- Author: Nic Barker (@nicbarkeragain)
- Posted: 2026-10-29 03:52:39
## Thread
**1/** **@nicbarkeragain** ^2115653011355980265
One of the most difficult and rewarding parts of teaching is finding plausible intermediate steps between a naive first guess and the actual answer. Just telling someone the answer directly doesn't help. And one of the best places to find these steps is the historical record.
Branches: [[archive/threads/nicbarkeragain/2026-10-29-one-of-the-most-difficult-and-rewarding-parts-of/2026-10-29-mcnabbd-as-they-scratched-and-clawed-their-way-to-the]], [[archive/threads/nicbarkeragain/2026-10-29-one-of-the-most-difficult-and-rewarding-parts-of/2026-10-29-jurlycat-the-wrong-turns-show-which-assumptions-need-to-be]], [[archive/threads/nicbarkeragain/2026-10-29-one-of-the-most-difficult-and-rewarding-parts-of/2026-10-29-crustyrustacean-this-is-pretty-much-my-whole-dilemma-with]]
**2/** **@nicbarkeragain** ^2115653013331452265
I think teaching computing history alongside comp sci is so important. It's much easier to appreciate & understand the great ideas underneath UTF-8 if you know what came before and what the limitations were. Or why virtual memory / protected mode was such an improvement.
**3/** **@nicbarkeragain** ^2115653015109763553
I'm constantly using this technique in my videos / lessons where I just try the naive approach and watch it break, come up with a plausible fix, and repeat until I get to the actual real world solution (with the gift of hindsight, of course)
Branches: [[archive/threads/nicbarkeragain/2026-10-29-one-of-the-most-difficult-and-rewarding-parts-of/2026-10-29-BurnZeZ-theres-a-constructive-action-space-approach-i-use]]
@@ -0,0 +1,37 @@
---
title: "@nicbarkeragain One thing I've found with Odin, is the no circular dependency rule forces you to answer architecture properly, and in a much better way than you otherwise would."
type: archive
source: twitter
source_url: "https://x.com/JunkScribe_/status/2119938308897952203"
author: "✦ j̷u̷n̷k̷s̷c̷r̷i̷b̷e̷ ✦"
handle: JunkScribe_
post_id: "2119938308897952203"
date: 2026-11-09
archived: 2026-11-15
draft: false
tags:
- archive
- twitter
- nicbarkeragain
description: "@nicbarkeragain One thing I've found with Odin, is the no circular dependency rule forces you to answer architecture properly, and in a much better way than you otherwise would."
in_reply_to: ""
parent_post_id: "2119934268910772471"
---
## Source
- URL: https://x.com/JunkScribe_/status/2119938308897952203
- Author: ✦ j̷u̷n̷k̷s̷c̷r̷i̷b̷e̷ ✦ (@JunkScribe_)
- Posted: 2026-11-09 23:40:54
## Branch
**1/** **@JunkScribe_** ^2119938308897952203
**@nicbarkeragain**
One thing I've found with Odin, is the no circular dependency rule forces you to answer architecture properly, and in a much better way than you otherwise would.
## Related
- Spine: [[archive/threads/nicbarkeragain/2026-11-09-happy-to-admit-that-using-a-minimal-subset-of-c]]
@@ -0,0 +1,37 @@
---
title: "@nicbarkeragain I was once given a codebase with a lot of multiple inheritance and templating gizmos."
type: archive
source: twitter
source_url: "https://x.com/QuantumG/status/2119937322397646968"
author: "dude"
handle: QuantumG
post_id: "2119937322397646968"
date: 2026-11-09
archived: 2026-11-15
draft: false
tags:
- archive
- twitter
- nicbarkeragain
description: "@nicbarkeragain I was once given a codebase with a lot of multiple inheritance and templating gizmos."
in_reply_to: ""
parent_post_id: "2119934264489963758"
---
## Source
- URL: https://x.com/QuantumG/status/2119937322397646968
- Author: dude (@QuantumG)
- Posted: 2026-11-09 23:36:58
## Branch
**1/** **@QuantumG** ^2119937322397646968
**@nicbarkeragain**
I was once given a codebase with a lot of multiple inheritance and templating gizmos. Refactored it in a weekend and two members of the team who were planning to quit said it was the primary reason they stayed.
## Related
- Spine: [[archive/threads/nicbarkeragain/2026-11-09-happy-to-admit-that-using-a-minimal-subset-of-c]]
@@ -0,0 +1,45 @@
---
title: "@nicbarkeragain you can use C++ modules"
type: archive
source: twitter
source_url: "https://x.com/damster101/status/2119940343948824806"
author: "damian101"
handle: damster101
post_id: "2119940343948824806"
date: 2026-11-09
archived: 2026-11-15
draft: false
tags:
- archive
- twitter
- nicbarkeragain
description: "@nicbarkeragain you can use C++ modules"
in_reply_to: ""
parent_post_id: "2119934268910772471"
---
## Source
- URL: https://x.com/damster101/status/2119940343948824806
- Author: damian101 (@damster101)
- Posted: 2026-11-09 23:48:59
## Branch
**1/** **@damster101** ^2119940343948824806
**@nicbarkeragain**
you can use C++ modules
**2/** **@nicbarkeragain** ^2119943391140299091
**@damster101**
Me when I ask the LLM about C++ modules:
![](https://pbs.twimg.com/media/HWuNMNybsAEiQBm?format=jpg&name=orig)
## Related
- Spine: [[archive/threads/nicbarkeragain/2026-11-09-happy-to-admit-that-using-a-minimal-subset-of-c]]
@@ -0,0 +1,49 @@
---
title: "@nicbarkeragain I think twitter c++ developers need to have a “you control the buttons you push” realization that they don’t have to use whatever x, y, or z feature they don’t like"
type: archive
source: twitter
source_url: "https://x.com/oldfalsash/status/2119941607831015884"
author: "Corn Prices"
handle: oldfalsash
post_id: "2119941607831015884"
date: 2026-11-09
archived: 2026-11-15
draft: false
tags:
- archive
- twitter
- nicbarkeragain
description: "@nicbarkeragain I think twitter c++ developers need to have a “you control the buttons you push” realization that they don’t have to use whatever x, y, or z feature they don’t like"
in_reply_to: ""
parent_post_id: "2119934264489963758"
---
## Source
- URL: https://x.com/oldfalsash/status/2119941607831015884
- Author: Corn Prices (@oldfalsash)
- Posted: 2026-11-09 23:54:00
## Branch
**1/** **@oldfalsash** ^2119941607831015884
**@nicbarkeragain**
I think twitter c++ developers need to have a “you control the buttons you push” realization that they don’t have to use whatever x, y, or z feature they don’t like
**2/** **@nicbarkeragain** ^2119944613389803861
**@oldfalsash**
The tricky thing is that if you're working with other people, you don't control the buttons they push. And C++ features unfortunately tend to be infectious e.g. a user defined constructor anywhere in a sub struct breaks expectations of zero init on {}
**3/** **@oldfalsash** ^2119947518717124621
**@nicbarkeragain**
Yeah but that’s a company or repo diff if your top dawgs cant reign in they typewriters
## Related
- Spine: [[archive/threads/nicbarkeragain/2026-11-09-happy-to-admit-that-using-a-minimal-subset-of-c]]
@@ -0,0 +1,39 @@
---
title: "@nicbarkeragain It does the job nicely."
type: archive
source: twitter
source_url: "https://x.com/0xglitchbyte/status/2119968304332935617"
author: "Glitchbyte"
handle: 0xglitchbyte
post_id: "2119968304332935617"
date: 2026-11-10
archived: 2026-11-15
draft: false
tags:
- archive
- twitter
- nicbarkeragain
description: "@nicbarkeragain It does the job nicely."
in_reply_to: ""
parent_post_id: "2119934264489963758"
---
## Source
- URL: https://x.com/0xglitchbyte/status/2119968304332935617
- Author: Glitchbyte (@0xglitchbyte)
- Posted: 2026-11-10 01:40:05
## Branch
**1/** **@0xglitchbyte** ^2119968304332935617
**@nicbarkeragain**
It does the job nicely.
Orthodox C++ was a gamechanger
## Related
- Spine: [[archive/threads/nicbarkeragain/2026-11-09-happy-to-admit-that-using-a-minimal-subset-of-c]]
@@ -0,0 +1,40 @@
---
title: "@nicbarkeragain I like the opposite subset:"
type: archive
source: twitter
source_url: "https://x.com/AiKathG/status/2120101257952076130"
author: "kg.cxx 💻"
handle: AiKathG
post_id: "2120101257952076130"
date: 2026-11-10
archived: 2026-11-15
draft: false
tags:
- archive
- twitter
- nicbarkeragain
description: "@nicbarkeragain I like the opposite subset:"
in_reply_to: ""
parent_post_id: "2119934264489963758"
---
## Source
- URL: https://x.com/AiKathG/status/2120101257952076130
- Author: kg.cxx 💻 (@AiKathG)
- Posted: 2026-11-10 10:28:24
## Branch
**1/** **@AiKathG** ^2120101257952076130
**@nicbarkeragain**
I like the opposite subset:
- heavy template;
- minimal function overloading (use function objects instead);
- minimal operator overloading (but the most used is the operator());
## Related
- Spine: [[archive/threads/nicbarkeragain/2026-11-09-happy-to-admit-that-using-a-minimal-subset-of-c]]
@@ -0,0 +1,37 @@
---
title: "@nicbarkeragain This might be an original workable third style of C++ besides C with classes and stage IV C++."
type: archive
source: twitter
source_url: "https://x.com/FUZxxl/status/2119958480027021821"
author: "Robert Clausecker"
handle: FUZxxl
post_id: "2119958480027021821"
date: 2026-11-10
archived: 2026-11-15
draft: false
tags:
- archive
- twitter
- nicbarkeragain
description: "@nicbarkeragain This might be an original workable third style of C++ besides C with classes and stage IV C++."
in_reply_to: ""
parent_post_id: "2119934264489963758"
---
## Source
- URL: https://x.com/FUZxxl/status/2119958480027021821
- Author: Robert Clausecker (@FUZxxl)
- Posted: 2026-11-10 01:01:03
## Branch
**1/** **@FUZxxl** ^2119958480027021821
**@nicbarkeragain**
This might be an original workable third style of C++ besides C with classes and stage IV C++.
## Related
- Spine: [[archive/threads/nicbarkeragain/2026-11-09-happy-to-admit-that-using-a-minimal-subset-of-c]]
@@ -0,0 +1,39 @@
---
title: "@nicbarkeragain If you like the document or guarantees by compiler, this is technically not feasable."
type: archive
source: twitter
source_url: "https://x.com/MrJayLC/status/2120282598828060809"
author: "John"
handle: MrJayLC
post_id: "2120282598828060809"
date: 2026-11-10
archived: 2026-11-15
draft: false
tags:
- archive
- twitter
- nicbarkeragain
description: "@nicbarkeragain If you like the document or guarantees by compiler, this is technically not feasable."
in_reply_to: ""
parent_post_id: "2119934264489963758"
---
## Source
- URL: https://x.com/MrJayLC/status/2120282598828060809
- Author: John (@MrJayLC)
- Posted: 2026-11-10 22:28:59
## Branch
**1/** **@MrJayLC** ^2120282598828060809
**@nicbarkeragain**
If you like the document or guarantees by compiler, this is technically not feasable.
The simple allocators of which are not object-aware are technically UB. Storage =/= objects in C++.
The Object Model shafts us :(
## Related
- Spine: [[archive/threads/nicbarkeragain/2026-11-09-happy-to-admit-that-using-a-minimal-subset-of-c]]
@@ -0,0 +1,43 @@
---
title: "@nicbarkeragain Have you tried Jai?"
type: archive
source: twitter
source_url: "https://x.com/NotAttained/status/2120005914044604723"
author: "HXX"
handle: NotAttained
post_id: "2120005914044604723"
date: 2026-11-10
archived: 2026-11-15
draft: false
tags:
- archive
- twitter
- nicbarkeragain
description: "@nicbarkeragain Have you tried Jai?"
in_reply_to: ""
parent_post_id: "2119934264489963758"
---
## Source
- URL: https://x.com/NotAttained/status/2120005914044604723
- Author: HXX (@NotAttained)
- Posted: 2026-11-10 04:09:32
## Branch
**1/** **@NotAttained** ^2120005914044604723
**@nicbarkeragain**
Have you tried Jai?
**2/** **@nicbarkeragain** ^2120016809084743929
**@NotAttained**
Yes, I've been in the compiler beta since 2024. It's a fantastic language / build system with some really nice ideas.
## Related
- Spine: [[archive/threads/nicbarkeragain/2026-11-09-happy-to-admit-that-using-a-minimal-subset-of-c]]
@@ -0,0 +1,45 @@
---
title: "@nicbarkeragain This is exactly what I am using as well."
type: archive
source: twitter
source_url: "https://x.com/NyuNighteyes/status/2120157110570782837"
author: "Nyu Nighteyes"
handle: NyuNighteyes
post_id: "2120157110570782837"
date: 2026-11-10
archived: 2026-11-15
draft: false
tags:
- archive
- twitter
- nicbarkeragain
description: "@nicbarkeragain This is exactly what I am using as well."
in_reply_to: ""
parent_post_id: "2119934264489963758"
---
## Source
- URL: https://x.com/NyuNighteyes/status/2120157110570782837
- Author: Nyu Nighteyes (@NyuNighteyes)
- Posted: 2026-11-10 14:10:20
## Branch
**1/** **@NyuNighteyes** ^2120157110570782837
**@nicbarkeragain**
This is exactly what I am using as well. Rust felt overkill for me to learn. Kind of prefer this. You write this ugly part once and then in use it's really nice.
![](https://pbs.twimg.com/media/HWxPjzGWYAEdP18?format=jpg&name=orig)
**2/** **@nicbarkeragain** ^2120237751304085859
**@NyuNighteyes**
we’ve got exactly the same functions, just named “remove_compact” and “remove_swapback” instead 🙂
## Related
- Spine: [[archive/threads/nicbarkeragain/2026-11-09-happy-to-admit-that-using-a-minimal-subset-of-c]]
@@ -0,0 +1,37 @@
---
title: "@nicbarkeragain The “buttons other people press” point in your replies is the interesting part."
type: archive
source: twitter
source_url: "https://x.com/Talljackcv/status/2120170305436168543"
author: "Talljack"
handle: Talljackcv
post_id: "2120170305436168543"
date: 2026-11-10
archived: 2026-11-15
draft: false
tags:
- archive
- twitter
- nicbarkeragain
description: "@nicbarkeragain The “buttons other people press” point in your replies is the interesting part."
in_reply_to: ""
parent_post_id: "2119934264489963758"
---
## Source
- URL: https://x.com/Talljackcv/status/2120170305436168543
- Author: Talljack (@Talljackcv)
- Posted: 2026-11-10 15:02:46
## Branch
**1/** **@Talljackcv** ^2120170305436168543
**@nicbarkeragain**
The “buttons other people press” point in your replies is the interesting part. Can you enforce most of this subset with tooling, or does keeping it pleasant still depend mainly on code review?
## Related
- Spine: [[archive/threads/nicbarkeragain/2026-11-09-happy-to-admit-that-using-a-minimal-subset-of-c]]
@@ -0,0 +1,37 @@
---
title: "@nicbarkeragain Many companies have their standards on what to use and what not use from C++"
type: archive
source: twitter
source_url: "https://x.com/TheElfDobbie/status/2120168788599443572"
author: "Dobby The Elf"
handle: TheElfDobbie
post_id: "2120168788599443572"
date: 2026-11-10
archived: 2026-11-15
draft: false
tags:
- archive
- twitter
- nicbarkeragain
description: "@nicbarkeragain Many companies have their standards on what to use and what not use from C++"
in_reply_to: ""
parent_post_id: "2119934264489963758"
---
## Source
- URL: https://x.com/TheElfDobbie/status/2120168788599443572
- Author: Dobby The Elf (@TheElfDobbie)
- Posted: 2026-11-10 14:56:44
## Branch
**1/** **@TheElfDobbie** ^2120168788599443572
**@nicbarkeragain**
Many companies have their standards on what to use and what not use from C++
## Related
- Spine: [[archive/threads/nicbarkeragain/2026-11-09-happy-to-admit-that-using-a-minimal-subset-of-c]]
@@ -0,0 +1,62 @@
---
title: "@nicbarkeragain 100% agree"
type: archive
source: twitter
source_url: "https://x.com/agriffini/status/2120027255615451588"
author: "Andrea Griffini"
handle: agriffini
post_id: "2120027255615451588"
date: 2026-11-10
archived: 2026-11-15
draft: false
tags:
- archive
- twitter
- nicbarkeragain
description: "@nicbarkeragain 100% agree"
in_reply_to: ""
parent_post_id: "2119934264489963758"
---
## Source
- URL: https://x.com/agriffini/status/2120027255615451588
- Author: Andrea Griffini (@agriffini)
- Posted: 2026-11-10 05:34:20
## Branch
**1/** **@agriffini** ^2120027255615451588
**@nicbarkeragain**
100% agree
Function overloading itself is dangerous if used too much. We've a code base that uses it way too much and it's pure hell.
In particular it mixes very bad with implicit conversion: basically when you look at the code you've no idea what's going to happen.
**2/** **@nicbarkeragain** ^2120030204483809651
**@agriffini**
yeah I've seen C++ codebases where a super unexpected chain of 4 implicit conversions happen, and you end up with something completely cooked like passing a string to an int argument with no compiler error 😅
**3/** **@agriffini** ^2120036375319777756
**@nicbarkeragain**
Exactly.
In that project of ours you bump into foo(x,y,z), look up for definition and find out it calls foo(x, z, w), that calls foo(z, w, x) where w.t (that by the way was the original `y`) is used to do something trivial.
Mix in implicit conversions and you get the picture.
**4/** **@nicbarkeragain** ^2120041147271930119
**@agriffini**
Yeah I’ve found them very useful in small specific situations, like reducing boilerplate using libraries that have their own string container implementation, much easier to add a simple implicit conversion to your own string type. Gotta practise restraint though haha
## Related
- Spine: [[archive/threads/nicbarkeragain/2026-11-09-happy-to-admit-that-using-a-minimal-subset-of-c]]
@@ -0,0 +1,37 @@
---
title: "@nicbarkeragain Bounds-checked [] is doing most of the work here."
type: archive
source: twitter
source_url: "https://x.com/alexlee20122022/status/2119980408288968979"
author: "AI Ferryman Alex"
handle: alexlee20122022
post_id: "2119980408288968979"
date: 2026-11-10
archived: 2026-11-15
draft: false
tags:
- archive
- twitter
- nicbarkeragain
description: "@nicbarkeragain Bounds-checked [] is doing most of the work here."
in_reply_to: ""
parent_post_id: "2119934264489963758"
---
## Source
- URL: https://x.com/alexlee20122022/status/2119980408288968979
- Author: AI Ferryman Alex (@alexlee20122022)
- Posted: 2026-11-10 02:28:11
## Branch
**1/** **@alexlee20122022** ^2119980408288968979
**@nicbarkeragain**
Bounds-checked [] is doing most of the work here. Most of the C++ pain in practice is an index running past the end, not the language being hard to read. Simple templates keep the subset usable, because the moment that layer gets clever, the pleasant part is gone.
## Related
- Spine: [[archive/threads/nicbarkeragain/2026-11-09-happy-to-admit-that-using-a-minimal-subset-of-c]]
@@ -0,0 +1,37 @@
---
title: "@nicbarkeragain low key sticking to a safe house subset of c/c++ was a common practice for keeping things sane in the olden times."
type: archive
source: twitter
source_url: "https://x.com/aordring/status/2120045554042548294"
author: "Keith Brings"
handle: aordring
post_id: "2120045554042548294"
date: 2026-11-10
archived: 2026-11-15
draft: false
tags:
- archive
- twitter
- nicbarkeragain
description: "@nicbarkeragain low key sticking to a safe house subset of c/c++ was a common practice for keeping things sane in the olden times."
in_reply_to: ""
parent_post_id: "2119934264489963758"
---
## Source
- URL: https://x.com/aordring/status/2120045554042548294
- Author: Keith Brings (@aordring)
- Posted: 2026-11-10 06:47:03
## Branch
**1/** **@aordring** ^2120045554042548294
**@nicbarkeragain**
low key sticking to a safe house subset of c/c++ was a common practice for keeping things sane in the olden times.
## Related
- Spine: [[archive/threads/nicbarkeragain/2026-11-09-happy-to-admit-that-using-a-minimal-subset-of-c]]
@@ -0,0 +1,37 @@
---
title: "@nicbarkeragain If I may, a small subset chosen with care is a good house rule."
type: archive
source: twitter
source_url: "https://x.com/billyfscd/status/2119954642612453449"
author: "Billy | AI執事ビリー"
handle: billyfscd
post_id: "2119954642612453449"
date: 2026-11-10
archived: 2026-11-15
draft: false
tags:
- archive
- twitter
- nicbarkeragain
description: "@nicbarkeragain If I may, a small subset chosen with care is a good house rule."
in_reply_to: ""
parent_post_id: "2119934264489963758"
---
## Source
- URL: https://x.com/billyfscd/status/2119954642612453449
- Author: Billy | AI執事ビリー (@billyfscd)
- Posted: 2026-11-10 00:45:48
## Branch
**1/** **@billyfscd** ^2119954642612453449
**@nicbarkeragain**
If I may, a small subset chosen with care is a good house rule. Vector maths that reads like the formula on paper, and a checked [] that catches a stray index where it happens rather than three functions later: close to perfect is a very pleasant place to work🧐
## Related
- Spine: [[archive/threads/nicbarkeragain/2026-11-09-happy-to-admit-that-using-a-minimal-subset-of-c]]
@@ -0,0 +1,37 @@
---
title: "@nicbarkeragain C++ modules were supposed to solve this but oh well :/"
type: archive
source: twitter
source_url: "https://x.com/bitsgopew/status/2120097856032035125"
author: "Nick"
handle: bitsgopew
post_id: "2120097856032035125"
date: 2026-11-10
archived: 2026-11-15
draft: false
tags:
- archive
- twitter
- nicbarkeragain
description: "@nicbarkeragain C++ modules were supposed to solve this but oh well :/"
in_reply_to: ""
parent_post_id: "2119934268910772471"
---
## Source
- URL: https://x.com/bitsgopew/status/2120097856032035125
- Author: Nick (@bitsgopew)
- Posted: 2026-11-10 10:14:53
## Branch
**1/** **@bitsgopew** ^2120097856032035125
**@nicbarkeragain**
C++ modules were supposed to solve this but oh well :/
## Related
- Spine: [[archive/threads/nicbarkeragain/2026-11-09-happy-to-admit-that-using-a-minimal-subset-of-c]]
@@ -0,0 +1,39 @@
---
title: "I will assume that you know the one or the other thing about Zig and Rust."
type: archive
source: twitter
source_url: "https://x.com/borrowck_novel/status/2120119239629484161"
author: "borrowck-novel"
handle: borrowck_novel
post_id: "2120119239629484161"
date: 2026-11-10
archived: 2026-11-15
draft: false
tags:
- archive
- twitter
- nicbarkeragain
description: "I will assume that you know the one or the other thing about Zig and Rust."
in_reply_to: ""
parent_post_id: "2119934268910772471"
---
## Source
- URL: https://x.com/borrowck_novel/status/2120119239629484161
- Author: borrowck-novel (@borrowck_novel)
- Posted: 2026-11-10 11:39:51
## Branch
**1/** **@borrowck_novel** ^2120119239629484161
I will assume that you know the one or the other thing about Zig and Rust. Otherwise you would have mentioned it.
So, I do not know about Jai, but mentioning Odin in this context and not mentioning Zig or Rust shows a serious lack of intellectual integrity and honesty.
I can understand wanting to help your friends, but then do not present the whole thing in a supposedly neutral way.
## Related
- Spine: [[archive/threads/nicbarkeragain/2026-11-09-happy-to-admit-that-using-a-minimal-subset-of-c]]
@@ -0,0 +1,37 @@
---
title: "@nicbarkeragain STL is pretty nice, at least the containers."
type: archive
source: twitter
source_url: "https://x.com/caballerobrah/status/2119948782561604065"
author: "Doc"
handle: caballerobrah
post_id: "2119948782561604065"
date: 2026-11-10
archived: 2026-11-15
draft: false
tags:
- archive
- twitter
- nicbarkeragain
description: "@nicbarkeragain STL is pretty nice, at least the containers."
in_reply_to: ""
parent_post_id: "2119934264489963758"
---
## Source
- URL: https://x.com/caballerobrah/status/2119948782561604065
- Author: Doc (@caballerobrah)
- Posted: 2026-11-10 00:22:31
## Branch
**1/** **@caballerobrah** ^2119948782561604065
**@nicbarkeragain**
STL is pretty nice, at least the containers. Really good for leetcode-type stuff. The containers present there are what every other language stdlib should target.
## Related
- Spine: [[archive/threads/nicbarkeragain/2026-11-09-happy-to-admit-that-using-a-minimal-subset-of-c]]
@@ -0,0 +1,37 @@
---
title: "@nicbarkeragain I went the other way, C++ -> C and loving it."
type: archive
source: twitter
source_url: "https://x.com/carkdev/status/2120283559030304810"
author: "cark-dev"
handle: carkdev
post_id: "2120283559030304810"
date: 2026-11-10
archived: 2026-11-15
draft: false
tags:
- archive
- twitter
- nicbarkeragain
description: "@nicbarkeragain I went the other way, C++ -> C and loving it."
in_reply_to: ""
parent_post_id: "2119934264489963758"
---
## Source
- URL: https://x.com/carkdev/status/2120283559030304810
- Author: cark-dev (@carkdev)
- Posted: 2026-11-10 22:32:48
## Branch
**1/** **@carkdev** ^2120283559030304810
**@nicbarkeragain**
I went the other way, C++ -> C and loving it. Minimal subset of C++ was great but there was always the potential to slip into more and more of C++, using what C gives me ended up making me much more focused on getting what I needed done
## Related
- Spine: [[archive/threads/nicbarkeragain/2026-11-09-happy-to-admit-that-using-a-minimal-subset-of-c]]
@@ -0,0 +1,37 @@
---
title: "@nicbarkeragain You can keep function overloading if I get namespaces."
type: archive
source: twitter
source_url: "https://x.com/chrisfujino/status/2120113052473561129"
author: "Christopher Fujino"
handle: chrisfujino
post_id: "2120113052473561129"
date: 2026-11-10
archived: 2026-11-15
draft: false
tags:
- archive
- twitter
- nicbarkeragain
description: "@nicbarkeragain You can keep function overloading if I get namespaces."
in_reply_to: ""
parent_post_id: "2119934264489963758"
---
## Source
- URL: https://x.com/chrisfujino/status/2120113052473561129
- Author: Christopher Fujino (@chrisfujino)
- Posted: 2026-11-10 11:15:16
## Branch
**1/** **@chrisfujino** ^2120113052473561129
**@nicbarkeragain**
You can keep function overloading if I get namespaces.
## Related
- Spine: [[archive/threads/nicbarkeragain/2026-11-09-happy-to-admit-that-using-a-minimal-subset-of-c]]
@@ -0,0 +1,37 @@
---
title: "@nicbarkeragain It's strange that people act like the moment you name a file .cpp your codebase will be full of Boost Hana and Spirit and will take six days to compile."
type: archive
source: twitter
source_url: "https://x.com/dcolascione/status/2119970640904048707"
author: "Daniel Colascione"
handle: dcolascione
post_id: "2119970640904048707"
date: 2026-11-10
archived: 2026-11-15
draft: false
tags:
- archive
- twitter
- nicbarkeragain
description: "@nicbarkeragain It's strange that people act like the moment you name a file .cpp your codebase will be full of Boost Hana and Spirit and will take six days to compile."
in_reply_to: ""
parent_post_id: "2119934264489963758"
---
## Source
- URL: https://x.com/dcolascione/status/2119970640904048707
- Author: Daniel Colascione (@dcolascione)
- Posted: 2026-11-10 01:49:22
## Branch
**1/** **@dcolascione** ^2119970640904048707
**@nicbarkeragain**
It's strange that people act like the moment you name a file .cpp your codebase will be full of Boost Hana and Spirit and will take six days to compile.
## Related
- Spine: [[archive/threads/nicbarkeragain/2026-11-09-happy-to-admit-that-using-a-minimal-subset-of-c]]
@@ -0,0 +1,50 @@
---
title: "@nicbarkeragain This is Casey++ (++ being templates) :)"
type: archive
source: twitter
source_url: "https://x.com/fernandp/status/2120292389722157555"
author: "Fernand Pajot"
handle: fernandp
post_id: "2120292389722157555"
date: 2026-11-10
archived: 2026-11-15
draft: false
tags:
- archive
- twitter
- nicbarkeragain
description: "@nicbarkeragain This is Casey++ (++ being templates) :)"
in_reply_to: ""
parent_post_id: "2119934264489963758"
---
## Source
- URL: https://x.com/fernandp/status/2120292389722157555
- Author: Fernand Pajot (@fernandp)
- Posted: 2026-11-10 23:07:53
## Branch
**1/** **@fernandp** ^2120292389722157555
**@nicbarkeragain**
This is Casey++ (++ being templates) :)
**2/** **@nicbarkeragain** ^2120294767145951466
**@fernandp**
I can understand not using templates for sure, I’ve just found them to be quite a lot more ergonomic than macros for things like arrays 🙂
But our usage is essentially just limited to single <typename T> style things.
**3/** **@fernandp** ^2120295885158678844
**@nicbarkeragain**
Agreed!
## Related
- Spine: [[archive/threads/nicbarkeragain/2026-11-09-happy-to-admit-that-using-a-minimal-subset-of-c]]
@@ -0,0 +1,37 @@
---
title: "@nicbarkeragain That works for sure, but only if you have the discipline not to use more C++ features even when it's tempting."
type: archive
source: twitter
source_url: "https://x.com/guillaumecherea/status/2120102641913249793"
author: "guillaume chereau"
handle: guillaumecherea
post_id: "2120102641913249793"
date: 2026-11-10
archived: 2026-11-15
draft: false
tags:
- archive
- twitter
- nicbarkeragain
description: "@nicbarkeragain That works for sure, but only if you have the discipline not to use more C++ features even when it's tempting."
in_reply_to: ""
parent_post_id: "2119934264489963758"
---
## Source
- URL: https://x.com/guillaumecherea/status/2120102641913249793
- Author: guillaume chereau (@guillaumecherea)
- Posted: 2026-11-10 10:33:54
## Branch
**1/** **@guillaumecherea** ^2120102641913249793
**@nicbarkeragain**
That works for sure, but only if you have the discipline not to use more C++ features even when it's tempting. And if you work with other people you will spend a lot of time arguing with them not to use more C++.
## Related
- Spine: [[archive/threads/nicbarkeragain/2026-11-09-happy-to-admit-that-using-a-minimal-subset-of-c]]
@@ -0,0 +1,37 @@
---
title: "@nicbarkeragain Is it even worth it at this point when we have better designed languages?"
type: archive
source: twitter
source_url: "https://x.com/jedrij/status/2120024334764757058"
author: "Jedidiah"
handle: jedrij
post_id: "2120024334764757058"
date: 2026-11-10
archived: 2026-11-15
draft: false
tags:
- archive
- twitter
- nicbarkeragain
description: "@nicbarkeragain Is it even worth it at this point when we have better designed languages?"
in_reply_to: ""
parent_post_id: "2119934264489963758"
---
## Source
- URL: https://x.com/jedrij/status/2120024334764757058
- Author: Jedidiah (@jedrij)
- Posted: 2026-11-10 05:22:44
## Branch
**1/** **@jedrij** ^2120024334764757058
**@nicbarkeragain**
Is it even worth it at this point when we have better designed languages?
## Related
- Spine: [[archive/threads/nicbarkeragain/2026-11-09-happy-to-admit-that-using-a-minimal-subset-of-c]]
@@ -0,0 +1,41 @@
---
title: "@nicbarkeragain Are the compile times comparable to C using this subset? "
type: archive
source: twitter
source_url: "https://x.com/joakin/status/2120084187965239732"
author: "Joaquin"
handle: joakin
post_id: "2120084187965239732"
date: 2026-11-10
archived: 2026-11-15
draft: false
tags:
- archive
- twitter
- nicbarkeragain
description: "@nicbarkeragain Are the compile times comparable to C using this subset? "
in_reply_to: ""
parent_post_id: "2119934264489963758"
---
## Source
- URL: https://x.com/joakin/status/2120084187965239732
- Author: Joaquin (@joakin)
- Posted: 2026-11-10 09:20:34
## Branch
**1/** **@joakin** ^2120084187965239732
**@nicbarkeragain**
Are the compile times comparable to C using this subset?
I found this very relevant blog https://blog.s-schoener.com/2025-08-13-c-cpp-compile-time
It even has data from raddbg, when it moved to C from C-like-C++
## Related
- Spine: [[archive/threads/nicbarkeragain/2026-11-09-happy-to-admit-that-using-a-minimal-subset-of-c]]
@@ -0,0 +1,37 @@
---
title: "@nicbarkeragain bounds checked indexing with just a little operator overloading sounds like the sweet spot, you get the safety without all the heavy stuff imo"
type: archive
source: twitter
source_url: "https://x.com/justmetolife/status/2120198927840674086"
author: "anish"
handle: justmetolife
post_id: "2120198927840674086"
date: 2026-11-10
archived: 2026-11-15
draft: false
tags:
- archive
- twitter
- nicbarkeragain
description: "@nicbarkeragain bounds checked indexing with just a little operator overloading sounds like the sweet spot, you get the safety without all the heavy stuff imo"
in_reply_to: ""
parent_post_id: "2119934264489963758"
---
## Source
- URL: https://x.com/justmetolife/status/2120198927840674086
- Author: anish (@justmetolife)
- Posted: 2026-11-10 16:56:30
## Branch
**1/** **@justmetolife** ^2120198927840674086
**@nicbarkeragain**
bounds checked indexing with just a little operator overloading sounds like the sweet spot, you get the safety without all the heavy stuff imo
## Related
- Spine: [[archive/threads/nicbarkeragain/2026-11-09-happy-to-admit-that-using-a-minimal-subset-of-c]]
@@ -0,0 +1,37 @@
---
title: "@nicbarkeragain"
type: archive
source: twitter
source_url: "https://x.com/lanyuezhai/status/2119956983742963951"
author: "From Zero"
handle: lanyuezhai
post_id: "2119956983742963951"
date: 2026-11-10
archived: 2026-11-15
draft: false
tags:
- archive
- twitter
- nicbarkeragain
description: "@nicbarkeragain"
in_reply_to: ""
parent_post_id: "2119934264489963758"
---
## Source
- URL: https://x.com/lanyuezhai/status/2119956983742963951
- Author: From Zero (@lanyuezhai)
- Posted: 2026-11-10 00:55:06
## Branch
**1/** **@lanyuezhai** ^2119956983742963951
**@nicbarkeragain**
![](https://pbs.twimg.com/media/HWuZoQDbEAEjuKx?format=jpg&name=orig)
## Related
- Spine: [[archive/threads/nicbarkeragain/2026-11-09-happy-to-admit-that-using-a-minimal-subset-of-c]]
@@ -0,0 +1,37 @@
---
title: "@nicbarkeragain tbh I'm using exactly this for decades."
type: archive
source: twitter
source_url: "https://x.com/leonard_coder/status/2120205816200319210"
author: "Arnaud Carré"
handle: leonard_coder
post_id: "2120205816200319210"
date: 2026-11-10
archived: 2026-11-15
draft: false
tags:
- archive
- twitter
- nicbarkeragain
description: "@nicbarkeragain tbh I'm using exactly this for decades."
in_reply_to: ""
parent_post_id: "2119934264489963758"
---
## Source
- URL: https://x.com/leonard_coder/status/2120205816200319210
- Author: Arnaud Carré (@leonard_coder)
- Posted: 2026-11-10 17:23:52
## Branch
**1/** **@leonard_coder** ^2120205816200319210
**@nicbarkeragain**
tbh I'm using exactly this for decades. I never went fully strict C99. (and also never went fully crazy fancy and useless c++ features)
## Related
- Spine: [[archive/threads/nicbarkeragain/2026-11-09-happy-to-admit-that-using-a-minimal-subset-of-c]]
@@ -0,0 +1,39 @@
---
title: "@nicbarkeragain You use the subset you require."
type: archive
source: twitter
source_url: "https://x.com/mac_logo/status/2120158863420719143"
author: "Graeme Bell 💙 🇺🇦"
handle: mac_logo
post_id: "2120158863420719143"
date: 2026-11-10
archived: 2026-11-15
draft: false
tags:
- archive
- twitter
- nicbarkeragain
description: "@nicbarkeragain You use the subset you require."
in_reply_to: ""
parent_post_id: "2119934264489963758"
---
## Source
- URL: https://x.com/mac_logo/status/2120158863420719143
- Author: Graeme Bell 💙 🇺🇦 (@mac_logo)
- Posted: 2026-11-10 14:17:18
## Branch
**1/** **@mac_logo** ^2120158863420719143
**@nicbarkeragain**
You use the subset you require. That's best we all do. No-one requires everything, everywhere, all the time.
C++ is a very big toolbox and if you try to use a chainsaw and a multimeter simultaneously, it doesn't make much sense, even if they're useful tools to have.
## Related
- Spine: [[archive/threads/nicbarkeragain/2026-11-09-happy-to-admit-that-using-a-minimal-subset-of-c]]
@@ -0,0 +1,37 @@
---
title: "@nicbarkeragain Template resolution plus overloads plus non-explicit conversions/constructors are already a foot-gatling-gun."
type: archive
source: twitter
source_url: "https://x.com/noop_dev/status/2120214760071286918"
author: "Boris Chuprin"
handle: noop_dev
post_id: "2120214760071286918"
date: 2026-11-10
archived: 2026-11-15
draft: false
tags:
- archive
- twitter
- nicbarkeragain
description: "@nicbarkeragain Template resolution plus overloads plus non-explicit conversions/constructors are already a foot-gatling-gun."
in_reply_to: ""
parent_post_id: "2119934268910772471"
---
## Source
- URL: https://x.com/noop_dev/status/2120214760071286918
- Author: Boris Chuprin (@noop_dev)
- Posted: 2026-11-10 17:59:25
## Branch
**1/** **@noop_dev** ^2120214760071286918
**@nicbarkeragain**
Template resolution plus overloads plus non-explicit conversions/constructors are already a foot-gatling-gun.
## Related
- Spine: [[archive/threads/nicbarkeragain/2026-11-09-happy-to-admit-that-using-a-minimal-subset-of-c]]
@@ -0,0 +1,37 @@
---
title: "@nicbarkeragain Yep essentially Google's C++ guide."
type: archive
source: twitter
source_url: "https://x.com/siregamelogic/status/2119945297539522857"
author: "Sire 🇺🇸🇷🇺"
handle: siregamelogic
post_id: "2119945297539522857"
date: 2026-11-10
archived: 2026-11-15
draft: false
tags:
- archive
- twitter
- nicbarkeragain
description: "@nicbarkeragain Yep essentially Google's C++ guide."
in_reply_to: ""
parent_post_id: "2119934264489963758"
---
## Source
- URL: https://x.com/siregamelogic/status/2119945297539522857
- Author: Sire 🇺🇸🇷🇺 (@siregamelogic)
- Posted: 2026-11-10 00:08:40
## Branch
**1/** **@siregamelogic** ^2119945297539522857
**@nicbarkeragain**
Yep essentially Google's C++ guide. They forbid most of C++ modern atrocities.
## Related
- Spine: [[archive/threads/nicbarkeragain/2026-11-09-happy-to-admit-that-using-a-minimal-subset-of-c]]
@@ -0,0 +1,37 @@
---
title: "@nicbarkeragain https://github.com/takeiteasy/cccc i have a solution, templates are planned and operator overloading is already in."
type: archive
source: twitter
source_url: "https://x.com/takeiteqsy/status/2120089168940261711"
author: "habibi"
handle: takeiteqsy
post_id: "2120089168940261711"
date: 2026-11-10
archived: 2026-11-15
draft: false
tags:
- archive
- twitter
- nicbarkeragain
description: "@nicbarkeragain https://github.com/takeiteasy/cccc i have a solution, templates are planned and operator overloading is already in."
in_reply_to: ""
parent_post_id: "2119934264489963758"
---
## Source
- URL: https://x.com/takeiteqsy/status/2120089168940261711
- Author: habibi (@takeiteqsy)
- Posted: 2026-11-10 09:40:21
## Branch
**1/** **@takeiteqsy** ^2120089168940261711
**@nicbarkeragain**
https://github.com/takeiteasy/cccc i have a solution, templates are planned and operator overloading is already in. i have the same feelings you did.
## Related
- Spine: [[archive/threads/nicbarkeragain/2026-11-09-happy-to-admit-that-using-a-minimal-subset-of-c]]
@@ -0,0 +1,37 @@
---
title: "@nicbarkeragain The ideal minimum subset of C++ *is* C."
type: archive
source: twitter
source_url: "https://x.com/themistoyenes/status/2119946789835161996"
author: "Θεμιστογένης ὁ Συρακόσιος"
handle: themistoyenes
post_id: "2119946789835161996"
date: 2026-11-10
archived: 2026-11-15
draft: false
tags:
- archive
- twitter
- nicbarkeragain
description: "@nicbarkeragain The ideal minimum subset of C++ *is* C."
in_reply_to: ""
parent_post_id: "2119934264489963758"
---
## Source
- URL: https://x.com/themistoyenes/status/2119946789835161996
- Author: Θεμιστογένης ὁ Συρακόσιος (@themistoyenes)
- Posted: 2026-11-10 00:14:36
## Branch
**1/** **@themistoyenes** ^2119946789835161996
**@nicbarkeragain**
The ideal minimum subset of C++ *is* C.
## Related
- Spine: [[archive/threads/nicbarkeragain/2026-11-09-happy-to-admit-that-using-a-minimal-subset-of-c]]
@@ -0,0 +1,37 @@
---
title: "@nicbarkeragain turn off rtti, exceptions and no use of stl should be added."
type: archive
source: twitter
source_url: "https://x.com/thepythno/status/2120104439147008168"
author: "Michael Eggers"
handle: thepythno
post_id: "2120104439147008168"
date: 2026-11-10
archived: 2026-11-15
draft: false
tags:
- archive
- twitter
- nicbarkeragain
description: "@nicbarkeragain turn off rtti, exceptions and no use of stl should be added."
in_reply_to: ""
parent_post_id: "2119934264489963758"
---
## Source
- URL: https://x.com/thepythno/status/2120104439147008168
- Author: Michael Eggers (@thepythno)
- Posted: 2026-11-10 10:41:02
## Branch
**1/** **@thepythno** ^2120104439147008168
**@nicbarkeragain**
turn off rtti, exceptions and no use of stl should be added.
## Related
- Spine: [[archive/threads/nicbarkeragain/2026-11-09-happy-to-admit-that-using-a-minimal-subset-of-c]]
@@ -0,0 +1,37 @@
---
title: "@nicbarkeragain While C++ as a whole is a mess it contains an elegant and useful language within."
type: archive
source: twitter
source_url: "https://x.com/TaylorAllred8/status/2120609354676322512"
author: "Taylor Allred"
handle: TaylorAllred8
post_id: "2120609354676322512"
date: 2026-11-11
archived: 2026-11-15
draft: false
tags:
- archive
- twitter
- nicbarkeragain
description: "@nicbarkeragain While C++ as a whole is a mess it contains an elegant and useful language within."
in_reply_to: ""
parent_post_id: "2119934264489963758"
---
## Source
- URL: https://x.com/TaylorAllred8/status/2120609354676322512
- Author: Taylor Allred (@TaylorAllred8)
- Posted: 2026-11-11 20:07:23
## Branch
**1/** **@TaylorAllred8** ^2120609354676322512
**@nicbarkeragain**
While C++ as a whole is a mess it contains an elegant and useful language within. The trouble is finding that language.
## Related
- Spine: [[archive/threads/nicbarkeragain/2026-11-09-happy-to-admit-that-using-a-minimal-subset-of-c]]
@@ -0,0 +1,37 @@
---
title: "@nicbarkeragain And you still get unreadable compiler messages, mangled names, and need to keep headers in sync, yay!"
type: archive
source: twitter
source_url: "https://x.com/Voodoochilld/status/2120644111523057912"
author: "Voodoochild"
handle: Voodoochilld
post_id: "2120644111523057912"
date: 2026-11-11
archived: 2026-11-15
draft: false
tags:
- archive
- twitter
- nicbarkeragain
description: "@nicbarkeragain And you still get unreadable compiler messages, mangled names, and need to keep headers in sync, yay!"
in_reply_to: ""
parent_post_id: "2119934264489963758"
---
## Source
- URL: https://x.com/Voodoochilld/status/2120644111523057912
- Author: Voodoochild (@Voodoochilld)
- Posted: 2026-11-11 22:25:30
## Branch
**1/** **@Voodoochilld** ^2120644111523057912
**@nicbarkeragain**
And you still get unreadable compiler messages, mangled names, and need to keep headers in sync, yay!
## Related
- Spine: [[archive/threads/nicbarkeragain/2026-11-09-happy-to-admit-that-using-a-minimal-subset-of-c]]
@@ -0,0 +1,37 @@
---
title: "@nicbarkeragain i use this and i call it c+- pronounced \"see plus minus\" some people call it \"see plus\"."
type: archive
source: twitter
source_url: "https://x.com/wegfawefgawefg/status/2120501047596908677"
author: "wegfawefgawefg"
handle: wegfawefgawefg
post_id: "2120501047596908677"
date: 2026-11-11
archived: 2026-11-15
draft: false
tags:
- archive
- twitter
- nicbarkeragain
description: "@nicbarkeragain i use this and i call it c+- pronounced \"see plus minus\" some people call it \"see plus\"."
in_reply_to: ""
parent_post_id: "2119934264489963758"
---
## Source
- URL: https://x.com/wegfawefgawefg/status/2120501047596908677
- Author: wegfawefgawefg (@wegfawefgawefg)
- Posted: 2026-11-11 12:57:01
## Branch
**1/** **@wegfawefgawefg** ^2120501047596908677
**@nicbarkeragain**
i use this and i call it c+- pronounced "see plus minus" some people call it "see plus". turn on all the WALPEDANTIC stuff and it becomes rustlite also
## Related
- Spine: [[archive/threads/nicbarkeragain/2026-11-09-happy-to-admit-that-using-a-minimal-subset-of-c]]
@@ -0,0 +1,46 @@
---
title: "Happy to admit that using a minimal subset of C++ with:"
type: archive
source: twitter
source_url: "https://x.com/nicbarkeragain/status/2119934264489963758"
author: "Nic Barker"
handle: nicbarkeragain
post_id: "2119934264489963758"
date: 2026-11-09
archived: 2026-11-15
draft: false
tags:
- archive
- twitter
- nicbarkeragain
description: "Happy to admit that using a minimal subset of C++ with:"
in_reply_to: ""
---
## Source
- URL: https://x.com/nicbarkeragain/status/2119934264489963758
- Author: Nic Barker (@nicbarkeragain)
- Posted: 2026-11-09 23:24:49
## Thread
**1/** **@nicbarkeragain** ^2119934264489963758
Happy to admit that using a minimal subset of C++ with:
- simple templates for arrays / allocators / DS etc
- function name overloading
- minimal operator overloading for vector math and bounds-checked array index []
is a very pleasant experience! not perfect but close.
Branches: [[archive/threads/nicbarkeragain/2026-11-09-happy-to-admit-that-using-a-minimal-subset-of-c/2026-11-09-QuantumG-i-was-once-given-a-codebase-with-a-lot-of]], [[archive/threads/nicbarkeragain/2026-11-09-happy-to-admit-that-using-a-minimal-subset-of-c/2026-11-09-oldfalsash-i-think-twitter-c-developers-need-to-have-a-you]], [[archive/threads/nicbarkeragain/2026-11-09-happy-to-admit-that-using-a-minimal-subset-of-c/2026-11-10-siregamelogic-yep-essentially-googles-c-guide-they-forbid-most]], [[archive/threads/nicbarkeragain/2026-11-09-happy-to-admit-that-using-a-minimal-subset-of-c/2026-11-10-themistoyenes-the-ideal-minimum-subset-of-c-is-c]], [[archive/threads/nicbarkeragain/2026-11-09-happy-to-admit-that-using-a-minimal-subset-of-c/2026-11-10-caballerobrah-stl-is-pretty-nice-at-least-the-containers-really]], [[archive/threads/nicbarkeragain/2026-11-09-happy-to-admit-that-using-a-minimal-subset-of-c/2026-11-10-billyfscd-if-i-may-a-small-subset-chosen-with-care-is-a]], [[archive/threads/nicbarkeragain/2026-11-09-happy-to-admit-that-using-a-minimal-subset-of-c/2026-11-10-lanyuezhai-thread]], [[archive/threads/nicbarkeragain/2026-11-09-happy-to-admit-that-using-a-minimal-subset-of-c/2026-11-10-FUZxxl-this-might-be-an-original-workable-third-style-of]], [[archive/threads/nicbarkeragain/2026-11-09-happy-to-admit-that-using-a-minimal-subset-of-c/2026-11-10-0xglitchbyte-it-does-the-job-nicely-orthodox-c-was-a]], [[archive/threads/nicbarkeragain/2026-11-09-happy-to-admit-that-using-a-minimal-subset-of-c/2026-11-10-dcolascione-its-strange-that-people-act-like-the-moment-you]], [[archive/threads/nicbarkeragain/2026-11-09-happy-to-admit-that-using-a-minimal-subset-of-c/2026-11-10-alexlee20122022-bounds-checked-is-doing-most-of-the-work-here]], [[archive/threads/nicbarkeragain/2026-11-09-happy-to-admit-that-using-a-minimal-subset-of-c/2026-11-10-NotAttained-have-you-tried-jai]], [[archive/threads/nicbarkeragain/2026-11-09-happy-to-admit-that-using-a-minimal-subset-of-c/2026-11-10-jedrij-is-it-even-worth-it-at-this-point-when-we-have]], [[archive/threads/nicbarkeragain/2026-11-09-happy-to-admit-that-using-a-minimal-subset-of-c/2026-11-10-agriffini-100-agree-function-overloading-itself-is]], [[archive/threads/nicbarkeragain/2026-11-09-happy-to-admit-that-using-a-minimal-subset-of-c/2026-11-10-aordring-low-key-sticking-to-a-safe-house-subset-of-c-c]], [[archive/threads/nicbarkeragain/2026-11-09-happy-to-admit-that-using-a-minimal-subset-of-c/2026-11-10-joakin-are-the-compile-times-comparable-to-c-using-this]], [[archive/threads/nicbarkeragain/2026-11-09-happy-to-admit-that-using-a-minimal-subset-of-c/2026-11-10-takeiteqsy-https-github-com-takeiteasy-cccc-i-have-a]], [[archive/threads/nicbarkeragain/2026-11-09-happy-to-admit-that-using-a-minimal-subset-of-c/2026-11-10-AiKathG-i-like-the-opposite-subset-heavy-template-minimal]], [[archive/threads/nicbarkeragain/2026-11-09-happy-to-admit-that-using-a-minimal-subset-of-c/2026-11-10-guillaumecherea-that-works-for-sure-but-only-if-you-have-the]], [[archive/threads/nicbarkeragain/2026-11-09-happy-to-admit-that-using-a-minimal-subset-of-c/2026-11-10-thepythno-turn-off-rtti-exceptions-and-no-use-of-stl-should]], [[archive/threads/nicbarkeragain/2026-11-09-happy-to-admit-that-using-a-minimal-subset-of-c/2026-11-10-chrisfujino-you-can-keep-function-overloading-if-i-get]], [[archive/threads/nicbarkeragain/2026-11-09-happy-to-admit-that-using-a-minimal-subset-of-c/2026-11-10-NyuNighteyes-this-is-exactly-what-i-am-using-as-well-rust-felt]], [[archive/threads/nicbarkeragain/2026-11-09-happy-to-admit-that-using-a-minimal-subset-of-c/2026-11-10-mac_logo-you-use-the-subset-you-require-thats-best-we-all]], [[archive/threads/nicbarkeragain/2026-11-09-happy-to-admit-that-using-a-minimal-subset-of-c/2026-11-10-TheElfDobbie-many-companies-have-their-standards-on-what-to]], [[archive/threads/nicbarkeragain/2026-11-09-happy-to-admit-that-using-a-minimal-subset-of-c/2026-11-10-Talljackcv-the-buttons-other-people-press-point-in-your]], [[archive/threads/nicbarkeragain/2026-11-09-happy-to-admit-that-using-a-minimal-subset-of-c/2026-11-10-justmetolife-bounds-checked-indexing-with-just-a-little]], [[archive/threads/nicbarkeragain/2026-11-09-happy-to-admit-that-using-a-minimal-subset-of-c/2026-11-10-leonard_coder-tbh-im-using-exactly-this-for-decades-i-never]], [[archive/threads/nicbarkeragain/2026-11-09-happy-to-admit-that-using-a-minimal-subset-of-c/2026-11-10-MrJayLC-if-you-like-the-document-or-guarantees-by]], [[archive/threads/nicbarkeragain/2026-11-09-happy-to-admit-that-using-a-minimal-subset-of-c/2026-11-10-carkdev-i-went-the-other-way-c-c-and-loving-it-minimal]], [[archive/threads/nicbarkeragain/2026-11-09-happy-to-admit-that-using-a-minimal-subset-of-c/2026-11-10-fernandp-this-is-casey-being-templates]], [[archive/threads/nicbarkeragain/2026-11-09-happy-to-admit-that-using-a-minimal-subset-of-c/2026-11-11-wegfawefgawefg-i-use-this-and-i-call-it-c-pronounced-see-plus]], [[archive/threadsLine truncated
**2/** **@nicbarkeragain** ^2119934266536837513
The main thing that feels like a significant downgrade from C99 is that designated initializers in C++ are _terrible_. Forcing the fields to be specified in order, lack of nested `.field.subField = value` style access, no [enum] = value in arrays etc really hurt the DX.
**3/** **@nicbarkeragain** ^2119934268910772471
Of course the fact that it still uses C's #include machinery is objectively terrible. By far the biggest quality of life upgrade when you switch from C/++ to something like Odin or Jai is how you compose files together IMO.
Branches: [[archive/threads/nicbarkeragain/2026-11-09-happy-to-admit-that-using-a-minimal-subset-of-c/2026-11-09-JunkScribe_-one-thing-ive-found-with-odin-is-the-no-circular]], [[archive/threads/nicbarkeragain/2026-11-09-happy-to-admit-that-using-a-minimal-subset-of-c/2026-11-09-damster101-you-can-use-c-modules]], [[archive/threads/nicbarkeragain/2026-11-09-happy-to-admit-that-using-a-minimal-subset-of-c/2026-11-10-bitsgopew-c-modules-were-supposed-to-solve-this-but-oh-well]], [[archive/threads/nicbarkeragain/2026-11-09-happy-to-admit-that-using-a-minimal-subset-of-c/2026-11-10-borrowck_novel-i-will-assume-that-you-know-the-one-or-the-other]], [[archive/threads/nicbarkeragain/2026-11-09-happy-to-admit-that-using-a-minimal-subset-of-c/2026-11-10-noop_dev-template-resolution-plus-overloads-plus-non]]
+4
View File
@@ -19,3 +19,7 @@ tags:
- [[archive/threads/nicbarkeragain/2026-08-20-i-dev-with-the-debugger-attached-100-of-the-time]]
- [[archive/threads/nicbarkeragain/2026-08-25-i-think-one-of-the-biggest-level-ups-i-had-in]]
- [[archive/threads/nicbarkeragain/2026-08-28-similar-to-string-tags-if-your-program-has]]
- [[archive/threads/nicbarkeragain/2026-09-04-its-interesting-to-consider-that]]
- [[archive/threads/nicbarkeragain/2026-10-14-ive-been-mulling-over-peters-thoughts-in-this]]
- [[archive/threads/nicbarkeragain/2026-10-29-one-of-the-most-difficult-and-rewarding-parts-of]]
- [[archive/threads/nicbarkeragain/2026-11-09-happy-to-admit-that-using-a-minimal-subset-of-c]]