Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
reedlaw
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
7 ms
·
61.
▲
by
reedlaw
11mo ago
I feel sorry for juniors because they have even less incentive to troubleshoot or learn languages. At the same time, the sheer size of APIs make me relieved that I will never have to remember another command, DSL, or argument list again. Ru
62.
▲
by
reedlaw
11mo ago
The probability space of a 140-nucleotide chain is 10^84. The estimated number of atoms in the universe is 10^80. The hypothesized RNA self-replicator is far simpler than the 238,000 base pair archaeal genome. But how are they formed? Even
63.
▲
by
reedlaw
11mo ago
Agreed, in the sense that software development is an iterative process of trial and error similar to custom car building. But software isn't like a car. It's more like a car factory. It should produce something of value repeatedly
64.
▲
by
reedlaw
1y ago
https://news.ycombinator.com/item?id=44220245 on the same topic was posted four months ago and has more substance than this short 2007 post.
65.
▲
by
reedlaw
1y ago
My point was that AI-produced code is not so foreign than no human could produce it, nor do any two humans produce the same style of code. So I'm not sure exactly what the idea of "engineering brand" is meant to protect.
66.
▲
by
reedlaw
1y ago
Isn't the intelligence of every other person alien to ourselves? The article ends with a need to "protect our own engineering brands" but how is that communicated? I found this [ https://meta.discourse.org/t&#x
67.
▲
by
reedlaw
1y ago
I found this: https://joeyh.name/blog/pics/solar_fence_with_powerracks.jpg
68.
▲
by
reedlaw
1y ago
Not an Old Testament reference, but a Gnostic belief. 1 John addresses Gnosticism by refuting some of its claims such as Jesus only appearing to be human. While John doesn't reference the Demiurge, the concept goes hand in hand with th
69.
▲
by
reedlaw
1y ago
Similar prompts didn't work for me.
70.
▲
by
reedlaw
1y ago
I couldn't get past Level 42: Reverse Turing using my own responses. Gemini's responses fooled it, which makes it a poor judge of humanness.
71.
▲
by
reedlaw
1y ago
Ironic to see this juxtaposed with another front-page story, "We put agentic AI browsers to the test – They clicked, they paid, they failed". The closing thoughts in the linked article ("feeling the AGI" and "very b
72.
▲
by
reedlaw
1y ago
I switched from Ubuntu to Arch because of the quality of the Arch wiki. Ubuntu search results were filled with so much misinformation that I realized it was a culture problem. Arch, with its higher barrier to entry, attracts users who maint
73.
▲
by
reedlaw
1y ago
Same here. I ordered from the link in the sibling comment. Mine took a few tries to connect to wifi.
74.
▲
by
reedlaw
1y ago
Emacs is practically an operating system (the vim joke being that it lacks a good editor). With git integration through magit, LSP server for language integration, and Projectile for project management, it very much acts like an IDE.
75.
▲
by
reedlaw
1y ago
It depends. Rails is great for rapid application development. You probably don't want to decouple at that early stage. But when your app struggles with slow responses and job management becomes increasingly difficult, you begin to see
76.
▲
by
reedlaw
1y ago
Does it drive sales though? Windows XP remained popular long after Vista was introduced. It was seen as stable and familiar. I guess the difference is that in the Apple ecosystem, you don't have much choice over which OS version you us
77.
▲
by
reedlaw
1y ago
App store software should have minimal warehousing and shipping costs compared with physical software. I was just trying to come up with a ball-park comparable figure.
78.
▲
by
reedlaw
1y ago
Amazon charges 15% for physical media, including software [1]. Since Apple is not shipping physical media, it seems something less than 15% would be more reasonable. 1. https://sellercentral.amazon.com/help/hub/ref
79.
▲
by
reedlaw
1y ago
"Dataset" connotes training data, but this seems to generate sample data, maybe for testing an application. Is there any use for synthetic datasets in ML?
80.
▲
by
reedlaw
1y ago
The use case in the article is relatively simple. For more complex structures, BAML ( https://www.boundaryml.com/ ) is a better option.
81.
▲
by
reedlaw
1y ago
The authors of those works (Pride and Prejudice, War and Peace, and The Brothers Karamazov) were greatly influenced by Greek classics, Milton, Shakespeare, or their contemporaries (such as Charles Dickens) who were also shaped by some of th
82.
▲
by
reedlaw
1y ago
In case you're wondering what it takes to run it, the answer is 8x H200 141GB [1] which costs $250k [2]. 1. https://github.com/MiniMax-AI/MiniMax-M1/issues/2#issuecomme... 2. https://www.ebay.
83.
▲
by
reedlaw
1y ago
Sounds like https://delta.chat/
84.
▲
by
reedlaw
1y ago
So this implements SD3 inference and fine-tuning?
85.
▲
by
reedlaw
1y ago
I'm not sure what this means. If it means the Stable Diffusion 3.5 model, why is it fetching that here: https://github.com/yousef-rafat/miniDiffusion/blob/main/enco... The training dataset is very
86.
▲
by
reedlaw
1y ago
I had a similar reaction to Jeffries' Gilded Rose solution in Ruby [1]. After 13 blog posts he ended up with something [2] a lot less elegant than Victor Shepelev who also used TDD but came up with a one-shot solution [3]. Just like No
87.
▲
by
reedlaw
2y ago
Writing software is more akin to building an engine factory than a single engine. At least, much of business software is used over a long period with lots of changes needed. Maybe writing a single game is more comparable to the race engine.
88.
▲
by
reedlaw
2y ago
I used something similar based on https://karl-voit.at/folder-hierarchy/ which I found helpful.
89.
▲
by
reedlaw
2y ago
Social is the key word. The author presumes the end goal of programming is overcoming all sorts of technical obstacles to deliver a working solution in a problem domain. But who defines working? The stakeholders. Ultimately, programming wor
90.
▲
by
reedlaw
2y ago
If developers factored the true cost of features into estimates then a lot less would get built. Instead, SMBs with tight budgets and long lists of feature requests go with the lowest bid or shortest time frame. Developers are incentivized
More ›