mvkel

13 ms·

241.
by mvkel 10mo ago
I largely agree. The product decision deficit at Apple is cultural, not due to any one person missing. Stephen Lemay, Dye's replacement, has been at Apple since 1999, yet was passed over for the design lead position multiple times. The
242.
by mvkel 10mo ago
One of the most egregious issues with macOS 26 is the accessibility/usability regression. Apple prided itself on making their operating system accessible. Good ux is inherently accessible. There are so many parts of the os that flagran
243.
by mvkel 10mo ago
Whenever I see headlines like this, I ask: what happened in 1994? It was post-Cold War and central banks were trimming USD reserves to test alternatives. Then, crises hit (tequila, Asian, Russian, dot com) and the world reconsolidated aroun
244.
by mvkel 10mo ago
It forces me to rotate the shapes of ideas; my beliefs become more durable when I'm willing to explore differing views through the act of writing (or it exposes how weak my initial take is)
245.
by mvkel 10mo ago
Given the near-impossibility of predicting something as "simple" as a stock market due to its recursive nature, I'm not sure I see how it would be possible to simulate an infinitely more complicated "world"
246.
by mvkel 10mo ago
Time heals all wounds, even self-inflicted ones. I don't sense much remorse from this guy.
247.
by mvkel 10mo ago
I have yet to see a webview that doesn't feel like a webview. In a native app, it inevitably ends up feeling like oil and water, empirically. Worth the extra effort to make it proper native, despite whatever promises the webview people
248.
by mvkel 10mo ago
the main summary of 1984: "neighbors are encouraged, via telesecreens, to spy on one another to enforce conformity." There thing to fear isn't some higher state; it's each other. We happily will surveil each other under
249.
by mvkel 10mo ago
Like any new social network, the first question to answer is: how do you make it useful -without- the social layer? Instagram did photo filters. LinkedIn did digital résumés. Strava did activity tracking. It's not zero sum. But if you&
250.
by mvkel 10mo ago
Should we also arrest computer co execs because computers are used to hack into things?
251.
by mvkel 10mo ago
The latter
252.
by mvkel 10mo ago
Is it because it's not on Amazon?
253.
by mvkel 10mo ago
The people who are inside the machine are usually the least qualified to predict the machine's future
254.
by mvkel 10mo ago
They absolutely are. NSA is obsessed with post-quantum projects atm
255.
by mvkel 10mo ago
This is like disqualifying Banksy because they use stencils
256.
by mvkel 10mo ago
This work is so critical. Read an article that was published just 10 years ago, and witness the bit rot as most external links will 404, gone forever. I think it's worth questioning the value of preserving -everything-, but it seems li
257.
by mvkel 10mo ago
Karpathy himself has used "vibe coding" to describe "usage of LLMs for coding," so it's fair to say the definition has expanded. https://karpathy.bearblog.dev/year-in-review-2025/
258.
by mvkel 10mo ago
On the tabs vs spaces battleground there are no winners; we just need to lower our expectations :)
259.
by mvkel 10mo ago
One only has to go read the original vibe coding thread[0] from ...ten months ago(!) to see the resistance and skepticism loud and clear. The very first comment couldn't be more loud about it. It was possible to create things in gpt-3.
260.
by mvkel 10mo ago
> In this world view, nano banana is a first early hint of what that might look like. What is he referring to here? Is nano banana not just an image gen model? Is it because it's an LLM-based one, and not diffusion?
261.
by mvkel 10mo ago
Haven't promoted it at all, beyond having a marketing site live on the web with competent SEO. I should say, it's not hugely popular or anything. But considering that I only built it for my own needs, I look at it more as art than
262.
by mvkel 10mo ago
These are separate internal teams, not one team shipping all these games. Your square list, for example, crosses 8 different studios. My original question was poorly worded; I meant how many games -per studio- are being shipped by these big
263.
by mvkel 10mo ago
Fascinating to see the increasing acceptance of AI generated code in HN comments. We've come a long way since gpt-3.5, and it's rewarding to see people who are willing to change their cached responses
264.
by mvkel 10mo ago
How many games has Valve shipped in the last decade? Square Enix? Bethesda? Epic?
265.
by mvkel 10mo ago
I think you'll find that most development -teams- ship about one game every decade. It's hard to find examples of that not being the case.
266.
by mvkel 10mo ago
Nice! Curious, what metrics did you land on? IF, VI, duration, etc., or an interval detector? That is the unfortunate thing about the serious cyclist industry; it's tiny! Not worth pursuing as a business unless the TAM is expanded broa
267.
by mvkel 10mo ago
TR's "AI" claims are a bit of a stretch. It basically just monitors fatigue and will ramp down a prescribed workout to ensure you don't overtrain. The workouts themselves are templates chosen from a list and do not adapt
268.
by mvkel 10mo ago
https://www.repth.com , an AI cycling coach that I vibe-coded back when gpt-3.5 was hot and vibe coding wasn't yet a thing. We've come a long way! Free for athletes, but I license the underlying "coach" logic
269.
by mvkel 10mo ago
This leaderboard feels incredibly accurate given my own experience.
270.
by mvkel 10mo ago
Correct, objectivity is not possible. Human observation is never perfectly neutral. What we call "objective" is usually just invisible judgment that aligns with our priors. An observer's choices about what to include, exclude

More ›