Viewing profile — mr_toad
mr_toad
HN member- Joined
- Tue, Oct 31, 2017, 9:23 PM UTC
- HN karma
- 7,559
- Public activity
- 4,537 items
- HN profile
- View on Hacker News ↗
About mr_toad
No profile information was provided.
Recent public activity
-
comment
Comment #49241601
I think music videos, then MTV, then the internet/youtube all gave ways for fans to see a band without actually going to see a band. Live gigs are expensive, require a bit of plann…
-
comment
Comment #49228164
“But they can’t match Intels performance” used to be the rejoinder of the Intel crowd. Now we’ve come full circle.
-
comment
Comment #49167979
It’s not like stock photos of infeasibly attractive people eating ludicrously overstuffed burgers were ever interesting or believable.
-
comment
Comment #49167367
The US bombed North Vietnam into dust, but because they didn’t occupy it with ground forces they simply rebuilt and kept coming. You can’t win a war with just air power.
-
comment
Comment #49154532
> What the hell is the point of human interaction if the "human" just forwards a chat bot's response Forwarding an AI response (or any canned response) sends the message that they …
-
comment
Comment #49145977
Mods might be amused to be described as fairly godmothers.
-
comment
Comment #49144526
Glid vs glided, slid vs slided are examples of the inconsistency of the English language. Glided is more common than glid, and slid is more common than slided. Linguistics says tha…
-
comment
Comment #49137944
If you constrained the inputs of Apple or Nvidia then their sales and profits would tumble (just look at what happened to Apples stock price in the last week because of the memory …
-
comment
Comment #49126276
It’s an odd kind of bubble that slowly deflates over the course of a year.
-
comment
Comment #49122312
> Quantum experiments you're always looking at some medium that detects or measures or whatever the original interaction. So you're only interrogating the detector, not the origina…
-
comment
Comment #49122252
Observations keep poking holes in those hypotheses (I’m not sure they even count as theories). We keep finding black holes with anomalous masses, or black holes that formed too ear…
-
comment
Comment #49122045
Headline is unnecessary clickbait, the mean was still signaling that something was wrong.
-
comment
Comment #49073179
> What's there, that politicians don't know already? They need to be ‘reminded’ of the issues that ‘campaign contributions’ are contingent upon.
-
comment
Comment #49059392
Have you seen Microsoft’s price over the last year. If AI doesn’t pan out they’re going to be the next IBM.
-
comment
Comment #49059331
Installed by default is their moat. Trying to arrange a video conference on other platforms means making sure that everyone has the client installed, which can be a real headache. …
-
comment
Comment #49059271
> But you don´t get to to that. You can, and for tax purposes you must. The bean counters hate that you can’t deduct capex.
-
comment
Comment #49053015
We’re gonna need a bigger wall.
-
comment
Comment #49050476
“ It has not occurred to you that your ancestors were survivors and that the survival itself sometimes involved savage decisions, a kind of wanton brutality which civilized humanki…
-
comment
Comment #49047083
https://en.wikipedia.org/wiki/ArianeGroup
-
comment
Comment #49046971
> Non-exclusive use of farmland or wasteland (or even - at least until maintenance complexity is considered - offshore) isn't a major barrier Not now, but with solar power output d…
-
comment
Comment #49035504
> The reason they couldn't because they didn't have the metallurgy and the machining to create pistons, valves, etc. It was a turbine. You don’t use pistons in a turbine. They lack…
-
comment
Comment #49034843
It’s not just government money. Factory workers wages in China are still a lot lower than western countries, and they have a lot fewer labor rights. Also, the government hinders as…
-
comment
Comment #49015592
> On this benchmark, a pure LLM generated an accuracy score of zero. Adding RAG, prompt engineering, and agentic AI raised accuracy to the 10+% range. That's awful. The AI in Datab…
-
comment
Comment #49015319
The rush hours are the worst. You're lucky if you hear two songs in 30 minutes it's so crammed full of ads.
-
comment
Comment #48996428
> We can then add in an existing dataset of _known_ incidents, indexed by the same common features, as a training/validation set. Beware of the Anna Karenina principle. Well behave…