Viewing profile — nonethewiser
nonethewiser
HN member- Joined
- Thu, Mar 25, 2021, 2:44 AM UTC
- HN karma
- 8,095
- Public activity
- 4,494 items
- HN profile
- View on Hacker News ↗
About nonethewiser
No profile information was provided.
Recent public activity
-
comment
Comment #49249403
It seems like theyd have incentive to
-
comment
Comment #49242697
Any universal system would. The problem is there isnt a universal system.
-
comment
Comment #49237927
A tabletop adventure game generator. Instead of tediously rolling for disconnected content in limited ways (enemy to fight, loot after fight, etc), you generate an adventure and pr…
-
comment
Comment #49228408
Seems reliable and effective
-
comment
Comment #49224648
Half of women leave by age 35. How many men leave by age 35? Whats the baseline? Why doesn’t the article state this? Is 50% low?
-
comment
Comment #49200395
Countries have militaries.
-
comment
Comment #49200374
[flagged]
-
comment
Comment #49200358
Except pushing it to the logical extreme is what makes it wrong. Nevermind that he literally was equating the two and explicitly says so.
-
comment
Comment #49197898
[flagged]
-
comment
Comment #49167273
It also says they declined to say how many the US has left. And saying “person familiar with the matter” is doing a lot of work. It could be anyone with no direct information. This…
-
comment
Comment #49167237
One begins by looking at the sources which the article dies not substantiate.
-
comment
Comment #49167212
With stories like this you always have to ask “Who are the sources?” “People who are familiar with this matter” is doing a lot of work. This reeks of laundering lies.
-
comment
Comment #49162523
Who is more productive? A python dev who learns about x86 registers in his free time or one that uses AI at work?
-
comment
Comment #49162507
Learn what? The thing the tool can do instantly? Take the win and spend your energy on bigger problems.
-
comment
Comment #49162504
Why do you think thats worth learning? Doing a few times manually sure… but to force yourself to solely rely on it is foolish. Just use the calculator.
-
comment
Comment #49162495
For many cases like the one described, its true.
-
comment
Comment #49123895
Yes. And all the “context engineering”, document/decision frameworks etc just make it way harder to work with the ai. You put in a lot of work just for an illusion of control.
-
comment
Comment #49123874
No because the CLAUDE.md is included in the first prompt which is given priority in the context. On prompt 20, it will remember message 1 better than message 12.
-
comment
Comment #49122067
Did it ever tell you not to optimize what you should redesign?
-
comment
Comment #49119854
Quite literally the most authentic version. Not my preference but it’s the canonical form.
-
comment
Comment #49119801
But Warsh wants to unload the Feds balance sheet so this is very much in agreement with his policy.
-
comment
Comment #49119756
[dead]
-
comment
Comment #49119500
How is this different than creating a feature branch off main then branching off that?
-
comment
Comment #49098230
This is not what the article is talking about. Its talking about policy documents not it forgetting something 5 prompts ago. In fact you adding things to CLAUDE.md is more what its…
-
comment
Comment #49098197
In my experience, the more structure you enforce on models, the worse they perform and the less they actually do what you want.