From Newsgroup: sci.physics



Hi,

So StackOverflow has already fallen, and
GitHub will be the next one. StackOverflow
was eclectic, insuinated a high Signal
quality but repelled its newcomers by
strick language rules and deletism.

StackOverlow is suplaned by ChatGPT, etc..
They are more tolerant and can deliver
excellent Signals, much beter than
StackOverflow. ChatGPT and other assistants
flipped the model: No downvotes.

No rCLduplicate questionrCY shaming. Conversational,
exploratory, and often faster than Googling +
scanning SO threads. Most importantly: they donrCOt
punish incomplete knowledge, which is where
most human learning happens.

LLMs give a more forgiving learning curve.

Bye

Mild Shock schrieb:

Henri Poincar|- believed that mathematical
and scientific creativity came from a deep,
unconscious intuition that could not be

captured by mechanical reasoning or formal
systems. He famously wrote about how insights
came not from plodding logic but from sudden

illuminations rCo leaps of creative synthesis.

But now we have generative AI rCo models like GPT rCo that:

- produce poetry, proofs, stories, and code,

- combine ideas in novel ways,

- and do so by processing patterns in massive
-a-a datasets, without conscious understanding.

And that does seem to contradict Poincar|-'s belief
that true invention cannot come from automation.

--- Synchronet 3.21a-Linux NewsLink 1.2

From Newsgroup: sci.physics

Hi,

Why have even a classification of code
into Python, JavaScript, Java, etc.. ?
They are just some form. Not content.

But what is content? The last 30 years
saw it that repository spurred
collaboration, because I can go home
forget everything and next morning,

first do a repo refresh and pick a
ticket and dig into something. So the
no code companion repository will do
the same, like it does now.

It will not be solely intent based.
Goals can be highly ambigious. I do
not think the future AGI based repos
will be intent based. They have to grasp

the full BDI, believe desire intent
of their users. So with a cognitive twin,
and some trust, I might even delegate some
work completely or multiply my ego a 100

times and delegate. Delegate to swarm
was always my dream.

Bye

Mild Shock schrieb:

Hi,

So StackOverflow has already fallen, and
GitHub will be the next one. StackOverflow
was eclectic, insuinated a high Signal
quality but repelled its newcomers by
strick language rules and deletism.

StackOverlow is suplaned by ChatGPT, etc..
They are more tolerant and can deliver
excellent Signals, much beter than
StackOverflow. ChatGPT and other assistants
flipped the model: No downvotes.

No rCLduplicate questionrCY shaming. Conversational,
exploratory, and often faster than Googling +
scanning SO threads. Most importantly: they donrCOt
punish incomplete knowledge, which is where
most human learning happens.

LLMs give a more forgiving learning curve.

Bye

Mild Shock schrieb:

Henri Poincar|- believed that mathematical
and scientific creativity came from a deep,
unconscious intuition that could not be

captured by mechanical reasoning or formal
systems. He famously wrote about how insights
came not from plodding logic but from sudden

illuminations rCo leaps of creative synthesis.

But now we have generative AI rCo models like GPT rCo that:

- produce poetry, proofs, stories, and code,

- combine ideas in novel ways,

- and do so by processing patterns in massive
-a-a-a datasets, without conscious understanding.

And that does seem to contradict Poincar|-'s belief
that true invention cannot come from automation.

--- Synchronet 3.21a-Linux NewsLink 1.2

From Newsgroup: sci.physics

On 7/15/25 1:58 PM, Mild Shock wrote:

Hi,

So StackOverflow has already fallen,

Those petty thief-Pajeets were already as good as dead back when my
alias was Excelfreak.

How did your kibbutz fare, Hanson? Any of them with Moses right now by "Iran-backed" donations knocking on their doors?


--- Synchronet 3.21a-Linux NewsLink 1.2

From Newsgroup: sci.physics

On 7/15/25 1:58 PM, Mild Shock wrote:

Hi,

So StackOverflow has already fallen, and
GitHub will be the next one. StackOverflow
was eclectic, insuinated a high Signal
quality but repelled its newcomers by
strick language rules and deletism.

StackOverlow is suplaned by ChatGPT, etc..
They are more tolerant and can deliver
excellent Signals, much beter than
StackOverflow. ChatGPT and other assistants
flipped the model: No downvotes.

No rCLduplicate questionrCY shaming. Conversational,
exploratory, and often faster than Googling +
scanning SO threads. Most importantly: they donrCOt
punish incomplete knowledge, which is where
most human learning happens.

LLMs give a more forgiving learning curve.

Bye

True. I quit COLA as a result of the same thing. A DeepSeek (and any
other Chinese AI services) replaced those useless dorks of the first
kind in that forum with excellent Linux assistance and troubleshooting
for me.

How many kids did you get to kill in Ghaza, Hanson? You sound more
positive compared to the wreck that you were before such opportunities
for you came up.




--- Synchronet 3.21a-Linux NewsLink 1.2