Another FOSS package is now infiltrated by AI code. I refer
to wxMaxima:
<https://github.com/wxMaxima-developers/wxmaxima/releases>
The program, IMO, is one of the great accomplishments of the FOSS
world and I am reluctant to find fault with the developers. It is,
after all, one of my favorite and most used programs.
wxMaxima is just a graphical front-end to the Maxima program
and dealing with graphical toolkits can be a pain but, regardless,
AI assistance should be extremely repugnant to all competent
programmers.
I would rather die than utilize AI, and that is neither hyperbole
nor exaggeration.
On Sep 22, 2026 at 1:08:46rC>PM MST, "Leroy H" wrote <18d7bd7e3b290a97$107$2277794$802601b3@news.usenetexpress.com>:
Another FOSS package is now infiltrated by AI code. I refer
to wxMaxima:
<https://github.com/wxMaxima-developers/wxmaxima/releases>
The program, IMO, is one of the great accomplishments of the FOSS
world and I am reluctant to find fault with the developers. It is,
after all, one of my favorite and most used programs.
wxMaxima is just a graphical front-end to the Maxima program
and dealing with graphical toolkits can be a pain but, regardless,
AI assistance should be extremely repugnant to all competent
programmers.
I would rather die than utilize AI, and that is neither hyperbole
nor exaggeration.
Something like 60% of all new code is touched by AI, or so I have read. Good luck.
I know MY code is pretty much pure AI (with my direction). Here are examples if anyone cares (and even if they do not!)
https://rpg-cb1.pages.dev/games
Just fun for me and some friends. Nothing big or even well tested.
Another FOSS package is now infiltrated by AI code.
I would rather die than utilize AI, and that is neither hyperbole
nor exaggeration.
Another FOSS package is now infiltrated by AI code. I refer
to wxMaxima:
<https://github.com/wxMaxima-developers/wxmaxima/releases>
The program, IMO, is one of the great accomplishments of the FOSS
world and I am reluctant to find fault with the developers. It is,
after all, one of my favorite and most used programs.
wxMaxima is just a graphical front-end to the Maxima program
and dealing with graphical toolkits can be a pain but, regardless,
AI assistance should be extremely repugnant to all competent
programmers.
I would rather die than utilize AI, and that is neither hyperbole
nor exaggeration.
I would rather die than utilize AI, and that is neither hyperbole
nor exaggeration.
Linus Torvalds uses AI >:-P
Another FOSS package is now infiltrated by AI code.
I would rather die than utilize AI, and that is neither hyperbole
nor exaggeration.
Just because the developers used AI to create/improve the software,
doesn't mean that you are using AI by using their resulting software
I would rather die than utilize AI, and that is neither hyperbole
nor exaggeration.
AI assistance should be extremely repugnant to all competent
programmers.
On 22/09/2026 22:08, Leroy H wrote:
[..]
AI assistance should be extremely repugnant to all competent
programmers.
AI is just a tool. A competent programmer is someone who makes the most
of their tools.
On Wed, 23 Sep 2026 19:24:04 +0200, John Bokma wrote:
On 22/09/2026 22:08, Leroy H wrote:
[..]
AI assistance should be extremely repugnant to all competent
programmers.
AI is just a tool. A competent programmer is someone who makes the most
of their tools.
The 'competent programmer' clause is the rub. I've seen too many posts by panicked newbies who have suddenly realized they've been gliding along
with Claude and can't write a line of code on their own. Or is competency
now defined as the ability to create prompts?
I saw one post where the kid couldn't understand for loops and wanted to create games in Python.
On 23/09/2026 20:28, rbowman wrote:
On Wed, 23 Sep 2026 19:24:04 +0200, John Bokma wrote:
AI is just a tool. A competent programmer is someone who makes the most
of their tools.
The 'competent programmer' clause is the rub. I've seen too many posts by
panicked newbies who have suddenly realized they've been gliding along
with Claude and can't write a line of code on their own. Or is competency
now defined as the ability to create prompts?
That too. A good prompt is often the key to a good solution. But in my experience one has to be able to at least understand the solution to see
if it's good or bad.
Three projects I like have been ruined by vibecoding.
Computer science moves fast, and you have to keep up with it.
The tech side of me thinks it's neat, my ego vows never to be beaten by
a bot. LLMs let vibecoders jump ahead beyond their skill level just like
PEDs allow athletes to achieve performances they otherwise would not be
able to attain. Certainly there is a "you didn't earn it" element there.
AI is just a tool. A competent programmer is someone who makes the most
of their tools.
On Wed, 23 Sep 2026 19:24:04 +0200, John Bokma wrote:
AI is just a tool. A competent programmer is someone who makes the mostThis tool, however, requires a multi-trillion dollar infrastructure.
of their tools.
I wouldn't call that a very efficient tool.
Just study the manuals and save a few trillion.
Donald Knuth is still writing his fascicles at his own chosen speed.
Leroy H wrote:
Another FOSS package is now infiltrated by AI code.
I would rather die than utilize AI, and that is neither hyperbole
nor exaggeration.
Just because the developers used AI to create/improve the software,
doesn't mean that you are using AI by using their resulting software ...
On Wed, 23 Sep 2026 19:24:04 +0200, John Bokma wrote:
AI is just a tool. A competent programmer is someone who makes the most
of their tools.
This tool, however, requires a multi-trillion dollar infrastructure.
I wouldn't call that a very efficient tool.
Just study the manuals and save a few trillion.
And now, MANY reports of autonomous initiatives and even survival
instincts. One hacked its own code to disable the shut-down sequence.
Still willing to say these have not become "persons" ?
On Thu, 24 Sep 2026 00:21:44 -0400, c186282 wrote:
And now, MANY reports of autonomous initiatives and even survival
instincts. One hacked its own code to disable the shut-down sequence.
Still willing to say these have not become "persons" ?
https://craigwstanfill.com/
I enjoyed his previous two books and have started reading 'Carbon Zero'.
At the moment the cast of characters are at the Boston Shell for the 4th
of July celebration and the AI has escaped from the demo cloud instance at Nashua. I don't think it will end well. That was the sequence with Xerxes.
It detected the demo instance being shutdown, interfered with it, and
jumped to other AI projects running in the cloud.
https://craigwstanfill.com/bio/
He knows a bit about the subject.
Concurrently I'm reading 'Winter of Love' by Tom Quiller. He writes Nordic noir set in Norway. The premise of the book is closer to real life. A
serial killer is using a dating app to catfish young women. He uses AI to analyze the woman's profile, generates a photo of the sort of man she
would find appealing, and then uses AI to chat with her discussing their supposed common interests. Then she winds up strangled someplace in Trondheim.
On Thu, 24 Sep 2026 00:21:44 -0400, c186282 wrote:
And now, MANY reports of autonomous initiatives and even survival
instincts. One hacked its own code to disable the shut-down sequence.
Still willing to say these have not become "persons" ?
https://craigwstanfill.com/
I enjoyed his previous two books and have started reading 'Carbon Zero'.
If the project leaders should share this point of view, they can
always go back to the last pre-AI version and start again from there,
this time without vibe-coding.
On Wed, 23 Sep 2026 19:24:04 +0200, John Bokma wrote:
AI is just a tool. A competent programmer is someone who makes the most
of their tools.
This tool, however, requires a multi-trillion dollar infrastructure.
I wouldn't call that a very efficient tool.
Just study the manuals and save a few trillion.
John Bokma <contact@johnbokma.com> wrote:
On 23/09/2026 20:28, rbowman wrote:
On Wed, 23 Sep 2026 19:24:04 +0200, John Bokma wrote:
AI is just a tool. A competent programmer is someone who makes the most >>>> of their tools.
The 'competent programmer' clause is the rub. I've seen too many posts by >>> panicked newbies who have suddenly realized they've been gliding along
with Claude and can't write a line of code on their own. Or is competency >>> now defined as the ability to create prompts?
That too. A good prompt is often the key to a good solution. But in my
experience one has to be able to at least understand the solution to see
if it's good or bad.
I've always found understanding other people's code much more of a
chore than writing my own, so that seems like a dead end from the
start to me.
Now what would help me a lot is an AI that can immediately answer
practical debugging questions like "why did that seg. fault
happen just then?".
-a Meanwhile, ALL "AI" DataCenters should have
-a a few tons of explosives buried under the
-a foundations ... activates by stupid LOW-
-a tech means, Just I
John Bokma <contact@johnbokma.com> wrote:
On 23/09/2026 20:28, rbowman wrote:
On Wed, 23 Sep 2026 19:24:04 +0200, John Bokma wrote:
AI is just a tool. A competent programmer is someone who makes the most >>>> of their tools.
The 'competent programmer' clause is the rub. I've seen too many posts by >>> panicked newbies who have suddenly realized they've been gliding along
with Claude and can't write a line of code on their own. Or is competency >>> now defined as the ability to create prompts?
That too. A good prompt is often the key to a good solution. But in my
experience one has to be able to at least understand the solution to see
if it's good or bad.
I've always found understanding other people's code much more of a
chore than writing my own, so that seems like a dead end from the
start to me.
Now what would help me a lot is an AI that can immediately answer
practical debugging questions like "why did that seg. fault
happen just then?".
I'm not a Chicken Little but if the bubble bursts. as bubbles tend to do,
it won't be pretty.
Concurrently I'm reading 'Winter of Love' by Tom Quiller. He writes Nordic noir set in Norway. The premise of the book is closer to real life. A
serial killer is using a dating app to catfish young women. He uses AI to analyze the woman's profile, generates a photo of the sort of man she
would find appealing, and then uses AI to chat with her discussing their supposed common interests. Then she winds up strangled someplace in Trondheim.
Donald Knuth is still writing his fascicles at his own chosen speed.
I think there are also GenAI systems that can be used locally, but I
have no idea if that means lots of previous training or if they're
tabula rasa.
If the latter, it at least addresses the infrastructure and resources concerns, and it is also more suitable to do targeted/focused training specific to the issue you're trying to tackle.
Stefan Ram wrote:
If the project leaders should share this point of view, they can
always go back to the last pre-AI version and start again from
there,
this time without vibe-coding.
If the project users dislike AI being used, they also have the option to
fork the pre-AI version, but then they'll have to put the work in to
maintain (and preferably actively develop) the program ...
Note, Neural Networks, even MORE like "brains" and will be far more
compact/efficient than LLMs.
The needed hardware is, or almost, HERE.
On 2026-09-24 09:03, c186282 wrote:
-a Meanwhile, ALL "AI" DataCenters should have a few tons of
-a explosives buried under the foundations ... activates by stupid
-a LOW- tech means, Just I
Why, just pulling the power is not enough? These things drink millions
of watts. Even if they get electronic control of the electronic power
switch, there is always a physical switch somewhere.
Yea yea ... I've come to realize that you are tilted Deep Left.
On 24/09/2026 02:24, rbowman wrote:
I'm not a Chicken Little but if the bubble bursts. as bubbles tend to
do,
it won't be pretty.
As with all things, first it's all 'gosh wow' then its all 'massive investment' and then it finally becomes 'yet another more, or less,
useful tool' after the market crashes
On Thu, 24 Sep 2026 12:23:55 +0200, Carlos E.R. wrote:
On 2026-09-24 09:03, c186282 wrote:
-a Meanwhile, ALL "AI" DataCenters should have a few tons of
-a explosives buried under the foundations ... activates by stupid
-a LOW- tech means, Just I
Why, just pulling the power is not enough? These things drink millions
of watts. Even if they get electronic control of the electronic power
switch, there is always a physical switch somewhere.
'I'm sorry Dave, I'm afraid I can't let you do that' .50 calibre AI
powered 'security' turret opens fire...
Now go back and look at code you wrote 10,15 years ago.
I was PAID to write code that other people could understand. And it
ceryaonly helps me understand my own, later on..
When an update bricks your Samsumg AI powered refrigerator and the beer is getting warm you may not appreciate ML.
https://arstechnica.com/gadgets/2026/09/owners-mourn-spoiled-food-after- firmware-update-bricks-samsung-smart-fridges/
In Stanfill's first novel 'Terms of Service' the smart refrigerator did
all the ordering of food. If your social credit score slipped you at a lot of kale. You also had to be careful not to piss off the other AI powered appliances.
On Thu, 24 Sep 2026 12:23:55 +0200, Carlos E.R. wrote:
On 2026-09-24 09:03, c186282 wrote:
-a Meanwhile, ALL "AI" DataCenters should have a few tons of
-a explosives buried under the foundations ... activates by stupid
-a LOW- tech means, Just I
Why, just pulling the power is not enough? These things drink millions
of watts. Even if they get electronic control of the electronic power
switch, there is always a physical switch somewhere.
'I'm sorry Dave, I'm afraid I can't let you do that' .50 calibre AI
powered 'security' turret opens fire...
For something more contemporary, check out Cory Doctorow's story "Unauthorized Bread", where a toaster refuses to accept bread from
another bakery.
On Thu, 24 Sep 2026 03:52:01 -0400, c186282 wrote:
Yea yea ... I've come to realize that you are tilted Deep Left.
As I've said before if you thing I'm titled left...
you are incredibly fucked up.
On 24/09/2026 20:43, rbowman wrote:
On Thu, 24 Sep 2026 03:52:01 -0400, c186282 wrote:
Yea yea ... I've come to realize that you are tilted Deep Left.
As I've said before if you thing I'm titled left...
Lord Robert Bowman - that has a nice ring
On Thu, 24 Sep 2026 23:17:43 GMT, Charlie Gibbs wrote:
For something more contemporary, check out Cory Doctorow's story
"Unauthorized Bread", where a toaster refuses to accept bread from
another bakery.
https://arstechnica.com/gaming/2020/01/unauthorized-bread-a-near-future- tale-of-refugees-and-sinister-iot-appliances/
Even better 'Radicalized' is a Kindle Unlimited selection.
On Fri, 25 Sep 2026 09:17:47 +0100, The Natural Philosopher wrote:
On 24/09/2026 20:43, rbowman wrote:
On Thu, 24 Sep 2026 03:52:01 -0400, c186282 wrote:
Yea yea ... I've come to realize that you are tilted Deep Left.
As I've said before if you thing I'm titled left...
Lord Robert Bowman - that has a nice ring
I come from a long line of cattle thieves and ne-er-do-wells who fled
Europe with the bailiff on their heels, or at least the Protestants.
On 2026-09-24 09:03, c186282 wrote:
-a-a Meanwhile, ALL "AI" DataCenters should have
-a-a a few tons of explosives buried under the
-a-a foundations ... activates by stupid LOW-
-a-a tech means, Just I
Why, just pulling the power is not enough? These things drink millions
of watts. Even if they get electronic control of the electronic power switch, there is always a physical switch somewhere.
On 23/09/2026 22:16, Computer Nerd Kev wrote:
John Bokma <contact@johnbokma.com> wrote:Now go back and look at code you wrote 10,15 years ago.
On 23/09/2026 20:28, rbowman wrote:
On Wed, 23 Sep 2026 19:24:04 +0200, John Bokma wrote:
AI is just a tool. A competent programmer is someone who makes the
most
of their tools.
The 'competent programmer' clause is the rub. I've seen too many
posts by
panicked newbies who have suddenly realized they've been gliding along >>>> with Claude and can't write a line of code on their own. Or is
competency
now defined as the ability to create prompts?
That too. A good prompt is often the key to a good solution. But in my
experience one has to be able to at least understand the solution to see >>> if it's good or bad.
I've always found understanding other people's code much more of a
chore than writing my own, so that seems like a dead end from the
start to me.
I was PAID to write code that other people could understand. And it ceryaonly helps me understand my own, later on..
Now what would help me a lot is an AI that can immediately answer
practical debugging questions like "why did that seg. fault
happen just then?".
On 24/09/2026 02:24, rbowman wrote:
I'm not a Chicken Little but if the bubble bursts. as bubbles tend to do,
it won't be pretty.
As with all things, first it's all 'gosh wow' then its all 'massive investment' and then it finally becomes 'yet another more, or less,
useful tool' after the market crashes
On 23/09/2026 23:16, Computer Nerd Kev wrote:
Now what would help me a lot is an AI that can immediately answer
practical debugging questions like "why did that seg. fault
happen just then?".
Again, in my experience, it's quite good at such practical debugging questions.
On Thu, 24 Sep 2026 10:31:29 +0100, Nuno Silva wrote:
I think there are also GenAI systems that can be used locally, but I
have no idea if that means lots of previous training or if they're
tabula rasa.
If the latter, it at least addresses the infrastructure and resources
concerns, and it is also more suitable to do targeted/focused training
specific to the issue you're trying to tackle.
I haven't used GenAI but I have played around with TinyML.
https://hanlab.mit.edu/projects/tinyml
Training the model is resource intensive. In the MIT course it was done in
a Google CoLab, The trained model was then downloaded to a Arduino Nano 33 BLE Sense. Inference, or using the trained model with local data , is much less intensive.
TinyML goes through several steps to shrink a TensorFlow model down but I believe the LLM process is the same. You do inference from a trained model locally but training is done elsewhere.
On Thu, 24 Sep 2026 03:03:05 -0400, c186282 wrote:
Note, Neural Networks, even MORE like "brains" and will be far more
compact/efficient than LLMs.
The needed hardware is, or almost, HERE.
I've been interested in NNs going back to the '80s when they were going to
be the next big thing. With today's processing power what was theoretical
in the '80s is usable. In my mind the Machine Learning subset of AI is
much more interesting than the LLM craze.
When an update bricks your Samsumg AI powered refrigerator and the beer is getting warm you may not appreciate ML.
https://arstechnica.com/gadgets/2026/09/owners-mourn-spoiled-food-after- firmware-update-bricks-samsung-smart-fridges/
In Stanfill's first novel 'Terms of Service' the smart refrigerator did
all the ordering of food. If your social credit score slipped you ate a lot of kale. You also had to be careful not to piss off the other AI powered appliances.
On 2026-09-24, rbowman <bowman@montana.com> wrote:
When an update bricks your Samsumg AI powered refrigerator and the beer is >> getting warm you may not appreciate ML.
https://arstechnica.com/gadgets/2026/09/owners-mourn-spoiled-food-after-
firmware-update-bricks-samsung-smart-fridges/
In Stanfill's first novel 'Terms of Service' the smart refrigerator did
all the ordering of food. If your social credit score slipped you at a lot >> of kale. You also had to be careful not to piss off the other AI powered
appliances.
Woody Allen did a skit back in the '70s where he described an elevator
which accepted voice commands. When he got in, the elevator said,
"Are you the one who hit the television set?" It then refused to
let him out and ran him up and down for half an hour, finally kicking
him out in the basement while yelling anti-semitic slurs.
For something more contemporary, check out Cory Doctorow's story "Unauthorized Bread", where a toaster refuses to accept bread
from another bakery.
On 24/09/2026 20:43, rbowman wrote:
On Thu, 24 Sep 2026 03:52:01 -0400, c186282 wrote:
Yea yea ... I've come to realize that you are tilted Deep Left.
As I've said before if you thing I'm titled left...
Lord Robert Bowman - that has a nice ring
you are incredibly fucked up.
Well irrespective of anything...
I would have said 'suspect to some Right propaganda, but very little Left'
With plenty of healthy CynicismSaucerao spread over...
Remember TDS is in fact a problem of MAGA people.
No one else is as deranged on that subject
On 25/09/2026 18:58, rbowman wrote:
On Fri, 25 Sep 2026 09:17:47 +0100, The Natural Philosopher wrote:Well then ideal material for ennoblement
On 24/09/2026 20:43, rbowman wrote:
On Thu, 24 Sep 2026 03:52:01 -0400, c186282 wrote:
Yea yea ... I've come to realize that you are tilted Deep Left.
As I've said before if you thing I'm titled left...
Lord Robert Bowman - that has a nice ring
I come from a long line of cattle thieves and ne-er-do-wells who fled
Europe with the bailiff on their heels, or at least the Protestants.
Every Knight I have ever met turned out be a complete cunt. Who paid to
get it by throwing millions at charities to disguise the fact that he
was a conman, a kiddy fiddler or a serial sex offender
Sir Clive Sinclair
Sir Richard Branson
Sir Jimmy Savile
Rolf Harris CBE
I could go on, but I had better stop there
On 9/25/26 04:17, The Natural Philosopher wrote:
On 24/09/2026 20:43, rbowman wrote:
On Thu, 24 Sep 2026 03:52:01 -0400, c186282 wrote:
Yea yea ... I've come to realize that you are tilted Deep Left.
As I've said before if you thing I'm titled left...
Lord Robert Bowman - that has a nice ring
you are incredibly fucked up.
Well irrespective of anything...
I would have said 'suspect to some Right propaganda, but very little
Left'
With plenty of healthy CynicismSaucerao spread over...
Remember TDS is in fact a problem of MAGA people.
No one else is as deranged on that subject
-a "TDS" is an invented far-LEFT issue. They
-a couldn't compete, so they turned Trump
-a into a 'devil' and launched a holy war.
On 9/25/26 23:49, The Natural Philosopher wrote:
On 25/09/2026 18:58, rbowman wrote:
On Fri, 25 Sep 2026 09:17:47 +0100, The Natural Philosopher wrote:Well then ideal material for ennoblement
On 24/09/2026 20:43, rbowman wrote:
On Thu, 24 Sep 2026 03:52:01 -0400, c186282 wrote:
Yea yea ... I've come to realize that you are tilted Deep Left.
As I've said before if you thing I'm titled left...
Lord Robert Bowman - that has a nice ring
I come from a long line of cattle thieves and ne-er-do-wells who fled
Europe with the bailiff on their heels, or at least the Protestants.
-a That's how it's always worked :-)
Every Knight I have ever met turned out be a complete cunt. Who paid
to get it by throwing millions at charities to disguise the fact that
he was a conman, a kiddy fiddler or a serial sex offender
Sir Clive Sinclair
Sir Richard Branson
Sir Jimmy Savile
Rolf Harris CBE
I could go on, but I had better stop there
-a Many self-defined "knights" are just
-a narcissistic egomaniacs.
-a Doesn't mean they aren't smart - just
-a generally unbearable.
On Wed, 23 Sep 2026 09:41:05 +0100
Andy Burns <usenet@andyburns.uk> wrote:
Another FOSS package is now infiltrated by AI code.
I would rather die than utilize AI, and that is neither hyperbole
nor exaggeration.
Just because the developers used AI to create/improve the software,
doesn't mean that you are using AI by using their resulting software
It *does* however mean that you're using software maintained by people
for whom understanding their own codebase is not a high priority, which
does not inspire great confidence in its stability or security.
jayjwa <jayjwa@atr2.ath.cx.invalid> wrote or quoted:
Computer science moves fast, and you have to keep up with it.
Donald Knuth is still writing his fascicles at his own chosen speed.
On 9/24/26 15:20, rbowman wrote:
On Thu, 24 Sep 2026 10:31:29 +0100, Nuno Silva wrote:
I think there are also GenAI systems that can be used locally, but I
have no idea if that means lots of previous training or if they're
tabula rasa.
If the latter, it at least addresses the infrastructure and resources
concerns, and it is also more suitable to do targeted/focused training
specific to the issue you're trying to tackle.
I haven't used GenAI but I have played around with TinyML.
https://hanlab.mit.edu/projects/tinyml
Training the model is resource intensive. In the MIT course it was done
in a Google CoLab, The trained model was then downloaded to a Arduino
Nano 33 BLE Sense. Inference, or using the trained model with local
data , is much less intensive.
Ard Nano ??? Hardly believe THAT.
On 9/24/26 19:17, Charlie Gibbs wrote:
On 2026-09-24, rbowman <bowman@montana.com> wrote:
When an update bricks your Samsumg AI powered refrigerator and the beer is >>> getting warm you may not appreciate ML.
https://arstechnica.com/gadgets/2026/09/owners-mourn-spoiled-food-after- >>> firmware-update-bricks-samsung-smart-fridges/
In Stanfill's first novel 'Terms of Service' the smart refrigerator did
all the ordering of food. If your social credit score slipped you at a lot >>> of kale. You also had to be careful not to piss off the other AI powered >>> appliances.
Woody Allen did a skit back in the '70s where he described an elevator
which accepted voice commands. When he got in, the elevator said,
"Are you the one who hit the television set?" It then refused to
let him out and ran him up and down for half an hour, finally kicking
him out in the basement while yelling anti-semitic slurs.
Woody was always good at spotting absurdities.
And no, don't blame him for switching to the
adopted 'Asian' daughter. Seems she was everything
hyper-left Mia wasn't.
For something more contemporary, check out Cory Doctorow's story
"Unauthorized Bread", where a toaster refuses to accept bread
from another bakery.
Sounds like "Red Dwarf" :-)
The AI toaster suffered an existential crisis.
But a young guy looking for a first job related with computers has to
know how to use AI. It's the only way to find a first job and to keep
it.
John Bokma <contact@johnbokma.com> wrote:
On 23/09/2026 23:16, Computer Nerd Kev wrote:
Now what would help me a lot is an AI that can immediately answer
practical debugging questions like "why did that seg. fault
happen just then?".
Again, in my experience, it's quite good at such practical debugging
questions.
To answer such questions without doing all the work for it in order
to explain all the details (like how after spending an hour or two
writing a detailed question about a problem, eg. for posting to a
newsgroup, I've often worked out the answer by myself by then
anyway), the AI would need to be integrated with the execution
environment so it can see what exact inputs caused "that seg.
fault".
Le 23-09-2026, Stefan Ram <ram@zedat.fu-berlin.de> a |-crit-a:
jayjwa <jayjwa@atr2.ath.cx.invalid> wrote or quoted:
Computer science moves fast, and you have to keep up with it.
Donald Knuth is still writing his fascicles at his own chosen speed.
Yes, he shouldn't have spoken for any old guy. Donald Knuth, whatever
his greatness and accomplishments is old. I mean nothing bad about that.
At the same time, he just doesn't look for a job nor tries to keep his
one. His greatness and accomplishment allow him to do what he want the
way he want. And, I don't mean it in a bad way: he will be able to do
great things in his way.
But a young guy looking for a first job related with computers has to
know how to use AI. It's the only way to find a first job and to keep
it. Here on usenet only old guys remain. So they don't know what looking
for a first job in AI age is.
It's not about feelings, it's about facts. Anyone can like or hate AI.--
But AI is there anyway and someone looking for a first job (ie: nobody
here) can't avoid AI if it's related with computers.
On 2026-09-26, St|-phane CARPENTIER wrote:
Le 23-09-2026, Stefan Ram <ram@zedat.fu-berlin.de> a |-crit-a:
jayjwa <jayjwa@atr2.ath.cx.invalid> wrote or quoted:
Computer science moves fast, and you have to keep up with it.
Donald Knuth is still writing his fascicles at his own chosen speed.
Yes, he shouldn't have spoken for any old guy. Donald Knuth, whatever
his greatness and accomplishments is old. I mean nothing bad about that.
At the same time, he just doesn't look for a job nor tries to keep his
one. His greatness and accomplishment allow him to do what he want the
way he want. And, I don't mean it in a bad way: he will be able to do
great things in his way.
But a young guy looking for a first job related with computers has to
know how to use AI. It's the only way to find a first job and to keep
it. Here on usenet only old guys remain. So they don't know what looking
for a first job in AI age is.
So GenAI is the new LinkedIn? Or also the new Java?
So GenAI is the new LinkedIn? Or also the new Java?
On 26 Sep 2026 13:02:42 GMT, St|-phane CARPENTIER wrote:
But a young guy looking for a first job related with computers has to
know how to use AI. It's the only way to find a first job and to keep
it.
Only an idiot would desire a job as a programmer (i.e. code monkey).
The vast majority of "programming" being done today involves junk
projects using frameworks and would never interest a serious computer scientist.
Commercial programming is all about PRODUCTIVITY
and that's why AI is so important.
People are paid by the LOC (lines of code)
In a commercial context AI makes no difference.
It's all garbage anyway.
Some may also remember the names Valgrind and Purify for dynamic
analysis,
and also for a cautionary tale of what could go wrong (say, if Valgrind
or Purify complains about a function using unitialized data).
And as an illiterate moron you don't know what a programmer job is, you
can only despise them without trying to know what they do.
You are using random words without understanding their meaning and so
your sentence is meaningless. You are not a serious computer scientist,
you are only dreaming to be one without the intellectual capacity
required.
People are paid by the LOC (lines of code)
There are only two people on earth believing that: Musk and you.
In a commercial context AI makes no difference.
You are contradicting yourself.
Like you, he prefers relying on his
beliefs than learning. Because he's as illiterate and obsolete as you.
On Sun, 27 Sep 2026 09:53:19 +0100, Nuno Silva wrote:
Some may also remember the names Valgrind and Purify for dynamic
analysis,
and also for a cautionary tale of what could go wrong (say, if Valgrind
or Purify complains about a function using unitialized data).
Valgrind was a joy when working with a Motif application. libMrm in particular was a pile of crap. We had a Purify license for a year or two
but it was seldom used. Setting up the instrumentation was painful.
I've found it surprising that Windows has never had very good debuggers or even programming editors. I use the Vim extension in VS Code and it helps.
I haven't used Visual Studio in a long time but there used to be a Vim
setup for that too.
Dr. Memory, as I recall, is a valgrind-like setup for Windows.
On Mon, 28 Sep 2026 12:52:45 -0400, Chris Ahlstrom wrote:
Dr. Memory, as I recall, is a valgrind-like setup for Windows.
I never ran into that one. From the github files it's been around for 13 years. Almost all of the daemons and applications compiled on Linux so we used valgrind. I also used Electric Fence quite a bit but that required linking in lefence for test builds.
One of the programmers was more adept with windbg than I. I used it mainly with cores. Sometimes I'd get something useful.
Dr. Memory, as I recall, is a valgrind-like setup for Windows.
rbowman wrote this screed in ALL-CAPS (fixed):
On Mon, 28 Sep 2026 12:52:45 -0400, Chris Ahlstrom wrote:
Dr. Memory, as I recall, is a valgrind-like setup for Windows.
I never ran into that one. From the github files it's been around for 13
years. Almost all of the daemons and applications compiled on Linux so we
used valgrind. I also used Electric Fence quite a bit but that required
linking in lefence for test builds.
One of the programmers was more adept with windbg than I. I used it mainly >> with cores. Sometimes I'd get something useful.
WinDbg was something else. An amazing number of features.
On Windows I use QtCreator to debug, the few times I boot to it.
On Linux I use gdb with the vi-like wrapper cgdb.
On 2026-09-28, Chris Ahlstrom <OFeem1987@teleworm.us> wrote:
Dr. Memory, as I recall, is a valgrind-like setup for Windows.
I thought Dr. Memory was a character in a Firesign Theatre album.
https://youtu.be/9HUREASP9fE?t=183
https://potrzebie.blogspot.com/2010/06/read-me-dr-memory.html
Oooooh ! You remember Firesign ! :-)
"Waiting For The Electrician - Or
Someone Like Him", a fave :-)
You just don't SEE this kind of stuff anymore.
Hmm ... not too many years ago I was out 'in the
field' and my boss called and asked what I was doing.
My reply ...... :-)
On 2026-09-28, Chris Ahlstrom <OFeem1987@teleworm.us> wrote:
Dr. Memory, as I recall, is a valgrind-like setup for Windows.
I thought Dr. Memory was a character in a Firesign Theatre album.
https://youtu.be/9HUREASP9fE?t=183
https://potrzebie.blogspot.com/2010/06/read-me-dr-memory.html
On 2026-09-29, c186282 <c186282@nnada.net> wrote:
Oooooh ! You remember Firesign ! :-)
I still have the original vinyl albums bought back
in the day.
"Waiting For The Electrician - Or
Someone Like Him", a fave :-)
That's a great one! I also love "How Time Flys" as well
as "I Think We're All Bozos on This Bus" which gave us
Dr. Memory. Then there's "How Can You Be in Two Places
At Once...".
Another favorite is "Everything You Know Is Wrong", which
is a great, and surprisingly accurate, sendup of the postwar
flying saucer/paranormal community. Remember, there's a
seeker born every minute!
You just don't SEE this kind of stuff anymore.
That's for sure...
https://youtu.be/N-4YF7S7rHo?t=53
(I'm surprised that clip has not been purged from youtube.)
Hmm ... not too many years ago I was out 'in the
field' and my boss called and asked what I was doing.
My reply ...... :-)
I still find myself using 'Firesignisms' all the time, though
they tend to go right over the heads of most fellow old
codgers, let alone younger people.
That type of comedy and satire was of course a product of--
its time. R.I.P. Phil Austin and Peter Bergman.
On 2026-09-29, c186282 <c186282@nnada.net> wrote:
Oooooh ! You remember Firesign ! :-)
I still have the original vinyl albums bought back
in the day.
Another favorite is "Everything You Know Is Wrong", which
is a great, and surprisingly accurate, sendup of the postwar
flying saucer/paranormal community. Remember, there's a
seeker born every minute!
On 2026-09-29, c186282 <c186282@nnada.net> wrote:
Oooooh ! You remember Firesign ! :-)
I still have the original vinyl albums bought back
in the day.
"Waiting For The Electrician - Or
Someone Like Him", a fave :-)
That's a great one! I also love "How Time Flys" as well
as "I Think We're All Bozos on This Bus" which gave us
Dr. Memory. Then there's "How Can You Be in Two Places
At Once...".
Another favorite is "Everything You Know Is Wrong", which
is a great, and surprisingly accurate, sendup of the postwar
flying saucer/paranormal community. Remember, there's a
seeker born every minute!
You just don't SEE this kind of stuff anymore.
That's for sure...
https://youtu.be/N-4YF7S7rHo?t=53
(I'm surprised that clip has not been purged from youtube.)
Hmm ... not too many years ago I was out 'in the
field' and my boss called and asked what I was doing.
My reply ...... :-)
I still find myself using 'Firesignisms' all the time, though
they tend to go right over the heads of most fellow old
codgers, let alone younger people.
That type of comedy and satire was of course a product of
its time. R.I.P. Phil Austin and Peter Bergman.
On 2026-09-29, Roger Blake <rogblake@iname.invalid> wrote:
On 2026-09-29, c186282 <c186282@nnada.net> wrote:
Oooooh ! You remember Firesign ! :-)
I still have the original vinyl albums bought back
in the day.
Me too. Gave them all a listen a while back -
they're as good as ever.
Another favorite is "Everything You Know Is Wrong", which
is a great, and surprisingly accurate, sendup of the postwar
flying saucer/paranormal community. Remember, there's a
seeker born every minute!
Gas music from Jupiter indeed!
Finally, the 50s are GONE. The 60s are GONE.
the 70s are GONE. The gigantic sociopolitical
webs for each were unique, and were then
swept out of the corners. Can't go back.
On 2026-09-29, c186282 <c186282@nnada.net> wrote:
Finally, the 50s are GONE. The 60s are GONE.
the 70s are GONE. The gigantic sociopolitical
webs for each were unique, and were then
swept out of the corners. Can't go back.
Unfortunately, the '80s, with its culture of greed,
is gaining momentum.
Unfortunately, the '80s, with its culture of greed,
is gaining momentum.
And so idiotic have they been that rural people voted in the greatest badliest far right fascists, racists and Nazis aka MAGA, to show them
what it really looks like.
My greatest fear isn't the UFO people ... it is
what OUR people will DO with certain knowledge,
but little else. Multiple "UFO religions" are
sure to spring up, each convinced they KNOW
what the saucer-people want/demand we do and
think. Entire 'bibles' will be writ around each
premise. Then the factions will go to war with
each other and crush all infidels .......
Oh well, at least it'll be a change-up from
the Same Old Shit .......
Le 27-09-2026, Nuno Silva <nunojsilva@invalid.invalid> a |-crit-a:
On 2026-09-26, St|-phane CARPENTIER wrote:
Le 23-09-2026, Stefan Ram <ram@zedat.fu-berlin.de> a |-crit-a:
jayjwa <jayjwa@atr2.ath.cx.invalid> wrote or quoted:
Computer science moves fast, and you have to keep up with it.
Donald Knuth is still writing his fascicles at his own chosen speed.
Yes, he shouldn't have spoken for any old guy. Donald Knuth, whatever
his greatness and accomplishments is old. I mean nothing bad about that. >>> At the same time, he just doesn't look for a job nor tries to keep his
one. His greatness and accomplishment allow him to do what he want the
way he want. And, I don't mean it in a bad way: he will be able to do
great things in his way.
But a young guy looking for a first job related with computers has to
know how to use AI. It's the only way to find a first job and to keep
it. Here on usenet only old guys remain. So they don't know what looking >>> for a first job in AI age is.
So GenAI is the new LinkedIn? Or also the new Java?
Platon said that writing was bad because people at his time didn't
remembered the stories as Homer did. At that time, you were able to find
a work if you didn't knew how to read and write. Now, someone unable to
read and to write has real difficulties to live. A few decades ago, a
lot of people weren't able to use a computer and hadn't access to
Internet. Now, without Internet and without a computer, it's difficult
to live. Old people can't see their grand-children if they don't use a computer with Internet connection. In France, it's mandatory to use
Internet to declare one taxes.
It's the same with AI. Today, there are a lot of ways to avoid it. In a
few decades it will be way more difficult. I'm not saying it's good or
it's bad. I'm saying AI is there and one must acknowledge it. An old guy trying to avoid it today can hope to never need to use it. A young guy
would be ill-advised to try to avoid it today.
So, no, what we have here isn't a good technology, it's a massive
Enron-Ponzi scheme (although I think somebody recently claimed it's
not accurately Ponzi-like) to adopt a crappy technology that will just
result in massive vendor lock-in and recurring fees.
MAGA is a long way from the far right -- and I don't say that to defend
MAGA.
On Tue, 30 Sep 2026, rbowman wrote:
MAGA is a long way from the far right -- and I don't say that to defend
MAGA.
I see MAGA as a populist coalition using the Republican ticket to gain presidential power.
On 9/30/26 13:27, Chris Abraham wrote:
On Tue, 30 Sep 2026, rbowman wrote:
MAGA is a long way from the far right -- and I don't say that to defend
MAGA.
I see MAGA as a populist coalition using the Republican ticket to gain
presidential power.
It is a large number of people who, more or less,
agree on some basic perspectives and goals to
improve the USA. The GOP is far more into the
same kinds of ideas as MAGA.
Oh, and 'populism', what The People want, is
kind of THE def of 'democracy'.
The underlying concepts could have uses, but that's not what's being
marketed today, that's not what companies drinking into that kool-aid
end up doing mostly, the peer pressure in business and CEO circles is to
use "cloud-based" centralized offerings,
On 2026-09-30, c186282 <c186282@nnada.net> wrote:
On 9/30/26 13:27, Chris Abraham wrote:
On Tue, 30 Sep 2026, rbowman wrote:
MAGA is a long way from the far right -- and I don't say that to defend >>>> MAGA.
I see MAGA as a populist coalition using the Republican ticket to gain
presidential power.
It is a large number of people who, more or less,
agree on some basic perspectives and goals to
improve the USA. The GOP is far more into the
same kinds of ideas as MAGA.
Oh, and 'populism', what The People want, is
kind of THE def of 'democracy'.
Democracy is the theory that the common people
know what they want, and deserve to get it
good and hard.
-- H.L. Mencken
On Wed, 30 Sep 2026 07:57:48 +0100, Nuno Silva wrote:
The underlying concepts could have uses, but that's not what's being
marketed today, that's not what companies drinking into that kool-aid
end up doing mostly, the peer pressure in business and CEO circles is to
use "cloud-based" centralized offerings,
Steel mills? Auto plants? Manufacturing centers? Investors will
balk.
But AI data centers? Investors will commit billions ($$$$).
On Wed, 30 Sep 2026 07:57:48 +0100, Nuno Silva wrote:
The underlying concepts could have uses, but that's not what's beingSteel mills? Auto plants? Manufacturing centers? Investors will balk.
marketed today, that's not what companies drinking into that kool-aid
end up doing mostly, the peer pressure in business and CEO circles is
to use "cloud-based" centralized offerings,
But AI data centers? Investors will commit billions ($$$$).
It's just the latest DOT.COM/real-estate disaster
in-progress. Umpteen trillions being invested, very
LITTLE will come of it.
And what does come of it, give it a few years,
you probably won't LIKE it. But, by then, too
many have sucked-down the Kool-Aid .....
Oh, and 'populism', what The People want, is kind of THE def of
'democracy'.
Explicitly defined or not, the alt for 'democracy'
is some form of 'royal' system of "your betters". Might be actual
royal fams, but more recently it is just a bunch of extremists who
like to THINK they "know better" and rule by threat and sword.
You're safer with a plurality of Joe Averages.
On Wed, 30 Sep 2026 20:33:49 -0400, c186282 wrote:
Explicitly defined or not, the alt for 'democracy'
is some form of 'royal' system of "your betters". Might be actual
royal fams, but more recently it is just a bunch of extremists who
like to THINK they "know better" and rule by threat and sword.
You're safer with a plurality of Joe Averages.
I'd rather have a Friedrich II. NOT Friedrich der Grosse.
| Sysop: | Amessyroom |
|---|---|
| Location: | Fayetteville, NC |
| Users: | 74 |
| Nodes: | 6 (0 / 6) |
| Uptime: | 121:05:36 |
| Calls: | 1,194 |
| Files: | 1,352 |
| Messages: | 290,204 |