The OFFICIAL programming thread
-
@Gators1 said in The OFFICIAL programming thread:
This should be in every programmer’s toolbox
Jeezus, I dunno whether I want to bring this to the attention of the Unix admins or not.
-
I wonder whether ChatGPT can correct syntax errors in MySQL statements?
One may write statements that are 90% correct, but they don’t work due to one’s not being an expert and their being an overlooked syntax error. We’ve found that even when accessing Oracle MySQL support, the process involves “try this.” Next day when that did not work, “OK, now try this instead.”
I haven’t tried AI for for anything yet, personally. It’s about time to experiment a bit.
I was listening to a report the other day which discussed AI being trained to interpret MRI data and this does seem like a good use. Use of AI in the medical field seems to have been established for reading breast exam tomography, along with a clinician’s opinion, of course. AI has not been permitted to fly solo yet in the medical MRI/CT field for results interpretation. Perhaps Kyle has seen this in action?
-
@Jam ChatGPT can both help you write the sql and help you fix it. If you give it the SQL and tell it “It gave me an ‘FU-666 - Your a dumbass’” error it can often tell you why. Or even if you tell it why the results are wrong it can help fix the logic. It’s often an iterative process though rather than just giving you the right answer at the beginning. Also Github copliot will help you write the code as you are first doing it in VSCODE. You can make a comment in your code and it can autogenerate the whole section or just start typing and it suggests what to type next for the rest of the line or whole SQL, function or whatever you are writing. Copilot has a chat option as well or you can highlight something and hit ctrl-i to get a box where you can tell it what to do contextually in your code.
I have had days where I was tired or braindead and just had the AI write the code and let me run and test it, then iterate until it’s right.
-
@Jam said in The OFFICIAL programming thread:
I wonder whether ChatGPT can correct syntax errors in MySQL statements?
One may write statements that are 90% correct, but they don’t work due to one’s not being an expert and their being an overlooked syntax error. We’ve found that even when accessing Oracle MySQL support, the process involves “try this.” Next day when that did not work, “OK, now try this instead.”
I haven’t tried AI for for anything yet, personally. It’s about time to experiment a bit.
I was listening to a report the other day which discussed AI being trained to interpret MRI data and this does seem like a good use. Use of AI in the medical field seems to have been established for reading breast exam tomography, along with a clinician’s opinion, of course. AI has not been permitted to fly solo yet in the medical MRI/CT field for results interpretation. Perhaps Kyle has seen this in action?
I’m sure it is in active assessment but it’s not regular practice. I would expect the UT people are working it.
-
I thought Linus Torvalds had mellowed in recent years but this is was just a week ago:
as a result your code IS GARBAGE. AGAIN.
lol.
From Linus Torvalds <>
Date Fri, 26 Jan 2024 12:25:05 -0800
Subject Re: [PATCH] eventfs: Have inodes have unique inode numbers shareSteven,
stop making things more complicated than they need to be.And dammit, STOP COPYING VFS LAYER FUNCTIONS.
It was a bad idea last time, it’s a horribly bad idea this time too.
I’m not taking this kind of crap.
The whole “get_next_ino()” should be “atomic64_add_return()”. End of story.
You arent’ special. If the VFS functions don’t work for you, you don’t
use them, but dammit, you also don’t then steal them without
understanding what they do, and why they were necessary.The reason get_next_ino() is critical is because it’s used by things
like pipes and sockets etc that get created at high rates, the the
inode numbers most definitely do not get cached.You copied that function without understanding why it does what it
does, and as a result your code IS GARBAGE.AGAIN.
Honestly, kill this thing with fire. It was a bad idea. I’m putting my
foot down, and you are NOT doing unique regular file inode numbers
uintil somebody points to a real problem.Because this whole “I make up problems, and then I write overly
complicated crap code to solve them” has to stop,.No more. This stops here.
I don’t want to see a single eventfs patch that doesn’t have a real
bug report associated with it. And the next time I see you copying VFS
functions (or any other core functions) without udnerstanding what the
f*ck they do, and why they do it, I’m going to put you in my
spam-filter for a week.I’m done. I’m really really tired of having to look at eventfs garbage.
Linus
-
I don’t know if I’m getting wiser or have just lost the energy to fight it (either way, I’m getting old) but re this comment he made:
Because this whole “I make up problems, and then I write overly complicated crap code to solve them” has to stop,.
I wouldn’t waste any energy on getting angry about that because it’s probably not in the guy’s nature to recognize he’s even doing it. Some people just do that - overcomplicate the fuck out of everything - and they couldn’t recognize the beauty in simplicity if it hit them in the head. Or they just don’t have the talent to see the real problem and they just keep furiously writing more and more code on top of failed approaches praying that the last fix on their fixes does the trick. I don’t know if you can change that either.
-
I didn’t understand any of that but still, fuck that guy!
-
@Hog said in The OFFICIAL programming thread:
I don’t know if I’m getting wiser or have just lost the energy to fight it (either way, I’m getting old) but re this comment he made:
Because this whole “I make up problems, and then I write overly complicated crap code to solve them” has to stop,.
I wouldn’t waste any energy on getting angry about that because it’s probably not in the guy’s nature to recognize he’s even doing it. Some people just do that - overcomplicate the fuck out of everything - and they couldn’t recognize the beauty in simplicity if it hit them in the head. Or they just don’t have the talent to see the real problem and they just keep furiously writing more and more code on top of failed approaches praying that the last fix on their fixes does the trick. I don’t know if you can change that either.
The two patterns I recognize in this direction are:
-
Require indirection everywhere. Solve a problem by writing a universal class that solves all possible problems and specializing it for the actual problem at hand. It may work, but you will never use it for something else, you shouldn’t use it for something else, and the next person that comes along will have no idea what the hell it does in the first place.
-
Think the problem is very simple. Write a logically wrong solution. Poke that solution with a stick until it is 7 times as long as you anticipated, calls itself recursively, and uses 3 external libraries someone on stack overflow has told you about. The right solution was to scrap it, doodle a bit on paper, and write completely new, short, and correct code.
-
-
Linus and Steve should try solving this in a cage fight.
-
@Pakoon said in The OFFICIAL programming thread:
Linus and Steve should try solving this in a cage fight.
lol, no joke, this is Steve:

-
@Hog said in The OFFICIAL programming thread:
@Pakoon said in The OFFICIAL programming thread:
Linus and Steve should try solving this in a cage fight.
lol, no joke, this is Steve:

Hmm in code fu likely Linus wins, but not actually fighting. Unless the penguins mass for him.
-
Linus’s wife was a black belt karate instructor, maybe she has taught Linus some tricks.
I attended a few of her lessons when I was 14 or something.
-
@Kilemall said in The OFFICIAL programming thread:
Unless the penguins mass for him.
I hate that penguin. I wish they’d stop infantilizing the Linux brand with that stupid thing.
Steve doesn’t hate it apparently since this is his GitHub avatar:

I’m feeling sorry for Steve now that I’ve looked him up. He’s obviously not a newbie nor a complete incompetent since he’s had hundreds or more commits merged into the kernel over the years by Torvalds. I’d struggle to just take destructive criticism like that in any setting, but to have my code called garbage (again) by someone as respected as Torvalds, in a public mailing list read by many thousands and then go viral so it was the first or second hit anytime someone searched my name…
-
Fucking Indians!
-
Would you like an orange squishy with that?
-
Maybe related but I watched a vid a couple of days ago about how a lot of younger programmers think “contributing to open source” is a necessary box you have to check off if you want to get hired these days. That attitude is attracting a lot of low value contributions from people who otherwise have no interest in the project they are contributing to.
Re Indian programmers: The client I work for hired one of the many guys I interviewed for them late last year and he started a couple of weeks ago. He’s turning out to be great. He’s pretty switched on and although still a little green in some areas, he really gets it and he learns quick. It’s such a fucking relief after having to work with the so many mostly low quality programmers from the IT managed services firm they use. I used to try to train those guys up but they’d fucking quit all the time because the conditions they worked under were so bad and they had no real connection or loyalty to the client. It’s nice to be able to mentor this new guy and not be afraid I’m wasting my time. He’ll likely take over some of the work I’ve been doing when my contract is up so it will also be nice to know that stuff will be in good hands.
-
@Hog said in The OFFICIAL programming thread:
Maybe related but I watched a vid a couple of days ago about how it’s considered by a lot of younger programs that “contributing to open source” is a necessary box you have to tick off if you want to get hired these days. That attitude is attracting a lot of low value contributions from people who otherwise have no interest in the project they are contributing to.
Re Indian programmers: The client I work for hired one of the many guys I interviewed for them late last year and he started a couple of weeks ago. He’s turning out to be great. He’s pretty switched on and although still a little green in some areas, he really gets it and he learns quick. It’s such a fucking relief after having to work with the so many mostly low quality programmers from the IT managed services firm they use. I used to try to train those guys up but they’d fucking quit all the time because the conditions they worked under were so bad and they had no real connection or loyalty to the client. It’s nice to be able to mentor this new guy and not be afraid I’m wasting my time. He’ll likely take over the work I’ve been doing when my contract is up so it will also be nice to know that stuff will be in good hands.
Ya succession is on my mind nowadays.
-
@Hog Apparently the blow up started with the “influencer” below. He was bitching about people trying to contribute to open source to buff their resumes or even get a free tshirt. So he posted a video about it that didn’t target Indians, but that spawned other videos and comments about Indians because they are significant offenders apparently. So then the influencer had to make another video saying “no racis” and then showing an Indian video teaching these people to go make contributions but half assed, like commenting their name in the code to get credit for the commit. There are some good Indian coders that I have worked with, but at the upper end I tend to hate working with their management and consultants (actually from India, not Americanized). It’s more a cultural issue than technical.
-
We dealt with the Tata people in the 80s when they were first starting out, and there is a SE Asian sub village in one of our analyst kingdoms. You end up dealing with their caste type stuff no matter what you do to step on it.
-
I deal with tatas daily
