the "not worthy of it's own thread" thread

What do you use the Camo truck for?
 
My guess would be "logistics", if we ever have to invade Mexico again.
 
Reached into the fridge tonight and realized I have leftover beer from OSH, Texas, and 6Y9. Selected a "Wisconsin Amber".
 
Reached into the fridge tonight and realized I have leftover beer from OSH, Texas, and 6Y9. Selected a "Wisconsin Amber".
Bring ‘em all to 3M0. The only beer that’s escaped 6y9 with me were the NA IPAs.
 
Annual budget of 168k$ to operate.. a mechanic for $15-30 per hour? As an owner that's news to me.
... IDK... Thinking Google is going to need to go ahead and leave that Generative "AI" in the oven a while longer, not quite done yet... :rolleyes:

1694153470682.png
 
Last edited:
Annual budget of 168k$ to operate.. a mechanic for $15-30 per hour? As an owner that's news to me.
... IDK... Thinking Google is going to need to go ahead and leave that Generative "AI" in the oven a while longer, not quite done yet... :rolleyes:

View attachment 120430
"Info quality may vary" between hogwash, balderdash, and outright nonsense. Such has been my experience.
 
probably scraping the "wisdom" from a certain tools on youtube
 
One of the more significant problems with AI is that it will give an authoritative sounding answer, complete with justifications, that is completely false. Similar to that of the TV, Internet, or random person at a bar. I did a search for 'ww2 aircraft with lowest stall speed' and it gave the answer as Me-262, for a variety of reasons.

To me, it's not really AI. It's intelligent language processing. Again, similar to sending someone to school for journalism, they are excellent at presenting information in a clear way that they don't understand at all.
 
One of the more significant problems with AI is that it will give an authoritative sounding answer, complete with justifications, that is completely false. Similar to that of the TV, Internet, or random person at a bar. I did a search for 'ww2 aircraft with lowest stall speed' and it gave the answer as Me-262, for a variety of reasons.

To me, it's not really AI. It's intelligent language processing. Again, similar to sending someone to school for journalism, they are excellent at presenting information in a clear way that they don't understand at all.
I think that's the best articulation of the problem I've heard, yet. Hopefully you didn't use GPT to write it! Jk.

Whenever using GPTs API in a programmatic setting I always have a validation layer specifically to catch the BS. It's not bulletproof but it's something. And it's amazing how frequently it can catch itself spewing nonsense and making poor choices.
 
"Info quality may vary" between hogwash, balderdash, and outright nonsense. Such has been my experience.

Sometime ago, in another thread, I mentioned that some of this new AI stuff was simply trash i.e. GIGO (garbage in/out) and was promptly told that I had no idea what I was talking about. It may get there someday, but that ain't these days!

JOMHO ...
 
Chat GPI says that lift is generated by Bernouli's principle which involves the equal transit nonsense and that an airfoil has more curve on the top than the bottom.
Google maps knows you need to make a Michigan left but as you get to the U turn (while, for example, on Fort St.) it's like "Make a sharp left to stay on Fort St." HUH? It's a U turn for crying out loud. A left puts me in the Kroger parking lot.
 
Sometime ago, in another thread, I mentioned that some of this new AI stuff was simply trash i.e. GIGO (garbage in/out) and was promptly told that I had no idea what I was talking about. It may get there someday, but that ain't these days!

JOMHO ...
Yeah. I asked ChatGPT a few questions to which I knew the answers, and those answers are readily available and have been available with simple web searches for many years. The answers ChatGPT came up with were absolutely false in every respect.

So I told CGPT that was the wrong answer; it responded along the lines of, “Oh yeah, you’re right, here’s the correct answer” - followed by different but equally wrong nonsense.

It’s good at generating text that LOOKS like an answer. That seems to be about it.

I also asked it to write a function in C to do a particular task that I’ve written more than once. The code it generated was workable, but just about the least efficient way to do it. First year programmer quality stuff, and I’m in no way a professional programmer.

We hear the term “Zero Trust” used a lot these days… that’s how I feel about AI. Zero trust.

And get off my lawn.
 
Yeah. I asked ChatGPT a few questions to which I knew the answers, and those answers are readily available and have been available with simple web searches for many years. The answers ChatGPT came up with were absolutely false in every respect.

So I told CGPT that was the wrong answer; it responded along the lines of, “Oh yeah, you’re right, here’s the correct answer” - followed by different but equally wrong nonsense.

It’s good at generating text that LOOKS like an answer. That seems to be about it.

I also asked it to write a function in C to do a particular task that I’ve written more than once. The code it generated was workable, but just about the least efficient way to do it. First year programmer quality stuff, and I’m in no way a professional programmer.

We hear the term “Zero Trust” used a lot these days… that’s how I feel about AI. Zero trust.

And get off my lawn.
I've found it helpful for writing routine boilerplate python stuff that I've written countless times over the last 10+ years. But I don't trust it with anything I can't verify directly.
 
On that note one more thing... If you want to see something fun with coding... use GPT to write regular expressions. Before even checking the regex tell GPT it's answer is glaringly wrong (without providing a reason) while asking it to fix it. Do this 5x and you'll send it down a rabbit hole where it fixes things that aren't broken while breaking new things to the point it's unrecognizable
 
Yeah. I asked ChatGPT a few questions to which I knew the answers, and those answers are readily available and have been available with simple web searches for many years. The answers ChatGPT came up with were absolutely false in every respect.

So I told CGPT that was the wrong answer; it responded along the lines of, “Oh yeah, you’re right, here’s the correct answer” - followed by different but equally wrong nonsense.

It’s good at generating text that LOOKS like an answer. That seems to be about it.
Sounds like a politician! :biggrin:
 
I'm starting to believe in the "we're living in a simulation" theory.

This all being written by an AI would explain everything.
 
I'm starting to believe in the "we're living in a simulation" theory.

This all being written by an AI would explain everything.
I think we slipped into that dimension when they spooled CERN up
 
I've never seen a prop like this before with these parts (in red) extruding. Does anyone know what those are for, or what these props are called?
1694365762543.png
 
I've never seen a prop like this before with these parts (in red) extruding. Does anyone know what those are for, or what these props are called?
View attachment 120497

sorry, but that's thread-worthy

 
Got a call from the MKE FSDO yesterday but who answers their phone these days? They tried twice. Somehow they also have my email address and asked me to call. I give them credit for persistence.
 
Got a call from the MKE FSDO yesterday but who answers their phone these days? They tried twice. Somehow they also have my email address and asked me to call. I give them credit for persistence.
What did you do?!
 
A pilot who ran afoul of our local D airport tower ATC had to recently call not the FAA but the airport’s control tower

Was doing some pattern work when a NORDO entered the D airspace, and a few other aircraft had to maneuver to avoid. The guy finally connects with the tower - “I had the wrong frequency”. Yeah - even if you did you can’t enter D airspace without talking to someone.

Thought it was interesting that he wasn’t given the FAA 800 number, but instead told to call the D tower number.
 
A pilot who ran afoul of our local D airport tower ATC had to recently call not the FAA but the airport’s control tower

Was doing some pattern work when a NORDO entered the D airspace, and a few other aircraft had to maneuver to avoid. The guy finally connects with the tower - “I had the wrong frequency”. Yeah - even if you did you can’t enter D airspace without talking to someone.

Thought it was interesting that he wasn’t given the FAA 800 number, but instead told to call the D tower number.

um, I thought a NORDO aircraft could enter. What if the guy really thought his radio was tango uniform (rather than trying to cover up a brain ***t?

(yes, the guy didn't have to fly into the class delta airspace, maybe he could have diverted and sorted things out on the ground)
 
Back
Top