

Could you import the column in as text to preserve it?
Could you import the column in as text to preserve it?
deleted by creator
My theory is that there’s a tonne of push back online about people coding without understanding due to llms, and that’s getting absorbed back into their models. So these lines of response are starting to percolate back out the llms which is interesting.
Lmao, when have age checks worked on any site ever
Tea has theanine which sort of counteracts the effect of caffeine, which is why tea generally has a milder effect. In theory if you switch to decaf coffee long enough you shouldn’t get withdrawals (which is dependent on the concentration/frequency of caffeine use).
2024 lol. Maybe senior dev is an overstatement, he was just more senior than me. He also left a database where the main table had one varchar, freetext column that users wrote multiple fields into because it was a ‘simpler user experience’ . Was a pain to extract all those fields with regex…
Lmao. I once had a senior dev put database passwords into documentation, and then was about to email those out to interview candidates with the passwords ‘blacked’ out. I caught it quick enough before it could be sent thankfully.
Yea it’s way too optimistic to presume these companies will not lash out in their death throes. They’d destroy as much as possible on their way out.
They probably eventually plan to bring back forced labour. They ‘deport’ people, and when that’s too arduous, they’ll round them up in camps and make them work their ‘sentence’.
I don’t think they actually stupid enough to deport their low wage vulnerable workforce when they can just further exploit them.
Draw.io is really good for diagrams.
Love my kobo libra h20, would highly recommend. Easy to sideload epubs with calibre.
No save summing on dice rolls anymore though 😢
I could imagine it working well if you give different characters models trained on different data based on what they know as characters, but keep their responses railed to sections of dialogue that are more ambiguous or not key to the plot. Could really add to the depth of the world without compromising to much on cutting human written dialogue. But then again knowing game developers I doubt they’d use a nuanced approach.
Most of the time they only suggest that as a proxy for telling the temperature. You don’t actually need oil at that temp.
Another fucking stupid outlook thing is that you can only schedule send emails from the Web client, not the desktop app. If you try to do it on the desktop app it sits locally in your outbox and only sends if the PC holding the email is on. And if it’s off when the scheduled time passes, the email just sits in the outbox. So incredibly stupid.
There are people in my workplace (which uses Gmail) who willingly use Outlook to access it then complain about Outlook problems. Outlook is the biggest piece of trash email client software I’ve ever used.
At one place I worked at, it was OBLITERATING any mails from one particular person because of a completely unrelated filter I had. Email notifications from them would appear for a few seconds, then the email would completely disappear (not even in deleted, or in any filter folders). Everyone else cced could see the emails except me, and IT couldn’t figure it out. I had to disable ALL my filters which trashed my inbox with stupid work circulars. Fuck Outlook.
People will just use it to perv on each other.
You do not wanna mess with narcissists though, it’s not worth the trouble.
Don’t get me wrong, it’s incredible. But it’s still a variation of the Chinese room experiment, it’s not a real intelligence, but really good at pretending to be one. I might trust it more if there were variants based on strictly controlled datasets.
But… bash snippet extensions already exist. The only difference is maybe it doesn’t auto name your variables for you. I’d take that over non-deterministic LLM outputs.