michaelskinner.bsky.social
Relevant stuff: Husband/Father | Sacramentan | Army Veteran | Tech Worker
Kings and general funny/interesting stuff. Used to mix it up in politics, but have reasons for public restraint now.
Keep Bluesky positive! I block liberally.
729 posts
390 followers
489 following
Regular Contributor
Active Commenter
comment in response to
post
Not as significant number as you would think, since Russia is primarily a Pepsi country. I'll see my way out.
comment in response to
post
And the asterisk reveal? 🤌🏼
comment in response to
post
I finally just subscribed to Optery (there are tons of options out there). It's wild when they do the scan and show you what all these data brokers have on you. It's whack-a-mole for the rest of your life though.
comment in response to
post
That's fair. This conversation has morphed from, "is it useful?" to, "is it good?"
People who think it's bad have good points and I'm not going to the mattresses on that one. I can tell you, however, that it has been very useful for me, which is why I'm agreeing with Hank on this.
comment in response to
post
Yeah, another Bronze Age Collapse is definitely in the cards. I am completely clear-eyed about that. Just trying to control my controlables at this point and hope there is enough left for my kids to have an enjoyable life.
comment in response to
post
Yep, with you 💯. I was "waiting" and now I'm "watching." 😬
comment in response to
post
I've been sitting around since the 90s waiting for the internet to break down our institutions even worse than the printing press did., so maybe I've just pre-chewed this all in my own mind.
comment in response to
post
Yeah, it is definitely ethically dubious. Things like blogs may not make it. Like, personal expertise may no longer be lucrative. That will have knock-on effects for sure. We've seen this before though, right?
comment in response to
post
There isn't an abstraction layer here, though. If your code is publicly available, SOMEONE is going to use it to learn how to code. In this case, that someone is an algorithm. That algorithm then goes and helps others.
In fact, it probably reduces the amount of ctrl-C/ctrl-V that is happening.
comment in response to
post
You can't steal my content from Reddit! That is for ADVERTISER USE ONLY!
comment in response to
post
It's ironic that the closer to being a communist you are, the more you hate LLMs. To steal information, doesn't that imply private ownership of information? Isn't that the backbone of capitalism? LLMs are simply the voice of the proletariat! Everything is the commune!
comment in response to
post
"Hey, you stole my incline plane idea!"
comment in response to
post
Yeah, I am a big "pay for your software people!" person. Good code is very valuable and this idea that we should get all of this magical stuff for free is toxic and has crazy negative impacts.
And the fact that AI sucks at it is sort of a self-correction right now. It's a tool, not a replacement.
comment in response to
post
Yes, I wish that the hugging face models, which are FOSS, were more capable. If I had more time I would be using chat GPT to get a good model spun up on my local server so I didn't have to pay one of the big tech companies 😆
comment in response to
post
Touch grass, friend
comment in response to
post
Did I?
comment in response to
post
Looooool. You are so anti-LLMs that you become pro software patents?
comment in response to
post
I actually totally agree with basically everything in that article. Why does everyone assume maximalist opinions on the internet?
I'm agreeing with Hank that I have found a lot of usefulness in AI. And I'm being called an idiot for that by anti-LLM crusaders.
comment in response to
post
First of all, that was pretty mean.
Second of all, I don't want to put the time and money into becoming an expert at everything I need to do throughout the day. I am not a Linux admin. I just want to get my Raspberry Pi to do things for me.
You don't know what I'm doing with that saved time.
comment in response to
post
In terms of how much power they use, yeah, that's definitely a bad thing. But it's also a cat that's not going back in the bag, and humanity's compute requirements have been on this trajectory for a while.
comment in response to
post
No, but the person I'm replying to did. There are myriad use cases and Hank apparently has some that are good for him.
comment in response to
post
They are really good at helping me meal plan and cook and also do a bunch of stuff in Linux that I wouldn't be able to do otherwise.
They're also really good at explaining complex topics and most of the time they are not wrong. Think Wikipedia in the early years.
comment in response to
post
Yes, everything they "create" is utterly banal. No argument there.
comment in response to
post
Convincing people that "LLMs are not pure evil actually," is going to be about as hard as convincing A Boomer that crime is lower now than when they were growing up.
They are just too morally and emotionally invested in that not being true.
comment in response to
post
Hank is not talking about using them to write things. If all you know they can do is write things, then you're not really informed enough to wade into this topic.
comment in response to
post
Update: got one anyway!
comment in response to
post
Still in line at Best Buy. People have been posting wins though.
comment in response to
post
Not this kind 😄
comment in response to
post
Oh, N/M just read a couple skeets
comment in response to
post
This is the correct take
comment in response to
post
Yeah, a lot.