13,114 karma · joined March 24, 2008
> Every post is either political ragebait, recycled "funny" cat videos, "Am I the asshole for divorcing my husband after he killed our two children while drunk and high?"-type slop, or tired wojack memes.
The trick to Reddit is to audit your subreddits. There are thousands of interesting, well run, topic specific subreddits. Find communities around your interests and only subscribe to them. Get rid of the default ones that are mostly just cesspools these days.
Over the years I've cultivated mine to include several book series and authors I like, 3d printing, homebrewing, etc.
The other trick is to avoid the Reddit app. Use https://old.reddit.com, even on mobile. It's still the best way to use Reddit.
Also take into account the generational adoption of older styles for a season or two before it fades again. I just saw a young man the other day wearing ginkos. Haven't seen those since the 90s. I asked him, and he said his dad had passed semi-recently and he really like the jeans when he was going through his dad's stuff.
Yesterday a clothing maker on YouTube I occasionally watch shared a video about making "medieval 1970s mod culture" clothes. https://m.youtube.com/watch?v=g0uOmf-wHag
She used some patterns last used in the 70s and even wondered if the original dress(es) might still be around.
"Style" is subjective. The artifacts influence future style, and should be recognized as separate but linked things.
A lot apparently goes into establishing the authenticity of portraits. As a conservator, he does a lot of work in the background that isn't always conveyed in the videos.
Sometimes the outcome of his restoration efforts is to communicate to the client that the painting is either a copy, or may have history but the subject isn't who they thought it was.
No actual comment on this particular story, but the little bit I've learned of the art community is that they put a huge amount of effort into confirming provenance. The history of the painting.
So if a lot of experts in the field are raising doubts, it's probably worth examining their statements.
Like archaeologists, most want to make a major find. It's career making. But that also requires being highly sceptical of new claims.
In the general sense, notepad and TextEdit should just be less nerdy nano's. They always have been and that's what they were meant for.
If you need something to write reports, a book, etc then you use MS Docs, Google Docs, or whatever Apple provides. Those are the tools where adding AI might be useful as a feature, like the ribbon in Office.
Let purpose made tools just be that.
So we're trying to use tools like this currently to help solve deeper problems and they aren't up to the task. This is still the point we need to start over and get better tools. Sharpening a bronze knife will never be as sharp or have the continuity as a steel knife. Same basic elements, very different material.
I did have a eureka moment the other day with deepseek and a very obscure bug I was trying to tackle. One api query was having a very weird, unrelated side effect. I loaded up cursor with a very extensive prompt and it actually figured out the call path I hadn't been able to track down.
Today, I had a very simple task that eventually only took me half an hour to manually track. But I started with cursor using very similar context as the first example. It just kept repeatedly dreaming up non-existent files in the PR and making suggestions to fix code that doesn't exist.
So what's the worth to my company of my very expensive time? Should I spend 10,20,50 percent of my time trying to get answers from a chatbot, or should I just use my 20 years of experience to get the job done?
So far Apple has kept it as a toggle in the settings, but it's easy bloat for it to keep spreading. Does anyone need AI in a text editor? No.
Edit: I'm being downvoted for expressing ignorance and asking for more information? Really?
Details are important.
My first was 1999 in Florida. Dude had a shop selling jewelry on eBay. His autistic brother (not kidding, dude's brother LOVED computers and our first conversation was him telling me how to deal with himself). Very introspective but really cool guy to get to know.
The owner had a small team of Vietnamese jewelers working in the shop to make and customize jewelry. There was no scamming going on. I had to work on photography of new items and to update the eBay listings. If a customer had a problem with the jewelry they received, we would correct it for free and send it back.
The downside is that despite running a legit business, the owner was an asshole and just kept kicking out one employee after another. Including half the jewelry team.
After a few months I even quit because he just started randomly bitching and physically threatening me because the focus on a couple pieces of jewelry was slightly off. Easily a 10 minute fix.
My last day when he did this again I told him to fuck off and I was out. I still vividly remember him physically spitting from his mouth on my driver side window as I rolled it up as a pulled away.
Now that I think about it, fuck you Jeremiah. Like, 9 fucks to wherever and back. You had somehow built a really good business early on and completely fucked it up just by being yourself. It's been 26 years, but seriously fuck you. I hope your brother finally got the help he needed and you clearly were not providing.
At some point you have to stop thinking in your head why you can't walk through the window and start testing and figuring out what keeps breaking your nose.
However we now have at least the basics to hypothesize and perform experiments. So it's no longer philosophy and is in the realm of understanding what actually are the mechanisms of consciousness.
Unless you can rephrase your argument as something testable, it's philosophy and thereby not relevant.
You make a polite but still ad hominem "attack" about me instead of addressing my points with demonstrations of evidence.
Make a video or blog article actually showing how your use of LLMs in coding is making you more productive. Show what it's doing to help you that has a multiplier effect on your productivity.
It doesn't take billions of dollars and all human knowledge to make a single human level intelligence. Just some hormones and timing. So LLMs are mostly a dead end. AGI is going to come from differenst machine learning paradigms.
This is all mostly hype by and for investors right now.
For the last year, I've tried all sorts of models both as hosted services and running locally with llama.cpp or ollama. I've used both the continue.dev vscode extension and cursor more recently.
The results have been frustrating at best. The user interface of the tools is just awful. The output of any models from Deepseek to quen to Claude to whatever other model is mediocre to useless. I literally highlight some code that includes comments about what I need and I even include long explicit descriptions etc in the prompts and it's just unrelated garbage out every time.
The most useful thing has just been ChatGPT when there's something I need to learn about. Rubber ducking basically. It's alright at very simple coding questions or asking about obscure database questions I might have, but beyond that it's useless. Gotta keep the context window short, or it starts going off the rails every single time.
Overall you aren't incorrect. However to point out, you don't have to pay for the service if you're ok with what comes with the free tier.
For me, I rarely use my phone for the purpose of phone calls. 99% of what Hiya gets from me are spam calls plus my current location. (Wooo I'm in SF, such a big data loss). Occasionally my mom calls me, though we're usually on WhatsApp. (Yes FB blah blah).
Make your own choices. When/if Apple implements actually high level call blocking I'll switch to it (goulash monster willing). For now I'm personally ok with Hiya. If you want to build another service that addresses your privacy concerns then I'll check it out and consider switching.
For me the utility of not getting annoyed all the time is greater than the lack of important data any 3rd party might be collecting.
Actually to address even that, go into the Hiya settings and just turn off things like geolocation sharing.
Install and enable Hiya or similar for calls.
Adjust your notification settings for apps like Lyft, Uber, and Uber Eats. They like to spam notifications as ads.
For adblocking, 1Blocker and AdGuard work great.
For YouTube on mobile web, use Vinegar.
I am 99% protected from ads with this setup
For the remaining, use old.reddit.com.
For websites with annoying subscription popovers, use reader mode. This doesn't work everywhere but does in a lot of places.
After all this filtering, I have absolutely no interest in anything that gets through.
I miss the days of html and little js.
CssZenGarden always had it right. That's where I learned early modern css during that transition from web1 to 2.
Whatever it's a new thing from Google that will make it more difficult to use whatever their product is.
Just disregard that millions of people know how to use things as-is, but to justify their jobs people need to change things that are working ok, instead of trying to do minor changes to improve the experience for the people on the edge.
Thanks for the links. Going to check them out this morning.