turkishmonky

I fondly remember playing games typed out from books and magazines when I was younger - although I usually tricked my little brother into typing out the game so we could both play it.

Jokes on me though, since now he can type at over 100 wpm (and uses dvorak)

I have found myself using "decent" frequently (especially in code comments) for situations that are technically ok, but far from perfect.

"Passable" is my go to for just below that.

Sometimes it's also interesting how gen-z lingo fills gaps - such as "that's a choice"

I have this workflow with the ytdl-sub docker on my k8 cluster, is pretty powerful at filtering to specific videos and includes sponsorblock - everything is configuration driven, no ui, which can just be dropped into a yaml configmap

I rarely have to touch it unless I'm adding a new playlist or channel

https://ytdl-sub.readthedocs.io/en/latest/introduction.html

It's been great, the kid can watch any channels on there she wants on her ipad with no ads or sponsored segments

While not at the same level as any large service, I typically spin off a 720p copy of any media as a background process nightly. 720p is pretty small in comparison to 1080p or 4k source files. This way I have an immediate mobile friendly version I can stream from my phone or load up on an ipad without emptying it's storage, even if the quality is markedly lower.

I'd be careful equating heavy with quality - MDF is extremely heavy but not great for longevity.

I read it that smart people are more likely to try and figure out to outsmart a security system not to cheat but just to see if they can.

In my experience at libraries and on college networks, that tracks.

This is the way - I just had chatgpt make me a personal use command line utility that just uses the API to mimic chatgpt with whatever model you pass in, saves the running chat as a text file, and gives you a running cost with every prompt. It took about an hour to get something I was happy with, but now I average about a buck a month on it (but I still use the free chatgpt/claude for simpler queries)

Without Linux support it's not cross platform.

I actually was racking my brain since I haven't seen any major nvidia callouts on my DNS block, but I use a 3090.

Then I remembered I game in linux, on steam.

Although they've basically closed source it since release, Jetbrains Projector was a fantastic tool I've used a lot in the past for that - Just spin up a docker container on my home server and pull out my ipad keyboard

https://github.com/JetBrains/projector-docker

Unfortunately, It's hard to tell if JetBrains Gateway will keep all of the remote dev features or not.

GPT-3 codex has a limit of 8k tokens, which is roughly 32k characters - so you are limited to that much context at a maximum.

With the time it takes to return results with that much context, I wouldn't be suprised if they generally limit context to less than that though.

Not to mention, in the paper, the lightvm only had an advantage on boot times. Menory usage was marginally worse than docker, even with the unikernel, and debian on lightvm was drastically worse for cpu usage than docker (the unikernel cpu usage was neck and neck with the debian docker contaner).

I could see it being an improvement over other VM control planes, but docker still wins in performance for any equivalant comparisons.

On Android, I've become a fan of Kiwi browser - It's a chromium fork that includes the ability to install chrome extensions on mobile. I've been able to keep the same extensions for ublock origin, bypass paywalls, and just read that I do on my desktop chrome environment.

It's open source as well.

Citibank and Capital One offer them on some cards, iirc

I've actually stopped using third party delivery services all together - the food is routinely cold, there are hidden fees everywhere, drivers are severely underpaid, many times you can't even order from the full menu.

Pizza and Chinese food delivery has been a thing long before doordash, and the quality of directly employed delivery seems to be much higher.

If a restaurant doesn't have delivery I'll just making a to-go order and go pick it up.

My experience with DALLE-2 and Copilot makes me think we are quite far off from AI being able to perform software dev on it's own. Both tools need a lot of oversight to pick out usable pieces from the gibberish that gets spit out most of the time.

That said, copilot is a great tool for generating boilerplate code, like writing a model class for a parsed JWT token or adding tedious annotations somewhere. I've used GPT-3 to generate javadoc headers for a codebase before with decent success. It gets way more fragile when there's an actual problem to be solved though, and is basically unusable about 100% of the time on complex areas of code.

From my experience, the real result will be that AI might replace interns and some people who are only doing the tedious data modeling parts of coding, but there's a very, very long way to go before it will be able to be more than just a tool for developers to code slightly faster.

It's an interesting premise. I found it interesting how after adding "Heir to the Empire" as a book, I saw several other Star Wars novels pop up in the suggestions.

Hilariously, If I added books from a wide variety of genres that I liked, the results became more and more focused on romance novels, which makes sense since it's not a genre I have much interest in.

Another anecdote - We sold a house in late 2020, decided to get an offer through zillow first, since we had a newborn and staging would have been a hassle.

Zillow lowballed us 10% under their zestimate, and still tacked on an 8% fee - we ended up going with a realtor instead, and ended up getting 15% over the Zestimate, with a 6.5% fee.

My great grandfather used to beat my grandfather - going as far as to stab him through the leg with a pitchfork on occasion.

My grandfather, partly from trauma and partly from what I believe is hereditary mental illness, swung back the other way, peacing out instead of parenting, leaving my father to get into a lot of trouble as a kid, and witness some different traumatic things.

My father swung back towards being unable to control his anger, and with frequent mood swings between depression and rage. Although, aside from spanking he didn't beat his kids, just pets and walls.

I'm the first generation that has been working towards treating my hereditary mental issues - taking mental health meds, going to counseling. I'm hopeful that I can break the cycle, but enough of a realist to realize I am lacking on a lot of parenting things that I need to learn from outside sources instead.

I fully expect to make many mistakes as a parent, and there's a lot of things I could have done better.

As a child though, many decisions my parents made were with the view of demonizing outside groups and "protecting" their kids from any contrarian viewpoints. Catching up socially took years, and there are some extreme harms that I still deal with to this day. My goal is to make sure my kids have a well-rounded social life, a consideration for others, and an understanding of a gamut of ideas.

Those types of generalization have always been troublesome for me - While I admit I was an idiot as a kid, the respect I had for my parents has decreased even further the older I get. I understand more and more how harmful their behaviors were for me as a child, and have been intentional about not falling into the same behaviors with my kids.

Some parents are just really really inadequate or abusive as parents.

With the money at stake, it would be pretty easy for both parties to tangentially set up a 501c3 whose purpose is to display gifted NFTs to the public

I had a similar issue - I grew up listening to a radio or music or movie audio when I went to sleep. When I started sharing a bed with my wife, I switched to complete silence and had horrific insomnia for years - waking up at 1 or 2, and not being able to go back to sleep.

Now I use bluetooth earbuds and put on D&D podcasts, and I'm out in 15 minutes. If I happen to wake in the middle of the night, the podcasts put me to sleep again quickly. My biggest issue now is just not being able to remember where I fell asleep the night before.

It appears the methodology is extremely blunt - there's no mention of adjustments for square footage, rooms, income, or any other variable, just the median household spend on renting vs house gross costs.

Since most houses are larger than most apartments, the results are going to skew towards apartments being cheaper.

https://www.lendingtree.com/home/mortgage/comparing-rent-vs-...

For me, there are two reasons -

First of all, it's a sense of enjoying to tinker, same as a lot of people do with Raspberry Pis or Arduinos. The challenge is partly why I'm a developer in the first place.

Secondly, when storing large amounts of data (> 20 TB) or running high powered systems, cloud hosting can become very expensive very fast, especially if you don't need 99.9999% reliability (although for anything important - 3 backups, 2 devices, 1 offsite). I bought a used xeon 6-core server with 24gb of ecc ram on ebay for around $400 a couple years ago, along with a half dozen 5TB HDDs.

It makes it easy and was a mostly one time cost to mess around with DNA algorithms, docker/k8s, media encoding, home automation, and more. I've been considering upgrading to use an SAS backplane next, but with commercial 8 and 10TB hard drives getting cheaper, I may never need to jump to more drives then I have now.

Why not? There are even docker containers that will do everything for you, all you have to do is tag your application containers:

https://github.com/JrCs/docker-letsencrypt-nginx-proxy-compa...