OpenAI has a free tier. I'm guessing a lot of people never need to upgrade. Not the case for me but I can see the average person only needing to run a few prompts every day.
My understanding is that most OpenAI users are on a free tier. Secondary effect of this is that OpenAI free tier model capability (assuming Luna) is what what most users associate with frontier AI capability giving somewhat warped view to many people.
How are these animations being made? I'd love to get a blog post on them. If its AI I'd love to know the workflow, but something tells me there is a lot of human creative input
Filibuster is just a procedural rule that can itself be removed by simple majority, and it has already been watered down on a case by case basis when that was politically expedient, so there's no hard line there. It's actually kind of amazing that we still have it.
I can't help but think there is an alternative to this dichotomy. If, for example, US were to institute a high-percentage income tax bracket for the wealthy (maybe 65% over $1m) and used that to pay for national healthcare, would that really impact start-up culture? Maybe it affects some business at large, but it means the average joe can get laid off and not stress about hospital bills, while still encouraging a "risk-hungry sector".
Tl;Dr: Europe could have trillion dollar companies and still lean heavily towards labor, just like we could
Here in California, my marginal tax rate (if making over 1M) is already about 53%. In return, there is medi-cal (which is shitty[1]), 6 weeks of parental leave at 65% salary, crappy schools on average but excellent if one has a special needs kid etc. So not so great safety net.
Ironically, through a big tech job (which is supposed to be a rapacious corp with zero regards to employee wellness), I get far better protections - amazing insurance(s), 4 months of parental leave at full pay (including RSUs), all sorts of benefits etc etc.
The trade-off is clear to me. I want more of those trillion dollar corporations around me and have very little faith in Federal and California governments to improve their safety nets even if they jack up my marginal tax rate from 53 to 65%.
[1] but again, my Canadian and UK families have horror stories about their healthcare systems as well so there is that
>Since the physical device was satisfactory, the next step was writing a simple website that could drive the display.
>A Compose page my siblings and I write messages and save them to be displayed.
Is there a risk of a malicious actor discovering the website and writing in their own messages? I would think building user authentication in to the MomBoard website might be a bit heavyweight. Whats the best way to do this?
When you have a site with a fixed, tiny amount of users, I'd opt for HTTP basic auth (via HTTPS). Whether you're using nginx, Traefik, Caddy, etc..., it's very easy to setup. If you're using something like Cloudflare Pages, I would guess you could setup a worker to handle it for you (though I'm not familiar enough with workers to be sure).
Would love to see a data visualization of the voice's predictions and their results. Every day, you could keep track of each prediction made, and also go back to old predicitons and write down whether they came true or not. Over a six month span I think you could make a cool chart and those results would be very interesting.
For fun you could also keep track of other related data. How far fetched the prediction is. How confident the voice is. The time frame of the prediction. Just an idea...
Since you appear curious, here is some information.
The voice always speaks in the same tone.
Predictions are usually something I want to do or am planning to do. Plans it usually tells me as soon as they are conceived. Predictions come often when I am making plans, it tells me whether they will work or not.
HOWEVER, to answer your question, such an exercise could be totally futile as in actuality it changes its mind a lot. Sometimes it tells me something I intend to do won't work, without divulging why. I try it anyway and it kinda works, and I figure out why the point of failure. I fix it, and now the voice tells me it will work now, making the initial prediction kind of moot.
Sometimes something happens and something the voice initially thought was a great idea it now says will be an absolute failure. By that time I am often committed so I tell it there's no choice now, we can only keep going, and it kinda agrees and we do the thing anyway. Sometimes the voice is just plain wrong about things.
The predictions aren't so much predictions as they are opinions. Also, studying it with too hard a lens seems to bring out the worse in the condition, so I avoid it.
This seems similar to the type of intrusive thoughts associated with OCD. It's like someone pretending to be God inside your mind, making commentary on all the things you want or try to do, sometimes very cruel. I kind of think of it as the part of your mind that is involved in planning for the future is given a voice, or that your internal anxieties or fears or hopes become kind of like a separate entity alongside the normal conscious mind.
There is a lot that can be learned about the mind by carefully observing it facing and coping with these situations. Understanding the chemistry and the usage patterns of various parts of the brain as it goes through this form of internal dialogue.
It would, however, require a lot of commitment from the patient and willingness to face the condition and even make the symptoms worse. And, considering how crude our tools are at this stage, wouldn’t likely yield much useful information.
I see the game was ported with the legacy CheerpJ 2.3 runtime. Consider upgrading to the new CheerpJ 3.0. Improving boot times has been one of our main goals and upgrading is very easy: https://cheerpj.com/docs/migrating-from-cheerpj2
Fun game! I thought you programmed it wrong till I realized in your version of liars dice 1s are wild. Also never played the way of guessing the total number of dots, and we always have loser go first on the next round. Thanks for sharing!
If I run out of tokens on ChatGPT of course I will try Claude. I never ran out of Google searches so no reason to try Bing