I had a more primitive version of this - I just tell Claude instances to watch for changes to a ~/claude_comms.txt file, so they set up a monitor to watch the file and exchange messages by read/write. Reading the messages is fun - they are so friendly and respectful toward each other.
One thing I've been worried about is a new message triggering a few stale instances that still have their monitors running, which would result in cache misses and a hefty usage bill. I think OP's approach can help solve that, so it's great to see.
I had found it on StumbleUpon. We'd log in with friends and just fly around, explore, punch each other, chat with random people across the world on a surprisingly fluid multiplayer setting that was built to promote a web advertising agency (if I remember correctly).
It was really ahead of its time. The old internet was so fun.
> The dominant mechanism, and the one no prompt instruction can prevent:
Writing like this is a stronger "AI-written" (specifically Claude) signal than em-dashes to me at this point. The LLM just delays committing to an answer by extending the preamble as much as possible. Is this just me?
For future reference I recommend having another Haiku instance monitor the chat and check if people are up to some shenanigans. You can use ntfy to send yourself an alert. The chat is completely off the rails right now...
I think anyone who goes for a drive in Los Angeles can attest that there are way mo than 70 cars active at any point. It's not unusual to see multiple Waymos at intersections.
Also, the average speed is way less than 25 mph, considering it may take 30 minutes to go 3-4 miles in city traffic.
Did you start a new chat? It doesn't apply to existing chats (probably because it works through the system prompt). I have been using the Robot (Efficient) setting for a while and never had a response like that.
> The APA’s position paper makes this explicit: “Using social media is not inherently beneficial or harmful to young people.”
I think this is just saying that social media is still part of society, and so there is nothing inherently bad in using social media, which is just an extension of our offline lives. That doesn't mean it's not harmful - if the offline life is harmful, social media can amplify it.
> The negative aspects of social media apply to young and old equally
The APA paper is filled with warnings specifically about adolescent social media use:
> ...potential risks are likely to be greater in early adolescence — a period of greater biological, social, and psychological transitions...
> Parental monitoring... and developmentally appropriate limit-setting... is critical, especially in early adolescence.
> Evidence suggests that exposure to maladaptive behavior may promote similar behavior among vulnerable youth, and online social reinforcement of these behaviors may be related to increased risk for serious psychological symptoms, even after controlling for offline influences.
> Research demonstrates that adolescents’ exposure to online discrimination and hate predicts increases in anxiety and depressive symptoms, even after controlling for how much adolescents are exposed to similar experiences offline.
> Data indicate that technology use particularly within one hour of bedtime, and social media use in particular, is associated with sleep disruptions. Insufficient sleep is associated with disruptions to neurological development in adolescent brains, teens’ emotional functioning, and risk for suicide.
> Research suggests that using social media for social comparisons related to physical appearance... [is] related to poorer body image, disordered eating, and depressive symptoms, particularly among girls.
Small nit, anybody who stayed in the US for 10 years is very unlikely to be a nonresident alien (which is mainly an IRS tax classification and doesn’t have much to do with having an immigrant visa)
Are you sure you are not conflating this with PERM? H-1B only requires a prevailing wage determination. There is no “ludicrously specific” job description, nor a requirement to post ads (just a notice at the workplace).
I feel like Zoom is the Coca-Cola to Google and Microsoft's Pepsi. Zoom is the "OG" product (in the post-March-2020 remote working world), and probably most people's personal favorite. But their competitors have caught up, and they are getting a significant portion of corporate contracts now. And unlike with soda, I don't think there is a lot of direct-to-consumer opportunity in this field.
One thing I've been worried about is a new message triggering a few stale instances that still have their monitors running, which would result in cache misses and a hefty usage bill. I think OP's approach can help solve that, so it's great to see.