Hacker News: orasis

New comment by orasis in "AI is going to kill app subscriptions"

orasis — Sun, 15 Feb 2026 16:07:30 +0000

This is a terrible take. Competition in the app space is competition for distribution/attention. Subscriptions are just how apps convert that attention to dollars. Consumers almost never price shop apps.

New comment by orasis in "Tao Te Ching – Translated by Ursula K. Le Guin"

orasis — Sat, 24 Jan 2026 18:37:40 +0000

This is most likely a copyright violation. I follow these translations and I’ve seen no evidence that the publisher put it in the public domain.

New comment by orasis in "Show HN: Bible translated using LLMs from source Greek and Hebrew"

orasis — Thu, 22 Jan 2026 18:48:58 +0000

Is this translation public domain?

New comment by orasis in "Apple loses UK App Store monopoly case, penalty might near $2B"

orasis — Sun, 26 Oct 2025 15:44:20 +0000

Of course. Apple removes their cut first before paying out. I’m in the US.

New comment by orasis in "Apple loses UK App Store monopoly case, penalty might near $2B"

orasis — Fri, 24 Oct 2025 05:10:23 +0000

As a professional iOS app publisher, I feel 15% is fair.

New comment by orasis in "New tools and features in the Responses API"

orasis — Wed, 21 May 2025 18:01:54 +0000

Reasoning models can now call tools during the reasoning process.

New comment by orasis in "Apple App Store guidelines remove ban on encouraging external payments in US"

orasis — Fri, 02 May 2025 15:19:23 +0000

There have already been class action lawsuits against both Apple and Google that have paid out to developers so who knows.

New comment by orasis in "Temu pulls its U.S. Google Shopping ads"

orasis — Tue, 15 Apr 2025 13:39:41 +0000

App Store ranking is determined heavily by install volume.

New comment by orasis in "Intentionally Making Close Friends (2021)"

orasis — Sat, 12 Apr 2025 13:19:46 +0000

This could be. At this point in my life I doubt I’ll become close with anyone who is incurious about psychedelics as there seems to be a strong dividing line in mindset between pre-psychedelic and post-psychedelic humans.

New comment by orasis in "Intentionally Making Close Friends (2021)"

orasis — Sat, 12 Apr 2025 13:12:12 +0000

Do it during the day, properly dosed, don’t mix with alcohol, and get plenty of sleep that night and this is not a problem. The MAPS studies have shown this again and again that the Tuesday blues is related to partying and not intrinsic to the medicine.

New comment by orasis in "Intentionally Making Close Friends (2021)"

orasis — Sat, 12 Apr 2025 13:06:52 +0000

Before I tried this process I would have agreed with you, however, we focus on longevity when it’s really intimacy that leads to deep lasting connections. It’s just that with the decades long friendships there is more time and opportunity for intimacy, vulnerability, and going through some shit together.

MDMA + the 36 questions is deep intimacy, vulnerability, and often going through some shit together.

We’re also not seeking friends to do MDMA with. On a few occasions, we’ve met interesting people that there is a strong initial spark of connection with and let them know about the practice. These friendships are now very deep and loving even if only a year or 3 old. They are anything but shallow.

As for short lived, only time will tell, but even if they were to end tomorrow I will have deeply cherished the present experience of these connections. I have one 10+ year friendship that is fading and while it breaks my heart from time to time, I still love this person and would do it all over again.

New comment by orasis in "Intentionally Making Close Friends (2021)"

orasis — Sat, 12 Apr 2025 02:51:56 +0000

I will second using “the 36 questions that lead to love” to create close connections quickly.

Combine this with MDMA to supercharge the intensity and speed of connection. This can also be done with a group of 3 to create a tight group that instantly has each other’s backs.

New comment by orasis in "Apple Intelligence comes to Apple Vision Pro in April"

orasis — Fri, 21 Feb 2025 20:15:26 +0000

I haven’t found much use for AVP besides watching movies on an airplane.

New comment by orasis in "Ketamine for Depression: How It Works (2024) [video]"

orasis — Fri, 07 Feb 2025 17:48:50 +0000

The low dose lozenges very rarely lead to negative experiences.

New comment by orasis in "20k federal workers take "buyout" so far, official says"

orasis — Wed, 05 Feb 2025 22:56:03 +0000

On a larger level, if the active & skilled people are taking the deal, doesn’t it mean they’re likely moving on to something that is a better fit and thus good for society anyway?

New comment by orasis in "Lines of code that beat A/B testing (2012)"

orasis — Fri, 17 Jan 2025 15:34:17 +0000

It’s best for immediate rewards. If you have delayed rewards there is a paper on sampling from the “delay distribution” that solves this.

New comment by orasis in "Lines of code that beat A/B testing (2012)"

orasis — Fri, 17 Jan 2025 15:32:43 +0000

lol. My first MAB implementation also used HyperLogLog for tracking unique conversions. We over engineer alike it seems.

New comment by orasis in "Lines of code that beat A/B testing (2012)"

orasis — Fri, 17 Jan 2025 15:26:34 +0000

Having significant experience with bandits in production, I strongly recommend only using them for immediate feedback. If the rewards are at all disconnected from the action you likely won’t be happy with the results.

New comment by orasis in "Lines of code that beat A/B testing (2012)"

orasis — Fri, 17 Jan 2025 15:24:53 +0000

Thompson Sampling is trivial to implement, especially with binary rewards. ChatGPT can do it reliably from scratch.

New comment by orasis in "Lines of code that will beat A/B testing every time (2012)"

orasis — Tue, 14 Jan 2025 05:31:01 +0000

Check out improve.ai if you want to see this taken to the next level. We combined Thompson Sampling with XGBoost to build a multi-armed bandit that learns to choose the best arm across context. MIT license.