tuna-piano

Thanks for the healthy skepticism.

I still think there's a lot to learn from those articles for most folks uninvolved in this area, even if some of their immediate optimism has additional complications.

I think what I mostly took away is a combination of technologies is likely to dramatically change how we have babies in the future.

1. We'll make sperm/egg from skin cells. This has already been done in mice[1], so it is not science fiction to do it in people.

2. When we're able to do this inexpensively, we could create virtually unlimited embryos. We can then select the embryos that have the most optimal traits. Initially, this may be simple things like not choosing embryos with certain genes that give higher risk of certain diseases.

This may involve selecting traits like intelligence and height (there are already companies that offer this embryo selection capability [2]).

3. Instead of creating a lot of embryos and selecting the best ones, we could instead create just one embryo and edit the DNA of that embryo, which has already been done in humans [3]. Alternatively, we could edit the DNA of the sperm/egg prior to creating the embryo.

The fact that none of this is science fiction is just wild. All of these steps have already been done in animals or people. Buckle up, the future is going to be wild.

[1] https://www.npr.org/sections/health-shots/2023/05/27/1177191...

[2] https://www.theguardian.com/science/2024/oct/18/us-startup-c...

[3] https://www.science.org/content/article/chinese-scientist-wh...

If someone in the year 2050 was to pick out the most important news article from 2025, I won't be surprised if they choose this one.

For those who don't understand this stuff - we are now capable of editing some of a body's DNA in ways that predictably change their attributes. The baby's liver now has different (and better) DNA than the rest of its body.

We still are struggling in most cases with how to deliver the DNA update instructions into the body. But given the pace of change in this space, I expect massive improvements with this update process over time.

Combined with AI to better understand the genome, this is going to be a crazy century.

Further reading on related topics:

https://www.lesswrong.com/posts/JEhW3HDMKzekDShva/significan...

https://www.lesswrong.com/posts/DfrSZaf3JC8vJdbZL/how-to-mak...

https://www.lesswrong.com/posts/yT22RcWrxZcXyGjsA/how-to-hav...

What's funny is the tariffs announced a couple day sago explicitly excluded semi conductors and pharmaceuticals. We are literally about to tariff t-shirts but not those more strategic industries.

Spending Buffet's staggering fortune in a manner nearing anything efficient feels like a herculean task that even the most capable would struggle with.

Objectively, his kids have not demonstrated any capability to do something of this magnitude. It's really quite disappointing to see Warren mostly backtrack on what he's always said (around nepotism and leaving staggering sums to future generations among other things).

You'd think even if this is related to Gate's Epstein connection, he would still recognized the importance of using his net worth for maximum good.

Maybe the human instinct to leave your possessions to your children is too strong for even Warren to fight.

It's really quite depressing watching the fiasco of how Mackenzie Bezos is donating her fortunes and what seems likely to come from Buffet's children. So many resources created+collected by brilliant people and then squandered by their loved ones.

Without a doubt "Endurance: Shackleton's Incredible Voyage" by Alfred Lansing.

Read it, read it now.

It's the kind of book that you once you read, you'll never forget. 4.8 stars with 23k reviews on Amazon. Similar survival story vibes as books like "Into Thin Air" or "Into the Wild" but just on another level. It follows the story of a journey to Antarctica in 1914 that goes wrong and ends up with the ship trapped in ice for many months, and follows the crews absolutely insane attempts at survival.

I am waiting for time to forget enough of it so that I can read it again.

Legitimately there are AWS resellers. Paying for cloud services through a middleman just seems so bizarre, but yet it exists.

https://aws.amazon.com/partners/programs/distribution-resell...

https://aws.amazon.com/partners/programs/solution-provider/

rtings.com has great reviews for electronics, measuring detailed metrics and putting them through various tests. (for example, their AirPods Pro review: https://www.rtings.com/headphones/reviews/apple/airpods-pro-...)

Outdoor Gear Lab for outdoor product reviews: https://www.outdoorgearlab.com/topics/camping-and-hiking/bes...

No 17 year old college applicant has power.

What would you call treating a 17 year old different based on their skin color? If we can't use the word racism anymore, then I'll call it bigoted.

I'd love to see a 2019-2021 salary change by height study. Or Zoom interview vs in person interview acceptance rate by height. Did short people benefit more compared to tall people given the remote work+hiring situation?

Amazon claims cameras automatically track what you take and put back. But if that's the case, then I'm left confused as to why receipts take so long (15-20 mins+ sometimes) to be generated after leaving the stores?

Is it possible it's actually a manual process of people watching the videos, or more likely identifying what happened at each "take off shelf/put on shelf" point?

They're still just building up the customer base, training data and models and it's not as magical as it all seems?

Just an interesting sidenote. It does feel like the letter is shorter than it used to be, so I plotted the number of pages by year in the PDF fils on their site. Looks like there was a decrease from 20 pages to 15 starting in 2017.

Graph:

https://imgur.com/a/TQ2oewY

The best humidity solution I've found is to set my shower up as a humidifier. Puts out a lot of humidity that spreads throughout the house. Doesn't need any cleaning, refilling, etc. Total cost on Amazon, like $40 (probably cheaper from Home Depot). Picture of my setup: https://imgur.com/a/Oca4USg

I did that with 4-5 products: A shower head splitter, a shower->hose adapter, a mister, and anti-leak tape.

Splitter: https://www.amazon.com/gp/product/B0064TXLHA/

Adapter: https://www.amazon.com/gp/product/B00GYV8G7S/

Elbow connector: https://www.amazon.com/gp/product/B003BZD03K/

Mister (I used just the nozzle and discarded the hose) https://www.amazon.com/gp/product/B004INGNPG/

Anti-leak tape: https://www.amazon.com/gp/product/B003D7K8E0/

Alternatively tried this fogger nozzle which put out too much water and wet my bathroom floor, may work in some shower setups though. https://www.amazon.com/gp/product/B07FNYNZZM/

My first reaction to any news from a large company that people question the legality of is "I'm sure they have a team of highly qualified lawyers who signed off on this plan."

But then I remembered about the RH checking account that clearly wasn't approved...

It would be easy to check that...

Any thoughts on what made him special (either from your perspective or from those choosing to promote him)?

Seems like a good method and actually more accurate than they do... seems like they just do a hash of IP.

Looked for a few minutes and couldn't find the full answer. How does Plausible calculate unique users if it can't store some type of identifier on the page?

I see this... "We do not generate any persistent identifiers either. We generate a random string of letters and numbers that is used to calculate unique visitors on a website and we reset this string once per day."

But where is that ID stored?

There is just conveniently space next to the box for him to get out of the box? Would think cargo holds are generally pretty stuffed full.

Is it really a novel data stream?

I believe you can already purchase billions of anonymized credit card transactions... https://datarade.ai/data-products/us-consumer-card-payments

Can you elaborate on the "framed for extortion"... what is the evidence that Trevor framed him?

This is a somewhat tough problem for companies at this point in time. To executives focused on customers, sales, product - this is an even more interesting issue because it has really nothing to do directly with the company itself.

Most customer segmentation problems can be solved with an optional feature or a new product line - make both chunky marinara sauce and a smooth variety.

Most employee problems can be solved similarly - optional programs, different roles for different folks, etc.

But this problem is unique because a certain segment of the employee+customer base is asking the complete company to take their side in certain matters. Of course the company taking that stand alienates the other segment of the population.

However, rationally, it becomes much easier to deal with this than what Coinbase did.

It seems though that the vocal side (liberal) is vocal because they care about companies stances on these matters, while the silent (conservatives) are silent because they don't seem to care as much.

Therefore, rationally, companies generally take the liberal position or no position at all.

When conservatives listen to politically-left company seminars, see liberal company statements, etc - they mostly just ignore and move on with their day. I don't think many conservatives would be motivated to quit or boycott a company due to a liberal company seminar that they disagree with. I get the feeling (due to the walkouts, etc) that liberals are much more likely to sever relationships due to differences in political beliefs.

Ya, the second dose of the Pfizer vaccine appears to make a pretty large majority of younger people feel fatigue. a safe vaccine does not mean a fun vaccine.

Agree for short term effects, even for young people. Due to the trial size, I think we can be confident that if there are any short-term safety issues with this vaccine they are extremely rare.

1,769 people 25-34 have died of Covid this year. There are 23m people in that age range. .0077% have died. If the vaccine had an adverse effect at a similar rate as covid has killed young people, we'd expect it should have showed in the trials in 1-3 people.

I haven't seen that, but I'd be careful to draw the meaningfulness of that metric, as there has also never been an mRNA vaccine deployed until this year.

That said, the large size of these studies should be extremely comforting to people.

I'm curious if anyone knows why Pfizer's mRNA vaccine still needs this while Moderna's mRNA vaccine does not?

Has Pfizer just not gone through the proper temperature testing or did they go through that testing and it showed that their vaccine needs to be stored so cold?

My theory is that anti vaxxers have poisoned the public conversation (if you say anything questioning a vaccine you shall be shunned). I plan to get this vaccine and generally trust our scientific institutions.

But of course there are legitimate safety concerns with injecting stuff into people - which is why they ran a huge study to show that there are no short term safety concerns.

I'm just asking (not implying anything) a real medical question I don't know the answer to - how do we know this vaccine doesn't increase the chances of some condition 20 years later when the median time people have been vaccinated is two months? And with a brand new vaccine platform that has never been used before...

I don't believe that's an accurate statement. Problems have been rare, but not unheard of.

A list here for the US (most of the listed vaccines on the page were later found not concerning):

https://www.cdc.gov/vaccinesafety/concerns/concerns-history....

-The results of the investigations showed that RotaShield vaccine caused intussusception in some healthy infants younger than 12 months of age who normally would be at low risk for this condition... the manufacturer voluntarily withdrew RotaShield from the market in October 1999.

-...that people who received the 1976 swine influenza vaccine had an increased risk for developing GBS .

-And the very recent Dengue vaccine made by Sanofi, which was given to 700k people. "Sanofi had found evidence that the vaccine increases the risk of hospitalization and cytoplasmic leakage syndrome in children who had no prior exposure to dengue, regardless of age."[1][2]

[1]https://www.npr.org/sections/goatsandsoda/2019/05/03/7190377...

[2]https://en.wikipedia.org/wiki/Dengvaxia_controversy

Looks like for fever, flue vaccine is more like 1-2%. So maybe 10-20x more fever with this vaccine than flu vaccine.

https://www.washingtonpost.com/health/flu-shot-facts/2020/09...

-Great great news. Can't wait to get the needle in the arm ASAP.

-Seems extremely effective at limiting covid disease.

-Very curious as to how long will the vaccine immunity will last?

-Seems with the second dose for 18-55 year olds, perhaps 75% will feel fatigue, 50% have chills and 20% will have a fever.[1] This hasn't been widely shared yet and I'm a bit concerned how the public will react.

-Safety: Given the huge trial size, we can be pretty confident that there aren't any significant negative short term effects. But yet the phase three trials will follow the participants for two years. Presumably, this is to find out about longer term safety+efficacy... but yet of course we will be vaccinating most of the world prior to the end of those two years.

How are we confident that these vaccines don't cause any long term safety concerns? Other vaccines have been pulled from the market for safety reasons - what happened in those cases that we're sure won't happen in this case?

I think given the fact that "anti-vaxxers" exist makes some people take the complete opposite position to distance themselves from the conspiracy theorists- "anything a doctor injects in your arm is safe". Of course the truth is that injecting stuff in people's arms is not always safe - which is why we have trials. Were these trials long enough to know with a high level of confidence if there are any long term effects of the vaccines?

[1] https://pubmed.ncbi.nlm.nih.gov/33053279/

Prediction markets seem to be doing a really terrible job at predicting an election that already happened...

Right now predictit.org has Biden winning at 89% and Trump winning at 15%[1].

But if you look at it by "electoral margin victory" and add up all the Republican win margins, you get that Republicans have 28% chance of winning and Democrats have 95%.[2]

This is two weeks after the election that is pretty clear a Joe Biden/Democratic victory. These markets may get things right, but they also get things really, really wrong.

[1]Yes, adds up to over 100. You could bet against both for a guaranteed profit, however I believe the markets aren't allowing new traders and there are also fees and counter party risk to consider [2] https://www.predictit.org/markets/detail/6653/What-will-be-t...