leplen

I think one of the biggest differences between comp benchmarking providers and RealPage is that two companies given access to the same benchmarks often make very different compensation decisions.

Some companies decide to pay at the 90th percentile of the market and some decide to pay at the 50th.

RealPage was much more proscriptive which is part of what tips the balance over into collusion vs just providing market statistics.

Zillow provides a zestimate for rents, which serves a similar market summary purpose but without the mechanism to enforce compliance and therefore is totally fine.

English sentences need a subject and a verb. Often a sentence fragment missing a verb (like this).

Price discrimination in the form of scholarships and the "show us exactly how much money is in your wallet" scam they call "need-based-aid" allowing universities have essentially perfect price discrimination and to gobble up the entire consumer surplus is definitely a big part of it.

Imagine any other service where they demanded your bank statements before they told you what they would charge you, and the impact that would have on sticker price.

If a hurricane kills a thousand people is it evil? It's not really clear whether we're better off trying to treat corporations as moral actors and constantly being surprised when they aren't, or whether we should treat them as amoral quasi-natural processes no one has full control over and construct regulations around them the same way we do for earthquakes.

The state of Mississippi ran Parchman Farm on this principle for decades, but as you might imagine it looked a lot like slavery. A lot of people suddenly got arrested for loitering come harvest time.

This is actually pretty interesting. Generic drug manufacturing has relatively high barriers to entry, but relatively low margins because there's no way for brands to command a premium. Standardized insurance costs work to prevent price discovery, and consequently you end up with supply shocks, especially for drugs that are difficult to store.

Even given dictatorial power over the FDA I'm still not sure exactly what I would do to untangle the swirl of incentive problems here.

The theory behind invite only is that it increases the density of the network. You invite people with similar interests as this the network can nucleate around some niche, rather than being 10,000 people with nothing in common. It's one of the standard plays for overcoming cold-start problems.

American options can be exercised at any time at the investor's discretion. This means the instrument has a maximum duration, but the actual duration is up to choice of the option holder, which you can't model with an equation.

The safety problem with Li-ion batteries is the electrolyte is a flammable organic solvent. It is unrelated to the amount of energy stored in the battery under normal operation.

The data is only useful in aggregate, but different people/use cases require different types of aggregations. Using pre-aggregated data is difficult, because it almost certainly hasn't been aggregated in the way that's convenient for whatever analysis you're trying to do with it.

It looks like there is a theory that dermal bone in early tetrapods was used this way, but it's still fairly speculative. https://www.nbcnews.com/id/wbna47176272

If you don't want to do something, why is it on your to-do list?

It feels like people have multiple competing systems trying to set priorities, and this is arguing that one of those systems is the only one worth listening to.

I procrastinate for the same reason I have a backlog. Because not everything I'd like to do is urgent. But I don't experience it as some inner war with myself and I don't think the key is to totally ignore the signal from my feelings, and buy into the tyranny of the the piece of paper I arbitrarily made up yesterday.

If you feel bad about a goal, maybe it's a bad goal?

Advice is like vitamins, useful for specific deficiencies and mostly harmless for the rest of is. I'm sure there are people who need to hear the advice in this post, but I think the counterpoint that maybe your intuition about a task isn't a thing to be silenced is also worth making.

They never found the plane. Most likely the lead pilot crashed it intentionally, but we don't really know. The person writing this blog post is more certain about that than the Malaysian authorities, but bloggers are allowed to speculate more than government commissions.

There's some neat math people used to try and find the plane that gets alluded to, but not explained in depth, and the post is a reasonable summary, but didn't really change my mind much and my previous awareness was based on overhearing cable news, so it's not really anything new.

It's written well enough to be entertaining, but isn't really enlightening.

The linked paper doesn't obviously support the claim you're making here. Maybe I've missed something?

"The median age at menarche was 14.5 years (95% CI=14–15; Nindividuals=128; Nobservations=199), which is within the range of variation of modern populations."

I use jupytext for this. For the cost of some header boilerplate I have python code that is seamlessly mirrored as a notebook.

A lot of the stuff described here sort of doesn't make sense. It requires idiot savant AIs that can figure out how to manipulate humans, but that we can only train with simple integer based reinforcement learning techniques.

The introduction of artificial agents is definitely a step increase in the complexity of society, and it will introduce new, and potentially catastrophic failure modes, but I find this story of AIs conspiring to rob a bank to be a fairly unimaginative take. Even if we grant the possibility of conspiracy of AIs, something that looks like insider trading is a lot more likely.

It shows that you bounced off one of the artificial rails openAI added to minimize PR risk. The underlying model is certainly flexible enough to answer that question, it's just been deliberately crippled to try and make it seem less anthropomorphic.

There's a lot of convergent evolution towards the crab body plan, a phenomenon known as Carcinisation. (https://en.wikipedia.org/wiki/Carcinisation)

Data science as a title has bifurcated several times. A lot of the engineering heavy DS roles are now data engineer or machine learning engineer. Probably those are the roles you should be looking for. Many of the DS jobs now are essentially analyst jobs, where SQL and stakeholder management is 75% of the role and writing code is just a bonus.

Data science isn't going away. Leadership is always going to need numbers explained to them, but DS roles have never been as numerous or as well paid level for level as software engineering.

Technically most Li ion batteries are created in the discharged state, with the lithium in the battery provided by the cathode at the time of construction, and then moved over to the LiC6 anode via charging.

The description in your link has the thickness: Freestanding unpolished diamond single crystal synthesized by heteroepitaxy on Ir/YSZ/Si(001). The thickness of the disc is 1.6 ± 0.25 mm and its weight is 155 carat.

Was anyone under the impression that Eric Schmidt wasn't investing in AI companies? Tech billionaire invests in tech companies seems pretty expected to me, even if he was doing consulting work for the Feds during the same time period.

A negative feedback loop is one that tends to keep things in a steady state. The thing being described here is actually a positive feedback loop with negative consequences.

This article is 5 months old, and talks more about NFT trade volumes than NFT prices. Having volumes decline from peak speculation is more a sign of market maturity than a market crash.

Blind copying is highly adaptive behavior and is the engine of human culture. Inventing things and reasoning correctly is really difficult, and copying widely adopted behaviors is usually much safer and more efficient.

Reading 'Not by Genes alone' really changed my perspective on this and helped me understand the blind copying as a very successful heuristic that sometimes gets things wrong.

Technically the speed of light in fiber is 2/3c because fiber has a higher index of refraction. Some HFT funds have actually moved away from fiber in some instances because they can pick up a few microseconds by going through the air instead.

Lithium batteries typically explode because the electrolytes we use are highly flammable. It's actually entirely incidental to the amount of energy stored in the battery. A Li-ion battery with a different electrolyte material could have the same energy density, but a greatly reduced risk of fires and explosion.

Primarily the app store, which has a number of placements with sponsored content.

Maybe this transmutation by nostalgia is the true miracle of Christmas. https://www.smbc-comics.com/comic/2012-12-03

I had assumed FB bought Giphy in part because Google bought Tenor. It's interesting that the FB acquisition has seen more antitrust scrutiny relative to the Google acquisition.