johbjo

> Calcifediol becomes active more quickly than cholecalciferol, enough so that if I were in the business of defending the Spanish study I might argue that it takes a few weeks for cholecalciferol to work, so giving it to someone who will already be dead or recovered by then is meaningless.

Yes, this is one possible conclusion of these experiments. It sounds to me like the mechanisms/pathways of vitamin D are well understood and not controversial. It is known that there can be very long delays (weeks) until blood levels increase from food supplements, and it is blood levels at the time of infection that matters.

The results confirm all this.

> It would make me more comfortable if someone who did understand Vitamin D biochemistry would confirm this, but I'm not going to hang the whole argument on it.

Ok, then. Thanks anyway.

The crashed bike at 0:29 looks like a Honda CBR 400RR NC29, from 1990, as best I can find out.

It sounds like the new teammates are too comfortable if their current learning rate is 2-4 years. That number could also be an overestimation, which reinforces the relative rank in the team.

This points to another "field of tension": when the value of the company is in the minds of the programmers (not in the code), how should an organization handle rank such that all team members become stakeholders in the most effective progress?

I suspect more like academic rank rather than "team lead" or "lead dev".

How does it do differential updates if everything is encrypted?

It would mean the server needs to keep an eternal log of operations, and each client must replay the log to perform a full sync.

This works with lightweight data like contacts and calendars.

The "more expensive" goods and services are not relocatable, or otherwise have special limitations on supply. College textbooks, for example, are priced monopolistically to maximize revenue as opposed to quantity. Housing is priced based on mortgage rates.

Improved manufacturing and (essentially) cheap Chinese labour is responsible for the cheaper goods.

The connection here is that "inflation" as it is measured by central banks is artificially lowered by improved and cheap manufacturing. This allows them to keep very low interest rates while being congruent with their "charter" of keeping stable average prices.

Artificially low interest rates can be seen as the cause of the most of the "more expensive" goods. It inflates the price of assets relative to income.

And there are many stake holders in asset inflation.

> How is Matrix a threat to Google?

A matrix user identity will eventually compete with a google account.

When google accounts are considered as important as myspace accounts, then much of their surveillance loses relevance.

> The time value of money can be negative

The time value is positive by definition, unless you somehow get penalized by having money in the future. Inflation and purchasing power is a separate phenomenon.

Negative interest rates is entirely artificial. Not in any way a market phenomenon.

Seems disingenuous.

Looking at the their data (p. 13), it does not look like a linear trend. There are two obvious change points; 1450 where rates dropped around 5%, and then 1680 it dropped 5%.

Without the extreme outliers in 1905 and 1930, the trend would be constant at 5% from 1680 onwards.

> Assuming that votes represent general quality

Maybe but if we let voting/karma/moderation be "perspectives" of the content rather than lossy filters, then it won't matter much.

> you need to choose who's votes you trust.

If users can freely choose their "root" in a trust tree, then it might be interesting. Each upvote confers trust and "karma".

The reason Reddit is a better user experience than usenet is 1) voting on posts and karma, and 2) continually polished user interface.

Design a decentralized protocol that can handle voting/karma, while also incentivising developers of clients. The problem is that this is not easy.

Since marginal cost is essentially zero in software, productivity as measured in revenue per head is not really relevant.

The business climate might just be better in the US.

A "placebo surgery", if it involves incisions to the point of being indistinguishable to the patient, isn't really a placebo since it will still require healing and recovery.

Managers in US software companies base their seniority on experience in software engineering. Up-valuing software engineers also up-values their own position.

Most large employers of engineers in Europe are rather old industrial companies. Management generally do not have software engineering backgrounds.

Profit-per-employee is not really meaningful for comparison since very few European companies are pure software, and therefore have cost structures involving much more than salaries. Relatively few employees are software engineers in Volkswagen, Daimler or Siemens, for example. Still huge employers.

The markets might be more competitive, which means less profits.

He believes he is paid what he "deserves".

He /should/ be paid commensurate with the value he produces.

In reality, he is paid the market rate.

> Why not (partly) you?

The short answer is that he is replaceable, while the owners are not.

> No one would like to take a risk unless it's absolutely necessary.

It's mostly a rational calculation. Because of crab mentality and stigma of failure, the downside is worse in Europe. The expected value of risk is lower than in America.

Somehow, we need the mindsets of Zara and H&M inside VW, Mercedes etc.

These have been different universes.

I'm probably using boomer to refer generally to "stale mindsets".

But the current state is the result of sequences of decisions made in the preceding twenty years. Until it was literally demonstrated by Apple, they entirely dismissed the idea that most value in phones would be generated through an open ecosystem.

There are plenty of people with stale mindsets in the European industrial giants struggling to grasp how electrification and AI will affect their legacies.

Apple and Huawei are very recent phenomenons in Europe.

GSM, and today's mobile networks, are essentially products of subsidized European national telcoms.

The problem is rather that Europeans have historically been ineffective or slow at understanding end-user "cultural" products. American pop culture still dominates, while locals usually seem like poor imitations.

But the value produced by FANGs is essentially cultural, rather than technological or industrial. This thinking is suspicious and foreign to European engineers. Boomer engineers and industrialists really struggle to grasp this and it's still a point of contention even though few know how to articulate it.

Algorithms would try to maximize expected value.

As far as I can tell, here it seems humans value choices over expected value. In other words, humans pay for the perception of freedom.

A vaccine prepares the immune system for the virus. It does not completely prevent infection or exposure. The infection is just so inconsequential that it's unnoticeable.

Early Google used commodity hardware to the point of being comical. Somewhere there is a "historical" display of commodity motherboards spread out on cork trays in a rack. Maximizing density/minimizing cost.

The ability to scale in this way with commodity hardware was new and important.

Wikipedia makes a point that contributions do not correspond to rank. It seems the article addresses typical ways in which people come in conflict with this.

Most workplaces make the rank ladder part of the incentive structure. Usually, the organizational structure is explicitly part of the rank system. Efforts and contributions are rewarded with higher rank, which affords the right to boss others.

People who derive their ego/rank from other sources besides the org structure can be humble in the way you describe. It's not surprising if this group intersects with senior academics.

Procedures don't have arrays/structs/classes/objects. This means they can't share data unless through parameters.

Polymorphism is possible, but not really encapsulation. One could argue that arrays/objects would cause consistency problems, and therefore don't belong in a data language.

Someone expecting python-like OO programming will go mad.

The performance argument depends on the whether the database is limited by CPU or bandwidth and what the load looks like. It's shown in benchmarks that the round-trip between db/network/orm/app takes many orders of magnitude longer than the procedures themselves.

There can be many databases. They can be running in parallel with the same SQL inputs. You can have provisions for this in your middle ware, but there is also stuff for this in the db replication. Replaying SQL logs on test instances. It is possible to run quite sophisticated testing in this sort of setup.

Relying on ORM doesn't make a difference, since you are still sending DDL commands to a live database.

Using stored procedures, views, and triggers makes it easier to add abstraction between your middle ware and the database structure.

But on the whole, I agree it's probably not recommendable to anyone feeling hesitant about it.

Money is not an issue at all. It's about protecting the perception of quality that comes from having Google co-authors, and ultimately the Google brand.

Otherwise, everyone and their mother would milk the Google name by submitting papers in every direction.

The walled gardens are using polished UIs and free hosting to win over open alternatives, and both of these costs money. From users' perspective, it's free and convenient.

To break these, we need technology and infrastructure that enables distributed decentralised hosting, and that decouples UIs from servers. But UI developers still need to be rewarded, and hosting needs to be profitable, otherwise they won't be competitive.

It would involve some form of micropayments system, cryptographic identities, and so on. If these things had been in place 20 years ago, IRC would be what whatsapp is today.

The matrix project sounds promising, but I'm guessing it underestimates what is actually needed to replace walled gardens.

A doctor treating a patient is not a scientist.

Neither is an academic economist that passively permits government officials to lie using scientific language.

The issue here is not the lying. The issue is the status of economics as a science. Once that status is gone, no one is obliged to listen to them any more than astrologists.