It stands for "diversity, equity, and inclusiveness."

In this case, though, the complaints were spurious because there are no protections to circumvent - YouTube doesn't have DRM. Github said that they reinstated the youtube-dl repo specifically because it doesn't do any circumvention.

It's even worse than that. Those numbers for the flu are actually model estimates of total deaths, correcting for the observation rate. On the other hand, for COVID-19 we have a count of confirmed patients who died in hospitals. The apples-to-apples comparison is to actual confirmed flu deaths - just 3,448 to 15,620 per year in the last 6 seasons.

Oracle and their friends continue to try to gaslight the court with this strawman. No one - least of all Google - is trying to eliminate copyright protection for software. They are arguing for the status quo that interface definitions, like forms in general, are not copyrightable.

Oracle's position is that they alone can name the function that takes two integers and returns the larger "max," because "int max(int a, int b)" is in the Java API.

The first thing I noticed in TFS is how nice Linus is being (and helpful - seriously).

Contracts aren't magical pacts. They have meaning only because there is a force monopoly that can enforce them. What's in question here is whether or not these contracts follow the force monopoly's rules for contracts, and thus whether they'll be enforced. This morality play about "words" and "consequences" is a non sequitur.

Bill's argument doesn't make sense. Any institution has finite investment resources. If they divest from industry A, they can apply those resources to industry B. The vast majority of proceeds from divestment are being immediately reinvested in something else. Unless you can print money to invest without selling first, by all means divest from fossil fuels - and reinvest in renewable energy, or whatever you think is best.

Drivers also idle in random places in the road, generating more congestion, and tend to be unfamiliar with local areas, leading them to drive at inconsistent speeds. I also suspect the app in line/shared/carpool mode of directing drivers out of their way through busier areas in order to fish for another passenger.

I've also talked to quite a few drivers who commute into San Francisco from Sacramento, Stockton, and other relatively distant areas, these jobs generate both local and commute traffic.

Uber and Lyft receive large implicit and explicit subsidies, including corporate tax breaks, evading employment costs by skirting or bending labor laws, transferring overhead to drivers, depending on public roadways while not paying taxes paid by traditional taxis, and finally not having to pay for negative externalies. You need to estimate these as well in order to compare ride hailing to mass transit (including public and private mass transit offerings).

Do you live in San Francisco? In some districts, every third car on the road is an Uber or Lyft.

You misunderstand the process. Peer opinions are based on the results. They are also based on years of study leading to an appreciation of what results are actually 1) interesting and 2) useful. These are crude words for the distinction, but to illustrate, if AlphaFold were to work perfectly it would only be useful. It wouldn't improve understanding and thereby advance science beyond making some specific current task potentially easier. (Even if it might be really great for engineering).

There's good reason to think this training set doesn't contain all the magic rules. What the AlphaFold team should do is use structures solved before 2005 to train their model, and structures with novel folds solved after 2005 to test. If they can achieve very high absolute performance in that context, all critics will be silenced.

That's why they should use the historical validation approach! Train on structures solved before 2005, then predict only novel folds solved after 2005. Perform well in that context and I'll be impressed.

Unlike the physics-based and statistical potential methods, can the DeepMind approach ever contribute to understanding how proteins fold? IMHO that's an open question, and one that's critical to their presumably forthcoming publication. For example, do their features weights say something interesting about cation-pi interactions? Rosetta infamously ignores cation-pi because of overfitting concerns (even though cation-pi can be very structurally important).

That's not a correct reading of the results. First, previous efforts are based on putative understanding about how proteins fold. Obviously, this understanding is incomplete - or the physics based methods would perform better. (Even statistical potentials like in Rosetta are physics based in important ways). Second, DeepMind isn't even on the radar in the server component of CASP. The server competition is intrinsically more difficult because it requires robust software that isn't highly dependent on user parameters. Rosetta for example is ~20th in the general competition and 4th place in the server competition.

Finally, DeepMind has not demonstrated the historical performance of their approach. They should see how well novel protein folds solved after e.g. 2005 are predicted using only structures solved before 2005 to train. To the extent that Rosetta works, it works in such an environment. In fact, one its first results was a novel fold (Top7).

Who believes Ajit Pai actually cares about or is working on this?

Random name-calling, totally orthogonal to the comment thread, is an interesting use of your time.