Your point is invalid. The kinds of students only partly determine the pass/fail rate. The professor is also a factor, and perhaps the major one. Not all professors are the same; some are better (or harder) than others, and it's always been that way, because professors are only human (at least until AI takes over education completely). A grading curve (e.g. a mandatory 7% D/F rate) guards against grade inflation by professors, which is a far more common occurrence than some hypothetical class filled with nothing but geniuses.

Many of the top universities grade on a curve for two reasons: 1) to guard against grade inflation, and 2) to foster competition for grades. That second reason in particular irks some people, who think it might be unfair. Yes, you could get an F despite getting 90% on every exam and homework if everyone else gets 100%. On the other hand, you could get an A despite getting only 60% of all possible points if no one else could top 40%. Overall it tends to even out. Competing for grades tends to bring out the best in students who want a good grade---you won't be able to just coast to an A, you're going to have to outperform other students. Is that unfair? Life is unfair. Students should be prepared for the type of competition they'll face in the real world.

You're confusing departmental grading guidelines with some "course standards" that you've invented. Berkeley is saying that these courses should be difficult enough that 7% will get a D or F, and that's entirely reasonable. It's really the university's way of putting the onus on professors to make the courses sufficiently challenging, which they should be.

When I was a student at Berkeley the failure rate in courses was typically higher than 7%. I remember taking a physics course where about 30% got an F. I got a B and had never been happier in my life about getting a B.

It's not my fault you don't read the news.

Anti-Semitism is all the rage among Democrats.

I'm enjoying the mental gymnastics that liberals are going through to explain the reasons for the move. They are unintentionally hilarious.

Thank you for capturing perfectly the mentality of today's American liberal: "let's kill anyone who disagrees with us!"
See: Tyler Robinson.

Or just use the well-known numerically stable version of the quadratic formula.

There's no reasoning with the climate cult. They don't even know the difference between correlation and causation, so don't expect them to understand the need to eliminate wasteful spending.

One of those reasons being the numerical instability of the standard version of the formula.

I took a summer course at UCSC long ago, as a visiting student. I got a nice written evaluation, which converted to an A at my regular university. But I noticed how short the evaluation was, and it seemed generic, which made me wonder just how much effort was put into the evaluation. I'd imagine the sheer number of students each professor has to evaluate semester after semester could lead to some copy-and-paste evaluations, with only the student's name changed.

That got me thinking about some other potential flaws in the system. For example, my professor liked me, but he clearly did not like some other students, for various reasons. A written evaluation seems more prone to being influenced by personal/subjective feelings toward a student. I also thought about how many employers would want to sift through several dozen evaluations just for one job applicant, rather than seeing a GPA.

I don't know what the optimal solution is. Both grades and evaluations have their pros and cons. And with grades, a straight scale can lead to grade inflation, while a curve can unfairly hurt or help students for being in the wrong or right class, respectively, at the time. I'm glad I don't have to make this decision.

It really is that simple. It doesn't even have to do that much with cost, at least not for the vast majority of Americans. EV market share for new car sales in the US peaked at about 10.5% in 2025 only because the people who were on the fence about buying an EV took the plunge due to the impending phaseout of the EV tax credit. Now that the tax credit is gone (as Trump promised during the 2024 campaign), EV market share for new car sales is down to about 5.8%. And that 10.5% in 2025 was up from about 8.7% in 2024, meaning there weren't as many people on the fence as anticipated.

America's car culture is something that many Europeans and purple-haired Americans just don't understand. ICE vehicles are going to dominate the US market for a long time, because that's what most Americans want. Even among the small percentage who want an EV, it just isn't feasible for many (most?) of them to own one, for a variety of reasons. That's why Tesla (smartly) focused on the luxury segment, and have about 60% of the US EV market. EVs will remain a niche market in the US for a long time.

By the way, for people who say the US will be cut off from the trend elsewhere for EVs, well, so be it. In fact, ICE technology continues to improve. There is still lots of R&D going on to improve fuel efficiency, performance, etc. Some EV proponents seem to think ICE technology is just standing still, when nothing could be further from the truth.

Fedora is not affected. Even though Fedora has algif_aead compiled in the kernel, it is not vulnerable to this exploit: running the Python script in Fedora 43 and Fedora 44 results in being asked for the root password. Entering anything but the actual root password results in an "su: Authentication failure" error and no root shell (confirmed by running id).

...that Kamala would win easily in 2024? lol

Have you tried running the exploit in Fedora? When I tried it in Fedora 43 and 44, I got prompted for the root password, meaning the exploit doesn't work.

Even though Fedora has algif_aead compiled in the kernel, it is not vulnerable to this exploit: running the Python script in Fedora 43 and Fedora 44 results in being asked for the root password. Entering anything but the actual root password results in an "su: Authentication failure" error and no root shell (confirmed by running id).