raegis

I'm sure they will rethink this when Trump's term ends.

> I don't envy educators right now.

I think we have a solution for the evaluation part: the overwhelming majority of a student's grade is computed from in-person assessments only. Some instructors allow repeats on exams as well.

The teaching part is as hard, but it has always been hard. The students you get one year are completely different the next year, for various reasons. As long as you have competent administrators, all is good--this is what we sign up for.

> They have 1 billion active users.

I'm naive, but is this possible? The power required per user to run those GPUs seems like enough to run an air conditioner. For a billion people everyday or so seems like enough to overload power plants.

It can happen by accident. I assembled a system for my son (for "gaming"), and noticed that the window movement on his monitor was significantly smoother than on the cheaper monitors I'm accustomed to. It was just...nice. So I upgraded the monitor on my desktop too.

The IRS made it crystal clear to me long ago. The tax filer is responsible for everything on the tax return, regardless of the preparer. Of course, accountants can be liable in other ways, but the filer has to pay the fines if the accountant makes a mistake.

I was just correcting the statement as written. I see they meant to say something else. You are correct about my other comment--I was in a misdirected, math-correcting rage.

Apologies, I'm incorrect. I misread your math.

> One thing I feel like a lot of us are missing is that we no longer have to worry about looking for a missing semicolon, typo, or off by 1 error, or manually combing logs to try and understand a problem.

This is what I hope for, but every time I use ChapGPT and similar, I get results which take work to optimize, so it saves me little time, if any. I admit I don't use paid versions of the A.I. stuff, so perhaps this is the reason.

That last statement is not correct. (2^sqrt(n))^2 equals 2^(2 * sqrt(n)). If you just take n=1 in your expression you can see the error, your (2^sqrt(n))^2 yields 4, while 2^n is just 2.

I bet all my money that all his future kids (if any) will learn arithmetic. He's talking about everyone else's kids.

> Are those capabilities valuable if machines are better at all of them than you?

Machines have been doing arithmetic (add/subtract/multiply/divide) better and faster than humans for 70+ years. I would ask anyone making this claim, "are you OK if your children do not learn basic arithmetic?"

Before Windows XP, desktop Windows users were accustomed to frequent system crashes. Viruses in the 1990s would take out offices around the world. Nothing collapsed, and nobody switched to Linux. Everyone was just used to it. I think most people will accept the tradeoffs, because it used to be much worse than it is now.

In the educational universe, this "interviewing" is more common than not. Teachers are asked to give feedback on textbooks, calculators, and software, including lots of educational half-baked A. I. garbage. Some even offer us food for our feedback. Can you give an example of a type of software which is developed with "no clear way" to contact users?

I've been teaching math for 35 years, which I admit, doesn't say much, but I assure you the difficulties students have with complex numbers by and large have nothing to do with the name. It's the same with everything else: lack of preparation in arithmetic and basic algebra.

No offense, but they were like this in the 1980s, before the internet,

Federal criminal convicts usually serve 85-90% with good behavior, at least for men. The minimum is 85%. There is no federal parole. I think she's getting out early for a different reason.

Solve the Collatz conjecture in the next five years? If humans publish significant advances during that time, and A.I. copies it, then yes. Otherwise, I'd definitely bet money it won't happen. I'll give you 10,000 brownie points if I'm wrong.

Thanks for the laugh!

> I lived in DTLA for several years and regularly used Little Tokyo/Arts District station. It was not uncommon during peak hours to see people - myself included sliding down the railing.

The Little Tokyo stop had no stairs or escalators before 2023, since it was at grade then. Do you mean a different stop?

Also, the busiest stations with huge escalator length, like Union Station or Wilshire/Vermont, have so much open space that escalators are rarely a bottleneck in my experience. 7th St/Metro station is crazy crowded, but the escalators are short and people usually give you room to run down the escalator.

The only station I worry about for future congestion is the Grand Ave station, which has no escalators, but is so deep that you have to take an elevator since walking stairs is not reasonable except for emergencies.

This already exists. I got a Thinkpad T14s 2-in-1 last year from Lenovo, and it shipped with Ubuntu, which made the price around $50 cheaper if I recall. I replaced it with Debian, of course, but the battery life is fine and it is a quiet machine.

I also purchased the T14s Snapdragon two years ago, and it gets longer battery life (if you just use a terminal only, the battery life remaining will sometimes say 20 hours, but it's more like 14-15 realistically). However, it can get uncomfortably hot on the left edge at the speaker. The T14s 2-in-1 (Intel) does not have any hot spots, though.

The Thinkpad T14s 2-in-1 with Intel Core Ultra 5 has great battery life: 10+ hours with moderate usage. I also have the Snapdragon T14s which gets about 14 hours, but I'm still waiting to get full Linux support for this two year old laptop.

Note however, since I've been using my Thinkpad 2-in-1 on battery almost everyday for a year, it is somewhat degraded. upower shows

    energy-full:         50.61 Wh
    energy-full-design:  58 Wh

But I still get more than 10 hours on a charge.

Does the report say any of this? I only see "FAIL" on a few tests with ext4 and one with xfs.

When testing, you restrict to a LAN which simulates the real internet. This would not be hard for a company which already copied the entire space-time of the internet. The LAN should be physically disconnected from the real internet. This is the first thing off the top of my head, and I have zero credentials in this space. C'mon.

> The frontier LLM model makers have to push the edge to make new discoveries. You don't know what guardrails are needed until it hits you in the face (reusing walking in the dark analogy).

Guardrails? Restricting access to certain networks is supposed to be hard in 2026?

If I recall (too lazy to check) folks made slight improvements to Perelman's work and published it in mainstream journals, satisfying the "qualifying outlet" requirement.

Sorry, I did not specify the threat model. If I git push to an encrypted volume on a virtual server, like on Linode or Digital Ocean, anyone with access to the hardware can inspect unencrypted data on a live volume.

A much easier hack by their agents would be on their own systems, but I doubt we'll ever see an external message board full of openAI agents discussing their hacking of their own system. OpenAI not protecting itself from its agents would be irrational, but OpenAI not giving a shit about others is well known. You're giving them way too much credit.

Now that the revolution has apparently started, does anyone know of a version control system which uses encrypted storage? (I'm not concerned about performance, for I will use it for small projects only.)

Concerning [5], the article says he was rejected at UC Santa Barbara. Haven't checked in a while, but I believe UCSB historically accepts 40-50% of first time applicants. The schools were not using SAT scores at the time, and UC has banned race considerations in admissions for nearly 30 years. Surely there's more to this story.

Attorneys commonly do this.