I think a chart graphing a person's wealth against their level of Dunning-Kruger Syndrome will be an increasing upward curve.

They own this. Whatever disasters continue to unfold belong not to the president and his Cabinet alone, but to the entire Republican Party.

I make this comment so regularly I should probably pin it: I am a retired educator who spent most of his career working in ed tech, and based on 40+ years of experience I wholeheartedly support de-emphasizing technology in schools.

You’re asking the Republicans to choose between calling the president incompetent or a criminal. If their major concern is DJT’s ego wounds and how he’ll lash out, that’s not much of a choice.

I agree, Congress makes the final call either way, and impeachment has a numerically lower threshold. Why even discuss the 25th Amendment?

Actually - if the Cabinet invokes, he contests, and the Cabinet invokes again, he’s out until Congress votes. *Then* impeach him before the 25th vote.

The President can contest the invocation of the amendment. If he does so, both houses of Congress need to vote by 2/3 majorities to remove him. It’s a higher threshold than impeachment.

There’s a possible time gap. If the president submits a written declaration that he is competent, he is immediately reinstated. He remains president until the Cabinet submits a new declaration of his inability to lead. If there is *any* delay, DJT will do everything possible to disrupt the process.

I don't think people fully appreciate how apocalyptic things are for US science. I haven't had any new funding since 2024, but I'm still ok since typical grants are for three years. This means next year I will be completely out of funding and will have to fire everyone in the lab. It's not great.

U.S. Marine Corps Archival Footage: 5th Div. Cemetery Dedication on Iwo Jima YouTube video by Smithsonian Magazine

This is archival film of the Marines dedicating the cemetery on Iwo Jima, with the soldiers visiting the graves of their fallen comrades. Reflect on Graham's comments as you watch it.

youtu.be/P77sZWcYoZ4?...

A country that launched a war - and killed their head of state - in the middle of negotiations doesn’t seem to me to be actually concerned about “diplomatic credibility.”

U.S. Marine Corps Archival Footage: 5th Div. Cemetery Dedication on Iwo Jima YouTube video by Smithsonian Magazine

Who is “we”?

Here is archival film of the dedication the Marine Corps cemetery on Iwo Jima after the island was taken. I suggest the Senator and his ilk watch it before spouting off about sending more Marines to shed their blood.

youtu.be/P77sZWcYoZ4?...

What this administration did in Iran (both last June and this month) were sneak attacks. We were in active negotiations at the time of the attacks. It is just demonstrated once again that the president is a completely untrustworthy actor on the world stage, to either (former) friend or foe. 3/3

A sneak attack is when one country attacks another without first declaring war. Pearl Harbor was a sneak attack, because the US believed we were still in negotiations. When President Franklin called it a "day that will live in infamy," it was because it was a sneak attack, not a surprise attack. 2/

Reflecting on the president's horrific gaffe today, I haven't seen much discussion on the difference between a "surprise attack" and a "sneak attack." It's an important distinction.

A surprise attack happens when one combatant attacks another in an unexpected time or place. It's strategy. 1/

I can scarcely imagine how satisfying that is!

Speaking as someone who spent 25 years in ed tech, I would add the opportunity cost of widespread tech adoption. We have spent hundreds of millions (or billions) of dollars on tech that has had scarce meaningful impact on student achievement. What else could that money have been spent on?

He seems to think all of that part of the world is flat, too. A quick glance at a map will show you there's a mountain range on the east side of that peninsula, which sort of gets in the way.

Axios has been a consistent source this kind of dreck recently. It's gotten to the point that I know just from the headline it will be from them.

To be clear: the SAVE Act does not simply require the identification voting ALREADY requires. It demands proof of CITIZENSHIP— a birth certificate with a name that matches yours now (bye, bye, married women), or a passport— both of which are expensive and take a long time to obtain.

Not to mention the lead time. In California right now it's 5-7 weeks.

To me this sounds like someone bragging about doing the Boston Marathon on a scooter.

Tired of the AI summaries at the top of your Google search? Put -ai at the end of your search string. It doesn’t fix how crappy the search results are, but at least it shuts off the AI nonsense.

It’s excellent!

Nope! I modified a scan of the book illustration with Procreate on the iPad.

Image from Dr. Seuss’ Green Eggs and Ham, with new text: I do not like it, Sam-Alt-Man. I do not like Your AI scam

They Republicans were hammered in Congress in '18, and just barely won the majority in '24 - a tiny shift would have flipped it blue. Most of the swingiest states have Democratic administrations this year, so it's hard to picture (unless the SAVE Act passes) how the Rs can blunt the D's strength.

He's also not a "trillioniare" anyway - the trillion-dollar pay package will be awarded in the future, only if Tesla stock increases in value 850% and car sales increase 1,100%. I think it's reasonable to say that ain't gonna happen.

You can use an LLM to look at a set of data and ask it to look for patterns or make graphical representations, if the data is formatted in a manner that is compatible. However, what I have read is that LLMs still make many mistakes with these tasks; it’s simply not what they were designed to do.

The horrendous costs I’m referring to are enormous and growing energy and water consumption, air pollution, theft of content, third world labor abuses, and the economic damage to media and creators who are losing web traffic and income to AI models that scraped their content without payment.

People build the ML model, and then train it using a very focused set of data, such as mammogram images, or by having the ML run millions of simulations and evaluate variations, which is used in a wide range of research efforts. Humans design the simulations and define the evaluation criteria.