Ordinarily, court approval would not be required for a non-class action monetary settlement (even one by the government). I'm not sure of any particular reason to think one would be required here, but haven't looked closely.

Be that is it may, it's presumably not what Sen. Wyden was alluding to, which he said was a classified legal interpretation, not simply a novel use of (potentially flawed) technology or algorithms.

Yeah - I was just wondering if you had a sense of the magnitude of its impact. I anecdotally recall people being quite motivated to get it repealed, but wasn't familiar with the data as to how big an issue it actually was.

As I recall, a lot of inventory was lost to vacancy/max rent decontrol before the 2019 reform?

Ah, so it's not the rent increase process itself, just the inventory of rent-stabilized apartments. Got it.

Out of curiosity, how is that different from the NY Rent Guidelines Board process? The capital improvement exceptions that NY has?

E.g. the government trains a model on the 702 corpus & then queries it not for communications with/about Person B, but rather more broadly for info relevant to the given investigation (without reference to B as the target). If it then happens to return info about the target, that's plain view. (??)

Thanks for the detailed analysis. I must admit I'm still not seeing how LLMs make this a new problem (I seem to recall backdoor searches coming up the last time we had a 702 renewal as well). But what if the "one weird trick" is that the LLM removes the "personalized" subsidiary search altogether?

After all, the subsequent analysis could unquestionably be for a criminal investigative purpose, to establish the specific individual's guilt for an offense that had not yet occurred at the time of the initial collection, without any necessary predicate of probable cause. Is that a 4A search?

The act of collection may be an act of analysis, but is the act of analysis necessarily an act of collection? Take a simple, non-digital example: The gov't lawfully collects the fingerprints arrestees and maintains them indefinitely. Is there a subsequent collection/search when the DB is later used?

Oh, that I'm definitely not giving him. I was half hoping that the articles of impeachment introduced against him would include "And he has held himself out as holding an office he was not appointed to with the advice and consent of the Senate and which Congress abolished before he was born." Alas.

I am definitely not inclined to be charitable as a general matter to SecDef, but FWIW, this read as intentional paraphrase to me: that he was quoting someone who had inserted the modern concepts into the quote to create an evocative image, rather than being confused about the actual text. But YMMV.

(Changing on a term basis would reduce the risk of weird race effects where the shape of the law depends on the sequence in which panels issue their decisions.)

Yeah, that just seems impractical, although I'm not exactly sure why. But there doesn't seem to be any reason why they couldn't draw the panels either on a term basis (as I think was suggested earlier in the thread) or even on a sitting basis (as many of the courts of appeals do).

If the overall docket of the Supreme Ct were roughly consistent (rather than, e.g., using multiple parallel panels to increase it), those effects ought to be less pronounced. (And esp. if combined with the "blind" cert panel suggestion, you'd be less likely to get the "same" issue over and over.)

(It also gives trial courts more certainty for their dockets. If circuit precedent bounces around more, the dist cts are going to have more pre-trial rulings that have to be reconsidered b/c the governing law changed before the judg was final under R. 54, less certainty on jury instructions, etc.)

I wonder if that's a volume effect, however. Given the much larger docket of the circuits, you'd both develop the "conflict" more quickly and, as a practical matter, the court doesn't have the bandwidth to consider all legal issues de novo. The strict panel stare decisis rule mitigates that.

I haven't thought through the game theory (which probably also depends on the ideological balance of the overall pool of justices), but one might think that having less certainty about the composition of future panels would encourage the creation of stronger (but not inviolate) stare decisis norms.

The Court today assumes that its prior precedents have stare decisis effect (even when the composition of the court has changed in the interim), and also that it can for ("good") reason overrule those precedents. Why wouldn't the same be true for subsequent rotational/random "panels"?

I think it's possible this precise dispute was resolved by the comma clarification, but why would not having "en banc" necessarily imply "stricter" stare decisis than we have today?

The latter clause is about appellate jurisdiction (not original), and the holding of Marbury is expressly that Congress can't expand the Court's original jurisdiction beyond the limited set in the prior post. But certainly agree about Congress; I just thought that's what this thread was about!

Oh well.

And the Mets aren't even playing.

He does, however, need to follow the law himself and ensure his subordinates do, which is the gist of several of the other articles, and his conduct undoubtedly "warrants impeachment and trial, and removal from office." (Among many other civil officers of the United States of whom that is true.)

E.g. SecDef has not taken a "constitutional oath . . . to the best of his ability, preserve, protect, and defend the Constitution of the United States," and does not have a "constitutional duty to take care that the laws be faithfully executed." (The PCA would like a word w/r/t the latter.)

While I appreciate the spirit and attempt of introducing a resolution impeaching the Secretary of Defense (including as a proxy for wanting to impeach the President), I wish they had edited it a bit more carefully to remove what appears to be copy-pastes from prior Presidential impeachments.

Also, FWIW, there also can't be majority-minority districts in the single representative districts anyway, so that's not a new issue. And are there any two district states where one is a plausible VRA district? For some reason I can't put my finger on, it seems unlikely?

"So the Mets with their third hit have matched Miguel Rojas."

No, and so we'll above the minimum. But there's always tomorrow.

Can Alvarez get picked off?