EverythingIsFine
Well, is eventually fine
I know what you're here for. What's his bias? Politically I at least like to think of myself as a true moderate, maybe (in US context) slightly naturally right-leaning but currently politically left-leaning if I had to be more specific.
User ID: 1043
Honestly rather than arrive at a surface-level verdict one way or the other, I find it both more useful and more accurate to speak more broadly and say that opinions and values about inter-generational responsibility are more grounded in culture rather than a rationalist conclusion based on axiomatic principles.
For example, the modern United States has for a long time been one of the most significantly individualistic societies of all time. Many of us do indeed carry an attitude of 18-21 as being a hard border of personal self-responsibility and somewhat related to primacy of a conjugal bond over all other relationships, and the somewhat non-sequitur that parents have a strong and longer-lasting responsibility for their kids than vice versa. However, many other cultures do indeed view family units as having their own collective sense of honor, of responsibility, and of continuity.
As an aside, I find that the whole collectivized Social Security setup isn't really related to any actual American principles but rather was just a convenient kick-it-down-the-road approach. Well, okay fine, that's not entirely true. You might find somewhat interesting the Social Security Administration's own short history, which notably traces the true beginnings of this kind of collectivization to Civil War pensions -- which to me reflects a tacit admission that the State more broadly was somewhat responsible for so many people dying unnaturally, undermining traditional self-sufficiency. Then later, you had the forces of urbanization, more mobile workforces, longer life expectancy, and basic "nuclear family" stuff. But the main Social Security idea was originally one more similar to regular "insurance" against sudden unpredictable bad things, and as time passed on the system only got bigger and bigger and the principle debt burden shifted earlier and earlier. In fact some early participants were explicitly given kick-backs in acknowledgment that they might not have participated long enough to get vested.
I know what book I read it in but it’s packed I will have to dig it out if you’re still interested - this appears to be an okay overview of a few of the issues and causality concerns in the meantime. The name PredPol sounds familiar?
The core idea is that: no, not necessarily. It heavily depends on the behavior of the police force in question. While some police forces might deter crime by their presence, other forces might by their policies generate crime (reports) by their proactivity in making stops, and yet other forces might lie about their crime data in order to make themselves look better. After all, what actually makes it to a computer-usable report in what is fundamentally a profession which emphasizes discretion is quite variable.
I disagree. If we say that an assassination plan has at best a 70% chance of success, statistics suggests two of those events drop noticeably in probability because you multiply them (now suddenly you're under 50%) if you have the same chance both times. And I really doubt you're ever going to get a plan with over that chance, it's often less. Much less, if we're talking about a single plan/event that would kill both of them at once - that's way harder because they don't get together very often in the same place without way more security than normal (such as in the White House or in a foreign country with other heads of state).
Plus, at least the Speaker is from the party with the greatest popular mandate, because of how House elections work.
Unless they get discovered and then it's like, literal war? Sounds like a bad bet.
Presidential assassins aren't really all that rational. Famously, Lincoln being killed not only failed to win the Confederacy the war, but made Reconstruction worse for the South. The limited history we have all suggests it's often counterproductive.
What actual policy or other objective would killing Trump uniquely accomplish?
How the heck does "procuring a shooter" even work? Much less the coordination or the concealment, but I'm still stuck at step 1. Like, if you're going to arrange an assassination, wouldn't you be relying on someone other than someone barely out of his teens, both to be successful as well as to not tell anyone? As a matter of actual fact, assassins get cold feet at least somewhat often, especially random people you don't know, on top of all the honeypots that the government sometimes puts out. Again, step 1 - how do you even find a dude like this, much less decide they are the best option?
Historical example of course is the actual (outside) plot to kill Lincoln. You had an actor, a former soldier, an assistant pharmacist, and a German immigrant all with targets. Respectively they were successful, got into a knife fight and ran away partway into the fight (and was almost caught as a spy before the attempt even started), ran away when said fight started, and the third guy got chickened out and went to the bar instead and got caught the next day.
Plus, if you think of all people Victoria Nuland, at best a mid-level functionary in a massive bureaucracy, knew about some sort of plot, that implies a pretty large web of people that strains credulity.
I mean, if all we get is the evidence we have... he searched about "major depressive disorder" in April and "In the days after the rally was announced, he made searches related to Trump, President Biden, the former president's whereabouts on July 13 and the Democratic National Convention."
So if we never hear anything else about motive, my educated guess is that he wanted to shoot either Biden or Trump and chose Trump due to proximity. Would be nice to get more, though.
#1 meaning what? Just a general question if anyone notable made stock bets? #2 is a good question, though I'm not totally sure if they would disclose ROE to the public, for fear of malicious manipulation.
#3: Right in the middle of the veepstakes, battleground state, and near the convention. Also a bigger rally than expected. Specifically, he teased the VP pick in his rally only 2 days before, so I think some media thought it would happen at that rally so people would know going in to the convention. Though Vance wasn't there, so maybe not. Still doesn't seem too odd.
#4 is still being revealed. Per this timeline at least they have two sources including the sheriff who are claiming they didn't know he had a gun yet and the roof attention wasn't until 6:09, with less than 3 minutes to go. That one claims that video of "he has a gun!" and people running was only 4 or 5 seconds before the first shot.
If he did, I'm pretty sure there would be a record and we'd find out? Usually people who shoot guns regularly aren't very quiet about the fact. Of course, might depend on how cooperative the parents are. We do know that he had gone to the range with his dad's AR a few times before.
I should further add that he clearly wasn't a pro. If he were always planning on using the roof, why would he have shown up near the metal detectors in the first place?
I mean, relationships with the USSS usually come out to some extent in books, biographies, and memoirs, though still not completely.
I did a quick search of 2015 to 2023 just with google, this claims most of the USSS were pro-Trump and even pro-J6 in some cases, and anti-Biden. Apparently there's a lot of talk that the USSS strongly disliked Hillary specifically. Per rumors, Vanessa Trump even got into a relationship with one officer. Overall I think there's good reason to be skeptical of claims that the USSS deliberately didn't protect Trump, that doesn't seem to match with their actual alignment as far as we know.
But the leadership of course is another matter, not always the same as rank and file. One critical question right now is who exactly was the highest level person who signed off on the plan that lacked USSS on that roof.
I can't see how anyone wouldn't love that fist pump. It is badass and awesome and defiant in a good way. I can understand people not liking chants of "fight, fight" at the convention, but personally I don't mind it given context.
I don't think the problem is (mostly) money, it's that the job is ass/boring and morale is bad. Which I think is party uncontrollable, but still points to a major leadership failure I think. It looks like (ironically) House Republicans gave them more money in 2021 than they asked for, they made a plan to hire more people, and then... just didn't. At least from what I just now read.
This survey report is pretty damning. USSS ranked 413 out of 459 sub-agencies in satisfaction. To be fair, they've been near the bad end for a long time -- 2016 they were rated the absolute worst of any, bottoming out at a 33.8% "engagement and satisfaction" score, though despite their poor ranking that improved to 57.7% as of last year despite their poor relative placement. They were last in the top half of subagencies in 2005 (first year of data), 2007, and 2011 only. So I don't think the USSS problems are recent, but they clearly are severe. Pay also was bottom quartile at 57.7% satisfaction, so one does indeed wonder where the money went after all.
Worth also noting that this most recent score broke down supervisors (80.1%, still ranked 361 only) vs senior leadership (satisfaction only 49.6%, ranked 406). Literally everything in the bottom 25%, then, but the leadership score still stands out.
I think this data supports the idea that leadership is horrible and should be replaced at the very least.
To be clear, the article says they did attempt to find him, question him, and watch him after both of the first two incidents, they didn't ignore it. At least as far as we can tell, not totally.
Well put. Agree totally. I'm optimistic we'll get a sense for what level this ends up on in the next few leaks, if we continue to get leaks at the same rate. With enough info I think you can usually tell who is spending their time on CYA
The Secret Service probably does do this. But the Secret Service was not "in charge" of the building. At least allegedly. And you know that Officer Black is not going to be as diligent. We all know how often a lot of cops sit in their comfy air-conditioned cop car on their laptop... not to be a dick, but it happens. So yes, corner cutting, 1000%. The conversation is about how much and how bad and how foreseeable those cut corners were. We're getting a clearer picture but we haven't seen the report yet, if they've even put it together yet.
To add on, funding levels which directly reflect in staffing availability are usually determined far in advance. As we saw with the Capitol police being chronically short-staffed and low-morale even before Jan 6th, these things happen in the federal bureaucracy at least somewhat often.
Why I call the extensive use of local cops at events "outsourcing". It seems to work and save money quite often, and works until it suddenly doesn't, and the bigwigs act all surprised "oh we never could have foreseen this". Seen it way too often.
If 'meat shield' is high on the list of qualifications, hasn't the protection meaningfully failed at that point anyways? Analytically, the ability to 15% more effectively take a literal bullet with your body is less valuable than for example a 10% faster reaction time, so selecting for criteria like that makes more sense. Plus, the amount of people who want to join the USSS... there's a lot of prestige, but I'm pretty sure the job itself sucks ass. Super boring, super stressful, so-so pay, and you can't even party when you go to a foreign country anymore after that last scandal
The USSS does currently have anti-drone tech in its protection detail, I think they have some sort of roughly man-portable radio jammer/wave disrupter kind of deal IIRC, but yeah, using drones for law enforcement could very well be a thing. However, privacy concerns means that local PDs don't usually (and IMO probably shouldn't). It does seem like pretty low-hanging fruit, though. When in doubt, budget might be the issue.
I think part of why is that they already use extensive general-level jamming/GPS stuff at events, and that might mess with their own drones? Plus it's easier to do IFF if literally every drone is a "bad" drone, you can shoot down/react with complete confidence. If you have your own special drones, it makes IFF hard for regular dudes on the ground.
Edit: Reddit turned up this so they at least had one program in progress. They don't use the high-altitude spy drones like we use in Afghanistan over US soil usually at all, partly for privacy and legal reasons.
We don't actually know (as of right now) based on what I've read if the Secret Service actually saw him point a gun and yet did not shoot. It's certainly possible of course. But the language I've seen does not currently imply that (not that we have very precise language in the first place, I'm sure we will in the coming weeks)
I mentioned this previously, but the typical threat model is basically just three things: guy up close with pistol or other concealed weapon, guy far away with sniper, and a bomb. The USSS is always skittish about #1 and can almost never but fully dealt with as you mention (usually it's metal detectors and searches in controlled environments, and luck plus avoiding unplanned mingling otherwise, plus a little bit of distance to give some reaction time in the worst case), #2 and #3 are supposed to be mitigated with careful prep work beforehand. Which, clearly, was not done properly. Also, I pointed out above that the sitting president probably has both higher protection than any given candidate, and furthermore that local-level deterrence is not what any head of state not currently at war relies on. Iran is only considering an assassination because they know we would bomb the hell out of them if they did and we found out, the level of USSS protection is a factor but never the main factor.
I totally agree that for practical purposes, the rally was supposed to be a high water mark of security, at least according to the standards of the time. It's not good at all.
From Exhibit B:
Butler County Sheriff Michael T. Slupe told ABC News that he was informed by other law enforcement officials that a Butler Township police officer searching for the suspicious person was vaulted onto the AGR building's roof and briefly confronted the gunman, who turned his weapon to the officer, causing him to retreat back down. Shots rang out moments later.
This is consistent with one of my speculations I mentioned a few days ago -- that the police did not have a shoot-first mentality and were expected to be point on security outside the perimeter due to outsourcing. I also talked about communication challenges. Most everything so far is still consistent with this framework, which is an incompetence one (gross incompetence to be sure, but perhaps not willful). To be fair, I say mostly consistent because:
Investigators believe that could have led to confusion as Secret Service snipers were trying to determine whether there was a threat to Trump and where it might be coming from, according to the sources.
Secret Service counter snipers positioned on buildings close to the rally stage were aware the AGR building was being used as a staging area by law enforcement, sources said. Investigators believe that could have led to a delay in the Secret Service sharpshooters' reactions because they had to first figure out whether the suspect was a threat, sources said.
But once shots were fired at the former president, the Secret Service snipers could not wait any longer and took out the gunman, sources said.
... Secret Service agents were listening to radio traffic about a suspicious person police were looking for and heard local law enforcement talking about some sort of confrontation involving police, the sources said.
These quotes are sending mixed messages. The first part of the quote and the last quote makes it sound like there were some communication challenges and clear confusion. Did the Secret Service know he had a gun? One sniper team clearly oriented themselves in the right direction, but I would direct you to this analysis that states the slope of the roof was such that the snipers did not have line of sight until the last second. However, the second part of the quote seems to suggest that maybe the Secret Service hesitated to take a shot, and THAT would be malicious, or incompetent enough to be malicious, and in that scenario I'd say that we are into conspiracy land as totally worth considering.
And if Hanlon's Razor does bears out and it was in fact merely incompetence... then we apparently live in a world where this is the best the US Secret Service can do while on high alert, actively preparing to defend their protectee against an Iranian-backed assassination attempt.
I kind of do think this is the case. It's worth noting that despite possible claims otherwise, I think candidates, even of a major party and even ones projected to win, are never going to be Secret Service protected the same as a sitting president. Or at least, are not currently protected the same. I think this is bad, and should change. Because obviously one of the two people are going to be President soon for all practical purposes, but I'm not convinced that the system is set up to reflect that properly.
However, for that matter, I'd also guess with fairly high confidence that the Vice President's protection even worse than Trump's. That might matter when it comes to questions of fairness. (If this is not the case I'd be interested to know). So if the question is "how vulnerable are the rest of US leadership to enemy agents" I'd say the answer is at least moderately vulnerable, yes! After all, in theory and in practice, most of the deterrent effect, at least for foreign nations, is supposed to be a combination of norms and most importantly the threat of traditional retaliation when it comes to assassinations. The system is not currently run that way. Cost cutting happens even at the highest levels, after all. And we are considered to be in peacetime. Obviously standards are different if we are actually at war, and funding is too. Most of the other-people protection is for regular-level crazy people and not dedicated-level crazy people, and I guess normally that is enough? Otherwise we'd see Nancy, not Paul Pelosi being attacked, or things more like that. I don't know any random crazy who would go out of their way to assassinate Vice President Kamala Harris... attention is normally drawn straight to the top.
It also depends on how specific the attack warning is. At times, the US clearly has high specificity intelligence, and the evidence is that they do usually act. Hell, we even warned Russia, practically our global enemy, that they were going to have a terror attack at specifically a concert venue, warned our own visiting citizens of such, and we were right. The reporting hasn't been super clear about how specific or "actionable" the Iranian threat was, AFAIK.
One thing is very clear however...
And why in the name of all that is holy does Secret Service Director Kimberly Cheatle still have a job?!
Biden should have fired her almost immediately, quite frankly. Absolutely crazy incompetence even in the best-case scenario. And the buck has to stop somewhere.
Current 270towin map at least has Republicans projected to get to 50, Democrats getting 48, and two states (OH and MT) as toss-ups. Since presidency is tiebreaker, I have to agree with your analysis here. The two least safe seats outside of that are Texas and Florida... yeah, no chance in hell that happens. That means even if they pick up both toss-ups and all their leaners, they still lose a tiebreaker if they lose the presidential race.
Depends how you spin it. As the proverb goes:
The best time to plant a tree was 20 years ago. The second best time is now.
This. The whole point is not necessarily that he's in full and total decline right now, just that the decline has started and that the trajectory is very worrying. Like, he can probably cope with his current 10-4 workday just fine. But in 4 1/2 years at the end of a hypothetical second term, who's to say it won't be a 2 hour workday, or worse? With how sharp mental decline often is, the difference between 6 months from now and 54 months from now can be absolutely massive. Stepping down is just acknowledging that his 54-month prognosis isn't good enough or likely enough. We've seen signs of decline, with the benefit of some extra retrospect, for about 2ish years I'd say? I think that was when aides started to limit his interactions in some form, including with foreign leaders, if memory serves and recent reporting is accurate.
Man, writing it out like that, 54 months... holy cow. No way he would make it.
Case 3 I’m pretty sure is most common in Chinese dynasties.
More options
Context Copy link