Hacker Newsnew | past | comments | ask | show | jobs | submit | modriano's commentslogin

Are we talking about survivorship bias or are you comparing comparably important levels of papers?


It's hard to correct completely for survivorship bias, but it still holds very clearly for very prominent recent papers and for other papers that happen to be in the same journal issue as the old paper I went there looking for.


For the threat profile of top leadership of the US government, yes, Signal is not secure. Signal runs on phones and phones can be compromised or lost, which can grant non-authorized individuals the ability to read the messages.

Spyware like Pegasus [0] has been able to use zero-click exploits to penetrate target phones and read messages as though they were the phone's owner.

The US has the best SigInt capacity in the world. The leaders of the US government know that phones are not secure against sophisticated adversaries and they know that we have very sophisticated adversaries. It's deeply troubling that so many of our leaders were so comfortable discussing Secret level plans in such a reckless and illegal way, and it's extremely likely that hostile adversaries have fly-on-the-wall level access to extremely sensitive US planning.

[0] https://en.wikipedia.org/wiki/Pegasus_(spyware)


> The US has the best SigInt capacity in the world.

How can anyone, including the top SigInt people in the US, know that? It has surely always been part of the principles of good spycraft that, if you've got fantastic SigInt (or other -Int) capabilities, then the best way to take advantage of them might be to make sure that nobody else knows about them.


I'm not OP, but he used mermaid.js's Entity Relationship Diagram model [0]. GitHub renders it automatically [1].

[0] https://mermaid.js.org/syntax/entityRelationshipDiagram.html

[1] https://github.blog/developer-skills/github/include-diagrams...


Thank you for this excellent post! I've been developing [my own platform](https://github.com/MattTriano/analytics_data_where_house) that curates a data warehouse mostly of census and socrata datasets but I haven't really had a good way to share the products with anyone as it's a bit too heavyweight. I've been trying to find alternate solutions to that issue (I'm currently building out a much smaller [platform](https://github.com/MattTriano/fbi_cde_data) to process the FBI's NIBRS datasets), and your post has given me a few great implementations to study and experiment with.

Thanks!


How did it take this long. I was a regular reader of Car and Driver magazine from 2002 to 2006 when I was in high school (in a Detroit suburb) and I remember them absolutely dragging the BMW iDrive system that eliminated a ton of physical controls and took the driver's attention off the road. I've kept my 2010 car largely to avoid getting a system without knobs and buttons that I can learn and use without looking.

I don't know how any auto designer could both regularly drive a car and not immediately reject the idea of eliminating physical controls for wipers/lights/temp control/sound system.


> ... historically, the US is a car-centric suburban-centric culture, and that's just who we largely are.

While we now do have a suburban-centric subculture, the car is just a tool that enables that subculture to easily access economic opportunities without having to interact with the classes segregated out of desirable locations by contemporary (explicitly racist) federal housing policies [0]. Although, I guess now that I look at it in that light (and noting the fact that every major city is still very racially segregated), I suppose that is just who we largely are and historically have been.

[0] https://www.huduser.gov/portal/sites/default/files/pdf/Feder...


> Even if this thing is 30% more expensive than a non-repairable toaster, I bet many consumers would pick the cheaper one, despite it probably not being the long term financially optimal decision.

Toasters are so cheap that I can't imagine repairing them could be cost effective (ie "financially optimal") unless you assign no value to your time. A new, good-enough toaster costs in the ballpark of $30. I love taking things apart and fixing them, but if the repair involves figuring out the part I have to order or soldering anything, it will take far more than $30 of my time.

This kind of project is for people who love to tinker. The economics do not make sense.


Repairability is not necessarily a factor here. A better toaster may need less repair over its lifetime, and thus be more cost effective over its lifetime than a cheaper toaster that will break sooner and probably make worse toast while it works.

Tangentially related: https://en.wikipedia.org/wiki/Boots_theory


If the economics make no sense, it’s only because the economics do not account for the cost to the environment. If the retail cost included reclamation, so that there was no waste at all, this kind of toaster would be the cheapest.


In addition to looking like a terrible President, Trump is also very bad at governing, especially in a crisis. Last time he was President, he inherited a great economy that he was focused on looting (giving himself and other billionaires a massive tax cut and trying to kill the Affordable Care Act to steal even more). He left office with unemployment at 15%, the country $8T further in debt, crime skyrocketing to 30 year highs, cities were literally on fire, and everyone's quality of life was horribly harmed by his terrible management of pandemic.

What was Trump good at?


Then why did people elect him again?

Listening to people who did, it sounds a lot like their lives got better the last time.

Are you saying they're lying? Or just stupid?

Can you think of another explanation for the disconnect?


If the majority of voting Americans' lives got better in Trump's first term, then Trump would have won in 2020, but he got blown out losing by 7 million votes.

Cleaning up the mess Trump left was painful. The unavoidable consequences of global disruption of the economy (ie reduced supplies of goods) was a bit of inflation, and while Biden achieved the best recovery of any developed nation with the lowest inflation and very high wage growth among the bottom 40% of income earners, enough voters were turned off by inflation and by rightwing efforts to make voting harder that Donald Trump was able to just barely pull out a victory.

Trump is eliminating guardrails left and right and unsurprisingly is crashing things much faster this time around.


Nah, Trump was never going to offer Ukraine the security guarantee needed to end this conflict. Trump is just extending this "give us $500B in mineral rights in exchange for nothing" deal to manufacture an excuse to cease all aid to Ukraine and exonerate his administration from the holocaust that Putin will execute against Ukrainians after Trump gives Putin all of the intel the US military has accumulated aiding Ukraine to this point.


Trump announced DOGE on Nov 12th, 2024 [0] and the transition doesn't all just happen on Jan 20th. The Presidential Transition Act has provisions that allow some members of transition staff to get access to federal IT systems, and with both Congress turning over on Jan 3rd, 2025 and Biden already completely checked out, it wouldn't be even a bit surprising to me if Congressional staffers or DOGE staffers started opening things up as soon as they could.

I don't have any evidence that it's the case, but if the claim about a surge of govt infrastructure appearing on shodan is accurate, I don't have any other hypotheses to explain it.

[0] https://www.nbcnews.com/tech/tech-news/trump-says-elon-musk-...


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: