Hacker News new | past | comments | ask | show | jobs | submit | rblatz's comments login

Find your last entry before the rebase using the reflog and reset your local branch to that entry.

The work isn’t lost, it is sitting right there.


> Find your last entry before the rebase using the reflog and reset your local branch to that entry.

You shouldn’t ever need to go to the reflog unless you’re in an exceptional case, and fit makes it very very easy to get into that exceptional case.


> You shouldn’t ever need to go to the reflog unless you’re in an exceptional case

If "loosing your work and it's not in the git log" isn't an exceptional case, what exactly is a "exceptional case"?


> If "loosing your work and it's not in the git log" isn't an exceptional case, what exactly is a "exceptional case"?

Losing your work should be exceptional. It's very, very easy to fuck a rebase up, and rebasing is (like it or lump it) a common operation. You shouldn't one step away from an exceptional case from a daily operation.


this is solved by branching your feature again and rebasing from that or the feature original in case the rebase gets fucked.

you don't have to treat rebases or merges as this black hole of "i have no recourse for not getting it perfect the first time".


Powerful tools have powerful consequences, case in point: with `rm` you're always one mistake away from blowing away large parts of your work.

The devs shouldn’t have access to prod credentials in the first place. That’s the real issue.

Internal audit said the same thing.

Quelle surprise!


Commiting credentials is also a real issue, best to avoid doing both.

Then you need to hire someone else to manage the deployment of services though.

You’re concerned about your phone keeping data on your phone, that it derived from data on your phone? At that point I think you are so far outside of the norm as to be considered a fringe use case.

At that point I think you are so far outside of the norm as to be considered a fringe use case.

So what if he is?

Apple has three trillion dollars. It can put in a "Turn all off" button.


Can you trade money for software quality and complexity forever?

What was learned from the cancellation of Apple Car ($10B) and market rejection of Vision Pro?

Forever? No.

For our lifetimes? Yes.


> derived from data on your phone

By violating the application sandbox!!!

Some E2EE messenger apps have "data at rest" encryption of local messages. How is Apple protecting the spidered app data?

Dec 2024, https://www.nbcnews.com/tech/security/us-officials-urge-amer...

> Amid an unprecedented cyberattack on telecommunications companies such as AT&T and Verizon, U.S. officials have recommended that Americans use encrypted messaging apps to ensure their communications stay hidden from foreign hackers. The hacking campaign, nicknamed Salt Typhoon by Microsoft, is one of the largest intelligence compromises in U.S. history, and it has not yet been fully remediated.


Do you actually have any reason to think it reads notification contents? This feature has never done anything that made me think it is. (genuine question, I’m in the EU and we don’t have Apple Intelligence here)

To be precise, there are two functions, both defaulting on for Siri, each requiring manual, per-application, opt out:

1. "Learn from this app", uses NSUserActivity API: https://developer.apple.com/library/archive/documentation/Ge...

2. "Show content in Search", uses Apple Spotlight to index all app data. For an E2EE messenger, that is not only notifications, but all text/image/audio data in message history.

As stated at the top of this subthread, these menu options preceded Apple Intelligence by several years. Implementation would have changed over time.


This isn’t he best comment I’ve seen on HN, you should delete it, or stop gatekeeping.

You are greatly understating the value and improvement that these services brought. Especially around accessibility.

In the city I grew up in taking a taxi was basically not an option. You can now typically get a ride from Uber or Lyft within 10 minutes.

In the city I moved to for college there were plenty of Taxis but could take 30+ minutes for one to show up, or maybe it would never show up and you’d have to call another company and hope they would eventually send someone to you.

These are anecdotes, but the massive increase in for-hire rides after these companies were introduced does show that they did improve the experience and made for-hire rides more accessible and appealing.


I was vaccinated and boosted, ended up catching covid 3x so far.


Now, hop in your time machine and do the A/B test.


Intramuscular vaccinations do not significantly prevent infection of the upper respiratory immune compartment (mucosal IgA antibodies, not humoral covered by IgG). The efficiency against infection by sars-cov-2 is about 20%. That means 80% get infected.

Now, the intramuscular injection vaccines do provide tissue resident B (+IgG antibodies) and T cells in the body serum (humoral, not mucosal) immune compartment and this has been proven to prevent serious (hospitalization) by covid-19 with decent (>50%) efficiency. These IgG antibodies do not seep into the upper respiratory (mucosal compartment) and provide protection there though; only the lower lungs.

But infection? The only good mitigation, lacking a proper intranasal vaccine like India has, is wearing a well fit and sealed mask like an N95 with good practices.


Infection isn't the only concern.

Severity of disease course is a consideration.


Read my second paragraph.


Serious question: How many boosters, and how many more do you plan to take until you give up?


I'm not OP, but in my case the answer is the same as flu - it's likely to be a routine annual thing. Eventually it'll probably be part of the flu shot.


Why “give up”? It’s just like a flu shot, might as well take it every year. (Until Republicans decide to ban it outright, that is: https://amp.idahostatesman.com/news/politics-government/stat...)


Because the poster said that he had covid 3 times despite the shot.


It's not a prophylactic, but it can significantly reduce severity of symptoms as well as the chances of getting long COVID.


> but it can significantly reduce severity

Only for people at risk. For randos in good health it does not change anything.


Does less severe symptoms from being vaccinated make people spread fewer virus around? If so it's like people haven't heard of probability. With vaccines people have less severe symptoms so they spew fewer virus to people with higher resistance. A combination of these things can drive the R0 value below 1 making the pandemic fizzle out. Is my amateur non expert mental model correct?

The parent comment feels like a “Reddit” comment. It appears when taken at face value defamatory and potentially in violation of California law. It implies guilt and does not use words like allegedly for things that have not been proven.

This seems like an unwise post to make, and adds nothing to the discussion. It very well may also violate the rules of HN.


>> was arrested in 2021 for domestic violence for beating up his girlfriend.

> The parent comment ... does not use words like allegedly for things that have not been proven.

The 2021 arrest (and charges) seem plainly factual. Is your position that the arrest might not have happened?


In many/most jurisdiction police are pretty much forced to arrest _someone_ (usually by written policy) if there is a DV complaint, yet the conviction rate is way under 100%. It is probably the least reliable arrest record as a prediction of finding of guilt.


As someone who long was married to an abusive spouse, I'd offer that we spouses tend to view mistreatment as a workable, if difficult, part of a marriage. Our desire to put events behind us doesn't mean that the events didn't happen.


The public doesn't know what events happened if they don't make it to trial or some other public forum. An arrest record unfortunately can't distinguish between events put behind and those that never happened. A presumption of innocence barring further evidence is not an unreasonable approach.


> A presumption of innocence barring further evidence is not an unreasonable approach.

General calls to stop bullhorning about arrest details are best directed at police and justice depts first. While they get a pass, calling out anyone else makes little sense.


You think that me, a European, is violating Calfironia law by posting a comment on Hacker News that cites the linked article?

Don't you people have a thing for free speech or something?

> This seems like an unwise post to make

What are you threatening, exactly?


unwise?


So just keep scanning repeatedly until it gets you to the right menu?

Wouldn’t it be easier to just have the link the QR code points to show a language picker?


Or use the available language information from the device to guess the best language (and have a picker to change)


Are they concerned that after their CEO was gunned down in broad daylight over his supposed unethical behavior, and the killer is being celebrated as a folk hero. That by doubling down and doing unethical things they may be putting targets on their own back?

I know if I were at UHC I’d be looking to leave before I ended up on a hit list.


> I know if I were at UHC I’d be looking to leave before I ended up on a hit list.

UnitedHealth has half a million employees. If you think individual employees are at risk of getting on a hit list then you don’t really understand the scale of the company.

I’m constantly amazed by how some people adopt the uninformed narratives that sprang up in the wake of this murder. I’ve asked multiple people to guess how much cheaper our healthcare would get if we forced insurance company profits to zero and redistributed their C-suite’s compensation to their covered patients. The answer is always off by orders of magnitude.

It’s all very strange. The narratives around this murder and, by extension, the health care system are being invented by people informed about both yet widely accepted as fact. Even details about the scale of UnitedHealth seem lost on people. This is a very large publicly traded company with scores of employees.


> asked multiple people to guess how much cheaper our healthcare would get if we forced insurance company profits to zero and redistributed their C-suite’s compensation to their covered patients. The answer is always off by orders of magnitude

The media is conflating the far-right and -left wing fringes who are celbrating Mangione as a person and his literal crime with those holding him as a symbol for their frustrations with our healthcare system.

If you're celebrating the person and the crime, you need to speak to a professional. (You're also in a minority. No, your subreddit isn't the world.) If you're upset about the system, it's reasonable to not care if you're being fucked by Bob or Alice.


> You're also in a minority

[citation needed]


"A majority of voters (68%) think the actions of the killer of the United Healthcare CEO, Brian Thompson, are unacceptable. Seventeen percent find the actions acceptable, while 16% are unsure" [1]. Also [2].

[1] https://emersoncollegepolling.com/december-2024-national-pol...

[2] https://d3nkl3psvxxpe9.cloudfront.net/documents/econTabRepor... page 12


That seems like most people who are healthy and military age are atleast somewhat OK with it. Its also excluding people who don't vote and have had their voting rights taken away.


> seems like most people who are healthy and military age are atleast somewhat OK with it

About a quarter of under-30 men and a fifth of under-30 women. For comparison, about 20% of Americans "think holding neo-Nazi views is" acceptable [1].

> excluding people who don't vote and have had their voting rights taken away

Where do you see the 2% of Americans [2] who have had their voting rights taken away being excluded by either poll? (Not challenging. That's just methodologically impressive.) Either way, none of what you're citing is material to an 80/20 margin.

I have friends in that 17%. But I also know people who thought Kari Lake was a shoe in, "defund the police" would work or that every guy is an incel. Echo chambers are powerful, especially online, where they can convince a quarter of a single demographic that they're in the majority on celebrating a dude capping a stranger in broad daylight.

[1] https://www.the-independent.com/news/world/americas/us-neo-n...

[2] https://www.sentencingproject.org/reports/locked-out-2022-es...


You're selectively moving groups around with your statistics to try and use simple numbers to make it seem like you're always right. 20% of every single person in the US being sort of into Nazis has very little to do with the vast majority of people under 40 (military age) finding it ok with Brian getting shot. The first group includes people who are getting government funded healthcare on top of not having spent most of their lives dealing with insurance companies run by MBAs

Edit: Just noticed statistical manipulation is your job and you're maybe in a similar tax bracket

>> Trade private equity. Former aerospace investment banker, and before that, algorithmic equity derivatives trader. FinTech + Space + B2C angel & seed investor. Jackson Hole local; frequently in New York and the Bay Area.


> 20% of every single person in the US being sort of into Nazis has very little to do with

I’m saying 20% of Americans can be found who believe almost anything. 20% of Americans supporting something is basically background.

> selectively moving groups around with your statistics to try and use simple numbers to make it seem like you're always right

I’m slicing the data to present a number bigger than 17%. I’m trying to find a group in which a majority support Mangione who aren’t on the far fringes of society. If you’re upset about the slicing and dicing you’re confirming my point.

> the vast majority of people under 40 (military age) finding it ok with Brian getting shot

Where do you see this?

18 to 29 is the most favourable bloc for Mangione, and there favourability is 39%. Not a majority let alone a vast majority. And that’s being generous by combining in somewhat favourable.


> I’m saying 20% of Americans can be found who believe almost anything. 20% of Americans supporting something is basically background.

The Lizardman's Constant is lower than 20%, more like 4% for the survey result that gave rise to the name — and also, despite the name, not constant: https://gwern.net/note/lizardman

Sadly, this also means that your chosen example, instead of illustrating your claim, actually has something to say about the state of the American Overton Window when it was taken.


> The Lizardman's Constant is lower than 20%, more like 4%

Lizardman’s Constant is, per Gwern, “‘jokester’ or ‘mischevious responders’.” They’re being disingenuous; think: “deez balls” voters.

I don’t think the 17% of respondents who think favourably of Mangione are trolling. I don’t think the 10% flat earthers (or, including those who are “unsure” about the Earth’s roundness, 19%) are bullshitting [1].

There is simply about a fifth of the population—and it’s a moving fifth, there isn’t a permanently-braindead section—that tends to respond one way in just about any survey about fringe or stupid theories. So 17% supporting Mangione (while making it seem like it’s everyone online) isn’t particularly surprising. (Even if a good fraction of them are responding provocatively for fun.)

[1] https://carsey.unh.edu/publication/conspiracy-vs-science-sur...


Why do people keep quoting "military age" here? There's no a priori reason why the next CEO shooter might not be underage (common in school shooters) or retired (with time, pain, and guns on their hands)?


For those like me who had no idea what coomer means, it’s a weird meme about men that failed no nut November. It seems popular in alt-right circles.


It's so strange because that must be what people who don't know about the term think, but those who do all know it's not really an alt-right meme at all even if it's technical origins are as such.

It's like if someone moved to America and said "Oh Thanksgiving? Never heard of it but I looked into it now, and it's a weird holiday based on whitewashing the murder of Native Americans". Technically it's true but you aren't a supporter of that or blind to that fact when you celebrate it with your family, and the insinuations about you from that are likely incorrect.


Not sure where you're getting that alt-right circle from, it's just regular Internet lingo. Unless you consider people like Hasan alt-right too? Would be mind-blowing for sure, but hey, to reach their own.

Got popular during the pandemic, though it predates it

https://www.urbandictionary.com/define.php?term=Coomer


The idea itself is so regressive and shaming, I don't find it surprising at all that the political right is drawn to it.


The idea of being a horny guy online and content existing for horny guys online is regressive and shaming?


I think GP meant that stigmatizing the idea of a coomer is regressive and shaming, in that it's sex-negative -- if someone wants to masturbate excessively, there's nothing wrong with that, as long as it's not negatively impacting other aspects of their life. Not sure I agree with that take, but I think that's what they were talking about.


But there is something wrong with that. The excessive part. Thanks for clarifying


Do we have regressive and shaming names for excessive water drinkers? Excessive breathers? Excessive readers?


Those aren't common things to be. We have regressive and shaming names for the excessively wealthy, the excessively macho, the excessively progressive, etc.

All things in moderation.


No, no, and yes.

There are few, if any, shuttering themselves away from regular social interactions due to an addiction to water or air.

The appearence of doing so because of a porn or more general book compulsion is more commonn.


So what's the regressive and shaming name for excessive readers?


Aside from four eyed bookworm et al you mean?


Many people still believe that 4chan et al are Alt Right.

They’re just what 14-28 year old boys are up to, sans filter algorithms. It’s the whole spectrum.


Join us for AI Startup School this June 16-17 in San Francisco!

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

Search: