Hacker Newsnew | past | comments | ask | show | jobs | submit | pabl8k's commentslogin


That is the one, thank you!


I've heard a similar story from someone who was at HP when the donut carts ended, it signaled a significant culture change.


My doctor accidentally wrote a prescription for a three month supply of a medication for an acute illness that is generally not prescribed and certainly not filled for more than a month of quantity at a time, and generally not for more than 10 days.

When I placed my order, I thought the price was high, but figured it was because I hadn't hit my deductible and it was an expensive medicine. There was no obvious indication on the UI about the quantity of the prescription, just the price.

Amazon pharmacy accordingly sent me almost a small pharmacy-worth of this medicine. I called them and asked why the pharmacist didn't notice that this was a _highly_ unusual quantity of this medication (The quantity was *EIGHT* retail packages of the medicine) and raise a flag and check with the provider or me. The pharmacist I spoke to told me that they personally would have done so in this circumstance and told me they would refund the order and provide feedback to prevent this sort of thing from happening in the future. I later got an automated email that the refund request by the pharmacist was rejected by their finance department and could not be appealed.

I won't be using Amazon pharmacy again.


- Boss, can I pull "Warn about the unusually large order" task in the sprint?

- Sure, Sam, right after you deliver the Prime Video integration.


It takes time for a tick to transmit lyme disease, but you can also take a single dose of doxycycline as post-exposure prophylaxis if you think it was attached for a while and ask a doctor


They recommend that dose if it’s been attached more the 36 hours and it’s less than 72 hours since removed. I fell into this slot last fall and got a single dose of doxycycline.

https://www.cdc.gov/lyme/resources/FS-Guidance-for-Clinician...


Odd how the sibling comment got downvoted into oblivion for this, but you're right - prophylaxis is an option here, and last I checked it was the recommended course of action if you develop the bullseye ring. Note that the tests have high false negative rates, so still might be sensible if you have a negative test.

Of course, these antibiotics are not without risk too.


Some points correcting what's been posted here and above. A single dose of Doxycycline is not an effective prophylactic. It failed me and other people I know (after my doctor and I read over the CDC guideline that was misinformation). The original study supporting it was poor regarding how it was established as effective (required the bullseye rash). The bullseye rash does not appear for everyone. Depending on the attach point/strength, ticks pass along bacteria and/or viruses in minutes. Testing the tick and monitoring yourself for symptoms is always a good practice. I prefer clothing & topical solutions vs ingested chemicals.

It would be nice to see support for natural consumption of ticks by fowl (chickens, turkeys, quail, etc), marsupial, etc.


What I read was that the bullseye was considered sufficient evidence even without a positive test to administer antibiotics. I don't know the recommended dose, perhaps a single dose is not enough.

What exactly are you correcting here?


And possibly even if you do understand it! It seems like it might be a fundamentally intractable problem with LLMs, even if it can be made more difficult to do, no?


Yes, exactly: right now I still haven't seen a convincing reliable mitigation for a prompt injection attack.

Which means there are entire categories of applications - including things like personal assistants that can both read and reply to your emails - that may be impossible to safely build at the moment.


I watched a cross-country trip review of a Volkswagen EV and one of the big takeaways was that despite funding Electrify America, the experience of using their chargers, even with a Volkswagen, was terrible at the time of review. Not only were the chargers often out of service or underpowered, you had to use some app on your phone where you type in the charger ID # to turn it on, instead of the car negotiating it.


My personal experience has been bad enough to not want to use it again.

First CCS cable I plugged in was broken or something because it would keep erroring out before charging. This is after 1+ min of using their app to connect to it each time (3 tries). Switched cables and it happened again on the first try. Tried again and it says it errored out again but starts charging. Obviously that is concerning but after 5+ mins of trying to get it started, I gave in. When I get back, my phone says it’s done and a $10 charge but the charger says $16. Check the app after and it says I was charged $16. Why is there such a significant discrepancy?

It’s such a backwards experience after using a supercharger where you just plug in and walk away.


My experience is the opposite for what it is worth.

I used to own an ID.4 (which was a great car in my opinion, but I recently moved from Dallas to Detroit and, ironically, was able to go car-free in Detroit).

I made a round trip from Dallas to Denver without any Electrify America charger issues in 2021 (no waiting times either).

I also made a round trip from Dallas to Detroit in September 2022 and I only encountered one (1) slow charger where I had to move my vehicle to the next available charger (I had to wait at an already full charging station for about 30 minutes at one (1) stop also).

I was pretty surprised by both trips in terms of the lack of hassles.

Today, I have zero reservations about driving any EV over long distances.

I was worried on my first Dallas to Denver leg, but after the trip was successful, my charging/range anxiety is gone for good.

Perhaps my experience would be different elsewhere in the US, but for the Midwest, my experience had been good.

Just my two cents…


Zero?

I'm about as pro-EV as it gets and solar/wind.

Charging away from major interstates in the Midwest is very very very iffy if you aren't in a Tesla, and if you ARE in a Tesla it can get iffy if your destination doesn't have a home charger.

For US people, a 500 mile car really is where the sweet spot of convenience is becausee:

1) you're not going to charge it all the way up (unless you have LFP chemistry), so knock 5% off of it

2) you're battery will lose 10-20% range over the lifetime of the car, so we'll take off 10%

3) Winter can knock another 10-20% off of range

4) and of course since you need to plan ahead to the limited stations, you can assume you'll not want to get to the "vapors" and assume 10% is less.

5) fast charging is only to 80% anyway.

Suddenly, your 500 mile car is really a 300 mile effective range.

We really need some sort of range extending trailer or similar simple scheme.

I'm really disappointed there isn't any 50-100 mile all-electric range PHEVs ono the market. This is perfect for electrifying all my short and medium range trips, but makes the long distance a seamless experience until charging stations are up to snuff.

If the hydrogen lobby (not that I like them) had any sense they would have pushed fuel cells + 100 mile battery as an effective compact PHEV format (since IIRC fuel cells can use gasoline), that would have developed the fuel cell economies of scale, but since hydrogen and BEVs are mortal enemies, not likely to happen. I also had hopes that the "inside out rotary" patent from a few years ago or Mazda engineers would cook up a very compact rotary recharging engine, but alas that never came to be.

Toyota was the company best poised to do this format of car (arguably should have been working towards this since the Prius was introduced in 1997), but they were so ossified and fat from being at the top of the automotive industry for 40 years they sat on their hands as the entire BEV revolution passed them by. We'll see now that Toyoda is retiring...


I've done several 2000 mile road trips in a 320 mile range EV. That's a 225 mile range 10%-80%. IOW just over 3 hours. I know I can drive longer than 3 hours without stopping for food, coffee, a bathroom or sleep, but I don't wan't to.

On a short road trip you can drive for 5 hours without stopping because you recover at your destination, but on a long one it's so much nicer to stop regularly.

Adding the complexity of a range extender for trips in the 3 - 5 hour range just seems silly.


If you really want a power-dense engine, teh two stroke opposed piston opposed cylinder is an interesting option.[0]

As for range-extending trailers, Tesla cofounder J.B. Straubel had it right. The last time I linked to his "pusher trailer" the response was surprisingly close-minded for HN. The reasoning was (paraphrased) "I don't believe it because I don't trust the builder that it works because I don't believe it."[1]

[0] https://idaoffice.org/posts/rebirth-of-opposed-piston-engine...

[1] https://news.ycombinator.com/item?id=20747218


Mazda rotary is revealed, I don't know is it small vs I3 https://www.caranddriver.com/news/a42485033/mazda-mx-30-r-ev...


Very interesting. The chorus has the lyrics: "a message from my heart" and then "listen to this message" before going into the section with the morse code message.


I noticed that the job posting locations and the headline locations don't match. Are these for Canadian candidates only or are all roles also open to US remote candidates?


Some of us will be deliverators for NvidiaBlackrockDominoes


Nah, the deliverators will get automated by FritoRaytheon


The reason it's not the "best Signal" is that WhatsApp doesn't have reproducible builds or any guarantees that e2e isn't subverted on the client side or removed entirely. And it's run by a company with incentives that are misaligned to e2e encryption and a history of product updates that don't respect the privacy or preferences of the end user.


Otoh, maybe a small company like signal is easier to completely subvert by Cia, nsa, Mossad than meta is?


Both tor and signal are funded By one or another branch of the American intelligence community. If you want more info i recommend googling "signal radio free Asia" It's original function was to allow CIA agents posing as reporters for the CIA mouthpiece called Radio Free Asia to be able to report their discoveries. I use the app but in many ways we are actually providing cover for the military and state intelligence agencies of the United States government much in the same way as when people use tor which is openly a project of naval intelligence


> If you want more info i recommend googling "signal radio free Asia"

Could you provide a link? All I can find with DDG is a reddit post and irrelevant stuff and I don't feel like using Google.

> tor which is openly a project of naval intelligence

Again, do you have a source?

That's the first time I encounter these claims.


https://en.wikipedia.org/wiki/Tor_(network) in the first paragraph of History.


Oh, indeed. But funded by and subverted by isn't the same thing.


I went and looked at the application form for the first job linked and this is absolutely standard in the US at every company I've ever applied to. It's not just some Wikimedia thing. It's voluntary to fill it out and the company is not supposed to use it in deciding who to hire. The purpose is to have retrospective data a company can use to make sure they are not introducing bias into their hiring process.


Here’s the EEOC information page on the practice:

https://www.eeoc.gov/pre-employment-inquiries-and-race

From an applicant perspective, while physical tear-off sheets for mail-in applocations (one can imagine hand-delivery setups that solve this) are clearly imperfect, web forms are even worse.


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

Search: