You have all these labs spending billions on researchers and training clusters, not seeing much return on investment, meanwhile Amazon just partners with the labs, and provides inference for their models, and that seems to be fairly profitable for them.
I'm not sure, but my interpretation is that Gary is implying that Prem Qu Nair received $20 million from the deal, and that by posting this tweet, he has violated the terms of the agreement, which generally have non disparagement clauses, and Gary will see to it that he won't receive anything.
Some of the third party apps were quite good, certainly better than the reddit mobile site, but that's mostly because the reddit mobile site is just so deliberately awful.
There aren't really any major technical reasons why the mobile site couldn't be as good.
All other things equal, higher bandwidth links are inherently lower latency.
Propagation delay usually dominates latency, so it's generally not the biggest factor, but on a simple local network with two PCs and a switch, you can expect about 1ms latency with 100BASE-T, and 0.12ms latency with 1000BASE-T.
They're not sending out 100BASE-TX ONTs to lower-tier subscribers. You're all using the same hardware; they're just limiting the packet rate to match the plan.
Regardless, the latency difference between 100BASE-TX and 1000BASE‑T pales in comparison to the difference in modem speeds between different providers' network types.
Fiber providers with an ONT get LAN-like speeds, while cable providers' DOCSIS modems add tens of milliseconds of latency, LTE modems are similar to double that of DOCSIS, and satellite providers range from similar to LTE all the way up to hundreds of milliseconds of latency, depending on the orbit and if there are ground stations nearby.
Again, within any provider, you get the same latency with any the plan, but changing providers can have order-of-magnitude differences.
This is a simple misunderstanding of the Lambda execution lifecycle, but it's maybe not a great look for AWS support, that no one there was able to explain the problem to this guy.
I guess it's possible someone did explain it to him, and David hasn't mentioned that in his diatribe.
I actually witnessed almost exactly the same thing happen in my org recently. AWS support was far _too_ nice and actually did escalate to the service team (or claimed to).
There was a few days of back and forth which should have been resolved by saying "please read the how lambda works" section of the documentation.
It's been several years at this point, but I used to be a cloud support engineer for AWS (i.e. the first line of defense against support tickets). We were trained to always be kind to the customer, even if we're frustrated with them, so a response like "please read how the lambda works" probably would've gotten me reprimanded. Customers can rate your response from one to five stars, and if I remember correctly, anything less than four required a written explanation to your manager as to what went wrong.
That job was a glorified call center; I didn't last very long.
> I guess it's possible someone did explain it to him, and David hasn't mentioned that in his diatribe.
Not just possible. Quite likely. It seems very clear the answer was "yes, that's how it works on Lambda, and it is not going to change" and this person said "but I want it to work the other way on Lambda".
I'm not sure if it's related, but the operator of archive.today has done odd things with DNS in the past, such as blocking DNS requests from cloudflare because they don't forward edns information.
I'm familiar with the voting patterns of HN users.
For this reason I tend to browse HN using the https://news.ycombinator.com/active frontpage because it contains the flagged topics that certain users of this site attempt to hide, while also preserving the vast majority of interesting tech-related topics.
It's not about voting patterns. It's about the rules of HN:
>On-Topic: Anything that good hackers would find interesting. That includes more than hacking and startups. If you had to reduce it to a sentence, the answer might be: anything that gratifies one's intellectual curiosity.
>Off-Topic: Most stories about politics, or crime, or sports, or celebrities, unless they're evidence of some interesting new phenomenon. Videos of pratfalls or disasters, or cute animal pictures. If they'd cover it on TV news, it's probably off-topic.
There's no guarantee that posts related to tech that are also related to political stuff will avoid being flagged.
Even when there is a relationship with tech political posts are generally removed if they don't align with a zeitgeist of modern America that seems to chose to fall in line instead of resisting unseemly actions by the new government.
The vast majority of stuff on HN I've seen about Trump and DOGE has been negative.
You seem to be saying HN will promote pro-Trump stuff and delete anti-Trump stuff. That's simply false. I can give a ton of highly upvoted anti-Trump posts.
I've been using "freestyle libre" devices on my diabetic cat, but they are quite expensive(~$110 AUD), and typically don't last the whole 14 day period before failing. Although I imagine they are much more reliable on a human.
I've stopped using them because it's just a bit too expensive, and my cat's diabetes is more stable now.
I opened up a freestyle libre 2, it has a EM9304 bluetooth SOC, and a TI RF430 NFC microcontroller, chatgpt deepresearch estimated the bill of materials to be about $5 USD[0].
Some companies[1] are developing reusable CGMs, the electronics are reusable, but the glucose oxidase probe, and applicator needle are still consumable. I'm not sure if that will bring the costs down much, it doesn't seem like the BOM is the main factor in the price of existing CGMs anyway.
> it doesn't seem like the BOM is the main factor in the price of existing CGMs anyway
No. A huge reason for the cost is (1) costs in R&D but also (2) the customer service/replacement requirements/etc required for medical devices.
If you were diabetic, and had a prescription for these devices, you'd be able to call Abott/Dexcom and get a replacement to a failed device shipped to you overnight. If the device didn't last the fully spec'd lifespan, you'd be eligible for replacements... etc
And on R&D; the QC bar for software and manufacturing is much higher. Every process has validation testing. So it is expensive both in R&D and manufacturing.
Another thing is that these off the shelf CGMs don't really differ in quality to their prescription only counterparts, which have a pretty extreme risk profile (patient death or hospitalization if it spits out the wrong number). They use the same parts so those costs are inherited.
Why buy the cow if you can get the milk for free?