3JPLW

> some rare instances, such as if a renter commits fraud

That's still super weasely. Why not say: "if the renter fails to return your equipment?"

Oh, because that's horrifying and will scare everyone away.

That's astounding on another level, too — they encourage owners to get their own insurance policy! And yet they aren't aware of any such providers. Insane.

> A) On our insurance page kitsplit.com/insurance We explain in underlined terms that “we recommend that owners need to get their own annual policy with voluntary parting coverage in addition to the insurance provided by the renter.”

> The sad reality is that insurance companies simply do not cover voluntary parting. As a result we are, as of now, unable to offer voluntary parting because there are no insurance partners who provide it.

Now I'm sure there's a difference between ensuring a business as a whole like that and insuring yourself, but seriously?

> hope this blows up on them

It didn't last time: https://news.ycombinator.com/item?id=18332918 (7 months ago). It's the exact same story, except the victim there got his story published at a major photography site and roused enough rabble that KitSplit ended up reimbursing him for the low low cost of an endorsement at the end.

Cache/mirror for now: http://archive.is/6ipJB

Seems like that only makes it 3x harder.

They seem to be having load issues. Here's a cache: http://archive.is/45DFI

Note that this is the mobile edition. The desktop site is simply https://apolloinrealtime.org/11/

I would love to see something special in July for the 50th anniversary. I really wish I could start more than 1 minute before launch — and ramp up the anticipation for July 16 a bit more.

Edit: ah, you can manually scrub back to 20 hours before launch. I want to have this just running on my phone/computer/TV for the entirety of the mission between July 15-24.

They also had the last-changed date (which was from when I set up my account).

You can setup a SIM PIN, but I hadn't enabled it and I bet few folks do as it's not connected to the device lock code and not on by default: https://support.apple.com/en-us/HT201529

SIM PINs aren't enabled by default on iPhones and are independent of your device lock code.

It certainly took me quite some time to put all the pieces together — and I'd wager very few folks defrauded like this are able to realize exactly what happened.

I do wonder if it's still happening...

Other ways are to use phones with 2 SIM card slots or simply use a local carrier with decent international roaming support or (of course) carry a paperclip to just do it yourself.

I assumed that even if it was algorithmic they’d consider things like return rate that we cannot see.

Yes, that's why I qualified my "real-world tasks" with "random". What is clear is that:

* Ryzen has a longer branch prediction history than Intel's processors.

* This will give it an advantage on repetitive executions.

* It's a challenge to robustly measure tasks since using repeated executions to gain confidence intervals can interfere with the measurement itself.

What's not clear is to what extent real-world tasks are repetitive enough to benefit or random enough to be negatively impacted. It's likely a mix of both.

By no means am I attempting to spread FUD — I find it quite interesting and wanted to spark a bit of discussion on it.

The Ryzens have an absurdly long branch prediction history that make them much better at repetitive tasks than random real-world workflows. I wonder how much this is effectively "gaming" the Geekbench suite.

https://discourse.julialang.org/t/psa-microbenchmarks-rememb...

I know we all love the low-brow dismissal and hating on social science and paywalls (I certainly do!) but this comment is terribly worthless.

It's easy to find the paper (it's even linked below), and they of course do attempt to correct for many covariates including geographic locations and socioeconomic effects.

Same is true for any sort of science. We're trying to find some sort of a story, and thus we bias ourselves towards bugs that lead to a meaningful conclusion. Most experiments fail, but a bug can make them look like a success.

I find it especially important if you're applying supervised machine learning techniques. Far and away, the easiest sort of bug to make is mixing up training and testing and holdout data (or not doing it at all). It's such an easy bug to make because you're trying to improve the model accuracy... and a really easy way to do that is to train and test on the same data.

That article seems worthy of discussion on its own. I submitted it (expecting to find an existing discussion) but it turns out it hadn't been submitted yet.

https://news.ycombinator.com/item?id=20128029

Is the way this piece presents, e.g., "D-Day -21" and "D-Day +2" historically accurate? I had always thought it was D-21 and D+2, explaining D-Day as being the D±0.

> You could have all of the fuel at the back, behind the passengers and engines.

That would cause quite the difference in handling between takeoff and handling. The fuel tanks span the length of the fuselage in this design — and I'm sure that's intentional.

They've sold/given/transferred patents to Airbus.

We still cannot repair a spinal cord injury. A complete decapitation is a total spinal cord transection and more.

I personally didn't find it easy to consume, and others here didn't either (see the confusion about what it is, asking for context, etc). The links aren't clickable, there are multiple newline-separated links in a single cell, etc.

I'm not just objecting to the use of Google; part of an open dataset is its ease of use.

As for alternatives, I agree, it's hard to rival Google Sheets in terms of the creator's time. But again, for someone invested in curating open datasets, I'd hope for a bit of time invested in curating their own data — even if it remains within Google sheets.

They're also specifically designed to operate in absurdly low temperatures — one in fact requires temperatures under 7 Kelvin and so in addition to the sun shield there's a mechanical cooler.

https://en.wikipedia.org/wiki/MIRI_(Mid-Infrared_Instrument)

It's a little funny that someone so invested in public and open datasets would embed their own data in a Google Spreadsheet.

Yes, indeed. It doesn't talk about turning STAB TRIM back on, though, nor does it talk about turning STAB TRIM on with etrim inputs already applied — that's what I was calling your homegrown protocol.

My point is that it's easy to armchair quarterback the pilots whether they followed the checklist or went off checklist to do something better or worse. The problem started long before they got in the cockpit, though, and had been made evident once before at the cost of 189 lives. To continue to focus on pilots' actions at this point is missing the elephant for the flea.

I completely agree.

Here's the deal, though: this wasn't just one crash. It was two. The failure mode of the second was indeed different than the first: it was a failure of Boeing and the aviation authorities to appropriately respond to the first disaster.

To propose that pilots should have willy-nilly flipped MCAS on and off based on your reading of the news is just staggering. What if there was another bug where turning on MCAS when there was an AoA disagree led to an instant plunge, regardless of etrim inputs? I'd bet you'd be critiquing those dead pilots for not following the checklist — and keeping MCAS off — in such a case.

Ok, sure, but you're picking on the pilots for not following Boeing's procedure exactly. And at the same time, WalterBright is picking on them for attempting to follow the procedure. And at the same time, I'm arguing that it's unclear if Boeing's procedure would have saved them at all.

The LionAir crash was a tragedy. The Ethiopian Airlines crash was not just a tragedy; it was a complete systemic failure to appropriately react.

Yeah, should have said "commuter rail" instead of light rail. For some reason I have those terms jumbled in my head.

Yeah, we just set our tracks on fire. It's fine.

(Well, that's the ground-level light rail — Metra — that does that)