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

The STOCK Act mostly only had the effect of creating some inconveniences for lawmakers (e.g., their families could still trade based on their guidance), and even that was only the case for a year as Senator Reid's House bill S.716 effectively gutted the STOCK Act and restored the status quo, was passed by unanimous consent after 14 seconds of discussion, and was signed into law by President Obama:

https://en.wikipedia.org/wiki/STOCK_Act#Amendment

https://www.congress.gov/bill/113th-congress/senate-bill/716



You can with Molly[0], which is a much better client and a great team behind it.

[0] https://github.com/mollyim/mollyim-android


Clive Wearing comes to mind. He has a ~7-30s memory, but can perform and conduct complex pieces lasting minutes without interruption.

https://en.wikipedia.org/wiki/Clive_Wearing



There is this experiment in OpenAI Gym to land a spacecraft via reinforcement learning. Is reinforcement learning actually working for this, and are such models deployed to real rockets, instead of "hardcoded" math?

https://gymnasium.farama.org/environments/box2d/lunar_lander...


Reinforcement learning does work for this, but it's brittle. You almost always end up with a strategy that exploits inaccuracies in the physics simulation – from the perspective of the RL algorithm, useful quirks of the laws of physics – and so doesn't transfer to reality very well, if at all. Its behaviour outside the conditions observed in training is not guaranteed (or even expected) to be sensible, and even its behaviour within the training conditions is often hard to characterise.

An algorithm designed by people who know what they're doing is usually better. More effort, yes, but rockets are a lot of effort! We can afford to pay the cost for a reliable landing system.


Someone made a neat autopilot for an older version of the game actually! https://szhu.github.io/lunar-lander-autopilot/


If you go fast enough sideways, the lander goes off screen and you can fall past the ground. Then you can flip for unlimited reward. Unfortunately, you can't then crash to see a final score!

Best I could do was a 970 point crash.


Oh my goodness this is gold. I've never had so much fun failing.


Felt completely intuitive. All hours playing Kerbal Space Program re-loading the last save after crashing into the surface paying off....


If it's too easy, try doing a couple flips for more points


If it's too hard, just do many dozens of flips as you try to race away from the surface.


The real challenge is landing after you get really really high up


Unfortunately that particular test telling everyone their fingerprint is unique isn't very useful.

Try something like fingerprint.com, close your browser session, go back, and compare IDs.

On Tor Browser or Mullvad Browser, it should generally register each session as your first session, with unique IDs. If not, the useful life of the tracking should be short.

Relevant discussion: https://old.reddit.com/r/TOR/comments/qxt7di/tor_is_still_fi...


Thanks for the explanation


I have relevant industry experience, and I would advise anyone to never, ever use Priority or Express Mail for a high-value item, under any circumstances. Even with third-party insurance.

Theft is much more likely than with UPS/FedEx overnight services, and it's nearly impossible to arrange the delivery so that the parcel is on camera for its entire journey, something easily accomplished with some other carriers. And if the package lost, prosecution is almost an impossibility, as is therefore recovery.

In the event of a loss, Postal Inspectors are typically useless for pursuing package theft committed by USPS employees, even with very high value because they live in perpetual fear of the APWU (their explanation, not mine). They also hilariously attempt to pass the investigative buck to local authorities, who rightly note that it was a Federal crime committed by a Federal employee on Federal property.

The Postal Inspectors recently had a high-visibility thread on another site that very much overstated their value, and ever since that thread went up that's been the narrative. It's wholly unjustified.

Registered Mail is the only USPS service worth considering for such items, and it's actually incredibly good, and tightly controlled. It's arguably the most secure service of the major carriers, and probably where that recommendation comes from. As bad as the USPS is at delivering everything else securely, they're excellent at getting Registered Mail items where they belong. Insurance up to $50K can be purchased (and sometimes over that amount).

For someone with no experience doing a one-off, high-value shipment, that is what I would strongly recommend. The insurance also comes without the endless hidden caveats one finds with UPS that requires a complex decision tree to validate, or the BTW-it's-not-actually-insurance aspect of FedEx's declarations (for which they'll happily charge).

Third-party insurance services are worth their (very reasonable) cost, but can be difficult for the average person to secure. Services like Brink's enter the picture when you start talking about moving items above that level.

But Priority/Express is the absolute wild west in terms of secure package delivery. Temporary employees working with no oversight, with a union that will reflexively and aggressively defend people even when it's fully aware it's defending criminals (sound familiar?), and unmotivated inspectors. It's a recipe for high rates of package loss, which is exactly what it has.


Thanks for sharing your experience. USPS says 'Registered Mail is kept highly secured and is processed manually, which naturally slows the speed at which it travels.Registered Mail is not recommended if speed of delivery is important.'

So, UPS/Fedex overnight is better.


That's a good point. Registered Mail isn't fast. In fact, it is, at times, very slow.


Lately I've been seeing comments like this surprisingly often, but my firsthand experience tells me otherwise.


A relative had mail stolen from an outgoing mailbox. She filed a complaint and was contacted months later when they caught the guy. They found her checks (to charity) in his house. She was shocked to hear back so much later. Apparently they were on the case the whole time, since it was affecting a lot of people.


I don't doubt they solve cases of serial theft by ordinary citizens.

They are, however, habitually unwilling to investigate Postal employees in all but the most exceptional cases (and the value of the thefts is not sufficient to be exceptional).


In this case, it was actually an inside job, since some of the mail had been deposited in slots at the post office itself!



That's quite the URL.

Cooking in silicone is something I would probably also avoid, especially if one can't be certain of its composition: https://www.beuc.eu/sites/default/files/publications/BEUC-X-...


It's an affiliate link. (tag=hyprod-20) You can use https://www.amazon.com/dp/B07Z5SPRWM/ instead.


It's not an affiliate link... I just copied the URL my browser gave me.


Did OP make an affiliate link just to earn a few bucks from HN'ers?

If so, I suspect there are better uses for their time...


I was curious so I googled the text "hyprod-20"

It let to a forum from 5 years ago with someone accusing someone else of doing the same thing!

In the end, it seems like it's something google appends to their search

Fun little read for you:

https://www.forestriverforums.com/forums/f218/50a-trailer-30...


From the other stuff in the url I'm pretty sure OP just copied the link from an ad to Amazon, not created their own affiliate link (which would be crass but par for the course in the techbro hustle we live in I guess)


He could get commission from any sales in the next 30 days

not just this 11$ product


30 days?! That seems shockingly generous for Amazon. What happens when someone has a handful of overlapping affiliate clicks? Do the referrers all split the pool or most-recent wins?


last click wins.


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

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

Search: