Hacker Newsnew | past | comments | ask | show | jobs | submit | darknavi's favoriteslogin

So does Time Team, and they're back making new episodes.

https://www.youtube.com/playlist?list=PLtpbubYLW2fdf9ySyCS0f...


For those fans of The Shining, I feel obligated to share my favourite YouTube investigation into a remarkable and little talked about detail.

https://www.youtube.com/watch?v=Qr6PgWFs0Pw (Overlooked! A detail in The Shining that you’ve never seen)

It examines the fact that Jack keeps stealing glances straight at the camera throughout the film (no-one else does this). Probably the best explanation is that Kubrick wanted to unnerve the audience and it's very effective.


Zigbee is good, but for lights or anything else you really want to always work, I'd recommend going with something that also works independently of HomeAssistant and Internet access even if you then need a proprietary hub to integrate with HomeAssistant. Lutron's Caseta system is what I went with.

I've found HomeAssistant with Zigbee to be extremely reliable, but the idea that a botched software upgrade or failed VM could make it so I can't turn my lights on and off feels like a bit too much. Caseta doesn't use an open protocol, but it does allow all your switches, remotes, lights, etc, to work totally independently of HomeAssistant or Internet/WiFi, and the hub allows integration with home automation or remote control. I've played around with Philips Hue and other smart bulbs, but it feels too much like a toy rather than something I'd replace all of my lights with, and ultimately I ended up with an all Caseta light setup for my home.


I you are using googletest, you owe it to yourself to check out catch2 which I find much better and uses modern C++. There are a few other test frameworks in C++ that look better than google test as well, but catch2 is the one I settled on (and seems to be the best supported): feel free to check them out.

I've given up on mock frameworks. They make it too easy to make an interface for everything and then test that you are calling functions with the expected parameters instead of the program works as you want. A slight change to how I call some function results in 1000 failed tests and yet I'm confident that I didn't break anything the user could notice (sometimes I'm wrong in this confidence - but none of the failing tests give me any clue that I'm wrong!)


This video by the dude it happened to is, imo, more interesting than Rossmann: https://www.youtube.com/watch?v=zDAg1aVnnH4

I recently came across the videos of Oregon State horticulturalist Andrew Millison, who is an amazingly clear presenter about permaculture: https://www.youtube.com/@amillison/videos

For example, he has nice videos about chinampas in Mexico https://www.youtube.com/watch?v=86gyW0vUmVs , a medieval Indian canal tunnel of a style found throughout Iran https://www.youtube.com/watch?v=Kplvq0C-cdE , efforts to return a Hawaiian watershed to something like its historical state https://www.youtube.com/watch?v=q7q8friw1p8 , the Indian Paani Foundation's "water cup" to revitalize drought stricken villages https://www.youtube.com/watch?v=DXqkSh7P7Lc , and the difficulty of using ancient iron-saturated aquifers for irrigation in western Egypt https://www.youtube.com/watch?v=QBC5wOLF1hQ


I can recommend PicoShare which I've been self-hosting for a while. It's a couple of years old now and still maintained. Also, it's very simple and allows for guests too.

https://github.com/mtlynch/picoshare


You mean, like every other service anybody stands up ever? The ones that charge money are the ones that stand any chance of surviving.

https://archive.is/7LxAh


Fun fact! EVs use powerline networking for vehicle<->charger communication. Since multiple chargers and cars might share the same circuit (and L2 network), in order to identify the connected car's network endpoint, they've invented a "sounding" protocol to determine the physically closest network node based on signal strength.

Read ISO 15118 for more details. Here's a paper where they built a sniffer and explore the protocol: https://www.sstic.org/media/SSTIC2019/SSTIC-actes/v2g_inject...


The lawsuit[0] is worth a read and goes into quite a bit of detail about the claimed timeline. Roughly:

* Upgraded missile defense systems were deemed a national security priority around this time

* Navy ships compatible with these systems were 5 years out, so…

* Live missile testing happens at a compatible land base in New Jersey, under congested airspace

* Multiple civilians report seeing missile tests in & around the date of the TWA crash

* Missile test hits TWA jet. Instead of the NTSB running the investigation (like normal), the CIA and FBI are immediately put in charge

* CIA/FBI confiscate records, run PR campaign claiming the crash was “NOT A MISSILE”, mislead the families and general public about the incident

* Missile tests continue post-crash

Damning if true, to say the least.

[0] https://www.docketalarm.com/cases/Massachusetts_District_Cou...


> As soon as Sphero completed the acquisition, bringing Kelsus in as an app developer, two things happened: 1) Sphero told us they needed fully revamped iOS and Android apps six months later with no schedule wiggle room. 2) Sphero spent the first two of those six months organizing a team on their side and hashing out requirements for the new apps.

This exact pattern has preceded every mismanagement disaster I've ever seen:

1) Management gives an unreasonable deadline to developers before defining the deliverables. Clock starts ticking.

2) Management then fails to settle on the requirements for months and/or continuously moves the goalposts while people are trying to work, turning an unreasonable deadline into an impossible deadline

3) Developers start realizing that they're going to receive the blame when impossible deadline isn't achieved, despite not having any control in the matter. They start leaving for new jobs, and the death spiral begins.

It's the universal pattern of a management team that doesn't know how to do anything other than make demands and then apply pressure. Unless you can get someone in charge who has real leadership and planning skills, it's not recoverable.


Consider applying for YC's Winter 2026 batch! Applications are open till Nov 10

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

Search: