ADSBX used to be volunteer ran until JETNET paid the guy who controlled the domain name $20 million dollars to "sell" it to them and steal everyone else's source code and data. They now do selective filtering to appease their commercial clients.
This only covers container ships btw. For full coverage of all vessels, try the 'vessel presence' layer in Global Fishing Watch's interactive map, based on a feed from Spire: https://globalfishingwatch.org/map/
RLS is very useful and can solve multi tenancy and other problems, but it is complicated and can add a significant per row cost to queries if your policies get complicated.
The common path of comparing some constant like the role name to some column in the table is fine, and it's fast enough as the policy checker already has the row in hand when it does the check, but the natural tendency for people to want to abstract their policies into a function like has_permission() will blow up fast.
The best approach I've seen from pyramation's launchql [1] which precomputes policies into a bitstring and then masks that against a query constant bitstring of required permissions. Flexible policy definitions compiled into the row as bits so the check is as fast as possible.
- Download as many LLM models and the latest version of Ollama.app and all its dependencies.
- Make a list of my favorite music artists and torrent every album I can.
- Open my podcast app and download every starred episode (I have a ton of those that I listen to repeatedly).
- Torrent and libgen every tech book I value. Then, grab large collections of fiction EPUBs.
- Download every US Army field manual I can get my hands on, especially the Special Operations Medic manual, which is gold for civilian use in tough times.
- Download every radio frequency list I can for my area of the country.
- Download digital copies of The Encyclopedia of Country Living by Carla Emory, Where There Is No Doctor, and Where There Us No Dentist.
I already have paper versions of almost all of these but it’s handy to have easily-reproducible and far more portable digital copies.
Homeland Security tries to get local first responders to join the SHARES emergency radio network.[1] This is 5 HF channels at 5 MHz, and some more around 15 MHz. They test on Wednesdays around noon. Transmission is voice or PACTOR. It's ham-type technology for
government emergency response.
Although many local first responders are not on this net, the USCG, military, and Homeland Security monitor it. So it's a way to reach U.S. Government resources in emergencies. This isn't something you access with a handheld, since it requires at least a long-wire antenna. You can get hundreds or thousands of miles of range. The idea is to have something that can get through from a large disaster area.
LoRa is just a transport layer. You can do whatever encryption you want and LoRaWAN has some basic encryption built in. It's hard(er) to triangulate if you don't have constant traffic (like route updates in mesh, or heavy concentrated users like at a concert or protest), but be aware that AWS sidewalk (and all the alexa devices fielded in peoples homes, etc.) run LoRa antennas and traffic. So in the US, Amazon could do a reasonable job at triangulation of frequent emitters. In China and EU there is infrastructure in place (5G/SDR stuff overtly) that can do a pretty good job at triangulating a wide band of RF emitters.
Meshtastic is not really designed to avoid that, but more for resiliency and off-grid type scenarios. Your best bet of really avoiding triangulation by state or telco level infratructure is to get creative with frequency and even transport layer hopping. None of which is really consumer friendly.
[edit-to-add] another tactic to for low probability detection is to hide in noise on high traffic channels. basically figure out what their filter sensitivty is and see if you can go below that threshold and still maintain coherent channel, etc.
Kingston NV2 is in that "what you get may differ" category, and Kingston explicitly says that what you get may change. I have two NV2s with differing die count, for example. Their controller might be different too. They're external, short-use drives, so I don't care.
So, returning to previously mentioned ones, from their respective datasheets:
If you also happen to think SupCom was the most ultimate superior RTS ever made, because it's truly strategy scale with a thousand units PER PLAYER including on 8v8 maps, and the simulated ballistics of almost all weapons means you can defend your base from an incoming nuke by scrambling your jet fighters so that they run into the incoming warhead and airburst it "safely" above your base, and love that artillery shells can also hit those same planes, or that a plane shot out of the sky can fall on a tank and destroy it, or that you think a giant walking mech with a backpack nuke launcher is the metalist fucking thing,
Check out Forged Alliance Forever. It's a community made, open source, "launcher" for the Supreme Commander game, that patches it, mods it, allows you to interact with other players and find games and get ranked against them, and also is a replay viewer, so you can watch how the pros actually do the insane micro and economy play that earns them the 2000+ ELO score, and also a mod manager, and also a friends list.
If you are like me, and love all that but actually suck at playing strategy games in general, check out Gyle on youtube https://www.youtube.com/@GyleCast who has been casting SupCom games for at least a decade now, and really shows off some of the best that the game has to offer, including multi-hour "EPICs" that can involve ten thousand units controlled by several semi-pro players.
It's not exactly a single place, but this page [0] lists the Github organizations for many federal agencies. It's not comprehensive, notably it's missing the NSA organization [1].
Soldering required here is _INSANE_. There are industrial flying probe machines that can perform same task in fully automated manner with no soldering, but typical Chinese RE involves sanding the board down one layer at a time https://www.chinapcbcopy.com/pcb-reverse-engineering/
There are Chinese outfits offering this service at really low prices, we are talking hundreds of dollars per pcb.
The complete set of offline topographic maps in GeoPDF format with embedded satellite image layer (that can be enabled or disabled if your PDF reader can handle layers properly) is ~2.8 Tb.
Sounds a bit like the very common story, "We hired this STAR EMPLOYEE from BIGCORP to do BIG things and it was a DISASTER! Help!"
I have often thought about this throughout my career, both being in the reporting chain of said star, and once being that star. The fundamental thing is that a persons performance in a role is much less under their own control than even they often expect. So why is that? Especially in management roles?
People have a reputation that permeates the organization they lead. It will have plusses and minuses, fans and haters, but generally there is a tremendous amount of context about how they lead and how to interpret their actions. There is also a tremendous amount of history about their choices and how they have dealt with both good and bad outcomes from their choices.
Additionally, people who have been in an organization a "long time" (years) and especially if they have "grown up" through that organization, have a really solid network of other folks they can turn to who both know what they can do, and understand what they are asking.
When you change jobs there is a "honeymoon" period where what you do, or don't do, can always be ascribed to "they are new here." It is imperative that new leaders use this period to "course correct" to get to a point where they can effectively lead the new organization they are leading. This is the most important thing they need to do in their first 100 days. Even if you are dropped into a complete dumpster fire, understanding how to "wield" your organization is critical if you want to actually address the fires. On at least one occasion I watched a VP Eng parachute into an org and flame out horribly (even though they had the chops to get the job done) because they focused on "getting things back on track" before "getting to know the organization."
If you work for someone who is new and "previously experienced" (which is to say they got their experience somewhere else). Helping them learn the ins and outs of the organization will benefit both you and them. You because you won't be subjected to obviously stupid decisions which are only obvious if you know how the organization works, and the "new guy" (assuming they aren't a sociopath) will appreciate your efforts and look favorably on your future inputs.
If you're on iOS, I highly recommend Yattee[1], a free Youtube app with no ads, background playback and sponsor block built-in.
The way they get around App Store restrictions is by claiming that they aren't a Youtube app. They're technically a video-watching app which is supposed to be used with your own personal server. The server API they require, however, is supported by Invidious[2], an open-source, privacy-preserving Youtube front end. If you configure an Invidious instance in settings[3] (you don't have to host your own, there are plenty of them out there), you effectively get access to the entirety of Youtube.
It's not a replacement for the official app by any means, it doesn't have a recommendation feed, it doesn't let you log in with a Google account, so it doesn't sync watch history with your other devices, it doesn't let you cast to Youtube-enabled devices, post comments, scroll through shorts etc, but if you have a playlist or a specific, longer video in mind, it's quite good. It also supports Invidious accounts, which let you manage channel subscriptions.
> In a distributed system, it is not possible in practice to synchronize time across entities (typically thought of as processes) within the system; hence, the entities can use the concept of a logical clock based on the events through which they communicate.
It is interesting that they have a day devoted to embedded using MicroBit v2.
The older Rust-Embedded Discovery book [0] also used Microbit but the later edition [1] suggests a STM32F3DISCOVERY. For RISC-V and Extensa fans there is also Ferrous Systems' Embedded Rust on Espressif training material [2] which has a coordinated board [3] that has been "cloned" by Wokwi [4].
We're actually in process of building this for individual components (bolts, motors, gears, etc.) at: https://beta.govolition.com where you can search for components, filter by spec, download CAD/specs, and purchase the parts.
Some comments on a great article from an EE who has many axes to grind about stupid old decoupling myths (which this article gets mostly right!):
> A related trick is to put ferrite beads on MCU output lines
You must be extremely, extremely careful with this. In general, ferrite beads are hard to apply correctly and great care is required if you want them to work out. Do it wrong and you'll probably be making things worse, instead of helping. See for example:
Honestly, the datasheets are usually just wrong. The writer is an intern and the material is cargo-cult copy-pasted. If it works it's by accident, or because it wasn't critical in the first place. Read the datasheet, understand what it's trying to do, then go ahead and achieve that end in the most sensible manner.
> It is true that at very high frequencies — hundreds of megahertz — the capacitor’s residual inductance becomes a limiting factor. At that point, combining multiple different capacitors can offer somewhat better wideband noise suppression.
Nope. It's still wrong and dumb to do this with MLCCs unless you have simulated the hell out of the whole thing. You should practically never parallel different values of MLCC. Instead follow EMC wizard Henry Ott's advice: pick the smallest package you're willing to deal with, then the largest capacitor you're willing to pay for in that package, and just use that everywhere. This is called "big V" decoupling by Ott and decoupling master Bruce Archambeault and it is not the best way to do things, but every better way is much, much, much harder to do. "Big V" will work for everything that doesn't involve underfilled BGAs, and even most of those.
> tantalum polymer
I actually kind of hate these guys, I don't find much use for them outside DC-DC converter output capacitors. They have too little ESR to damp things that need it, too much ESL for high frequency use, and are just too expensive for general use. They're not bad or anything, they just don't really have a sweet spot. MLCCs plus a few cheap high-ESR aluminum electrolytics (often found these days as the high-temp long-life parts) is a really effective combination. Maybe a few solid tantalums for intermediate bypass if appropriate.
> Y5V
Thank the heavens that these are basically extinct. Good riddance.
Okay, enough comments, you might then ask, how the hell do I decouple things in production designs?
First, put one bulk capacitor, minimum, on every rail. Aluminum electrolytic if the rail leaves the board ever, maybe tantalum if it doesn't (or maybe not, solid tantalums have... reputations). My go-to series is Rubycon YXM or YXJ for through-hole electrolytics, Nichicon UCB/UCW or Chemi-Con MLE/MLF for surface-mount, and AVX TAJ for tantalums. These can physically be located anywhere.
Then sprinkle down one 1uF 0402 per part for the small parts, or one per power pin for the big digital chips. Place these at the power pins, no exceptions. Things like MCU analog rails usually don't need ferrites but might get pi filter type structures. It depends on how important they are, really. If it's a big or dense board, toss in another tantalum or two physically near the chips or chip clusters to help keep the electrolytics honest.
You can decouple 500MHz processors and pass radiated EMC at Class B with this approach. It's not hard to do, it's cheap (but not cheapest, this isn't going to get you to Shenzhen-special COGS), and it works great.
I work for a vector database company (Pinecone) and can confirm that most of the mind-blowing built-with-ChatGPT products you see launching every eight'ish hours are using this technique that Steve describes. That is, embedding internal data using an LLM, loading it into a vector database like Pinecone, then query the vector DB for the most relevant information to add into the context window. And since adding more context with each prompt results in higher ChatGPT costs and latencies, you really want to find the smallest and most relevant bits of context to include. In other words, search quality matters a lot.
Edit to add: This was an aside in the post but actually a big deal... With this setup you can basically use an off-the-shelf LLM (like GPT)! No fine-tuning (and therefore no data labeling shenanigans), no searching for an open-source equivalent (and therefore no model-hosting shenanigans), no messing around with any of that. In case you're wondering how, say, Shopify and Hubspot can launch their chatbots into production in practically a week.
Karpathy claimed they worked for years and could not reap the benefits from multiple sensors however hard they tried. He seemed really convinced and does not get to me as one who tells stuff just to justify cost reductions, like Elon sometimes is carried away.
Now, some of you may not ever write computer programs, but perhaps you cook. And if you cook, unless you're really great, you probably use recipes. And, if you use recipes, you've probably had the experience of getting a copy of a recipe from a friend who's sharing it. And you've probably also had the experience — unless you're a total neophyte — of changing a recipe. You know, it says certain things, but you don't have to do exactly that. You can leave out some ingredients. Add some mushrooms, 'cause you like mushrooms. Put in less salt because your doctor said you should cut down on salt — whatever. You can even make bigger changes according to your skill. And if you've made changes in a recipe, and you cook it for your friends, and they like it, one of your friends might say, “Hey, could I have the recipe?” And then, what do you do? You could write down your modified version of the recipe and make a copy for your friend. These are the natural things to do with functionally useful recipes of any kind.
Now a recipe is a lot like a computer program. A computer program's a lot like a recipe: a series of steps to be carried out to get some result that you want. So it's just as natural to do those same things with computer programs — hand a copy to your friend. Make changes in it because the job it was written to do isn't exactly what you want. It did a great job for somebody else, but your job is a different job. And after you've changed it, that's likely to be useful for other people. Maybe they have a job to do that's like the job you do. So they ask, “Hey, can I have a copy?” Of course, if you're a nice person, you're going to give a copy. That's the way to be a decent person.
There's other stuff like build tools, cross-browser, and other stuff, but that's likely to be confusing and not super necessary to begin with. The above should be enough to get you running with what it sounds like you want to do.
I appreciate you sharing those links. I'm trying to learn mechanical engineering stuff myself, if you have any further useful learning materials I would love to see them.
My wife loves to play it, she is still learning how to use the right stick to aim but is getting much better. Know any other girlfriend friendly co-ops?
This is an awesome view into the implementation details of python. Can anyone recommend other resources that go into the lower level python implementation?
I've had cause to research the topic of lead (and other metals) toxicity quite extensively.
A key factor seems to be the body's propensity to produce metallothionein [1], proteins that bind metallic compounds in the body, either to metabolise and utilise them (in the case of beneficial metals like zinc, copper, iron, selenium etc), or to remove them (in the case of toxic metals like lead, mercury, aluminium, cadmium etc).
This propensity seems to be determined by genetic predisposition and/or genetic expression in response to environmental stimuli.
This is why, in a group of people subjected to similar levels of environmental lead exposure (e.g., soldiers in the same military unit, children in the same town with contaminated water), only some will exhibit symptoms.
Here's my own experience:
I have not done any military service as an adult, nor used firearms more than a handful of times. (I participated in army cadets in high school but it was little more than boy-scout training - camping, survival skills, hiking, etc).
But I did spend a few years in my late teens/early 20s working in an electronics factory soldering circuit boards. I grew up in the last few years of leaded car fuel still being commonplace. I also grew up in an old house that may (or may not) have had lead water pipes.
For much of my life, I've experienced a complex set of vague, but moderately debilitating symptoms. Nothing apparently life-threatening, but always frustrating and often painful.
In childhood it manifested as respiratory problems, learning/cognition difficulties (including modest academic and sporting performance even though I tried hard at both and was often assessed as being of high natural ability), social difficulties, anxiety and mood instability, tinnitus, and digestive issues.
In adulthood it progressed into conditions such as depression, chronic fatigue syndrome, endocrine problems, painful muscle tension in the back, head, neck and hip regions, dizziness/confusion/fuzzy vision at times, among other things that have come and gone along the way.
I happened to do a home urine test just this morning, something I do from time to time. It indicated the presence of lead in my urine, which has routinely been the case for the past year or more.
I've also sent hair samples to a commercial testing lab every 6-24 months over the past 10 years, which have consistently suggested accumulations in my body of lead and other metals including aluminium, cadmium, mercury and others.
People may react with skepticism that home urine tests and commercial hair tests are of any clinical validity.
Fair enough.
But, as the linked article explains, you just can't easily get this stuff considered, tested and diagnosed by mainstream doctors.
I've seen multiple mainstream doctors over the years, and the best they can ever offer is anxiety or anti-depressant medication. They'll run tests on all the standard stuff - cholesterol, Vitamin D, iron... that's usually about it. Everything always shows up close enough to normal and they tell you you're fine and send you on your way.
But despite doctors' assurances, I haven't felt fine, and my ability to live a normal life, compared to most other people I observe, has not been fine.
So I do the only thing that seems to be available to me, which is home urine tests and commercial-lab hair tests. And they tell me something thing that makes sense when I read things like the linked article; there's just too much lead in my body.
In the 10+ years since this possibility became apparent and I started doing these kinds of tests, I've done a lot of work to get my physiology to a state where it could remove the metals from my system. I've considered doing chelation therapy but I understand that carries its own risks (i.e., removing important nutrients and moving toxic metals to areas where they can cause more harm), and that you're better off just supporting your body to detoxify at its own pace.
It seems to be working, slowly but surely. That would be why the lead is showing up in my urine now, after not doing so in the first few years; as alluded to in the article, it's been accumulated in my bones and is now coming out.
My health, and life, is steadily improving. People keep saying I'm looking healthier and functioning better. My fitness keeps improving. My muscle tension and pain keeps subsiding. My work performance and career seems to be on the up.
But boy, it's been tough to get through, and all the more difficult through the lack of mainstream medical recognition, and from the ridicule that inevitably comes your way as soon as you start talking about testing methods and healing techniques that are outside the mainstream.
That's probably the toughest part. I'm a scientifically-minded person from a family of mainstream science and medical professionals, yet due to the hand I've been dealt, in the absence of any other options, I've been forced to venture outside the mainstream, and have found myself being the subject of ridicule and scorn at times.
Anyone who has read this far might want to think about that next time you mock someone for seeking help from non-mainstream practitioners like naturopaths, acupuncturists, chiropractors, etc. You probably have little idea of just how much difficulty and frustration they've been through before winding up there, or what undiagnosed underlying issue has caused their predicament.
I know I'm not alone in having these challenges, and I know plenty of people have it a lot worse than me - particularly those lacking the financial means and family+social support that I’ve had.
With that in mind, it's high time that we started taking this problem seriously.
Knowing what I know now, it wouldn't surprise me if it turned out that lead and other metals toxicity is an explanation for a huge amount of the debilitating illness that mainstream medicine is currently unable to diagnose and treat.
Everyone has moved to https://globe.airplanes.live/ and https://app.airframes.io/flights now.
Here is the lawsuit from one former group of contributors: https://www.documentcloud.org/documents/23963235-golden-hamm...