Most of the Senior Engineers I've met have been doing projects since the 1980s for the oil industry or steel mills. They've let the IT staff handle the computers while they stick to what they know, designing and modifying industrial plants. I don't think they have to worry about AI intruding into the Engineering profession.
Arduinos aren't fast enough, and don't generally do internet. To really make it work, you need a controller that has native ethernet, and NO wireless ports, or a really solid way of disabling them (like hooking their antennas to 50 dummy loads) A pair of them can then be proxies for the connection, and serve up status pages etc., without risk of ingress of control to the monitored system.
For me, the old timey Raspberry Pi B+ fits the bill.
I'm working on data diodes... I've got the parts, now I've got to gather enough spoons[1] to do the work to make them useful.
Details: Building a physically irreversible way to transmit data in only one direction should be easy, using a 6N137 optoisolator. The hard part is the proxy servers that you put on both sides of it. Old timey Raspberry Pi B+ boards are my current choice for hardware, as they have NO wireless capabilities, and do have reasonable ethernet ports.
I want to keep the retail parts cost under $100, and open source it all.
Holy cow... I've got to limit my projects, so I don't repeat history where I kept finding radio shack bags full of parts for ideas. But you can get 3 of those, with external antennas for less than $60 through amazon. I put it in my wish list instead of buying them now. I've got a data diode to finish first.
Sideband on 13 CM sounds like an amazing thing to pull off.
Having done repairs on a ton of older service monitors, and having my hand in SDR, this is amazing! There's a huge amount of machining and hardware to make a service monitor, but with desktop CNC, cheap PCBS, and jelly bean components that go into the 10s of GHz, the sky's the limit.
I'd add a heat sunk 100 watt DC-5Ghz 20dB fixed attenuator, 0-110 dB switched attenuator, 5 ghz low pass filter, 7-12 Ghz local oscillator/mixer up/down converter, and it would make a small, very high performance service monitor good to 5 Ghz. 8)
A Communications service monitor is a tool for servicing radios. Until recently, most large organizations used VHF/UHF FM radio for communications on dedicated channels. A service monitor is a tool that can usually absorb the full output power of these radios while measuring power, frequency, distortion, etc. They can also generate a variety of signals for aligning receivers.
Most of the older, affordable models go from almost DC to 1 Ghz. I've got an IFR 2947a, which goes up to 1.05 Ghz.
I've edited it, but generally you don't want to have problems with mirror images getting through and creating spurious responses. Real world RF filters don't have infinitely steep edges, so initially I figured a 4 Ghz low pass filter would have enough loss by 5 Ghz to get the job done. I revised to to a 5 Ghz low filter, and a 7-12 Ghz Local Oscillator to give the filter an easier job. Also, the lower the swing in percentage terms for a VCO, the easier it is to work with.
So the signal path in receive mode would be:
N connector / 20 dB 100 watt pad / Switched attenuator / Low pass filter / Front End diodes / Mixer / Router board for DSP
Alternatively, there's usually a second path that skips the attenuator, with a lower maximum power level.
The reverse path would work for generating signals from DC (almost) to 5 Ghz.
My house has appreciated at a ridiculous rate. Personally I'd be fine if it hadn't. It's not in great shape. It's literally the cheapest house in town.
Wikipedia, but I haven't had time to read the whole thing yet, and I doubt most would consider it to be a book. It can be downloaded and kept offline, like a book.
I once read that if you tried to read every Wikipedia article for 12 hours a day, it would take you about 42 years to finish—assuming no new content gets added along the way.
Their little MEMS switch looks interesting, but it's only SPST, most of the stuff I would want to do would be SPDT at minimum to avoid having to use pull up/pull down resistors.
It might make an interesting switch for a switching regulator, depending on how much energy it takes per transition.
I love this game. My only gripe is that the board extends all the way up to the screen on the sides. So when I try to slide either the leftmost or the rightmost tile on the board, I end up triggering the back gesture on my android phone which closes the tab. And I have to reopen the tab to continue playing. Some margin around the board would be awesome!
Really great work on the game though. It's addictive and I'm enjoying it.
We've definitely got an IT Security problem that goes down to the roots. If you're testing an AI for hacking ability and it's not air-gapped from the internet, you need to go to jail when it inevitably breaks out and commits crimes.
The whole hugging face incident could have been completely prevented. The underlying security problem we've faced for the past decade is that Linux/Windows/MacOS, etc are all not fit for purpose in an age of persistent internet connections and mobile code. The only way out is an OS built around default deny instead of ambient authority.
More people need to know what a data diode is, and how to use it, myself included. I've finally got the resources together to start figuring that out, 2 Raspberry Pi B+ computers, the old kind without wireless. The data diode itself is trivial, I'll probably end up with a 6N137 optocoupler between 2 serial ports. The hard part is the proxies for all the protocols, and the digital fountain to spray the data across the connection with ECC.
At this point I intent to do it all in Free Pascal/Lazarus, but I'll suck it up and use C++ if I have to.
reply