The reason for the negativity is that this demonstrates the "Design by StackOverflow" mentality where the solution is like swatting a fly with a sledgehammer and no real domain knowledge. Plus the author didn't even train the neural nets: it's just a LEGO project. I'd higher this person to be a lab intern, but nothing above that. The fact the author couldn't solve it locally and had to invoke the CLOUD is... laughable. This problem has been solved for over two decades on lesser hardware.
TL;DR: reinventing wheels is a good way to learn a lot.
For trying out something in a few hours, of course you don't want to spend hundreds of hours setting it up, by definition. Yes, the result is "just about works, but doesn't scale" - but that's the point of experimenting. Sure, this is a LEGO-style experiment in reinventing the wheel, but exactly for that, an excellent way to start learning about this problem domain: power consumption? Latency? ML basics? Sure. That's hacking at its core - even though the project is rudimentary.
> Sure, this is a LEGO-style experiment in reinventing the wheel
If I'm not mistaken Larry Page used to be praised some time ago for building a printer out of LEGO pieces (that may just be an urban legend, I admit I never verified this information).
This is a great example of how so many programmers think they are the first to discover something that is "ancient". In reality this is a problem older than Linux (linux just had the advantage of a huge and diverse open-source userbase from which to draw data). On the flipside, lots of good information takes years to go from discussion groups to books, so it can be hard to find relevant research. Applied compsci is literally the cutting edge of reality.
He doesn't describe an attack, he describe literally what 802.11 was designed to do. An attack is forcing a deauth and then stealing the 4-way handshake data and, say, cracking WEP. Which is why WEP was decommissioned ... checks notes ... 15 years ago.
No need to edit aircrack-ng, WireShark does what he did natively (filter out and set channels), and a good realtek chipset allows you to set the scan interval so you can cover more channels (which is why the new ALFAs suck).
Also the DTIM and keepalive can be set such that the MCU can sleep while the phy link maintains a connection without a costly handshake, esp. if using TLS <1.3 to talk to the cloud. Reconnecting costs a shit ton of energy so they usually don't disconnect.
Hacking Wi-Fi has become exceptionally more difficult, as noted by the slow dating of materials at DefCon's WiFi Village over the past 8 years: cracking WPA2 is basically so hard no one bothers, even in CtF games.
Yeah, I agree with you. I had a sneaking suspicion.
Funny you mention Electron: my company switched from Qt to Electron because it opens up a larger pool of developers with instant cross-platform portability with a fraction of the debug overhead.
I've spent significant time in HCMC (about 6 months over four years). It is vibrant, friendly, booming, tech heavy, and goddamn cheap. I didn't need medical care, and didn't have to cook because food is amazing and cheap. I computed about 1/10th to 1/12th the cost of my life in Northern Cali. I hope when I'm older it is still cheap and not environmentally wiped out. (I'm also biased because twice I met a woman who I thought was waaaay out of my league and we ended up dating.)
Interestingly, at DefCon last year, a presentation was talking about how in Africa there are interposer SIM shims that slip over the top of the simcard and provide additional functionality. Citation needed, but I was impressed they could fit a small MCU in such a thin form factor!
There is no home in LA county that didn't see price rise in an arbitrary 10 year period. Flip that house is a real thing in the rat race real estate markets in the U.S.
Kinda reminds me of Transmeta. Except they had a product, a shitty one, but a product. Or Ampere, but they only have one product they inherited and no flashy names.
Silicon startup challenges are rough. RISC-V is probably the best hope for a new architecture (hint: its not a new architecture).
Transmeta was going to be the next big thing, code morphing! Linus was on board. News on slashdot every day! And then ... some product released but no needle moved. I wonder if they built something that got outpaced by intel, needed network effects to take off that never materialized, were ahead of their time, or just no market for their product?
These guys seem like they’ve identified a market and who their first customers are going to be.
Sort of... Transmeta was basically a VLIW chip optimized to recompile x86 code. The problem was by that point all x86 chips (even plucky little VIA/Centaur!) were recompiling x86 instructions into RISC-ish ones, and without sucking up RAM and more importantly memory bandwidth to do it.
So even if the VLIW chip was on paper more powerful/efficient (and I'm not sure it was), it just couldn't keep up with doing it in hardware.
