Share this page!

Enter URL or ID to Unroll

You can paste full URL like: https://x.com/threadreaderapp/status/1644127596119195649
or just the ID like: 1644127596119195649

How to get URL link on X (Twitter) App

  1. On the Twitter thread, click on or icon on the bottom
  2. Click again on or Share Via icon
  3. Click on Copy Link to Tweet
  4. Paste it above and click "Unroll Thread"!
  5. More info at Twitter Help
MG Profile picture
@_MG_
Jun 3, 2024 • 22 tweets • 7 min read • Read on X
Scrolly
I like to read replies to posts like this just to remind myself how misinformed the general public is about “USB-C”
So here is a thread looking at a few of them…
🧵1
First, USB-C is a specification for the physical connector. NOT the protocol. And it intentionally supports multiple protocols like USB, USB-PD, Thunderbolt, DisplayPort, HDMI, PCIe, etc.

Some protocols exclusively use USB-c, like USB-4, Thunderbolt 3 & 4, USB-PD.
🧵2
Now, because a high quality C to C cable can support ALL of these protocols, people incorrectly think the protocols are the same thing.
🧵3
Image
Image
And a lot of it has a belief that USB-C is somehow anti-Apple.
Reality: Apple (& Intel) designed USB-C.
The USB-Implementers Forum is responsible for USB-4, USB-PD, and many others. There are a lot of companies on the USB-IF, including Apple!
🧵4
A lot of people celebrate the cable “standardization” & low cost availability. It’s becoming common knowledge that there are 8 types of compliant cables. But people don’t understand that quality matters. Else, you get perceptions like this:
🧵5 Image
Most people with wired CarPlay that switched from a lighting cable to usb-c will notice how much more fragile the connection is if using cheap cables. That’s because the moving parts went from the socket (lightning) to the cable (USB-C). So cable quality matters more.
🧵6
It really is impressive how confidently wrong people are about this stuff. But also how it’s almost like it challenges their identity or something 😂

Anyway, if I left anything out, let me know.
🧵7

Image
Image
Image
To reiterate, a lot of companies are involved in the USB-IF now.

One of the most inspirational & enlightening people in the USB-C space, for me, is @Laughing_Man. He opened my eyes to the complexity, beauty, & horrors of USB-C.

🧵8
Correction: The idea that Apple/Intel invented C & gave it to USB-IF is based on an industry rumor citation. It’s believable with the timing & numerous similarities with Lightning, but only rumor.

Apple is still on USB-IF, so trying to say C is anti-Apple is silly either way
🧵9 Image
Lotta people getting defensive 😂

USB C (& even earlier USB) is a confusing mess, as my thread shows & tries to inform. Especially with all the protocols & cable types.

The screenshots of confident-but-wrong “correction” reply guys isn’t purely a spec problem though :p

🧵10
Yes, if you aren’t aware, there are 8 possible spec compliant C to C cables.
Lets not count the non-compliant ones 😂

USB-IF has a labeling proposal to “fix” this. But mfgs are also YOLOing it and making cables look like NASCAR logos
🧵11 people.kernel.org/bleung/
Image
Here are the logos that USB-IF released in late 2021. Only cables that have been certified by USB-IF will be allowed to have the logos. And they must have the logo to get certified.
Yet somehow the Apple cable I bought yesterday doesn’t have any of them. 🤷‍♂️
🧵12 Image
lmao, 12hrs later and people are acting like I personally designed these specs.

Quite the contrary. I make USB cables that abuse these specs, like the OMG Cable.



Anyway, there are some good questions too, so I’ll try to update this thread.

🧵13
For people asking for details on “moving parts”, here’s a graphic. This is NOT the only way a connector fails, but cheap metal loses its spring much faster.

Spec says the connectors should achieve 10k+ mating cycles. But bottom price cables generally aren’t to spec…

🧵14 Image
For people asking how to pick a good cable… ehhh 🤷‍♂️

Figure out which of the 8 possible cables fits your need (post #11) and then find one that was certified by USB-IF with a logo (post #12) or some trusted brands.

Or YOLO it, test them, & don’t be sad when they die.

🧵15
Somehow this turned into an AMA about USB-C. So let’s go with it. Next most common question I’m seeing: “why doesn’t <device> charge from USB-C?”

Answer: cause the maker didn’t bother to read the spec. But there is usually a fix! (Next post).
I see the is failure a lot. Ex:
🧵16
So, if your poorly designed device won’t charge or power up with a C to C cable, try converting it to USB-A and back again. Adapter MUST be on the Charger/Host side!

Also, try flipping the connector on the device side, just incase they really screwed it up.
🧵17 Image
@seb_mc2 Even if Apple’s only involvement was via USB-IF, it still makes the weird anti-Apple views pretty silly. I just want to know about the 2012 piece for personal curiosity.
@seb_mc2 Lighting: released in 2012, compact, reversible, orientation marker in cable, host/device responsible for reorienting, reconfigurable pins, etc

USB-C: same, but more pins and reversed mechanical. 2014

Could also be industry response to Lightning? 🤷‍♂️
I WANT TO KNOW!
@CinnamonComfy @calicodev There is no scenario that I can imagine where simply using a C to C cable what supports higher wattage will result in failure to power on when a lower wattage one works fine. So far more info is needed.
@CinnamonComfy But if we are making guesses with insufficient info, context clues suggest that @calicodev doesn’t actually have the theoretical 100w cable and is wondering why 100w is unsafe for a 60w device. (Which is not actually unsafe at all)
Sketchy cables aren’t just a risk of annoyance. Safety is an issue too, especially with the recent enhancements to USB-PD allowing up to 240w. Design considerations now include electrical arcing! ⚠️⚡️
Probably good if the mfg has read the specs!

(video unrelated)
🧵18

• • •

Missing some Tweet in this thread? You can try to force a refresh
　

Keep Current with MG

MG Profile picture

Stay in touch and get notified when new unrolls are available from this author!

Read all threads

This Thread may be Removed Anytime!

PDF

Twitter may remove this content at anytime! Save it as PDF for later use!

Try unrolling a thread yourself!

how to unroll video
  1. Follow @ThreadReaderApp to mention us!

  2. From a Twitter thread mention us with a keyword "unroll"
@threadreaderapp unroll

Practice here first or read more on our help page!

More from @_MG_

MG Profile picture
Mar 4
This is likely snake oil, but tons of people are boosting it.

Ultrasonic mic jammers are real & a fraction of the price. What they claim is new: using AI to detect mics.

There are ways to find hidden mics. The TSCM space (bug sweepers) has tons of tricks that seem like pure magic. Have you ever listened to transistors turn on & off inside of an electronic device? Seen what a non-wireless camera sensor sees from across the room because every copper trace on a PCB is still an antenna?

…I have 😎. Hell, the first time I heard the “heat beat” of one of my naughty little OMG Cables, it was kinda reminiscent of hearing the heartbeat of my literal unborn kids! 😂

… anyway, you also have thermals, magnetics, etc. But most of these tricks require that you either sweep a detection device within a few cm of the bug, or you have a bulky antenna pointed directly at the bug. This “Deveillance” device is a small stationary puck that you place in the center of the space you want to protect.

So what can you do with a small stationary object to detect mics? Well, anyone who’s used an ultrasonic jammer knows that most of the space is going to be filled with ultrasonic emitters, especially if you want the claimed 2 meter range. So that leaves a pretty small space for the detection electronics.

You could do wireless protocol discovery. WiFi, BLE, etc. This would be easy. But it’ll only find a fraction of hidden mics.

You could do wideband RF sweeps to detect any active radio emissions. Here, AI could actually help identify based on raw signal. But this already feels like a stretch for this product. Lots of legit wireless mics are going to slip through the cracks with the minimal hardware that fits in a small puck.

But let’s say we make it this far. What about every mic that is not actively transmitting? Saving to local storage for later retrieval, etc. Well, you could use your ultrasonic emitters to create saturated pulses into the mics, which in turn will create electrical impulses down the copper lines between the mic & whatever catches the signal. Every bit of copper, no matter the length, is also an antenna. So you catch those emissions and look for signals that match your own ultrasonic emissions. Packing equipment sensitive enough to do this inside a little puck though…. Ehhhh

And after all that, you are still blind to passive MEMS microphones.

And more so: there are already ways to defeat ultrasonic jammers too.

However, this device doesn’t claim to protect you against bugs and other hidden mics. It’s very tightly constrained to:

“prevents smart devices and AI recorders from picking up your voice”

That’s an incredibly narrow scope. Existing ultrasonic jammers cover that scope pretty easily.
I really want to emphasize that their own marketing copy has a VERY narrow scope of what devices they work against. And that does NOT include the very wide range of microphone-containing devices that everyone is assuming.

However, the replies from the product team don’t seem to help remind people of this limitation. In fact, it reads like dodges that let people’s minds run wild. I’ll touch on that as I add more posts to this thread.Image
Context: I rarely say something is 100% impossible. There is always a way. It’s a motto that I have built a lot of my recent work on!

I do think these claims are extreme enough that they warrant strong evidence and probably 3rd party testing. There’s a lot of “weird vibe” as well. And more importantly, the security/privacy space is filled with “solutions” that create dangerous situations for consumers by not clearly showing the limits.

Anyway, on to some of the replies from their team…
Read 6 tweets
MG Profile picture
May 2, 2025
People were concerned that gov record retention laws were being broken due to the use of Signal. This picture shows that’s not the case!

See the “TM SGNL” on screen? That’s a Signal wrapper specifically for maintaining archives of the messages.

telemessage.com/mobile-archive…

I can’t comment on security implications of the wrapper, but I suspect my prior criticism of “probably not using Signal securely” are fairly off base considering how this alters the threat model.Image
Yes. I’d love to know what “I have confirmation … its turned off” refers to as well :p
Looks like TeleMessage was probably procured and rolled out under Biden. There are public records for it. Image
Image
Read 9 tweets
MG Profile picture
Mar 24, 2025
I have a slightly different take on these leaked signal messages of the Trump admin planning the attack on Houthis…

Once again, it’s that nobody seems to use Signal SECURELY. If they had, this leak would have been less likely.
🧵1/n Image
Image
Image
The Atlantic article on this touches on policy of where & how these comms should happen, but completely misses the Signal failure that started this.

It starts right at the beginning. A new connection request was made by Mike Waltz, and then immediately added to a signal group. Waltz did NOT do anything to verify the identity of the person (else he’d quickly notice it was a journalist), and clearly did not verify the Signal Safety Number over a trusted channel (which means it’s susceptible to interception).

I doubt they are using things like Registration Lock either, which means anyone can hijack their Signal accounts with a simple SIM swap… which should be assume an automatic threat when the telcos have admitted China has access to everything.

The failure to do this also points to a strong indication that this is likely a recurring pattern of OPSEC failure when using Signal. The consequences could be much more severe than this leak.

🧵2/nImage
We constantly hear about “Signal being broken” and it’s always seems to be from people who don’t know how to securely use Signal.
🧵3/n
Read 7 tweets
MG Profile picture
Feb 27, 2025
Anyone can track the physical location of Bluetooth devices across the earth, & it flew under the radar.

All you need is the MAC, which is trivial to get by just using a BT scanning app neatly the target.*

Then you guess (90% success rate) a matching key that makes Apple’s FindMy network think this device happens to be a lost AirTag.**

That’s it! The FindMy Network then tells you the latest location. It was discovered by George Mason University & called nRootTag. Apple has started to roll out some sort of change, but it’s unclear how impacting it will be or how long it will take. GMU researchers worry it could take years to be effective.

*unless it’s Bluetooth 4.2+ and is using the optional privacy features.
**requires running the MAC address through a GPU cluster, which can also be done in advance by building a rainbow table.
Details are a bit sparse, but more will come in August, according to the GMU post:

cec.gmu.edu/news/2025-02/f…
Exactly. It doesn’t matter who the manufacturer of the Bluetooth device is, it’s findable by the FindMy network!

There are a few ways Apple can minimize this I believe. Heavier restrictions on how a devices is reported as lost. Blocking lookups of devices not rotating their key (which will probably break a lot of non-Apple devices using the FindMy network). Etc.
Read 4 tweets
MG Profile picture
Jan 12, 2025
“recognized as malware” is the end of the analysis? Bruh…

At least share the exe so others can check it out and either validate this or put the nail in the coffin.

There are so many ways something gets flagged without it being malicious itself. Down to being simply unsigned.

The chances of this being intentionally malicious are very low. And you haven’t done nearly enough to demonstrate otherwise.

That doesn’t mean it’s necessarily safe. You paid pennies above the cost of the hardware via AliExpress. That gets you the lowest effort software too, where security is not a concern.
Imagine buying DIY canned food from an alley and then pearl clutching when it’s not FDA approved… and then acting like the makers are spies trying to poison you. 🙃
Prove it!
Don’t get me wrong. China is an intentional adversary in many avenues. But the threat posed by AliExpress & Temu is economic.

If you are buying lowest cost hardware, you aren’t getting any effort beyond basic functionality. Safety and security aren’t part of that. Don’t confuse that for intentionality.

COULD these things be leveraged by a 3rd party to undermine your security? Very likely. But you are just about as intentionally complicit in that as the seller.

I’ve ranted about this before:
Read 7 tweets
MG Profile picture
Nov 26, 2024
When we added C2 capabilities to OMG Cable, people would say “But I’d notice it on my network!”

I said: yeah, but would you notice it on your neighbor’s wifi, free cafe wifi, etc? 😈

Also, here is a free nightmare: when wifi drops due to power loss, those battery powered IOT devices do ALL kinds of useful things if you’re in range.
Oh absolutely. Most places won’t notice. Especially with the added MAC spoofing and the C2 traffic looking like misc web traffic. There are lots of options before needing to use a nearby network, or supplying your own.
The guest net is often a great option too!
Read 4 tweets

Did Thread Reader help you today?

Support us! We are indie developers!

This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3/month or $30/year) and get exclusive features!

Become Premium

Don't want to be a Premium member but still want to support us?

Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal

Or Donate anonymously using crypto!

Ethereum

0xfe58350B80634f60Fa6Dc149a72b4DFbc17D341E copy

Bitcoin

3ATGMxNzCUFzxpMCHL5sWSt4DVtS8UqXpi copy

Thank you for your support!

Follow Us!

Send Email!

:(