You obviously need another agent that is the only one that can run perf tests.
Turtles all the way down.
822 karma · joined July 19, 2012
Turtles all the way down.
It plays into the generalisation that people aren't good about saying what is the right way to do something, but much more open to saying what something is doing the wrong way.
In a previous job we used to do prototypes in languages like Perl because we couldn't ship them and knocking something non-production ready was easy. We were able to quickly come up with the PoC, iterate on it, and then plan and write the whole thing in $LANGUAGE_OF_CHOICE.
Yes they are constantly listening, but this is just for the watch word (as a cue to start listening/sending) but they lack the on-device ability to convert anything else to text on their own. All they can do on their own is listen out for the watch word.
Certain smart TVs on the other hand, are known to constantly stream audio back to their HQ as they don't have on-board watchword detection.
I have all of my IoT devices on separate Wifi network(s) and VLANs and almost all of them are isolated so they can't talk to each other, plus I occasionally look at how much data they are sending/receiving from the Internet (some is expected obviously, and it differs by device).
Doing this requires a considerable amount of admin work and IT knowledge though. It also requires something a step above most consumer grade or supplier provided networking equipment.
I've never spotted anything egregious like the coffee maker in the OP but if I did I'd be making sure other people knew about it and the device itself is either firewalled off properly or replaced by a brand that isn't a security risk.
We had a similar thing, the other team wouldn't back down.
We ended up implementing a kind of rate limiting internally.
If the previous request (from that IP) was more than 4.5 seconds ago we let the check request through as normal.
If the previous request (from that IP) was more recent than that we just returned a cached "there is no update" payload that had a TTL of 60 seconds.
We told them this and left it up to them, they soon changed their polling frequency.
https://www.oralb.co.uk/en-gb/product-collections/oral-b-app for an example.
As I said it showed me I was neglecting certain areas over time and fixing that saved me more than the price of the toothbrush in under a year.
But feel free to continue to disagree with it as a gimmick. In my experience most people that do this are some of the lucky people that could brush their teeth once a week with a stick and still have no dental trouble, meanwhile a small percentage of people seem to have to take great care of their teeth, avoid a vast array of food/drink, and still end up paying out thousands each year to a dentist. Genetics plays a big part in it and you don't get to pick your genes.
https://www.royalmail.com/sending/barcoded-stamps
There are still some "Forever" stamps (those with pictures on or Christmas Stamps) but the old non-QR code non-forever stamps stopped being legal a while back (but you can always swap them for the new ones for free).
I've had people try and reuse a QR-coded stamp, you just get a postcard saying there's an item of mail with invalid postage for you to collect (and pay for). You get to look at it before you decide whether to pay the fine or not and then your argument is with the sender to reimburse you or not.
My $150 "smart toothbrush" (some Oral B iO model) has accelerometers to track its head position and the app lets you know if you're not covering certain areas well enough. Since using it properly (i.e. checking the app and adjusting my brushing) I don't have to go to the hygienist as often.
Needs changing at 100,000 to 150-000 miles or every 10 years.
The 10 year service is a spendy one.
SpaceX groundstations look a bit different, they are a bunch of white domes about 1-2m in diameter, usually 8 or 9 of clustered together.
Here's an example in the UK: https://maps.app.goo.gl/Zy61dfdyp5xrq4w1A
It's handled by Arqiva here in the UK it seems, so no SpaceX branding on anything.
SpaceX also applied for two new grounds station sites in the UK, one a few miles West of Telehouse, south of Rainham. The other site is up in Harlow, Essex.
(I don't think any of this is secret, all of the info above can be found with some googling and scrolling in maps.)
Proposed venue looks in keeping with this theme. Capacity of 600 and I'm sure whatever tickets will be priced to keep the hoi polloi away.
Sure there are things like an Elgato StreamDeck that you can do similar things with but they still require you to move a hand from the keyboard.
I suppose it is similar to using voice control things like Alexa. Setting a timer by hand in the kitchen isn’t exactly hard but it is so much more convenient to do it by voice if your hands are covered in raw chicken juice or the like.
My kids school puts some information that I need to see (not just "would like to see") on Instagram. Same with the sports team they play for.
I've complained to both about it but change is glacial. I'm lucky in that I know how to work around it, and I do my best to keep others who aren't so tech savvy updated about things they need to know but may have missed.
Irony is they moved most of this info from Twitter/X to Instagram.
The one I did (QbCheck) was a similar setup but you were presented with shapes on the screen, either a square or a circle, and either blue or red. All you had to do was press the spacebar when the shape on the screen was the same shape and colour as the previous one. No sound distracitons (or maybe I just had the volume muted.)
You also had to do the test on a laptop with a forward facing camera so the software could track eye movement.
I remember starting the test and thinking "This is exactly what my brain is wired for, I'm going to be awesome at this test!"
Within a minute my mind had already moved on to "I wonder how they score this test? I guess they're looking at accuracy, reaction times, how those trend over time, whether the accuracy/reactions goes up/down if they show the same shape/colour more than twice in a row?" etc...
It didn't take long for me to then get jolted into a "wait, did I just zone out a bit there and miss one?"
Then I became overly conscious of my eyes wandering around and not looking at the screen.
Then my mind moved on to "what if I'm too good at this and it means I won't get a diagnosis?"
Soon it was "When will this ever end? This must have been going on for ages!" only to find it had only been about 3 minutes.
The test took 20 minutes or so and I was mentally exhausted by the end of it.
My actual ADHD diagnosis was backed up by an 80 minute interview, basically doing the DSM-5 questions but linking it all back to childhood experiences too. I've no idea how much weight the QbCheck test had.
Glad I got the diagnosis, I've been masking both for years (diagnosed in my late 40s) but just got burnt out doing so. Medication helps but isn't perfect and since I'm probably AuDHD I find the ADHD medication amplifies some of the Autism traits and makes me into a bit of an emotional zombie, but that seems to be better than having zero motivation (aside from stress/deadlines and serious consequences).
Can't speak for the whole of the continent but UK here and I got my ADHD diagnosis a few years shy of 50 years old. Wasn't tricky at all, just a long-ish (8 month) wait for a non-private diagnosis (not full NHS but via an outsourced provider via Right-To-Choose).
When going through the DSM-5 questions (during an ~80 minute interview) I was asked to give examples from both current life experience and from childhood, and that was enough.
The traffic was separated so that anything on the public hotspot was not attributed to the private subscriber network.
Secondly for years I lived somewhere very remote so that meant that hardly anyone ever got to use my broadband connection via such a hotspot, but I got to use hundreds of other peoples hotspots when I was somewhere less isolated.
I was quite happy to opt in to this system as I benefitted hugely from it.
If the user data is worth that much then I wouldn't be surprised if some manufacturers put such radios in their non-smart TVs in order to try and gather such data.
I live in a densely populated part of a large UK city and (checks `nmcli dev wifi list`) there are 14 distinct wireless networks near me that aren't mine. There's a good chance that any IOT device could find something nearby, and there's even a whole load of LoRa nodes that it could use for a very slow uplink.
The robot decided to spin up an expensive setup prior to getting access, so the setup was sitting there costing money whilst it did nothing.
If it had designed the setup but not spun it up until it had authorisation to join the network then it would have been much less costly an exercise.
If your webapp has unfettered database access then don't be surprised if it is hacked and someone can do `select * from users` and then posts that dump somewhere.
The attack surface changes if your webapp can only do a REST call to pull a single user record at a time. That way you can put some auditing in, you can put rate limiting in to detect that, etc.
Obviously the user record REST api endpoint is still vulnerable, but it's a much smaller attack surface, easier to audit, and can be monitored a lot more closely.
Yes, ultimately, there will still be a set of vulnerable humans that have access to the database servers themselves and they can always walk out of the place with an SD card hidden in a Rubik's cube but there has to be an element of trust somewhere.
The problem is that too many people put that trust boundary way too far out into the big bad Internet. Or don't even consider it at all and just rely on the fact that other targets are more appealing.
It may keep out the bottom x% of spammers/hackers but it doesn't do much for the increasingly sophisticated scams that are appearing.
If the bit before the + ends up in your inbox anyway then it'll just get stripped off and used. Spammers seeing this kind of thing across several breach dumps:
bob+trello@example.com, bob+spotify@example.com, bob+chase@example.com
and will leverage that to target spam at you for other sites, or just email bob@example.com as there's a good chance that'll get through.
Years ago I did a test with my own domain where I created who unique aliases with plus addresses, e.g. steve.smith+iawer@example.com, bob.jones+wpoqe@example.com
It didn't take long for emails to start arriving to steve.smith@example.com and bob.jones@example.com even though that email address had never been used anywhere ever before.
As others have said, you're better off just creating unique emails with `pwgen -s 16` such as wmR5pNhGI8yidU7N@example.com and storing that in your password manager alongside a similarly random password. (Yes, this is roughly what those unique email address services provide.)
Also many services/sites/providers simply assume the username is immutable. $DEITY forbid you might have to change your email address at some point in the future.
The only extra bits I saw for the other emails on my domain was a plus address I'd used for last.fm which had been leaked. None of the other emails (wife, kid, family, etc) appear in any breach.
I'm slowly moving away from using my own personal domain as it's becoming an ever increasing burden. I'm also concerned that my wife/kid will be left with something they may not have access to, or would stop working at some point, if I suddenly dropped dead.
I'd planned on writing something myself to parse the HTML and write a suitable exporter but I thought I'd give Claude a chance.
In a sandboxed VM I gave Claude a single static HTML file of the status page from the printer, also in the directory was the equivalent of "hello world" in Go, literally just the minimum needed to do `fmt.Printf("OK\n")`. The directory was called `brother-exporter`. That was it. No other instructions or information. I hadn't told it what it needed to write. I hadn't said what it should do. I hand't told it what language it was supposed to use.
Just by doing a `/init` in that directory Claude decided that it needed to write a Prometheus exporter in Go that would fetch and parse the HTML file from a printer (defaulting to 192.168.1.1) and then present the associated metrics in a way that they could be scraped by Prometheus.
It did this flawlessly in about 10 minutes.
I could have done it in several hours but this was definitely an "oh shit" moment for me. I think the biggest thing was the fact that it guess/assumed so much (correctly) from so little information in the beginning.
In the end I just replaced each X540-T1 with a X520-DA2 which are pretty much the same price on eBay (under USD20) and then I can just use a DAC that's a fraction of the cost of the RJ45-SFP+ transceivers.