It's not enough to simply have a IoT VLAN that you put all your IoT devices on. Because those devices can see one another. In this case, if the coffee machine can see what type of smart fridge and smart toaster you're using, they can sell that data.
I default to adding IoT devices to a 2.4g "Guest" network where they can't see each other. Exceptions are IoT devices that need to see their friends to do what I bought them for, or devices I want to integrate with HomeAssistant. In those cases I create a separate IoT device per IoT brand. Excessive but necessary.
In those cases I create a separate IoT device per IoT brand.
That requires a lot of SSIDs though and AFAIK it reduces airtime for each SSID on the same router (which may be bad if you also use 2.4GHz for your regular devices).
So far two separate VLAN + SSID (IoT-Good and IoT-Bad) with client isolation on IoT-Bad has worked pretty well for me. In some cases mDNS advertisements have to travel at least from the IoT VLAN to the VLAN HA is on for the devices to be discovered.
There's ways to reduce the expense like changing DTIM interval (for less-frequent broadcasts) and increasing base data rate. Few folks in 2026 have anything at home that requires the 1Mbps base rate of 1999's 802.11b anyway, and slower DTIM can have other advantages, so these tend to work well at increasing available airtime for more SSIDs.
There's an additional trick that can also be used, though: Wifi client isolation. This lets the IoT widgets share an SSID and VLAN with which to talk to the HA machine and/or their cloud-based mothership(s) or whatever, but without being able to see eachother on the SSID.
It still needs set up right lest stuff walk right by it, but it's an available feature.
A smart coffee machine may need to talk to some kind of controlling endpoint (whether local or afar) in order to do whatever it does. It does not need to be able to enjoin in a conversation with my light bulb in order to get there. But that doesn't mean that they can't share an SSID. :)
There's an additional trick that can also be used, though: Wifi client isolation. This lets the IoT widgets share an SSID and VLAN with which to talk to the HA machine and/or their cloud-based mothership(s) or whatever, but without being able to see eachother on the SSID.
Indeed, this is what I do on our home network. I have two IoT SSIDs. One with devices I generally trust, where mDNS proxying is enabled and another that is excluded. I've found that the worst devices are generally not locally controllable anyway and always want to go through the cloud, even when you control them through HA or your phone.
I hope to be able to find the time and motivation to re-do the wifi stuff at home this weekend. I'll try to adopt the good-and-bad model of having two IoT SSIDs.
While I mostly use Opus for coding, I use Sonnet in my actual application, because it is far cheaper at scale. I have Sonnet parse user free-text input, understand it and return the meaning as structured json. It doesn't need advanced reasoning, just a good enough understanding.
If you start your own veterinary practice, and eventually build up a steady stream of happy customers, then sooner or later the PE firms will come knocking on your door, too.
Imagine you're drowning in student debt and worrying about keeping your business afloat, when someone offers you $5M, and says you'll still get to work with animals, which you love, while they take care of the financial and business side, which you didn't enjoy. It's a no brainer for most people.
No way. I think most people realize “they are offering me $5M because they know they’ll make a good profit on this business. That means it’s worth more.”
I think that's true to an extent. And, especially successful people, often would rather be their own boss for less money than to be a corporate cog. Having the freedom to make ethical judgements yourself and build local community is a real luxury. We are, mostly, just social animals that want to build our status by helping each other out in the end.
Two things, I think, might make the decision for them, in spite of that:
- The money is actually more than their business could be projected to make alone. This is because the PE firm isn't just buying the single practice but every practice in the area to kill competition. Get that captive market and you can charge more.
- The personal ownership freedom can be already lost on these independent people if they feel they are already tied down too much. I know this is especially the case in small medical offices where insurance and regulation can over-rule them. Feeling like you don't have control in this way can cause a lot of depression (tangent but look up how we condition mice in order to test treatments of antidepressants on them). At that point, the monied exit starts looking like a better option.
it is worth more, but you have to do 100% of the work to get 25% more upside.
Or take $5m and get no more upside and 70% less work. The "work" here is dealing with regulations, insurance companies, administration, payments, and accounting, not actually looking after dogs
the PE model is often to provide a competent regtech/administration core and then plug heaps of regulated businesses into it and centralise all the admin
imagine opening a vet in SF in 1950 in a cheap shoplot and people just paid you with hard cash or cheques in person, you wrote receipts with a biro, and filed taxes once or twice a year on a few sheets of paper... compared to now
Indeed. But remember a PE firm has economies of scale because they own many vet practices. They can negotiate better rates with suppliers than you can, and better payment terms. Maybe they even own some of those suppliers. They can divide their advertising costs by all the practices they own, whereas you can't.
Even if you were willing to be as ruthless as a PE firm, raising your prices and sacking half your staff, and even if you were somehow an expert in financial engineering and business optimization, you still won't make the business as profitable as they can.
Porsche is famous for underpromising and overdelivering.
For instance they claim a 0-60 time of 2.4s for this car. Car & Driver independently tested it and achieved 2.1s. They also claim 298 miles of EPA range (mixed highway & city driving). Car & Driver worked out 300 miles at a constant 75 mph (which is harder on an EV than city driving.) On the base model, which claims 303 miles range, MotorTrend achieved 350 miles of pure highway driving.
Pick any Porsche and you'll see that they lie to make things sound worse than they really are.
German brands in general tend to do that - BMW is notorious for under-reporting power, for example. They'd rather understate than have customers be mad at them for underdelivering.
Yeah, the difference between "reported and delivered" in the range of 350-298=62miles is actually that one special thing, that makes Porsches worth the price :-D
Had this done too to cure PVCs - a irregular heartbeat that wouldn't have killed me, but it did lead to an enlarged heart and low energy.
I remember feeling great during the procedure thanks to whatever meds they gave me. I could watch them on a screen threading the wire in through my femoral artery all the way up into the inside of my heart. Then I felt the exact moment I was cured. The irregularity I'd known my whole life just changed rhythm. Then they injected some stuff to stress test the heart at high and low speeds. 24 hours later I was back home.
Moving Synology-to-Synology, I was surprised at how easily I could just move the original HDDs (in their original order) from old enclosure to new enclosure, and it just worked.
I default to adding IoT devices to a 2.4g "Guest" network where they can't see each other. Exceptions are IoT devices that need to see their friends to do what I bought them for, or devices I want to integrate with HomeAssistant. In those cases I create a separate IoT device per IoT brand. Excessive but necessary.
reply