Hacker News Reader

>There's no "gotcha." That's the whole idea.

Strange. I have closely inspected the box my wifi router came in, and I so far have failed to find "This is worse than wired ethernet" printed anywhere on it.

Perhaps you could show me a single wifi AP, anywhere, where the ad copy says anything even remotely close to that.

I think most routers promote how their wireless technology is different than older models (since that's what might cause users to buy a new router).

But, for example, the Netgear Nighthawk[1] page spends a lot of ink advertising that it offers 1300+600 mb networking (whatever that means) and also notes that it offers 5 1000mb ethernet ports. Even without an understanding of the details of a common medium, 5000mb wired capacity is greater than 1900mb wireless.

[1]https://www.netgear.com/home/products/networking/wifi-router...

It advertises that it's faster, not that it's more reliable. A reasonable customer might extend their experience with other objects and conclude the faster mode is less reliable. A F1 car destroys its tires in hours, while wheels on a commuter car last for years, an overclocked processor is less stable than one at stock frequency, supersonic jet fighters kill their passengers at a rate thousands of times higher than subsonic commercial airflight, etc.

Understanding that ethernet is entirely different from wifi requires... domain expertise! The exact thing a Best Buy customer won't have.

>5000mb wired capacity

A suspicious man might note that switching capacity isn't stated on that page, is not mentioned at all in the manual, and is the exact thing a manufacturer would be tempted to cut corners on, since vanishingly few customers will attempt multiple simultaneous network transfers in a home LAN. A normal consumer would never notice a switch capacity higher than 1 gigabit.

> Wireless technologies are founded on exchanging the convenience of wirelessness for reduced performance and reliability. There's no "gotcha." That's the whole idea. Maybe that trade isn't for you, but it's not a secret. If you take any course on physical networks, they go over the advantages of not sharing the medium. [Emphasis added]

I mean, yes, if you think the OP is addressing the article to people who have taken courses on networking, then maybe the article isn't saying anything interesting.

If you read it as trying to make people who don't know this more aware of the tradeoff they're making, it makes a lot of sense.

My guess is that very few people understand how much better wired is over wireless, or in which situations (I know that I don't always think of it, and I have lots of experience in networking).

There's also the reality that USB isn't really that much less of a mess than Bluetooth. With Bluetooth you have interoperability problems between the controller and the target device. With USB, you have that + the interoperability of the cable.. and it's a mess. USB does overall work better than Bluetooth, but the problems he's highlighting in the article very much exist in the wired space (including the device polling screwing up latency).

>Wired standards are not free from complex protocols and error-filled implementations.

Sure, but once you figure it out for the set of devices you are using, that's it! It just works until it breaks for good. Wireless stuff... not so much. One day it works, the next it doesn't.

> It just works until it breaks for good.

...or until you move furniture around and then need a connection over here instead of over there.

For certain devices like headsets, even the wired ones I find randomly stop working, I am sure it is not purely random but typically due to opening up X, Y, and Z programs In a certain order. Issue resolved with restarting but that can be pretty disruptive in the middle of a work day.

His beef is with broken network and BT stacks, I suspect. Not with the actual underlying standards.