This sounds like a horrible feature.
"Honey, WiFis not working in the bathroom, could you wave over the router, that might fix something?"
Having said that a router like this is a huge step forward for the masses. Most people don't bother configuring their routers, changing admin passwords, etc. They never update the firmware. These new routers from Google solve that, which in itself is fantastic. Here's what I'd love to see happen in addition to all this.
- OpenWRT + automatic updates. Why don't we have this yet? I am contemplating building an x86 based router just because of the auto-update thing.
- Wi-Fi AP's that double as smoke detectors. I already have a bunch of those in the house on every floor. As a bonus they could probably use the wires they already have to set up a high speed backbone. I currently use a Ubiquiti Uni-Fi that looks like a smoke detector and it's fantastic.
- Ability to sign onto a Wi-Fi network that is less painful than sharing the passphrase. For example, if a guest comes to my house all we have to do is bump our phones to get them on my guest network. My router is hidden and my access point is on the ceiling of the second floor.
And yes, I'd prefer that the routers not have security holes in them in the first place. But I'm also realistic.
InstaWifi does this. I think you only need to have it installed on the sending device. It hasn't been updated in a while but it probably still works. Would be nice to have this feature built in to Android.
https://play.google.com/store/apps/details?id=net.jessechen....
If you use WPA-Enterprise, you can create a rolling username/password that's deterministically generated. [0] And/or you can load a passphrase into a Yubikey or similar for laptops that have a USB port and other devices that have NFC.
> - Ability to sign onto a Wi-Fi network that is less painful than sharing the passphrase.
If all smartphones came with a QR code reader (seriously, why doesn't every Android phone come with ZXing's Barcode Scanner??), you could print the passphrase and frame it or something. Additionally, if Google got their shit together, they could make a URI of the form
$wifi_crypt_method//$uriescaped_ssid:$uriescaped_username:$uriescaped_pass
I'm not sure why they've failed to do so for the past several years.
> I currently use a Ubiquiti Uni-Fi...
Ooh! Which one? I have a UAP-AC v1 and a UAP-AC-LR.
[0] Actually, you can do a lot more than just this. I've been having so much fun with FreeRADIUS over the past week. >:D
(From what I understand, the LITE and LR use the same (Qualcomm-Atheros) radio, but the LR has a much better antenna, and the PRO uses a 3x3 version of the radio used in the LITE and the LR.)
Channel congestion is a huge problem in urban environments. From my home office I'm seeing 20 networks right now. Even a rudimentary dynamic channel-switching implementation could help a lot, but that's not what happens. Instead, a power outage occurs, all devices reboot, and some pathological combination causes my wireless to degrade because my router came up faster and picked a channel that my neighbor's stupid ISP-provided POS decided to squat after-the-fact. Used to be I could sidestep a lot of this by using the 5GHz band, but it appears that ISP-provided devices are starting to show up there, too (though I _think_ the newer 802.11 protocols have better behavior around channel congestion? Maybe?)
Anyways, that's my $2: this is an issue, and involving the user will just result in magic-thinking/cargo-culting being taught to non-techies for problems that could be solved through better standards. It's 2015 and the best I can get from my 802.11n network is still ≈ 50mbps, even in the same room. If wireless is the future, we need solutions, not band-aids.
Odd. My 2007 laptop gets 160+ Mbps goodput in the same room using 11n @5GHz.
New Macbook Pro gets routinely 500 Mbps+ goodput using 11ac.
Maybe you should get a better router?
That's a pity. I have a two-antenna 802.11n client that gets between 160 and 200 mbps transfer rate one room away from the AP and between 100mbps and 160mbps two rooms away in the pipe-obstructed, tile covered bathroom.
Are you -by chance- trying to do 802.11n in the 2.4Ghz band? [0] If you are, and have any other networks around you in that band, that's a recipe for sadness:
* If you don't use 40Mhz channels with 802.11n, you'll -at best- only see slightly better than 802.11 a/g transfer rates.
* AIUI, if you try to use 40Mhz channels in the 2.4 Ghz band, you have to either use channel 1 or channel 11 as your base channel. You'll also be stomping on every other channel in the band. What's more, some APs will detect interference from other users in the band and automatically drop back to 20Mhz channels. (This is a long-winded way of saying that 40Mhz channels in the 2.4 Ghz band are often impossible for any urban or semi-urban site.) [1]
[0] If you're advertising the same SSID on both bands, and don't have band steering that you know works, you really can't rely on your client's software to be smart enough to prefer the 5Ghz band. WiFi client software is almost always oh so dumb and chooses signal strength over throughput.
[1] Adjust channel numbers and interference advice appropriately if you're not in the US.