US blacklists Chinese surveillance companies over human rights abuses

[pete_c]

China-based Hikvision and other companies were listed

By Colin Lecher
Oct 7, 2019, 6:00pm EDT

The United States is adding eight Chinese technology companies to its trading blacklist, barring some major producers of surveillance equipment from working with American companies.

In a filing released Monday, the Commerce Department said it was adding 28 entities to the list that have been “implicated in human rights violations and abuses in the implementation of China’s campaign of repression, mass arbitrary detention, and high-technology surveillance against Uighurs, Kazakhs, and other members of Muslim minority groups” in the country’s Xinjian region. The listed entities include Chinese state security bureaus, as well as the eight companies.

Those companies include Hikvision and Dahua Technology, massive international purveyors of video surveillance cameras. Also included on the list was SenseTime, a Chinese AI startup, and speech recognition company iFlytek.

The filing did not detail the allegations apart from saying they were “activities contrary to the foreign policy of the United States,” but China is known to have launched a high-tech surveillance apparatus to monitor and imprison minority groups in Xinjian.

The move is the latest US policy crackdown on Chinese businesses. Most prominently, the US blacklisted Huawei over security concerns earlier this year, a decision the company has pushed back against. Meanwhile, the US is continuing to pursue a trade deal with China, while President Trump has suggested that Huawei’s status could be considered in the discussions.

 
Update 09th of October, 2019

There is not a ban on imports. There is a ban on Hikvision & Dahua from buying parts from the USA.

See here ==> hxxps://www.reuters.com/article/us-usa-trade-china-exclusive/us-puts-hikvision-chinese-security-bureaus-on-economic-blacklist-idUSKBN1WM25M

 

[upstatemike]

Probably not a good time to buy into any automation ecosystem that depends on servers based in China to function.

 

[lanbrown]

Even if it weren't for this, I wouldn't be buying anything that relied on the cloud that was based in China.
 
If you want privacy, you need to use services based in the EU because of GDPR.

 

[pete_c]

Here have been tinkering with WiFi automation trinkets.  Never have enabled any of them with my cell phone / cloud.  Only have 6 running and all have been JTAG'd with Tasmota or Espurna to use for MQTT clients.

 

[upstatemike]

Even without the cloud concerns I think there are serious challenges to Wi-Fi based automation. Some of those include include:
 
Scale: If you took all the devices currently handled by other hubs like Insteon, Hue. Z-Wave, etc. and started pushing that communication through your wireless access points/wireless router they would bog down and stop. Even with an update to 802.11ax you probably won't see enough capacity improvement to balance the added load of IoT devices which just seem to keep multiplying. (What was 1 switch is now 10 independently addressed bulbs etc.)
 
Address Space: Consumer networking equipment is generally configured to provide 254 usable addresses which is plenty for computers, phones, plus a few set top boxes for streaming but the math does not work in a fully automated home. These days it is not just the set top bos but probably also the TVs, DVD players, Cable or Satellite boxes and AV receivers that will all need IP addresses. If you use Sonos or some other networked music system you can easily burn through two dozen addresses as each individual speaker needs one. A ten bulb chandelier that used to need one address for the light switch now needs 10 addresses so each bulb can be accessed for color and othe new features. Door and Window sensors that previously used proprietary communications will now want in on the Wi-Fi address pool as will leak sensors, motion sensors, wireless cameras, and Echo or Google Home digital assistants. And of course there are thermostats, weather stations, and miscellaneous modules to control appliances and display items not to mention modules to monitor power consumption. And to cover all that traffic you will need mesh networks or multiple access points which themselves consume addresses. The average consumer is not going to want to go in and change the defaults on their router to add addresses and router manufacturers don't seem to be changing those defaults to accomodate the new reality of an IoT based home.
 
Coverage: 2.4GHz does not have the penetration to get/hear a signal to every corner of a home and the cost to remediate signal issues like chimneys, stone walls, large metal appliances etc. gets impractical very quickly. 5 GHz is even worse. Practical HA signal penetration starts around 900 MHz and below.
 
Congestion and latency: Besides the Access points getting overwhelmed with too many wireless connections there is also the problem of a consumer router getting bogged down and not providing sufficient response time for real-time home automation events to be detected and responded to reliably. It is one thing to have your streaming music or movie glitch a little because your router is overloaded but much more serious when grandma walks into a dark room and the lights don't come on because congestion blocks the automation actions involved to make that happen.
 
No Mechanism for Local Association: If I want an Insteon switch to control a floor lamp I can set up a direct association and it works just as if it was wired that way. Same goes for a set of Hue bulbs linked to a Hue switch. In the Wi-Fi world I don;t see any way to accomplish this except maybe using something like Hubitat or Homeseer and learning how to write your own rules. This could work for HA enthusiasts but is it practical for an average consumer? Even the cloud based offerings seem pretty weak for the common scenario where you want a smart switch or keypad to directly turn on your smart color bulbs in a room.
 
A lot of folks are looking to Wi-Fi to take over as the new universal HA standard but I don't see that happening until there are better solutions available to address these concerns.

 

[LarrylLix]

<snipped>
 
A lot of folks are looking to Wi-Fi to take over as the new universal HA standard but I don't see that happening until there are better solutions available to address these concerns.

I have 25 WiFi RGBWW bulbs installed. WiFi can have it's problems with larger quantitiies.
 
As a side note:
Many newer routers can have IP addresses of two pages of IP addresses but there is problems with that.
WoL (wake on LAN) packets are typically send out to the highest IP address supported, and the devices on the lower IP address subnet will ignore them.
 
ie. WoL a device using IP 192.168.0.120
Router allows IP addresses of 192.168.0.1 to 192.168.1.255
WoL command sends out generic address of 192.168.1.255
Device at 192.168.0.120 finds WoL packet not on it's subnet (192.168.0.xx) and ignores the WoL packet.

 

[pete_c]

Here I never put all my automation eggs in to one protocal. 
 
The heartbeat of the home is managed by my OmniPro 2 panel which speaks X10 (100%), UPB (100%), ZWave (100%) and Zigbee (100%). 
 
Homeseer speaks X10, UPB, ZWave, Zigbee, WiFi (local modded firmware MQTT with WiFi as a transport with no dependencies on the cloud today) and RF.
 
I do not watch it, measure it et al...it all works fine for me. 
 
The panel automation is always up and running and 100% all of the time.  (using wired sensors, serial touchscreens (and IP touchscreens), serial thermostat, et al.
 
Homeseer (X2) is an addedum to the "heart beat of the home" automation.
 
 
I use all but primary stick to UPB for my wall light switches.  My electric here is conduit and metal boxes.  Last home build I paid extra for using conduit over romex.  In Florida went with Romex and customized it a bit for automation.  That said it was difficult to modify electric post contruction.
 
I would not today totally immerse myself in ZWave or Zigbee and will not do that with WiFI.   I like to tinker and see WiFi as a new methodology of automation.  Leviton is doing WiFi to their cloud. 
 
Here in the last few months added under kitchen counter and over counter 12VDC LED strip lighting lighting.  The work mostly was installing customized aluminum strips with diffusers (tiny) that connect.  These can connect to a dimmable 120VAC to 12VDC power supply.  Doing this today with my home office book shelves.  For the kitchen tried something new using small micro 120VAC to 12VDC transformers in electrical boxes (smallest I have ever seen) connected to a standard automated switch.  The add here is a firmware and hardware updated magic home controller which talks WiFi MQTT and a manual digital on off dimmer pot.  I can set the level with the tiny switch and shut off the main power and resume the same dim level or leave the automated switch on and control it the custom magichome controller via WiFi.  Doing it this way I do not see electrical boxes for LED transformers either below or above the counter and have way better dimming of the 12VDC lighting (although the dimming works fine with the dimmable 120VAC to 12VDC LED controller I am using).
 
Today I see WiFi (and phone apps) as just primarily a cloud based easy button (phones) mechanism of automation.  Folks that are tethered to their cell phones will flock to it like ants at a picnic because it is easy button automation.  No one really cares about the grey areas (cloud) rather just that it works with their smart phones.
 
Watching globablly most of the world went from X10 to Zigbee (not ZWave) to Wifi Cloud automation.    It is still the "wild west" out there relating to automation and automation protocals.  I do see the lean to WiFi automation mostly because it is easy button automation for the masses.  BUT this is my opinion.

 

[upstatemike]

I am doing multiple technologies for myself but is that practical for most folks? On the other hand I put 20 Lifx bulbs outside for fun Holiday colors and stuff simply replacing my porch lights, lamp posts, etc. (didn't even mess with the floodlights). Even just that was 20 bulbs right there. I don't see how it would ever be practical to expand on that. Living Room lamps and walls = 12 bulbs. Front Hall chandelier = 10 bulbs for 1 fixture... it adds up fast! Hue is impractical with the 50 device per hub limit because you can't use more than 1 hub with Alexa so Wi-Fi isn't the only technology with scaling problems. Other Zigbee alternatives have not solved the local association problem, and so on. Nobody is serious about building anything that scales well beyond the basic entry level starter kit, or if they do then the product offerings are so limited (Lutron) that you still need something else to fill in the gaps.

 

[pete_c]

@Mike
 
Understood.
 
is that practical for most folks?
 
No not at all.  My wife still questions why I have automated switches in place versus the $5 analog light switch or why the house talks (been talking since the 80's).
 
For me automation is a hobby so I tinker.
 
I don't see how it would ever be practical to expand on that.
 
It isn't and not scalable.  It's cheap, easy and fast.  Kids now a day don't care.
 
They just will impress each other showing their friends and neighbors how they can turn lights on with the smartphones or with Alexa or Google Home.

 

[LarrylLix]

Just when they get the masses onto cloud things they fold them up or start to charge subscription fees.
 
Even for non-cloud usage, when they get the masses onto WiFi switches, bulbs and door locks, 5G will be the new home comm system and WiFi will be a thing of the pa$t, like all good marketing plans. 
 
OTOH: My latest laptop is doing 866Mbps in WiFi. Kind of makes my 1 Gbit wired  Ethernet cables obsolete, but we always want more. Keep your wallet open.

 

[lanbrown]

Scale: If you took all the devices currently handled by other hubs like Insteon, Hue. Z-Wave, etc. and started pushing that communication through your wireless access points/wireless router they would bog down and stop.

 
The scale is not an issue.  Z-Wave could easily go over WLAN and not have an issue and the same for Insteon or even ZigBee.  Z-Wave is not a fast protocol; like 100Kbps (max).  Insteon is even slower.  Z-wave is 9.6Kbps or 40Kbps and Z-Wave plus goes up to 100Kbps.  Insteon is 20Kbps is 40Kbps.  ZigBee is 250Kbps.  Some (if not all) are RAW data; actual will be less.  So you could run all three on WLAN and not have an issue.
 
I have four 802.11ac AP's in the attic; one at each corner.  So 2.4GHz and 5GHz coverage is excellent.  I can roam from AP to AP since they are essentially just dumb; the controller is where the traffic terminates so a client moving from an AP to AP doesn't change their IP or anything since the controller just moves the traffic from AP to AP.
 
I do not like WLAN for many HA related tasks though.  WPA3 addresses some of the concerns but I wish devices supported WPA-Enterprise rather WPA-PSK.  I do wired where I can.  DoorBirds are all wired, Amazon Fire TV and Roku boxes are all wired as well.  Nest thermostat and protects are obviously wireless but they are part of the IoT SSID and on their own VLAN.  Computers are on a different SSID and VLAN.  Guest is on a different VLAN and SSID as well.  DoorBirds are on their own VLAN but I do have a few Fire Tablets on that network with a separate SSID.  So things are well segmented and the firewall doesn't allow cross communication except in specific instances.

 

[pete_c]

Update 09th of October, 2019 of original OP.

There is not a ban on imports. There is a ban on Hikvision & Dahua from buying parts from the USA.

See here ==> hxxps://www.reuters.com/article/us-usa-trade-china-exclusive/us-puts-hikvision-chinese-security-bureaus-on-economic-blacklist-idUSKBN1WM25M