Still the article and security implications are bullshit. If you can get access to installing your malware on the machine, than the physical domain of eavesdropping is irrelevant. It's not like there is a vendor selling TEMPEST secured equipment with headphone jacks but no mics (and that messing with audio drivers would pass). Switching signal direction on jacks has been a standard feature of audio chipsets since the AC'97 standard, it's just that the auto-detection routines in most CODECs would correctly direction the jack for what you plugged in.
What is interesting is that this "hack" is in the same realm of overblown and needing excessive access as the Cisco VoIP phone hack that everyone was fellating Ang Cui for a few years back. Yeah, if I can hang out physically connected to a diagnostic port on someone's phone for several minutes to flash the firmware, I can do much better as far as surveillance. Not to mention the frequency that VoIP VLANs can't reach the Internet to egress their eavesdropping.
I have a really simple system for reading and filtering reviews at Amazon. Many advocate for reading the 3-4 star reviews for honest criticism and balanced reviews, but besides some user promoted ones in this range, they're pretty bland and lacking in key information, it's easy to see the product's strengths in the first few 5 stars that appear at the top of the listing.
The reviews to pay attention to are the 1 and 2 star reviews and look for patterns of complaints. If out of 50 1 star reviews only a few of them share similar issues with the product, than it is probably bad luck or user error. It's easy to filter some of the derp derp what did you expect the product to do. If 25 of those 50 1 stars all had the same problem and this is an aspect of the product you find unfavorable, then it's a no-buy, move on.
I've also found this aggregate negative review process to be much faster than trying to average a skimming of 20-100 of 3-10,000 different reviews all over the place as far as their overall impression with the product. After all, at some point the price of the product exceeds my time lost to reading redundant, useless reviews. There is rarely much useful in positive reviews, I obviously already want the product for the good things it can do in my life if I am looking into buying it.
It is when nearly half are not "doers" in the system - non-clinical staff. That means for every doctor, nurse, gp, and hospital staff member, there is a bureaucrat jockeying a desk.
I work at a retailer with >100k employees. Our corporate office staff is under 5k.
The NHS clearly needs an efficiency expert to come slash office jobs.
The same guy who is complaining that TV / Movies at normal speed are too slow, probably couldn't comprehend and fully appreciate the average Aaron Sorkin sitcom dialog interchange slowed down by 20%.
Different writer / director combinations dump data at the viewer at different speeds and use a variety of compression algorithms (references, partial quotes, alliteration, anagrams, homonyms, puns, etc.) to embed additional metadata into the verbal stream.
Guo and others who are fans of accelerated speech are watching low data density dialogue as well as not intellectually interested in the full texture of well written material. They want the bullet points and laughs and to be done with it.
Also, Google customizes search result based on one's browsing history and past searches, so maybe those results are because of the searcher's history as well. You bring up an excellent point though.
It's nigh impossible to get a write-in a majority over the 2 offered candidates. A real "no confidence" option would likely win a majority more than any write-in option.
The point is to have an EFFECTIVE alternate option, of which write-in is not.
You have every right to complain about how people vote when there isn't a "no confidence in either candidate" option. Not voting is a growing sign of discontent which will hopefully result in violent revolution. We've proven consistently that grassroots and other forms of non-violent revolution make little long term change (e.g. the Tea Party) with minor exceptions for obvious, and frankly trivial from a legal standpoint social policy (suffrage, civil rights).
Real change to the movers and shakers, campaign contributions, cronyism, wealth distribution, general body of criminal, tax, budget, and welfare law are well beyond non-violent means.
I haven't found SmartTVs to be a significant cost increase over the regular ones since last fall with the rise of <$1000 4k televisions. There are lots of premium TVs which heavily exhort their SmartTV capabilities, but comparing actual comparable models (size, panel quality, etc.) seems to be very little difference in price. Where there is a huge difference is getting larger, premium quality panels with superior color reproduction and refresh rates. Those ones are already going to cost a nice premium, and universally that halo market includes Smart functionality.
Let's go through a few specific statements:
1) "media-center computers and DVRs are ubiquitous" - Umm nope, not even close. They're common with middle and upper class white techy males and their families. Heck, I have XBMC on a PC hooked up to my TV and still use the native "SmartTV" functionality for Netflix and Prime because dealing with those on XBMC is extra steps and inferior. XBMC is for playing stuff I've downloaded almost exclusively.
2) "smartphones have HDMI connectivity" - Most do NOT, more have Miracast/AllShare or similar wireless tech. They literally overheat or lose battery while charging and mirroring whether wired or wirelessly.
3) "Raspberry Pi is inexpensive and can play 1080 content at full framerate" - This is a lot of extra work for non-technical common folk. You might have gotten more mileage from saying AppleTV, Roku, Chromecast, or FireTV. Of those, only AppleTV and Roku have a user interface and setup which isn't vastly more painful than the average SmartTV. Chromecast and FireTV are actually a huge PITA for non-techy people, and even as an IT engineer, I found FireTV to be a piling steam of slow crap, eclipsed in speed and usefulness by a $600 black Friday special (read: cheaper components) 4K SmartTV's native software (Tizen?). Further the $100 price tag for the AppleTV or better model Roku is more than the SmartTV price jumps in most situations.
4) None of these devices are terribly expensive anymore, and the price jump from a non-smart TV to a smart TV makes it difficult to justify the expense. - See original point.
The Linux community is still in denial over the fact that claims of "year of the Linux desktop" starting around 1998 were quickly destroyed by Windows 2000 and XP. Windows 7 certainly put the final nails in that coffin.
Linux/GNU is a fine OS for monolithic servers with minimal update needs and it works fine as a base kernel to run a completely different environment and API on top, as is the case with Android.
Android is paid for by advertising and data mining. Further, Android is provided with virtually zero real support (e.g. direct support and ticketing, solutions not starting with full wipe, etc.). There is no reason I would pay for a product like that ever.
Now if Microsoft started making Windows ROMs for Android or Apple with iOS, I might consider paying just to try those out on Android hardware. Same would go for Linux distributions (or QNX or whatever) that came without OS level data mining / advertising, and came with real support.
If the primary method of paying for a product or service is trying to advertise to me or monetize my behavior, I will not pay for it. The same goes for television / cable content, music, and websites. I will pirate it or adblock it without guilt and without sitting through advertisements because I as a consumer already buy plenty of products from the companies paying for the content. I've already paid my share for the product.
My biggest gripe has been the fact that they're over priced in the US since you have to go through crappy eBay or similar overseas distribution channels. Then again, I have a soft cap of around $300 that I will spend on a smartphone. The difference in construction cost between a $700 flagship and $200 basic phone is about $70 ($70 vs $140 parts / build cost).
I'm certainly not spending $300-600 for a Mi5 where the $300 model is crippled with low RAM/ROM (2/16GB) when I can get something vastly more usable like an Asus Zenfone 2 (4/64GB+microSD) at the same price point over a year ago. As long as I can keep crapware and API (un)improvements like biometrics replacing my password and blowing through CPU away, the only reason I would need the fastest / newest CPU/GPU would be gaming....
Given how much goes to the hands of middlemen and just a pittance to the artists, I am happy obtaining the music I want through non-subscription means and alternate distribution channels.
Seriously, I can't see how people would stomach paying more than $20-40 a year for unlimited, well curated music and music suggestions. The current Spotify and Pandora pay models are just stupid overpriced.
Secondly, I already pay for most of the music. You know the products which the do the radio ads which pay for the music on the radio. We've all bought some of them. The margin on those products paid for the music, we've paid for it already.
You know what music I do pay for and buy all the time? Albums from local artists I enjoy who aren't on AM/FM radio. Some are marginally obtainable on online services, but I rather pay $10-15 for a CD and know that $8-14 went straight to the band / artist.
What an incredibly stupid way to blow through CPU cycles. Seriously, use my local processing power for things I want, like local search, voice interaction and navigation which can work offline / from cache consistently.
There is a second HUGE problem with this. Any app can gather sufficient biometrics to falsify a Trust Score. Even worse, unlike say an intentionally malicious app which could just replace your keyboard app and grab passwords by key logging, advertising and other agencies could request little pieces of biometrics and heuristics from different sources in innocuous ways until a complete picture for forging a Trust Score emerge.
Didn't we just go over the bit about RunKeeper recording and then passing along a fairly nice stack of location / movement statistics?
Except presumption or guilt / innocence along with a misuse of deliver versus create. That legal interpretation is logically flawed. The evidence, if a person is guilty in the first place, already exists in their memory as well as any physical evidence. The same logic would be to jail someone indefinitely for failing to tell the state where they ditched the knife / gun / big stick they used to kill someone.
The funny thing is I am getting used to hearing this special kind of rambling. It's an epidemic in dealing with nearly every IT or really any engineering resource in Southeast Asia, especially certain parts of China, all of India, and others. It appears to stem from two intellectual deficiencies. The first is big picture thinking. I've met plenty of engineers who were one trick ponies and couldn't even see how their expertise in X could be applied to Y, much less understood how their knowledge fit in the tapestry of technical architecture. The second is the need to fill time / space with words when they have nothing useful. A huge list of banal generalities spew out, either positive or negative based on what they think the audience wants to hear.
The outsourcing movement has a couple more decades of rubber band-like hysteresis. Most companies that send technical services out to Chennai or Bangalore save their short term cash for 2-4 years, pay the CEO and CIO their bonuses, and then bring them back to the US as soon as they can get out of the contracts they've signed.
"Your friendly reminder that without open standards, you're not "buying" smarthome hardware, you're renting it." - The F/OSS FUD is tiresome.
For nearly 20 years I have made the argument that software simply needs to fall under the same laws as any other consumer product for quality, reliability, safety, and doctrine of first sale. Bugs that make software unusable / crash prone should be the same as a coffee machine that only lasts a week or tends to start fires. The manufacturer can replace it until it works, or give me a full refund, regardless of if it puts them out of business.
But yes, you're right, people are assuming the robot has some expert or additional knowledge, e.g. it's wired into the building's fire alarm system and knows the safe place to go and similar.
On the downside, I keep thinking about the movie adaptation or I, Robot, and what a huge segment of the population are complete ignorant sheep.
Also pretty sure anything audio related in Linux land is going to be stillborn thanks to PulseAudio. From what I have run across, base ALSA isn't going to help much either.
Content creation whether, audio, video, 2D, 3D, or other at the professional level generally requires the ability to jump among a variety of software packages, plugins, and occasional one-jobbers. You're not going to see that anytime soon in Linux. Audio and 2D have a foothold in OSX because of ProTools and Adobe, but everything else runs on Windows workflows. 3D used to be a big one for industrial *nixes, but that ended in the late 90's / early 2000s when Microsoft bought SoftImage to port their software to NT 4.0. I run across a few Linux zealots talking about Blender and on the more professional side, Houdini, but none of those people are gainfully employed on a regular basis. Content creation is about the workflow and being able to accomplish things end to end. There's little hope there for Linux as a desktop OS unless you can restrict your use case to a subset of the software available there.
I was going to say, as much as the study of sociopaths occupying C-level positions and Boards of Directors has gotten increasing attention the recent few years, I wonder if studies of middle management following Milgram's studies similarly correlates.
Isn't this why we run our bots in VMs. Plus we can clone the template VM and run many, many copies of the software and bot as well as upload it to the cloud!
Still the article and security implications are bullshit. If you can get access to installing your malware on the machine, than the physical domain of eavesdropping is irrelevant. It's not like there is a vendor selling TEMPEST secured equipment with headphone jacks but no mics (and that messing with audio drivers would pass). Switching signal direction on jacks has been a standard feature of audio chipsets since the AC'97 standard, it's just that the auto-detection routines in most CODECs would correctly direction the jack for what you plugged in.
What is interesting is that this "hack" is in the same realm of overblown and needing excessive access as the Cisco VoIP phone hack that everyone was fellating Ang Cui for a few years back. Yeah, if I can hang out physically connected to a diagnostic port on someone's phone for several minutes to flash the firmware, I can do much better as far as surveillance. Not to mention the frequency that VoIP VLANs can't reach the Internet to egress their eavesdropping.
I have a really simple system for reading and filtering reviews at Amazon. Many advocate for reading the 3-4 star reviews for honest criticism and balanced reviews, but besides some user promoted ones in this range, they're pretty bland and lacking in key information, it's easy to see the product's strengths in the first few 5 stars that appear at the top of the listing.
The reviews to pay attention to are the 1 and 2 star reviews and look for patterns of complaints. If out of 50 1 star reviews only a few of them share similar issues with the product, than it is probably bad luck or user error. It's easy to filter some of the derp derp what did you expect the product to do. If 25 of those 50 1 stars all had the same problem and this is an aspect of the product you find unfavorable, then it's a no-buy, move on.
I've also found this aggregate negative review process to be much faster than trying to average a skimming of 20-100 of 3-10,000 different reviews all over the place as far as their overall impression with the product. After all, at some point the price of the product exceeds my time lost to reading redundant, useless reviews. There is rarely much useful in positive reviews, I obviously already want the product for the good things it can do in my life if I am looking into buying it.
It is when nearly half are not "doers" in the system - non-clinical staff. That means for every doctor, nurse, gp, and hospital staff member, there is a bureaucrat jockeying a desk.
I work at a retailer with >100k employees. Our corporate office staff is under 5k.
The NHS clearly needs an efficiency expert to come slash office jobs.
The same guy who is complaining that TV / Movies at normal speed are too slow, probably couldn't comprehend and fully appreciate the average Aaron Sorkin sitcom dialog interchange slowed down by 20%.
Different writer / director combinations dump data at the viewer at different speeds and use a variety of compression algorithms (references, partial quotes, alliteration, anagrams, homonyms, puns, etc.) to embed additional metadata into the verbal stream.
Guo and others who are fans of accelerated speech are watching low data density dialogue as well as not intellectually interested in the full texture of well written material. They want the bullet points and laughs and to be done with it.
Also, Google customizes search result based on one's browsing history and past searches, so maybe those results are because of the searcher's history as well. You bring up an excellent point though.
It's nigh impossible to get a write-in a majority over the 2 offered candidates. A real "no confidence" option would likely win a majority more than any write-in option.
The point is to have an EFFECTIVE alternate option, of which write-in is not.
You have every right to complain about how people vote when there isn't a "no confidence in either candidate" option. Not voting is a growing sign of discontent which will hopefully result in violent revolution. We've proven consistently that grassroots and other forms of non-violent revolution make little long term change (e.g. the Tea Party) with minor exceptions for obvious, and frankly trivial from a legal standpoint social policy (suffrage, civil rights).
Real change to the movers and shakers, campaign contributions, cronyism, wealth distribution, general body of criminal, tax, budget, and welfare law are well beyond non-violent means.
I haven't found SmartTVs to be a significant cost increase over the regular ones since last fall with the rise of <$1000 4k televisions. There are lots of premium TVs which heavily exhort their SmartTV capabilities, but comparing actual comparable models (size, panel quality, etc.) seems to be very little difference in price. Where there is a huge difference is getting larger, premium quality panels with superior color reproduction and refresh rates. Those ones are already going to cost a nice premium, and universally that halo market includes Smart functionality.
Let's go through a few specific statements:
1) "media-center computers and DVRs are ubiquitous" - Umm nope, not even close. They're common with middle and upper class white techy males and their families. Heck, I have XBMC on a PC hooked up to my TV and still use the native "SmartTV" functionality for Netflix and Prime because dealing with those on XBMC is extra steps and inferior. XBMC is for playing stuff I've downloaded almost exclusively.
2) "smartphones have HDMI connectivity" - Most do NOT, more have Miracast/AllShare or similar wireless tech. They literally overheat or lose battery while charging and mirroring whether wired or wirelessly.
3) "Raspberry Pi is inexpensive and can play 1080 content at full framerate" - This is a lot of extra work for non-technical common folk. You might have gotten more mileage from saying AppleTV, Roku, Chromecast, or FireTV. Of those, only AppleTV and Roku have a user interface and setup which isn't vastly more painful than the average SmartTV. Chromecast and FireTV are actually a huge PITA for non-techy people, and even as an IT engineer, I found FireTV to be a piling steam of slow crap, eclipsed in speed and usefulness by a $600 black Friday special (read: cheaper components) 4K SmartTV's native software (Tizen?). Further the $100 price tag for the AppleTV or better model Roku is more than the SmartTV price jumps in most situations.
4) None of these devices are terribly expensive anymore, and the price jump from a non-smart TV to a smart TV makes it difficult to justify the expense. - See original point.
The Linux community is still in denial over the fact that claims of "year of the Linux desktop" starting around 1998 were quickly destroyed by Windows 2000 and XP. Windows 7 certainly put the final nails in that coffin.
Linux/GNU is a fine OS for monolithic servers with minimal update needs and it works fine as a base kernel to run a completely different environment and API on top, as is the case with Android.
Android is paid for by advertising and data mining. Further, Android is provided with virtually zero real support (e.g. direct support and ticketing, solutions not starting with full wipe, etc.). There is no reason I would pay for a product like that ever.
Now if Microsoft started making Windows ROMs for Android or Apple with iOS, I might consider paying just to try those out on Android hardware. Same would go for Linux distributions (or QNX or whatever) that came without OS level data mining / advertising, and came with real support.
If the primary method of paying for a product or service is trying to advertise to me or monetize my behavior, I will not pay for it. The same goes for television / cable content, music, and websites. I will pirate it or adblock it without guilt and without sitting through advertisements because I as a consumer already buy plenty of products from the companies paying for the content. I've already paid my share for the product.
I was speaking of buying CD directly from the artists, like when they do live shows, not people with a signed distributor / contract.
My biggest gripe has been the fact that they're over priced in the US since you have to go through crappy eBay or similar overseas distribution channels. Then again, I have a soft cap of around $300 that I will spend on a smartphone. The difference in construction cost between a $700 flagship and $200 basic phone is about $70 ($70 vs $140 parts / build cost).
I'm certainly not spending $300-600 for a Mi5 where the $300 model is crippled with low RAM/ROM (2/16GB) when I can get something vastly more usable like an Asus Zenfone 2 (4/64GB+microSD) at the same price point over a year ago. As long as I can keep crapware and API (un)improvements like biometrics replacing my password and blowing through CPU away, the only reason I would need the fastest / newest CPU/GPU would be gaming....
Given how much goes to the hands of middlemen and just a pittance to the artists, I am happy obtaining the music I want through non-subscription means and alternate distribution channels.
Seriously, I can't see how people would stomach paying more than $20-40 a year for unlimited, well curated music and music suggestions. The current Spotify and Pandora pay models are just stupid overpriced.
Secondly, I already pay for most of the music. You know the products which the do the radio ads which pay for the music on the radio. We've all bought some of them. The margin on those products paid for the music, we've paid for it already.
You know what music I do pay for and buy all the time? Albums from local artists I enjoy who aren't on AM/FM radio. Some are marginally obtainable on online services, but I rather pay $10-15 for a CD and know that $8-14 went straight to the band / artist.
What an incredibly stupid way to blow through CPU cycles. Seriously, use my local processing power for things I want, like local search, voice interaction and navigation which can work offline / from cache consistently.
There is a second HUGE problem with this. Any app can gather sufficient biometrics to falsify a Trust Score. Even worse, unlike say an intentionally malicious app which could just replace your keyboard app and grab passwords by key logging, advertising and other agencies could request little pieces of biometrics and heuristics from different sources in innocuous ways until a complete picture for forging a Trust Score emerge.
Didn't we just go over the bit about RunKeeper recording and then passing along a fairly nice stack of location / movement statistics?
Wait, wait, maybe we should get them into Congress to enact laws and guidance to correct the dama... well crap.
Except presumption or guilt / innocence along with a misuse of deliver versus create. That legal interpretation is logically flawed. The evidence, if a person is guilty in the first place, already exists in their memory as well as any physical evidence. The same logic would be to jail someone indefinitely for failing to tell the state where they ditched the knife / gun / big stick they used to kill someone.
The funny thing is I am getting used to hearing this special kind of rambling. It's an epidemic in dealing with nearly every IT or really any engineering resource in Southeast Asia, especially certain parts of China, all of India, and others. It appears to stem from two intellectual deficiencies. The first is big picture thinking. I've met plenty of engineers who were one trick ponies and couldn't even see how their expertise in X could be applied to Y, much less understood how their knowledge fit in the tapestry of technical architecture. The second is the need to fill time / space with words when they have nothing useful. A huge list of banal generalities spew out, either positive or negative based on what they think the audience wants to hear.
The outsourcing movement has a couple more decades of rubber band-like hysteresis. Most companies that send technical services out to Chennai or Bangalore save their short term cash for 2-4 years, pay the CEO and CIO their bonuses, and then bring them back to the US as soon as they can get out of the contracts they've signed.
"Your friendly reminder that without open standards, you're not "buying" smarthome hardware, you're renting it." - The F/OSS FUD is tiresome.
For nearly 20 years I have made the argument that software simply needs to fall under the same laws as any other consumer product for quality, reliability, safety, and doctrine of first sale. Bugs that make software unusable / crash prone should be the same as a coffee machine that only lasts a week or tends to start fires. The manufacturer can replace it until it works, or give me a full refund, regardless of if it puts them out of business.
Inverse Dunning-Kruger?
But yes, you're right, people are assuming the robot has some expert or additional knowledge, e.g. it's wired into the building's fire alarm system and knows the safe place to go and similar.
On the downside, I keep thinking about the movie adaptation or I, Robot, and what a huge segment of the population are complete ignorant sheep.
Nice replacement for phosphorus rounds if you ask me.
Closest might be the HP Elite x3 if you get the laptop dock. I'm only half joking.
It's the vegan of phones. Terrible specs at the price point at that.
Also pretty sure anything audio related in Linux land is going to be stillborn thanks to PulseAudio. From what I have run across, base ALSA isn't going to help much either.
Content creation whether, audio, video, 2D, 3D, or other at the professional level generally requires the ability to jump among a variety of software packages, plugins, and occasional one-jobbers. You're not going to see that anytime soon in Linux. Audio and 2D have a foothold in OSX because of ProTools and Adobe, but everything else runs on Windows workflows. 3D used to be a big one for industrial *nixes, but that ended in the late 90's / early 2000s when Microsoft bought SoftImage to port their software to NT 4.0. I run across a few Linux zealots talking about Blender and on the more professional side, Houdini, but none of those people are gainfully employed on a regular basis. Content creation is about the workflow and being able to accomplish things end to end. There's little hope there for Linux as a desktop OS unless you can restrict your use case to a subset of the software available there.
I was going to say, as much as the study of sociopaths occupying C-level positions and Boards of Directors has gotten increasing attention the recent few years, I wonder if studies of middle management following Milgram's studies similarly correlates.
Isn't this why we run our bots in VMs. Plus we can clone the template VM and run many, many copies of the software and bot as well as upload it to the cloud!