Dropping OtherOS on the PS3 was the final straw for me and Sony. At least when I buy a Samsung product I expect to be screwed in some way (e.g., all my old Samsung TVs slowly dropping support for apps... easily fixed with a FireStick or similar).
Not MRIs, but for some reason the major genome sequencing instrument vendors generally require remote access to their instruments (Illumina and PacBio both do this - PacBio was just bought by Illumina, but they've been doing it since the beginning). Heck, there used to be a map that someone made that found Illumina instruments on the internet and plotted their physical locations based on the IP address (it doesn't seem to exist anymore). They also tend to run unpatched versions of Windows.
Sequencing data is even easier than MRI data is to mess with. The raw data is large (10-100s of GBs per run) and always processed by computational pipelines that are mix of scripts and random tools downloaded from the internet. Unlike with artifacts that would be detectable in an altered image, changing sequencing data is simply a matter of flipping a few characters.
We once wrote a script that scanned for a specific sequence related to a certain cancer. By flipping a few characters, it was possible to give the patient the variants that lead to a higher probability of developing cancer. We could have also done the opposite and made patients appear to have no pre-disposition when they in fact did. (our script looked at all the short reads right of the instrument and tweaked them, it didn't catch all cases, but more than enough that the variant caller gave our intended call)
An hypothetical extension of this would let scan for specific individuals based on previously sequenced samples. From there, one could write a script that only "gave" cancer to that person.
Related: I wonder how well humans familiar with folding motifs and all the confounding factors present in nature would do vs. the models. While most chemists rely on modeling, NMR, and crystallography, the techs running these systems all have intuitions built up from years of generating structures.
Would some of them outperform the models in the same way Google's approach did?
Because they are not safe and the companies are exploiting gaps in laws and enforcement to profit at their customer's expense.
I've been a bike commuter for 20+ years and am well aware of the hazards of the roads and how to bike in a city safely and lawfully. The vast majority of scooter riders are not being safe and are breaking laws on a regular basis.
Some examples (Austin as my reference point):
- Scooter users regularly dart through intersections when lights are red, often executing dangerous left turns. Proper action: act like you're a car and obey the stop lights or get off the scooter and use the cross walk. - Scooter users try to pass cars on the right when cars are making right turns. Proper action: act like a car and line up behind the car and wait your turn. - Scooter users weaving in and out of pedestrians on side walks. Proper action: don't do this. All it takes is a pedestrian waving their arm or shifting their path and both the scooter user and pedestrian are injured. - Lack of helmets. Proper action: Wear a helmet. Sure, they don't stop concussions (common argument used against helmets that's generally true), but they do stop your skull from getting cracked in low speed collusions (much worse than a concussion).
I can go on, but you get the point. Riding these scooters is not like walking or using a push scooter. They operate at speeds most users are not familiar with on infrastructure not designed for their use. There's also an entire population of other infrastructure users that are not conditioned to be aware of them.
Look, I love the idea of these scooters. I love that there's potential to help us all adapt to make it safer for all lower speed methods of transportation (bikes, skateboards, push scooters, running;) ). But, putting everyone at risk and pretending you're not is an incredibly irresponsible way to go about it.
As someone who's tried most VR tech since the early 90s, all my experience with the current generation suggests that the tech is indeed good enough.
In particular, I get lost in the Steam/HTC Vive setup my friend has every time I use it. Google Earth alone is a killer app, if you know where to go (I'm a climber: try Yosemite or Eldorado Canyon in Boulder, they've imaged the cliffs in both places to the point where you can actually see the handholds and make out routes). The paint programs are surreal as well.
I get motion sick easily. This is the first generation of VR gear that I've been able to spend 30+ minutes with the headset on and feel fine afterwards.
So, why don't I have my own VR setup? Two reasons: (1) cost and (2) I have half of it. For the latter, I purchased an X-Box One X for my son specifically because MS was setting it up as a VR platform. $500 for a game console was steep, but the hardware was right for good VR. Unfortunately, MS has now signaled that VR is not coming to the platform and I've overpaid for a gaming console. (yes, I should have just bought a PS4)
Cost, and to a lesser extent the hassle associated with that cost, is what I think is the real issue. Without a consumer friendly setup in the $500-700 range all-in that supports all VR content (PS4's problem is content), it's just too expensive to get started. I don't want to drop a few grand on a high end gaming PC, then the hundreds on the VR gear, plus the time it will take to setup and maintain the PC. It's just too expensive in money and time commitment.
The tech is there. There are compelling apps. It's just still too expensive to get started.
Not sure about the OPs case, but if you work around gear that’s sensitive to RF noise, you don’t want everyone brining their own transmitter to work. Even without a radio, electronics can interfere with sensitive equipment.
For the knowledge workers posting here insisting that they need 8 hours to be productive, how many of those 8 (or 10 or 12) hours that you insist you work are spent on the web, “training”, shopping, or hanging with co-workers?
I’ve worked in software since the beginning of the web as a developer, manager, founder, and executive with highly productive teams developing technical software. You know what’s been constant regardless of the company or team? “Down time” throughout the day, either via the web or smoking breaks or exercise breaks. No one who’s honest with themselves will claim to work a full day every day. Sure, some days you’re heads down in code and bang out 12 hours of solid work, but most days, if you’re honest about the downtime, you did much less.
I had one employee once who worked closer to 6 hour days. He’s the only person I’ve seen not slack off regularly. He came in and got to work. He was more productive than most of the 8 hour employees. N of 1 but I noticed it then and have since paid attention to how much people actually work. It’s closer to 6 hours.
Web Master was the hottest job 20 years ago. Right up until every realized that the position was better filled via a mix traditional IT techs and software engineers.
Data science will go the same way, but it will be software engineers and statisticians that replace the current crop of bootcamp trained data "scientists". (actually, all real data science shops already do it that way... the market will correct)
Exactly why a click campaign may work. At some point, advertising budgets have to be justified. While initially the clicks will be seen as a positive, after a quarter or two of no uptick in revenue relative to ad spend, those ad budgets will start to get questioned.
Procter and Gamble is continuing to cut online ad spend due to the ineffectiveness of their ads. They're on the leading edge here, it's just up to the community to help move everyone else down that path (http://www.adweek.com/brand-marketing/when-procter-gamble-cut-200-million-in-digital-ad-spend-its-marketing-became-10-more-effective/)
For tracking, the goal is to decrease the signal to noise ratio and make the tracking databases worthless.
Why can’t we do the same with auto play, gifs, and everything else that fights for our attention?
My gueriila suggestion: just start clicking on all ads that annoy you. Make the advertisers pay. And mess up their “valuable” tracking metrics in the process.
This is not about volitility or any other type of consumer protection. This is about protecting their points programs.
Basically, people were buying crypto coins on their credit cards, getting the points for the purchase, then immediately selling the coins and paying off the card. The points more than made up for the transaction costs.
Since all these cards were being paid off before any interest accrued, the credit card companies were losing a lot of money on the transactions in the form of reward payouts.
The CC firms will never admit this publicly, but given the number of people I knew who were doing this, I’d bet this is the real reason they’ll disallowing the purchases.
My kids can message all they want with people I approve of. How do I enforce this? All messaging happens through my phone while I'm present (Facetime with relatives, text messages to plan playdates, photo sharing, etc).
Anything else that tries to set up messaging accounts for my kids? If messaging/chat features can't be disabled, my kids can't use it.
Minor correction: cargo holds are pressurized at cabin pressure. The whole plane is a giant cylinder. Having two diffent pressure zones or even just an oddly shaped cabin is more dangerous than just pressurizing the whole thing.
Remember: animals travel in the cargo hold all the time.
And I should reply directly the the parent: statistically stringing together text is also not new. We just have better collections to start training algorithms with.
Handwriting, speech recognition, and image processing along with their machine learning foundations do not impress the older/. crowd because they are not new technologies.
Dragon has been doing speech recognition better than Siri for almost 20 years. Simple command-based systems that only recognize a few words have been around longer than that.
Handwriting recognition for constrained tasks is also not new. The US Postal Services has had zip-code OCR systems since the 1980s.
Feature detection in images is not new, either. The only thing that's really changed there is we have the processing power to do it at scale.
Going beyond the applications, all the modern "AI" systems are simply classifiers on steroids. Processing power and great storage capacity allows us work on larger data sets, but in the end, we're just creating complex hyper-planes to bin data in one bucket or another.
Machine learning algorithms are great tools and it's great that we have the compute resources to really leverage them, but there's nothing really new that wasn't obvious 30 years ago. The only question was when we'd have the compute power to start doing the cool things we knew they could be used for.
(ok, I'll give a little credit to the deep learning researchers for bringing neural nets back into vogue, since those were written off 30 years ago during the AI winter, but they're still just classifiers from the mathematical perspective).
Indeed, and it didn't need that fsck'ing HARDWARE cursor emulation that the PC needed, either!
(Yes, Hercules, CGA, EGA and VGA had the text-mode cursor in hardware (including the blinking)). VGA (and maybe EGA, I forget) also had a single "sprite" for the "hardware" cursor.
I was coming here to make exactly this point. Cursors used to be hardware sprites that required no additional CPU cycles. At some point windowing systems took over the task of rendering the cursor, but still typically used XOR'd sprites to keep things fast and efficient. Then they started using GPU-optimized code and software CPU emulators as a fallback and things went downhill from there...
I still keep my 80s-era graphics programming books on my shelf as a reminder of how to do fast graphics when all you have is the ability to draw pixels...;)
What I've always found funny about this interview process is that the assumption is always that the interviewer knows the correct answer(s) to the question. It's painful when they don't.
Years ago I interviewed at Google and was asked a question about bit counting (some variation on "given a bit vector, wat's the fastest way to count the number of 1s?"). I quickly answered, "well, if your processor has a population count instruction, stream your vector through that and accumulate the result in a register". Having just evaluated bit counting methods as part of my Ph.D. dissertation, I knew this was the fastest way to do it, assuming the instruction was available (it's not on x86, but is on Power/VMX and most DSPs support it as well).
After I got a blank stare back from the interviewee, I said, "Oh, you were looking for the lookup table answer". We could have left it at that, but he went on to explain using some very convoluted logic how the lookup table would actually be faster than using a dedicated instruction and that my answer was clearly wrong. I mentioned a little bit about the number of cycles required for each approach, but he had none of it. In his mind, I had the wrong answer, even though my second answer was what he was looking for.
It was at that moment that I realized Google was not going to be a good place to work.
As others are saying, don't live in the Bay Area if you can't afford it. But, if you want housing that's affordable and not too far away, it's not impossible...
There's the whole Central Valley within driving distance of the Bay Area. Sure, a 1-2 hour commute isn't ideal, but with a flexible work schedule and work-from-home options some days the of the week, it's totally doable. You can get a nice house with a pool in a small CV town for less than $250k. Hell, in New England "bedroom communities" are all over the place and feature similar price differences/commute times. (you can even throw in a few nights each month for a hotel and still come out ahead)
Fwiw, I grew up in the Central Valley. Day in and day out it's really no different than living anywhere else - you eat, sleep, and work, lather/rinse/repeat. Oh, and you're much closer to the Sierras than you are in the Bay Area, if mountains are your thing. An hour and half to the slopes is much nicer than the 6-12 hours it takes to get to Tahoe on the weekends from the Bay Area.
Of course, since this is in mail.app, which I use constantly, this is the first I've heard about it.
I wonder how many great features in Apple products people miss simply because Apple refuses to provide sensible documentation and instead relies on users to "discover" features organically or via message boards.
So, at the more hardcore geek conference (Supercomputing comes to mind), there has never really been an issue with booth babes for a simple reason: geeks are scared to talk to them. Every now and then a company will hire one, only to see a nice exclusion zone form around their booth. Sure, sales guys from other booths will stop by, but none of the intended audience will risk talking with an attractive female.
Slashdot Mirror
Dropping OtherOS on the PS3 was the final straw for me and Sony. At least when I buy a Samsung product I expect to be screwed in some way (e.g., all my old Samsung TVs slowly dropping support for apps... easily fixed with a FireStick or similar).
Not MRIs, but for some reason the major genome sequencing instrument vendors generally require remote access to their instruments (Illumina and PacBio both do this - PacBio was just bought by Illumina, but they've been doing it since the beginning). Heck, there used to be a map that someone made that found Illumina instruments on the internet and plotted their physical locations based on the IP address (it doesn't seem to exist anymore). They also tend to run unpatched versions of Windows.
Sequencing data is even easier than MRI data is to mess with. The raw data is large (10-100s of GBs per run) and always processed by computational pipelines that are mix of scripts and random tools downloaded from the internet. Unlike with artifacts that would be detectable in an altered image, changing sequencing data is simply a matter of flipping a few characters.
We once wrote a script that scanned for a specific sequence related to a certain cancer. By flipping a few characters, it was possible to give the patient the variants that lead to a higher probability of developing cancer. We could have also done the opposite and made patients appear to have no pre-disposition when they in fact did. (our script looked at all the short reads right of the instrument and tweaked them, it didn't catch all cases, but more than enough that the variant caller gave our intended call)
An hypothetical extension of this would let scan for specific individuals based on previously sequenced samples. From there, one could write a script that only "gave" cancer to that person.
TFA actually does a good job of discussing it in layman's terms. It's surprisingly devoid of hype and hyperbole.
As a starting point that's not TFA, Formal Verification is the sub-field of CS that this is based on: https://en.wikipedia.org/wiki/...
-Chris
I wish I had mod points for this.
Related: I wonder how well humans familiar with folding motifs and all the confounding factors present in nature would do vs. the models. While most chemists rely on modeling, NMR, and crystallography, the techs running these systems all have intuitions built up from years of generating structures.
Would some of them outperform the models in the same way Google's approach did?
-Chris
Because they are not safe and the companies are exploiting gaps in laws and enforcement to profit at their customer's expense.
I've been a bike commuter for 20+ years and am well aware of the hazards of the roads and how to bike in a city safely and lawfully. The vast majority of scooter riders are not being safe and are breaking laws on a regular basis.
Some examples (Austin as my reference point):
- Scooter users regularly dart through intersections when lights are red, often executing dangerous left turns. Proper action: act like you're a car and obey the stop lights or get off the scooter and use the cross walk.
- Scooter users try to pass cars on the right when cars are making right turns. Proper action: act like a car and line up behind the car and wait your turn.
- Scooter users weaving in and out of pedestrians on side walks. Proper action: don't do this. All it takes is a pedestrian waving their arm or shifting their path and both the scooter user and pedestrian are injured.
- Lack of helmets. Proper action: Wear a helmet. Sure, they don't stop concussions (common argument used against helmets that's generally true), but they do stop your skull from getting cracked in low speed collusions (much worse than a concussion).
I can go on, but you get the point. Riding these scooters is not like walking or using a push scooter. They operate at speeds most users are not familiar with on infrastructure not designed for their use. There's also an entire population of other infrastructure users that are not conditioned to be aware of them.
Look, I love the idea of these scooters. I love that there's potential to help us all adapt to make it safer for all lower speed methods of transportation (bikes, skateboards, push scooters, running ;) ). But, putting everyone at risk and pretending you're not is an incredibly irresponsible way to go about it.
They did signal VR and even put a lot of development effort into it. It's only recently that they decided against it:
https://www.cnet.com/news/here...
As someone who's tried most VR tech since the early 90s, all my experience with the current generation suggests that the tech is indeed good enough.
In particular, I get lost in the Steam/HTC Vive setup my friend has every time I use it. Google Earth alone is a killer app, if you know where to go (I'm a climber: try Yosemite or Eldorado Canyon in Boulder, they've imaged the cliffs in both places to the point where you can actually see the handholds and make out routes). The paint programs are surreal as well.
I get motion sick easily. This is the first generation of VR gear that I've been able to spend 30+ minutes with the headset on and feel fine afterwards.
So, why don't I have my own VR setup? Two reasons: (1) cost and (2) I have half of it. For the latter, I purchased an X-Box One X for my son specifically because MS was setting it up as a VR platform. $500 for a game console was steep, but the hardware was right for good VR. Unfortunately, MS has now signaled that VR is not coming to the platform and I've overpaid for a gaming console. (yes, I should have just bought a PS4)
Cost, and to a lesser extent the hassle associated with that cost, is what I think is the real issue. Without a consumer friendly setup in the $500-700 range all-in that supports all VR content (PS4's problem is content), it's just too expensive to get started. I don't want to drop a few grand on a high end gaming PC, then the hundreds on the VR gear, plus the time it will take to setup and maintain the PC. It's just too expensive in money and time commitment.
The tech is there. There are compelling apps. It's just still too expensive to get started.
-Chris
Not sure about the OPs case, but if you work around gear that’s sensitive to RF noise, you don’t want everyone brining their own transmitter to work. Even without a radio, electronics can interfere with sensitive equipment.
For the knowledge workers posting here insisting that they need 8 hours to be productive, how many of those 8 (or 10 or 12) hours that you insist you work are spent on the web, “training”, shopping, or hanging with co-workers?
I’ve worked in software since the beginning of the web as a developer, manager, founder, and executive with highly productive teams developing technical software. You know what’s been constant regardless of the company or team? “Down time” throughout the day, either via the web or smoking breaks or exercise breaks. No one who’s honest with themselves will claim to work a full day every day. Sure, some days you’re heads down in code and bang out 12 hours of solid work, but most days, if you’re honest about the downtime, you did much less.
I had one employee once who worked closer to 6 hour days. He’s the only person I’ve seen not slack off regularly. He came in and got to work. He was more productive than most of the 8 hour employees. N of 1 but I noticed it then and have since paid attention to how much people actually work. It’s closer to 6 hours.
Web Master was the hottest job 20 years ago. Right up until every realized that the position was better filled via a mix traditional IT techs and software engineers.
Data science will go the same way, but it will be software engineers and statisticians that replace the current crop of bootcamp trained data "scientists". (actually, all real data science shops already do it that way... the market will correct)
Exactly why a click campaign may work. At some point, advertising budgets have to be justified. While initially the clicks will be seen as a positive, after a quarter or two of no uptick in revenue relative to ad spend, those ad budgets will start to get questioned.
Procter and Gamble is continuing to cut online ad spend due to the ineffectiveness of their ads. They're on the leading edge here, it's just up to the community to help move everyone else down that path (http://www.adweek.com/brand-marketing/when-procter-gamble-cut-200-million-in-digital-ad-spend-its-marketing-became-10-more-effective/)
For tracking, the goal is to decrease the signal to noise ratio and make the tracking databases worthless.
Why can’t we do the same with auto play, gifs, and everything else that fights for our attention?
My gueriila suggestion: just start clicking on all ads that annoy you. Make the advertisers pay. And mess up their “valuable” tracking metrics in the process.
-Chris
'nuff said.
-Chris
This is not about volitility or any other type of consumer protection. This is about protecting their points programs.
Basically, people were buying crypto coins on their credit cards, getting the points for the purchase, then immediately selling the coins and paying off the card. The points more than made up for the transaction costs.
Since all these cards were being paid off before any interest accrued, the credit card companies were losing a lot of money on the transactions in the form of reward payouts.
The CC firms will never admit this publicly, but given the number of people I knew who were doing this, I’d bet this is the real reason they’ll disallowing the purchases.
(reposting as myself to get some mod points)
My kids can message all they want with people I approve of. How do I enforce this? All messaging happens through my phone while I'm present (Facetime with relatives, text messages to plan playdates, photo sharing, etc).
Anything else that tries to set up messaging accounts for my kids? If messaging/chat features can't be disabled, my kids can't use it.
-Chris
Minor correction: cargo holds are pressurized at cabin pressure. The whole plane is a giant cylinder. Having two diffent pressure zones or even just an oddly shaped cabin is more dangerous than just pressurizing the whole thing.
Remember: animals travel in the cargo hold all the time.
-Chris
And I should reply directly the the parent: statistically stringing together text is also not new. We just have better collections to start training algorithms with.
Handwriting, speech recognition, and image processing along with their machine learning foundations do not impress the older /. crowd because they are not new technologies.
Dragon has been doing speech recognition better than Siri for almost 20 years. Simple command-based systems that only recognize a few words have been around longer than that.
Handwriting recognition for constrained tasks is also not new. The US Postal Services has had zip-code OCR systems since the 1980s.
Feature detection in images is not new, either. The only thing that's really changed there is we have the processing power to do it at scale.
Going beyond the applications, all the modern "AI" systems are simply classifiers on steroids. Processing power and great storage capacity allows us work on larger data sets, but in the end, we're just creating complex hyper-planes to bin data in one bucket or another.
Machine learning algorithms are great tools and it's great that we have the compute resources to really leverage them, but there's nothing really new that wasn't obvious 30 years ago. The only question was when we'd have the compute power to start doing the cool things we knew they could be used for.
(ok, I'll give a little credit to the deep learning researchers for bringing neural nets back into vogue, since those were written off 30 years ago during the AI winter, but they're still just classifiers from the mathematical perspective).
-Chris
Was just coming to post the reasons to go to the Alamo as really the only reasons to go to a theater.
You left off my personal favorite: fried pickles
(in general, their food sucks, but there are a few bright spots on the menu).
Fried pickles, beer, and the great pre-preview content (including the PSAs) are what keep me going there.
Indeed, and it didn't need that fsck'ing HARDWARE cursor emulation that the PC needed, either!
(Yes, Hercules, CGA, EGA and VGA had the text-mode cursor in hardware (including the blinking)). VGA (and maybe EGA, I forget) also had a single "sprite" for the "hardware" cursor.
I was coming here to make exactly this point. Cursors used to be hardware sprites that required no additional CPU cycles. At some point windowing systems took over the task of rendering the cursor, but still typically used XOR'd sprites to keep things fast and efficient. Then they started using GPU-optimized code and software CPU emulators as a fallback and things went downhill from there...
I still keep my 80s-era graphics programming books on my shelf as a reminder of how to do fast graphics when all you have is the ability to draw pixels... ;)
-Chris
What I've always found funny about this interview process is that the assumption is always that the interviewer knows the correct answer(s) to the question. It's painful when they don't.
Years ago I interviewed at Google and was asked a question about bit counting (some variation on "given a bit vector, wat's the fastest way to count the number of 1s?"). I quickly answered, "well, if your processor has a population count instruction, stream your vector through that and accumulate the result in a register". Having just evaluated bit counting methods as part of my Ph.D. dissertation, I knew this was the fastest way to do it, assuming the instruction was available (it's not on x86, but is on Power/VMX and most DSPs support it as well).
After I got a blank stare back from the interviewee, I said, "Oh, you were looking for the lookup table answer". We could have left it at that, but he went on to explain using some very convoluted logic how the lookup table would actually be faster than using a dedicated instruction and that my answer was clearly wrong. I mentioned a little bit about the number of cycles required for each approach, but he had none of it. In his mind, I had the wrong answer, even though my second answer was what he was looking for.
It was at that moment that I realized Google was not going to be a good place to work.
-Chris
As others are saying, don't live in the Bay Area if you can't afford it. But, if you want housing that's affordable and not too far away, it's not impossible...
There's the whole Central Valley within driving distance of the Bay Area. Sure, a 1-2 hour commute isn't ideal, but with a flexible work schedule and work-from-home options some days the of the week, it's totally doable. You can get a nice house with a pool in a small CV town for less than $250k. Hell, in New England "bedroom communities" are all over the place and feature similar price differences/commute times. (you can even throw in a few nights each month for a hotel and still come out ahead)
Fwiw, I grew up in the Central Valley. Day in and day out it's really no different than living anywhere else - you eat, sleep, and work, lather/rinse/repeat. Oh, and you're much closer to the Sierras than you are in the Bay Area, if mountains are your thing. An hour and half to the slopes is much nicer than the 6-12 hours it takes to get to Tahoe on the weekends from the Bay Area.
-Chris
Of course, since this is in mail.app, which I use constantly, this is the first I've heard about it.
I wonder how many great features in Apple products people miss simply because Apple refuses to provide sensible documentation and instead relies on users to "discover" features organically or via message boards.
-Chris
So, at the more hardcore geek conference (Supercomputing comes to mind), there has never really been an issue with booth babes for a simple reason: geeks are scared to talk to them. Every now and then a company will hire one, only to see a nice exclusion zone form around their booth. Sure, sales guys from other booths will stop by, but none of the intended audience will risk talking with an attractive female.
There's an old saying in the visualization community: "3D is just bad 2D"
-Chris