Other optical sensors have proven wildly inaccurate when people use them for exercising or just wearing them all day. Maybe if the watch waits until you are stationary and not moving much it could get a better reading, but realistically trying to measure blood flow accurately using through-the-skin optics is never going to be very good.
On the upside it could probably detect if you'd had a heart attack or stroke and fallen down dead.
Now you may start shouting "All of this is ungodly! All of you are impure!" upon reading this and chant the operator precedence mantra from The Bible until you calm down and you'd be totally justified in doing so. Still it works.
Complains that SIMD isn't implemented in Chrome 62, and unfortunately it's not implemented in Firefox 57 either.
So the answer is probably no, not yet.
Last time I used SIMD I used the Intel intrinsics in C++. Which are OK, but Visual Studio's C compiler didn't do a particularly good job at generating efficient assembler. Mind you, a vectorizable algorithm will still run faster using the intrinsics than not using them, and the compilers get better with each release. And of course if you use the intrinsics wrapped in a Float32xN class instead of hand rolled assembler it's pretty easy to support other SIMD instruction sets. E.g. 128 bit SSE to AVX512 you just need to fill in the operators with the appropriate intrinsic function and then tell your code to use Float32x16 instead of Float32x4. With hand rolled assembler you'd have to rewrite everything. So they do make sense. AVX512 has gather too -
Silly business owner anyway. Who cares how far away the van is if it doesn't have stock of what you want, the warehouse location (or equivalent depending on the goods) of where the goods are actually coming from is far more informant to a customer.
The business in question was fixing or replacing equipment. So if someone calls and they've not busy they'll arrive in 15-30 minutes because the area they operate in is pretty small and all the places in it are within 15-30 minutes. If not they'll visit as soon as someone is available.
As far as fixing stuff goes they've got a couple of vans with a stock of the most common parts in the them, so there's a fair chance they can do it one visit. As far as replacing stuff goes, they'll need to order the parts and come round when they're in stock. They probably need a visit anyway to see how much space they have to do an installation and see if they should fix what's there or replace it.
So "15 minutes away" is a bit of an exaggeration but not that much of one. What's funny is that it's not that hard to do with geolocation and Google Maps. You just geolocate the customer, work out which map square they're in, pick a random point inside it and snap it to the road and say 'that's our nearest van to you'. It'd have nothing to do with where the nearest van was, but the distance between you and the point is accurate, give or take a bit of marketing spin.
It's kind of a hack, but it gets the job done. Basically you have types like float32x, int32x4 etc. There are methods to add and subtract them using non vector code but if the hardware supports them using vector instructions, those vector instructions get substituted for the non vector code.
Funny thing is I know a business in the UK. They wanted a realtime map on their webpage that shows how far away their nearest delivery van is. The conversation with the website developers went like this
Business Owner : Just say the van is fifteen minutes away from wherever they are. That way people are more likely to order. Developers : But how can that work? That means we need to fake a location and also that two users using the site in different places will see different locations for the van Business Owner : Just say the van is fifteen minutes away from wherever they are. That way people are more likely to order. Developers : How about we put a GPS in the van and show that? Business Owner : Just say the van is fifteen minutes away from wherever they are. That way people are more likely to order. Developers : How about we put a fake van at the midpoint of the locations of all the users currently asking. That way it will be consistent if multiple users in different locations compared notes. Business Owner : Just say the van is fifteen minutes away from wherever they are. That way people are more likely to order. Developers : How about we have multiple fake vans patrolling so that any one user is guaranteed to have one within fifteen minutes of their location. About fifteen should do it for the size of area you cover. Business Owner : Just say the van is fifteen minutes away from wherever they are. That way people are more likely to order.
And it went on in that general vein ad infinitum. I don't think the 'where's the nearest van?' realtime feature ever actually got implemented.
Some people are both simultaneously too stupid and too smart for this world.
Yeah, I changed the variable name when I was bowdlerizing the code. And broke it, because I'm sloppy.
Well but the deeper point is that Perl - at least the version I was using then - didn't properly support UTF-8. length($delim) returned the number of bytes like strlen in C by default.
Meanwhile trying the same thing in Python. All the dashes in my code at EN DASHes which Slashdot won't let me post.
foo='-' print ( len(foo) )
prints 1 as even if foo is a EN DASH.
Mind you, just now I tried this
use utf8;
my $d = '-'; my $d_len = length($d); print $d_len; print $d
Without the use utf8; line I get 3 when $hyphen is a EN DASH With it I get 1. It still complains about "Wide character in print".
A bit of Googling turns up binmode as the way around this. I.e. to get rid of the complaint I can do
use utf8; binmode STDOUT, ":utf8";
my $d = '-'; my $d_len = length($d); print $d_len; print $d
I.e. Perl 5 is a bit clunky for this stuff because by default it runs in a mode which isn't UTF-8 aware probably for compatibility reasons. You need to turn on support for utf8 strings in the string functions and utf strings in stdout and probably in other places too.
You can see how non automatic it is because slashdot, which is written in Perl, is pretty much guaranteed to corrupt unicode characters in comments even though slashdot is is utf8.
Of course Python doesn't have these issues. Then again Python doesn't really care about back compatibility - they probably turned on UTF aware mode at some point and just told people to suck it up if that broke their scripts.
Still these days I prefer Python to Perl. And not just or even primarily the UTF stuff - there's loads of useful things available to Python code for data manipulation like numpy I really like. Could I do the same thing in Perl? Yeah, I'm sure I could. It'd just take longer.
At least it seems like it's getting somewhat decent Unicode support. I remember writing a Perl script to strip out stuff from something because reasons.
my $hyphen = '-'; my $delim_len = length($hyphen); my $index=rindex($foo, $delim); if ( $index != -1 ) ...
Why the awkward $delim_len? Because if the hyphen is a U+ 2013 EN DASH the length is 2. If it's a regular ASCII hyphen it's 1. And the only reason I'm stripping them out is because they cause problems later.
You expect this sort of thing in C because strlen returns a length in bytes because it was invented in the 1970s, but it kind of sucks for a scripting language.
tl;dr - nothing particularly interesting. Archive links because FB will probably pull his account to protect us all from reading his rather empty, but basically harmless rants.
Just tell him the mass limit, any limitations on what can and can't be launched and, the odds of it reaching Mars and the odds of it blowing up on impact. And let him think for a week or so.
Elon Musk is basically an IRL Tony Stark but Robert Zubrin is more like Hans Zarkov. He'll come up with something clever.
Hey! Cut Rei some slack. It's not about karmawhoring. Colonising the solar system works a bit like fairies in Peter Pan. If we all believe it will happen, but all it takes is a few snarky comments on Slashdot and, like Tinkerbell, Elon Musk will wither away and so will the dream of human spaceflight
In which case why not launch something useful? I realise he doesn't want to risk a commercial satellite but he could launch fuel. Or he could get a bunch of cube sats and put those into orbit, assuming the launch works.
Or he could talk to Robert Zubrin and ask him what's the best thing to launch to keep in reserve for a future Mars Direct like program given a) it's a free launch but b) there's a high risk of mission failure
I bet if you asked him he could come up with something that'd be handy to have in Mars orbit and also wouldn't be a disaster if it got blown up on launch.
.... God's way of telling you you have too much money.
No, using one of your company's $90m rocket to launch one of your other company's $100k dollar sports cars into orbit around another planet is a God's way of telling you you have too much money.
"The reason that the rich were so rich, Vimes reasoned, was because they managed to spend less money.
Take boots, for example. He earned thirty-eight dollars a month plus allowances. A really good pair of leather boots cost fifty dollars. But an affordable pair of boots, which were sort of OK for a season or two and then leaked like hell when the cardboard gave out, cost about ten dollars. Those were the kind of boots Vimes always bought, and wore until the soles were so thin that he could tell where he was in Ankh-Morpork on a foggy night by the feel of the cobbles.
But the thing was that good boots lasted for years and years. A man who could afford fifty dollars had a pair of boots that'd still be keeping his feet dry in ten years' time, while the poor man who could only afford cheap boots would have spent a hundred dollars on boots in the same time and would still have wet feet.
This was the Captain Samuel Vimes 'Boots' theory of socioeconomic unfairness."
Thus demonstrating it's a bad idea for the regulators to force a company to make a product that the company think isn't commercially viable.
Look at the EV-1. The consumers who tried it liked it but not enough people were willing to pay a price that made it viable. GM were forced to produce it so they had a lease scheme. But that wasn't viable for them either. As soon as they could kill it they did.
Well that's the thing isn't it? If you look around you on the subway everyone is gazing at that their phones. Some are playing games, some are reading, some are chatting to their buddies, some are browsing the 'Net.
I'd say some of these - reading and chatting - are more worthwhile than others. And what if they're playing a game I approve of like Monument Valley or spurring an impassioned debate on slashdot by acting as a troll, advocatus diaboli or social gadfly? Is that worthwhile?
Saying "everyone around me is a dumb smartphone zombie" feels good, but how do you know they're not using their phones for something worthwhile.
However, there's no reason in principle that the same entity that controls the network also needs to know *who* corresponds to each IP address, or *whom* you're communicating with.
How else would they protect us from the tairists and the pedofiddlers?
It's kind of disingenuous to require scientific precision from journalism. That's not its role. It's for promoting awareness. The actual arguments and data have been around a while.
As a friend of mine, who was an actual peer reviewed published scientist, observed - '"Meta Studies are not science". I.e. as soon as you get someone doing a metastudy they can use ad hoc criteria to decide which paper they include and which one they exclude. In the case of "Skeptical Science" John Cook is not a scientist, he's an environmental activist.
I think you can make a case for some sort of absolute morality. E.g. Ben Shapiro made the case for the Republicans ditching Roy Moore, when moral relativists like me would say 'Accusations unproven in court. Fuck the Democrats'.
Well the odd thing is, as much as I disagree with the religious basis of Shapiro's morality you can see if it works. E.g. right now the GOP could offer a trade of Moore for Franken. If they'd have dumped Moore earlier, as Shapiro points out, they could have run someone else as a write in instead. Hell have Sessions stand down as AG and be nominated for Moore's seat - everyone knows he hates Trump and the feeling is mutual.
Right now sticking with Moore may lose them the contest. In fucking Alabama. Maybe Shapiro was right, even though I don't believe in the axioms of his moral system.
And his point is that if the GOP stick to principle they'll outperform their unprincipled opponents.
It's almost like my supposedly rational moral system performs worse than Shapiro's supposedly irrational one. Going back to Nietzsche you can make an argument that when he was saying 'God is dead' he wasn't saying this was a good thing.
Slashdot Mirror
Other optical sensors have proven wildly inaccurate when people use them for exercising or just wearing them all day. Maybe if the watch waits until you are stationary and not moving much it could get a better reading, but realistically trying to measure blood flow accurately using through-the-skin optics is never going to be very good.
On the upside it could probably detect if you'd had a heart attack or stroke and fallen down dead.
No they cannot.
You can take code with pointers and compile it to WebAssembly
https://stackoverflow.com/ques...
Now you may start shouting "All of this is ungodly! All of you are impure!" upon reading this and chant the operator precedence mantra from The Bible until you calm down and you'd be totally justified in doing so. Still it works.
They're working on it it Chromium
https://bugs.chromium.org/p/v8...
However this page
https://peterjensen.github.io/...
Complains that SIMD isn't implemented in Chrome 62, and unfortunately it's not implemented in Firefox 57 either.
So the answer is probably no, not yet.
Last time I used SIMD I used the Intel intrinsics in C++. Which are OK, but Visual Studio's C compiler didn't do a particularly good job at generating efficient assembler. Mind you, a vectorizable algorithm will still run faster using the intrinsics than not using them, and the compilers get better with each release. And of course if you use the intrinsics wrapped in a Float32xN class instead of hand rolled assembler it's pretty easy to support other SIMD instruction sets. E.g. 128 bit SSE to AVX512 you just need to fill in the operators with the appropriate intrinsic function and then tell your code to use Float32x16 instead of Float32x4. With hand rolled assembler you'd have to rewrite everything. So they do make sense. AVX512 has gather too -
https://en.wikipedia.org/wiki/...
And if you need to support NEON you can use the NEON intrinsics.
Silly business owner anyway. Who cares how far away the van is if it doesn't have stock of what you want, the warehouse location (or equivalent depending on the goods) of where the goods are actually coming from is far more informant to a customer.
The business in question was fixing or replacing equipment. So if someone calls and they've not busy they'll arrive in 15-30 minutes because the area they operate in is pretty small and all the places in it are within 15-30 minutes. If not they'll visit as soon as someone is available.
As far as fixing stuff goes they've got a couple of vans with a stock of the most common parts in the them, so there's a fair chance they can do it one visit. As far as replacing stuff goes, they'll need to order the parts and come round when they're in stock. They probably need a visit anyway to see how much space they have to do an installation and see if they should fix what's there or replace it.
So "15 minutes away" is a bit of an exaggeration but not that much of one. What's funny is that it's not that hard to do with geolocation and Google Maps. You just geolocate the customer, work out which map square they're in, pick a random point inside it and snap it to the road and say 'that's our nearest van to you'. It'd have nothing to do with where the nearest van was, but the distance between you and the point is accurate, give or take a bit of marketing spin.
It's kind of a hack, but it gets the job done. Basically you have types like float32x, int32x4 etc. There are methods to add and subtract them using non vector code but if the hardware supports them using vector instructions, those vector instructions get substituted for the non vector code.
https://github.com/tc39/ecmasc...
Funny thing is I know a business in the UK. They wanted a realtime map on their webpage that shows how far away their nearest delivery van is. The conversation with the website developers went like this
Business Owner : Just say the van is fifteen minutes away from wherever they are. That way people are more likely to order.
Developers : But how can that work? That means we need to fake a location and also that two users using the site in different places will see different locations for the van
Business Owner : Just say the van is fifteen minutes away from wherever they are. That way people are more likely to order.
Developers : How about we put a GPS in the van and show that?
Business Owner : Just say the van is fifteen minutes away from wherever they are. That way people are more likely to order.
Developers : How about we put a fake van at the midpoint of the locations of all the users currently asking. That way it will be consistent if multiple users in different locations compared notes.
Business Owner : Just say the van is fifteen minutes away from wherever they are. That way people are more likely to order.
Developers : How about we have multiple fake vans patrolling so that any one user is guaranteed to have one within fifteen minutes of their location. About fifteen should do it for the size of area you cover.
Business Owner : Just say the van is fifteen minutes away from wherever they are. That way people are more likely to order.
And it went on in that general vein ad infinitum. I don't think the 'where's the nearest van?' realtime feature ever actually got implemented.
Some people are both simultaneously too stupid and too smart for this world.
Just waiting for the cdreimer post extolling the virtues of Cheesy Pizza.
Electric ovens can't melt cheese beams! Wake up sheeple!
But then it isn't a hyphen.
Yeah, I changed the variable name when I was bowdlerizing the code. And broke it, because I'm sloppy.
Well but the deeper point is that Perl - at least the version I was using then - didn't properly support UTF-8. length($delim) returned the number of bytes like strlen in C by default.
Meanwhile trying the same thing in Python. All the dashes in my code at EN DASHes which Slashdot won't let me post.
prints 1 as even if foo is a EN DASH.
Mind you, just now I tried this
Without the use utf8; line I get 3 when $hyphen is a EN DASH With it I get 1. It still complains about "Wide character in print".
A bit of Googling turns up binmode as the way around this. I.e. to get rid of the complaint I can do
I.e. Perl 5 is a bit clunky for this stuff because by default it runs in a mode which isn't UTF-8 aware probably for compatibility reasons. You need to turn on support for utf8 strings in the string functions and utf strings in stdout and probably in other places too.
You can see how non automatic it is because slashdot, which is written in Perl, is pretty much guaranteed to corrupt unicode characters in comments even though slashdot is is utf8.
Of course Python doesn't have these issues. Then again Python doesn't really care about back compatibility - they probably turned on UTF aware mode at some point and just told people to suck it up if that broke their scripts.
Still these days I prefer Python to Perl. And not just or even primarily the UTF stuff - there's loads of useful things available to Python code for data manipulation like numpy I really like. Could I do the same thing in Perl? Yeah, I'm sure I could. It'd just take longer.
At least it seems like it's getting somewhat decent Unicode support. I remember writing a Perl script to strip out stuff from something because reasons.
Why the awkward $delim_len? Because if the hyphen is a U+ 2013 EN DASH the length is 2. If it's a regular ASCII hyphen it's 1. And the only reason I'm stripping them out is because they cause problems later.
You expect this sort of thing in C because strlen returns a length in bytes because it was invented in the 1970s, but it kind of sucks for a scripting language.
CBS have declined to explain but a bit of searching turns up this
https://www.facebook.com/RedXS...
https://archive.fo/eoZiN
https://www.facebook.com/Tracy...
https://archive.fo/IcXKV
https://www.facebook.com/notes...
https://archive.fo/ywhAk
tl;dr - nothing particularly interesting. Archive links because FB will probably pull his account to protect us all from reading his rather empty, but basically harmless rants.
s/size/mass/g
Or oxygen. You could imagine a future human mission to Mars might find an oxygen tank the size of Tesla roadster handy.
Personally I'd ask Robert Zubrin for suggestions because he's been thinking about Mars missions for a really long time
https://www.youtube.com/watch?...
Just tell him the mass limit, any limitations on what can and can't be launched and, the odds of it reaching Mars and the odds of it blowing up on impact. And let him think for a week or so.
Elon Musk is basically an IRL Tony Stark but Robert Zubrin is more like Hans Zarkov. He'll come up with something clever.
Hey! Cut Rei some slack. It's not about karmawhoring. Colonising the solar system works a bit like fairies in Peter Pan. If we all believe it will happen, but all it takes is a few snarky comments on Slashdot and, like Tinkerbell, Elon Musk will wither away and so will the dream of human spaceflight
Rei is helpfully pointing this out.
Repeat after me. I do believe in Elon! I do!
In which case why not launch something useful? I realise he doesn't want to risk a commercial satellite but he could launch fuel. Or he could get a bunch of cube sats and put those into orbit, assuming the launch works.
Or he could talk to Robert Zubrin and ask him what's the best thing to launch to keep in reserve for a future Mars Direct like program given a) it's a free launch but b) there's a high risk of mission failure
Zubrin gave a memorable presentation here
The Case For Mars | Robert Zubrin
I bet if you asked him he could come up with something that'd be handy to have in Mars orbit and also wouldn't be a disaster if it got blown up on launch.
Would progress happen any faster if we weren't allowed to make snarky comments about St Elon of Musk here?
.... God's way of telling you you have too much money.
No, using one of your company's $90m rocket to launch one of your other company's $100k dollar sports cars into orbit around another planet is a God's way of telling you you have too much money.
https://www.goodreads.com/quot...
"The reason that the rich were so rich, Vimes reasoned, was because they managed to spend less money.
Take boots, for example. He earned thirty-eight dollars a month plus allowances. A really good pair of leather boots cost fifty dollars. But an affordable pair of boots, which were sort of OK for a season or two and then leaked like hell when the cardboard gave out, cost about ten dollars. Those were the kind of boots Vimes always bought, and wore until the soles were so thin that he could tell where he was in Ankh-Morpork on a foggy night by the feel of the cobbles.
But the thing was that good boots lasted for years and years. A man who could afford fifty dollars had a pair of boots that'd still be keeping his feet dry in ten years' time, while the poor man who could only afford cheap boots would have spent a hundred dollars on boots in the same time and would still have wet feet.
This was the Captain Samuel Vimes 'Boots' theory of socioeconomic unfairness."
Thus demonstrating it's a bad idea for the regulators to force a company to make a product that the company think isn't commercially viable.
Look at the EV-1. The consumers who tried it liked it but not enough people were willing to pay a price that made it viable. GM were forced to produce it so they had a lease scheme. But that wasn't viable for them either. As soon as they could kill it they did.
Well that's the thing isn't it? If you look around you on the subway everyone is gazing at that their phones. Some are playing games, some are reading, some are chatting to their buddies, some are browsing the 'Net.
I'd say some of these - reading and chatting - are more worthwhile than others. And what if they're playing a game I approve of like Monument Valley or spurring an impassioned debate on slashdot by acting as a troll, advocatus diaboli or social gadfly? Is that worthwhile?
Saying "everyone around me is a dumb smartphone zombie" feels good, but how do you know they're not using their phones for something worthwhile.
The dark inside of the cubicle is a pathway to many abilities, some consider to be ... unnatural
However, there's no reason in principle that the same entity that controls the network also needs to know *who* corresponds to each IP address, or *whom* you're communicating with.
How else would they protect us from the tairists and the pedofiddlers?
It's kind of disingenuous to require scientific precision from journalism. That's not its role. It's for promoting awareness. The actual arguments and data have been around a while.
Well people disagree about that
https://www.thegwpf.org/matt-r...
If you are serious about learning more, I highly recommend http://skepticalscience.com/
As a friend of mine, who was an actual peer reviewed published scientist, observed - '"Meta Studies are not science". I.e. as soon as you get someone doing a metastudy they can use ad hoc criteria to decide which paper they include and which one they exclude. In the case of "Skeptical Science" John Cook is not a scientist, he's an environmental activist.
https://en.wikipedia.org/wiki/...
So his summaries of science include a big chunk of editorial bias. He's free to do that, but so is Matt Ridley. And both have the same credibility.
I think you can make a case for some sort of absolute morality. E.g. Ben Shapiro made the case for the Republicans ditching Roy Moore, when moral relativists like me would say 'Accusations unproven in court. Fuck the Democrats'.
Well the odd thing is, as much as I disagree with the religious basis of Shapiro's morality you can see if it works. E.g. right now the GOP could offer a trade of Moore for Franken. If they'd have dumped Moore earlier, as Shapiro points out, they could have run someone else as a write in instead. Hell have Sessions stand down as AG and be nominated for Moore's seat - everyone knows he hates Trump and the feeling is mutual.
Right now sticking with Moore may lose them the contest. In fucking Alabama. Maybe Shapiro was right, even though I don't believe in the axioms of his moral system.
And his point is that if the GOP stick to principle they'll outperform their unprincipled opponents.
It's almost like my supposedly rational moral system performs worse than Shapiro's supposedly irrational one. Going back to Nietzsche you can make an argument that when he was saying 'God is dead' he wasn't saying this was a good thing.