Yes. It's big news because it uses USB class 3 audio, which as you can tell by the number is something completely new and original, a first.
Joking aside,this class 3 stuff actually stopped me investing in audio gear before the summer. Class compliant is now ubiqutious in audio -- even iOS supports it through the camera kit -- so pretty much everything has it. But all the press about Class 3 has talked as though nothing came before it, so has incidentally failed to discuss backwards compatibility....
No human brain could ever hope to process a grammar as "big" as UNL. And no, I've never read Wittgenstein. Although I did take that as my name for a German class once.
The first link in the summary is to Goole's blog. The first translation roolout is Chinese to English -- all such translations from now on will be carried out by our new connectionist translation overlords, and I for one welcome them. What I'm curious about is how it will handle languages than had insufficient data for old-style Google Translate's statistical translation engine. Will this do better, or will it be even more sensitive to small datasets?
Given that it's a neural network, no-one's going to know precisely what the vector encodes, or quite what it "means", or what a vector would look like for a given word. I imagine that was what the researcher meant.
Personally I would build my own system to run linux on. It isn't as hard as some think it is. There are lots of tutorials on it and odds are there is a nerd some where close by that will help you.
Yeah, but I don't have a 3D printer to make the case out of, and if I make it out of wood in my garage it'll be a bit too heavy, not to mention hard to ool. (I'm sure you understand what form factor an "UltraBook" is.)
The Mac isn't an "appliance", it's a "computer". Apple's appliances are iOS devices, and they go out of their way to lock them down. But Microsoft calling Signature Edition an "appliance" is just sophistry -- Windows is a computer operating system and offers the power user near complete control of your system, whereas iOS is deliberately feature-starved and controlled. In fact, Windows Signature Edition offers the end-user even more control by giving them an OS free of vendor crapware. It's even more "computer" than OEM Windows.
There is no "workaround" for installing Linux on a Mac. In fact, they even included a multi-boot utility (Boot Camp Assistant) with the specific goal of allowing you to install an alternative operating system to give you flexibility. I'm not a Mac user -- this is a PC. But it's dual boot because I find that useful. I have a lightweight Linux install for browsing and Python dev work, and I boot Windows when I want to use apps that otherwise wouldn't be available.
To point a finer point on this issue, I think the deficiency in the study has to do with the groups involved.
They may have found two groups who have a common goal (weight loss), but I'm willing to bet that the group who went out of their way to consult with a health counselor have FAR more motivation to achieve the goal of overall health
I think that everyone in the study was consulting with a counsellor, and that's how they got onto the study. I may be wrong, but that's what the study seems to be describing....
That's cos there's no transmission of smell in the browser. With all that fluff, a dancing llama builds up a really stinky sweat dead quick, particularly given that they mostly dance to upbeat Latin American rhythms.
It is impossible to lose weight if you eat more than you burn, even if all of those calories are "healthy".
Not necessarily true. For a while, I was on a "grazing" regime, as it had been advised both by a doctor as a method of managing my heartburn (heartburn often being triggered by filling up the stomach, or by eating on an empty stomach) and by a trainer in my local gym as a means of losing weight. The trainer's assertion was that the human body only starts to metabolise food to bodyfat when the stomach is nearly empty. If your stomach isn't empty at all during the day (you need to let it empty at night, though) there's no trigger to the body to save some of the calories for later.
Now I can't say for sure that this is true -- it could be that I only lost weight doing this because I retrained my reaction to hunger -- but I lost a surprising amount of weight surprisingly quickly.
As much as this explanation appears to make sense, like anything in a social study, the results are likely virtually meaningless at the individual level. How any one person reacts to a tracker will not be predictable.
That's an overstatement. What the results cannot do is state categorically that "you won't get fitter with a fitbit". But they're a starting point for an informed decision. When you're engaged in behaviour modification, you have to first be aware of your behaviour, and that includes being aware of tendencies that you may (or may not) have which are likely to affect your behaviour. Knowing something about the psychological effects of wearing a fitness tracker allows you to modify your use of it/attitude towards it and (hopefully) achieve better results.
Sounds like Atkins, which has worked quite well for me.
That's a large leap. The grandparent was talking about cutting out high GI foods, Atkins promotes cutting out all carbohydrate. Ketosis is a pretty extreme state to get yourself in, and there's really no need to worry about low GI carbohydrates. And if exercise is part of your weight-loss regime, you do really need to have a reasonable carbohydrate intake.
You're talking about persistent state across devices? Sorry, with sites rendering in device-dependent configurations, that sound like a really bad idea.
The two-keypress thing is actually quite consistent with other switching operations -- alt-tab, ctrl-tab, ctrl-pgup/pgdn etc. I've internalised holding down a modifier key as a context-switch trigger, and then I see the other button as cycling one way or the other.
Accesible UI design is only needed in accessible UIs. Major browsers are now configurable enough that UIs can be customised to individual needs. I have a friend who drives cars with heavily modified controls. These controls would be illegal for fully able-bodied drivers as they make the car more dangerous, but they are permitted in his car as they make it possible for him to drive without using his feet, which have very limited movement.
A one-touch-back key isn't likely to kill anyone, but it's still an error hazard.
Haven't you cave dwellers heard of mouse gestures?
The GO referred to people with accessibilty issues, and not everyone with accessibility issues can use a mouse. Many who can cannot use it with enough accuracy to use gestures. So please, don't start flinging insults at people who physically can't use a computer the same way you would.
But this way they get inertia on their side. Google rightly wants to see an end to backspace-as-back-button, and through this action they have made sure it will never again be ubiquitous. Once people get used to Alt+left arrow, we can get rid of this abomination entirely.
But, I'm also annoyed that it's gone, because I used it frequently.
Instead of them being so binary, they could have just made it a configurable option.
They should simply be promoting Alt+left arrow hard as the alternative. It has existed in all browsers since the days of Netscape Navigator, it is not shared with any other common operation, and all-in-all is very difficult to do unintentionally. There are webpages out there that switch focus away from form elements unexpectedly, and that's where backspace-as-back-button gets very dangerous.
I recently used a site where if you delete all the text in a textbox, the keyboard focus goes back to the page. Major design flaw - if you delete by holding down backspace, you're almost certainly going to end up going back a page. It took me ages to buy that ticket....
Freezing the state of a page would break the intended functionality of a lot of pages, because either you save the state before doing the onexit and onunload events, and have data that may have been invalidated by them, or you save the state after them... when the state is absolutely nothing.
Applications? HTML was never meant for "applications". It was meant for web *pages*. Websites.
And it's not even very good for those, because the guys that wrote it were still thinking of paper publishing. What was the first major use for unnecessary JS bloat? Interactive menus. Even now, the least JS you can use for a menu hierarchy is a quick change of hidden/visible attributes in an onClick attribute.
But if they'd had a <menu> tag, we could have done the whole hierarchy in nested HTML and then only needed to invoke Javascript at the last minute. And even then, if the nodes at the end of the tree hierarchy were links (as was the case in c.2000 JS menus) there would be no script needed at all -- pure HTML.
On a different note, the nesting of procedures is generally another of the bad-ideas-that-work, because they introduce unnecessary context switches. Unless you have a decent compiler, in which case it optimises smaller procedures out as macros, in which case it is better all round. One of the things that niggles at the back of my head with Python is that when I try to write neat code, I'm always aware that ever single call is a call, so there's the efficiency hit in context switches. But I do it anyway, because I want to understand my code in a week's time.
Slashdot Mirror
Yes. It's big news because it uses USB class 3 audio, which as you can tell by the number is something completely new and original, a first.
Joking aside,this class 3 stuff actually stopped me investing in audio gear before the summer. Class compliant is now ubiqutious in audio -- even iOS supports it through the camera kit -- so pretty much everything has it. But all the press about Class 3 has talked as though nothing came before it, so has incidentally failed to discuss backwards compatibility....
No human brain could ever hope to process a grammar as "big" as UNL. And no, I've never read Wittgenstein. Although I did take that as my name for a German class once.
The first link in the summary is to Goole's blog. The first translation roolout is Chinese to English -- all such translations from now on will be carried out by our new connectionist translation overlords, and I for one welcome them. What I'm curious about is how it will handle languages than had insufficient data for old-style Google Translate's statistical translation engine. Will this do better, or will it be even more sensitive to small datasets?
To be fair, symbol-processing human intelligence is a dying dream -- just one of the mistakes that Piaget made.
Given that it's a neural network, no-one's going to know precisely what the vector encodes, or quite what it "means", or what a vector would look like for a given word. I imagine that was what the researcher meant.
Personally I would build my own system to run linux on. It isn't as hard as some think it is. There are lots of tutorials on it and odds are there is a nerd some where close by that will help you.
Yeah, but I don't have a 3D printer to make the case out of, and if I make it out of wood in my garage it'll be a bit too heavy, not to mention hard to ool. (I'm sure you understand what form factor an "UltraBook" is.)
The Mac isn't an "appliance", it's a "computer". Apple's appliances are iOS devices, and they go out of their way to lock them down. But Microsoft calling Signature Edition an "appliance" is just sophistry -- Windows is a computer operating system and offers the power user near complete control of your system, whereas iOS is deliberately feature-starved and controlled. In fact, Windows Signature Edition offers the end-user even more control by giving them an OS free of vendor crapware. It's even more "computer" than OEM Windows.
There is no "workaround" for installing Linux on a Mac. In fact, they even included a multi-boot utility (Boot Camp Assistant) with the specific goal of allowing you to install an alternative operating system to give you flexibility. I'm not a Mac user -- this is a PC. But it's dual boot because I find that useful. I have a lightweight Linux install for browsing and Python dev work, and I boot Windows when I want to use apps that otherwise wouldn't be available.
To point a finer point on this issue, I think the deficiency in the study has to do with the groups involved.
They may have found two groups who have a common goal (weight loss), but I'm willing to bet that the group who went out of their way to consult with a health counselor have FAR more motivation to achieve the goal of overall health
I think that everyone in the study was consulting with a counsellor, and that's how they got onto the study. I may be wrong, but that's what the study seems to be describing....
Smells like teen spelling....
That's cos there's no transmission of smell in the browser. With all that fluff, a dancing llama builds up a really stinky sweat dead quick, particularly given that they mostly dance to upbeat Latin American rhythms.
Thanks to Brexit and the tanking exchange rate, they cost about £6 more each week than the week before...
It is impossible to lose weight if you eat more than you burn, even if all of those calories are "healthy".
Not necessarily true. For a while, I was on a "grazing" regime, as it had been advised both by a doctor as a method of managing my heartburn (heartburn often being triggered by filling up the stomach, or by eating on an empty stomach) and by a trainer in my local gym as a means of losing weight. The trainer's assertion was that the human body only starts to metabolise food to bodyfat when the stomach is nearly empty. If your stomach isn't empty at all during the day (you need to let it empty at night, though) there's no trigger to the body to save some of the calories for later.
Now I can't say for sure that this is true -- it could be that I only lost weight doing this because I retrained my reaction to hunger -- but I lost a surprising amount of weight surprisingly quickly.
As much as this explanation appears to make sense, like anything in a social study, the results are likely virtually meaningless at the individual level. How any one person reacts to a tracker will not be predictable.
That's an overstatement. What the results cannot do is state categorically that "you won't get fitter with a fitbit". But they're a starting point for an informed decision. When you're engaged in behaviour modification, you have to first be aware of your behaviour, and that includes being aware of tendencies that you may (or may not) have which are likely to affect your behaviour. Knowing something about the psychological effects of wearing a fitness tracker allows you to modify your use of it/attitude towards it and (hopefully) achieve better results.
Sounds like Atkins, which has worked quite well for me.
That's a large leap. The grandparent was talking about cutting out high GI foods, Atkins promotes cutting out all carbohydrate. Ketosis is a pretty extreme state to get yourself in, and there's really no need to worry about low GI carbohydrates. And if exercise is part of your weight-loss regime, you do really need to have a reasonable carbohydrate intake.
You're talking about persistent state across devices? Sorry, with sites rendering in device-dependent configurations, that sound like a really bad idea.
The two-keypress thing is actually quite consistent with other switching operations -- alt-tab, ctrl-tab, ctrl-pgup/pgdn etc. I've internalised holding down a modifier key as a context-switch trigger, and then I see the other button as cycling one way or the other.
A one-touch-back key isn't likely to kill anyone, but it's still an error hazard.
Haven't you cave dwellers heard of mouse gestures?
The GO referred to people with accessibilty issues, and not everyone with accessibility issues can use a mouse. Many who can cannot use it with enough accuracy to use gestures. So please, don't start flinging insults at people who physically can't use a computer the same way you would.
But this way they get inertia on their side. Google rightly wants to see an end to backspace-as-back-button, and through this action they have made sure it will never again be ubiquitous. Once people get used to Alt+left arrow, we can get rid of this abomination entirely.
But, I'm also annoyed that it's gone, because I used it frequently.
Instead of them being so binary, they could have just made it a configurable option.
They should simply be promoting Alt+left arrow hard as the alternative. It has existed in all browsers since the days of Netscape Navigator, it is not shared with any other common operation, and all-in-all is very difficult to do unintentionally. There are webpages out there that switch focus away from form elements unexpectedly, and that's where backspace-as-back-button gets very dangerous.
I recently used a site where if you delete all the text in a textbox, the keyboard focus goes back to the page. Major design flaw - if you delete by holding down backspace, you're almost certainly going to end up going back a page. It took me ages to buy that ticket....
Freezing the state of a page would break the intended functionality of a lot of pages, because either you save the state before doing the onexit and onunload events, and have data that may have been invalidated by them, or you save the state after them... when the state is absolutely nothing.
Local function call != Context switch.
Still has overhead, but nothing like a context switch.
But macros only have overhead at compile-time. I was basically lamenting the lack of macros in Python. Python is far too dynamic.
Applications? HTML was never meant for "applications". It was meant for web *pages*. Websites.
And it's not even very good for those, because the guys that wrote it were still thinking of paper publishing. What was the first major use for unnecessary JS bloat? Interactive menus. Even now, the least JS you can use for a menu hierarchy is a quick change of hidden/visible attributes in an onClick attribute.
But if they'd had a <menu> tag, we could have done the whole hierarchy in nested HTML and then only needed to invoke Javascript at the last minute. And even then, if the nodes at the end of the tree hierarchy were links (as was the case in c.2000 JS menus) there would be no script needed at all -- pure HTML.
On a different note, the nesting of procedures is generally another of the bad-ideas-that-work, because they introduce unnecessary context switches. Unless you have a decent compiler, in which case it optimises smaller procedures out as macros, in which case it is better all round. One of the things that niggles at the back of my head with Python is that when I try to write neat code, I'm always aware that ever single call is a call, so there's the efficiency hit in context switches. But I do it anyway, because I want to understand my code in a week's time.