I don't own a Windows anything, but my partner bought one because of the camera. The UI is really nice, but it's let down by the complete lack of apps. I'd buy one if it had support from more third parties, and more third parties would support it if more of them were sold. It's not clear that there's a way out of this spiral for MS.
It's not completely clear cut. It might come under tying (you're not allowed to only sell unrelated products together if doing so would distort the market), but it looks as if it is most likely to just make me buy things from someone else. Most Amazon deliveries for me arrive in 1-2 days using the free super-saver delivery option, so paying extra for the same service doesn't sound very attractive. Most DVDs are cheaper on Rakuten (formerly play.com) than Amazon anyway, but if Amazon starts responding to searches telling me that they won't sell things to me then that's a good reason to not even bother searching for things there in the future.
eliminate ANY deductions (even for mortgage interest)
Eliminating deductions for mortgage interest would likely be regressive. People with more money will set up a company to own their house and then have mortgage interest (and any maintenance) be a tax-deductable loss, just like any other business expense. If you want to make things fairer, then you should ensure that individuals get all of the same tax advantages as companies.
Are they? Next time you see something from an Amazon Marketplace Seller (most of the goods on Amazon these days), try searching for the seller's name. I've found that you can often get the same thing cheaper directly from the seller's own web site. Lots of companies have an Amazon (and eBay) presence for the exposure, but because they have to pay these sites a cut of the sales they just bump up the prices.
There's also the issue on Amazon that it's impossible to find anything if it's not something very similar to their core market (books). For example, look at projectors on Amazon - they have a load of things you can filter on but most of the metadata is wrong. Filter on native resolution of 1080p and you'll get a load of results, most of which tell you in the description that the native resolution is 480x800.
The issue with Intel in low-price market segments is that they insist on artificial segmentation. When I built my NAS, you simply could not buy an Atom motherboard with more than two SATA connections. AMD was happy to sell E-350s with as many as motherboard makers wanted to put on them. Intel has slowly relaxed this, but they still put other artificial limitations (such as no hardware virtualisation support) on their low-price chips that mean that you can't get the slow-and-cheap Intel chip that does what you want, letting AMD undercut them: you'd have to buy a much faster (and more expensive) Intel chip than you need to get the other features.
The problem is that the legislators did not really understand the problem that they were trying to solve. The law was intended to require consent if you are tracking the user for longer than the current session. That's an entirely reasonable thing to do. The implementation was a complete disaster because it conflated a mechanism that's used for tracking (and more benign uses) with the act of tracking. To give a car analogy, it's like noticing that a lot of the people who drive dangerously drive red cars and then insisting that all red cars warn you of possible danger of accidents whenever they drive near you.
How easy is it to do that? When you turn on an Android phone for the first time, it insists that you either provide an existing Google account or create a new one. You have to create a throw-away account before you can even get to the part of the system that lets you disable using the Google stuff.
Try disabling all of the Google services on a typical carrier-tied phone. You can, eventually, get something that works, but it's far more effort than 99% of consumers will go to. Now look around for phones that are pre-installed in this configuration.
It's a good thing for the EU. It's a bad thing for us, as our authoritarian MPs will no longer be slapped down by the EU when they decide to go completely crazy.
It's a general problem with police forces in general. A police force can only function effectively if it has the consent and support of the population. To do this, it has to be seen as being on the side of the majority of the population. When you pass laws that criminalise the majority and when you cut funding for police programs that visibly assist the community, then this breaks down.
Mine's a late 2013 model and I'm a bit sad that it's coming to the end of its warranty period and there isn't something noticeably faster. The upgrade from the Core 2 to the Sandy Bridge i7 was a big jump (especially as I got an SSD at the same time). From Sandy Bridge to Haswell was noticeable, but not huge (though moving to the 1TB SSD meant that I didn't have to be too careful about space on the laptop). Skylake looks like another incremental jump.
Aside from what the other posters have said, text rendering is actually one of the most processor-intensive tasks that a typical desktop does. Each codepoint has to be converted to one or more glyphs. These glyphs are sequences of bezier curves that are rendered to raster images (which are typically cached). Next, you need some fairly complex calculations to work out the spacing between glyphs, which is starts as a fixed advance and is then subtly tweaked based on pixel alignment and shape of the next character. Now you have a set of glyph runs, but you want to render a paragraph of text, so you need to work out where to break the lines. If you're something that sucks less than MS Word, you then use a fairly simple dynamic programming algorithm to work out the place to break the lines for optimum readability, otherwise you use a greedy strategy. This is fairly easy in a rectangle, though gets more complex if there's a background. Now you know where the glyphs need to go, and all that's left is to alpha-blend them with the background (remember, antialiasing needs an alpha channel and sub-pixel AA means that each rasterised glyph will be in three colours). This last step is typically offloaded to the GPU, because the CPU hit of just that part is quite noticeable.
And then, if it's a web page, something tweaks the DOM, or a new CSS file finishes downloading, and you need to do the whole thing again.
I miss the drawing app on the Apple Newton. It could recognise a few shapes (square, triangle, circle, maybe a few others) and it could detect when lines were meant to be straight (or orthogonal) and spot arrows. You'd sketch a diagram and get a nice vector drawing out at the end. I'd love to have something similar on a tablet today. Oh, the Newton could also recognise some people's handwriting, but not mine (must humans struggle with mine too).
When I was an undergrad (back before most students could afford laptops - I think only two people in my class had them and they were both mature students), I found that taking hand-written notes and then typing them up was best. Taking notes didn't seem to do much, but transcribing them and expanding the short-form notes into something coherent was a good way of lodging the knowledge in my brain.
I don't get the coffee from a bag, I have a hopper on top of my grinder that stores about a week's worth of beans and grinds one cafetiere's worth at the touch of a button. Ground coffee loses its flavour very quickly. The main problem (ignoring the environmental aspects) with the Keurig machines is that the packaging is very expensive. To keep the costs down, they use cheap beans. For about a quarter of the price of a pod, you can buy an equivalent quantity of far higher quality beans. There's a similar problem with decaf: the process of decaffeination adds a noticeable cost, but companies don't want to sell decaf for more than the cost of normal coffee and so they use much cheaper beans for the decaf.
If you're going to spend a lot of money on coffee, you may as well drink decent coffee. Spending a lot of money on mediocre coffee seems silly.
I'm not from the US, so I can't comment on their instant coffee (it's not something that I've sought out to try when I've visited). Apparently the instant coffee that we get in the UK is pretty good by the standards of instant coffee, but it's still crap in comparison to the real thing.
A modern effects-heavy movie made with thousands of computers doing the rendering? Wow, that's totally unlike every other modern effects-heavy movie. Ever wondered why there are so many Dell computers in action movies with closeups of the Dell logo? Dell either loans, or gives a very steep discount on, the computers for the renderfarms in exchange for product placement.
It's not that bad - it's far better than instant coffee, which would merit that description. It's coffee, it's just not good coffee, though it's better than the over-roast crap you get in places like Starbucks that consider bitterness to be the only flavour that coffee should have.
Neither will be matched by '*.*'. Hidden files, . and.. would be matched by '.*', but a '*.*' glob pattern requires characters before the dot. A '.*' pattern is particularly dangerous for this reason - it looks like it will only match hidden files, but it will also match the current directory and the parent directory.
so gave a sales person root access who promptly ran rm -rf/*.*.
Running rm -rf/*.*. (or even rm -rf/*.*) is unlikely to cause much damage on a *NIX system as it's very unusual to have any files with dots in their names in the root directory and even more unlikely to have any directories that have dots in their names.
The very idea of a cancer cure is idiotic as it is a wide spectrum of different kinds of cellular malfunction
The treatment in TFA (assuming that TFA is about the same treatment that was in the news a couple of days ago) works by allowing the body's natural defences against cancer to work properly. Cells become cancerous all of the time but they're almost always killed immediately by your autoimmune response. Some mutate in a way that prevents the immune system recognising them as broken and can then spread quite quickly. This treatment allows the immune system to recognise the cancer cells, so is likely to work on most, if not all, forms of cancer. The catch is that it requires a specific implementation for each patient.
Swipe left - message. Swipe right - call. Press and hold - for menu. Not hard to learn, very comfortable to use.
How do you learn it? Discoverability is a core concept in UI design. It's one of the reasons that the menu bar, in spite of its other failings, is still a pretty good UI model: everything that you can do is visible in the menu bar and a user can explore the functionality of a new application by browsing the menu. Gestures tend to be terrible for discoverability. Do I need to go and read some documentation to discover that I can swipe those things (or read a pop-up hint that either pops up once and then I forget the gesture, or pops up repeatedly and annoys me)? Or do I just gesture at every widget and eventually find the ones where it works? You can bet, given Google's horrendous track record for HCI, that they aren't going to make swipeable controls visually distinctive in any way.
Slashdot Mirror
I don't own a Windows anything, but my partner bought one because of the camera. The UI is really nice, but it's let down by the complete lack of apps. I'd buy one if it had support from more third parties, and more third parties would support it if more of them were sold. It's not clear that there's a way out of this spiral for MS.
It's not completely clear cut. It might come under tying (you're not allowed to only sell unrelated products together if doing so would distort the market), but it looks as if it is most likely to just make me buy things from someone else. Most Amazon deliveries for me arrive in 1-2 days using the free super-saver delivery option, so paying extra for the same service doesn't sound very attractive. Most DVDs are cheaper on Rakuten (formerly play.com) than Amazon anyway, but if Amazon starts responding to searches telling me that they won't sell things to me then that's a good reason to not even bother searching for things there in the future.
eliminate ANY deductions (even for mortgage interest)
Eliminating deductions for mortgage interest would likely be regressive. People with more money will set up a company to own their house and then have mortgage interest (and any maintenance) be a tax-deductable loss, just like any other business expense. If you want to make things fairer, then you should ensure that individuals get all of the same tax advantages as companies.
Are they? Next time you see something from an Amazon Marketplace Seller (most of the goods on Amazon these days), try searching for the seller's name. I've found that you can often get the same thing cheaper directly from the seller's own web site. Lots of companies have an Amazon (and eBay) presence for the exposure, but because they have to pay these sites a cut of the sales they just bump up the prices.
There's also the issue on Amazon that it's impossible to find anything if it's not something very similar to their core market (books). For example, look at projectors on Amazon - they have a load of things you can filter on but most of the metadata is wrong. Filter on native resolution of 1080p and you'll get a load of results, most of which tell you in the description that the native resolution is 480x800.
The issue with Intel in low-price market segments is that they insist on artificial segmentation. When I built my NAS, you simply could not buy an Atom motherboard with more than two SATA connections. AMD was happy to sell E-350s with as many as motherboard makers wanted to put on them. Intel has slowly relaxed this, but they still put other artificial limitations (such as no hardware virtualisation support) on their low-price chips that mean that you can't get the slow-and-cheap Intel chip that does what you want, letting AMD undercut them: you'd have to buy a much faster (and more expensive) Intel chip than you need to get the other features.
The problem is that the legislators did not really understand the problem that they were trying to solve. The law was intended to require consent if you are tracking the user for longer than the current session. That's an entirely reasonable thing to do. The implementation was a complete disaster because it conflated a mechanism that's used for tracking (and more benign uses) with the act of tracking. To give a car analogy, it's like noticing that a lot of the people who drive dangerously drive red cars and then insisting that all red cars warn you of possible danger of accidents whenever they drive near you.
How easy is it to do that? When you turn on an Android phone for the first time, it insists that you either provide an existing Google account or create a new one. You have to create a throw-away account before you can even get to the part of the system that lets you disable using the Google stuff.
Try disabling all of the Google services on a typical carrier-tied phone. You can, eventually, get something that works, but it's far more effort than 99% of consumers will go to. Now look around for phones that are pre-installed in this configuration.
It's a good thing for the EU. It's a bad thing for us, as our authoritarian MPs will no longer be slapped down by the EU when they decide to go completely crazy.
It's a general problem with police forces in general. A police force can only function effectively if it has the consent and support of the population. To do this, it has to be seen as being on the side of the majority of the population. When you pass laws that criminalise the majority and when you cut funding for police programs that visibly assist the community, then this breaks down.
$499 of hardware? The list price of the processors is $281 - that doesn't leave much for the motherboard, screen, ram and SSD.
Mine's a late 2013 model and I'm a bit sad that it's coming to the end of its warranty period and there isn't something noticeably faster. The upgrade from the Core 2 to the Sandy Bridge i7 was a big jump (especially as I got an SSD at the same time). From Sandy Bridge to Haswell was noticeable, but not huge (though moving to the 1TB SSD meant that I didn't have to be too careful about space on the laptop). Skylake looks like another incremental jump.
Aside from what the other posters have said, text rendering is actually one of the most processor-intensive tasks that a typical desktop does. Each codepoint has to be converted to one or more glyphs. These glyphs are sequences of bezier curves that are rendered to raster images (which are typically cached). Next, you need some fairly complex calculations to work out the spacing between glyphs, which is starts as a fixed advance and is then subtly tweaked based on pixel alignment and shape of the next character. Now you have a set of glyph runs, but you want to render a paragraph of text, so you need to work out where to break the lines. If you're something that sucks less than MS Word, you then use a fairly simple dynamic programming algorithm to work out the place to break the lines for optimum readability, otherwise you use a greedy strategy. This is fairly easy in a rectangle, though gets more complex if there's a background. Now you know where the glyphs need to go, and all that's left is to alpha-blend them with the background (remember, antialiasing needs an alpha channel and sub-pixel AA means that each rasterised glyph will be in three colours). This last step is typically offloaded to the GPU, because the CPU hit of just that part is quite noticeable.
And then, if it's a web page, something tweaks the DOM, or a new CSS file finishes downloading, and you need to do the whole thing again.
I miss the drawing app on the Apple Newton. It could recognise a few shapes (square, triangle, circle, maybe a few others) and it could detect when lines were meant to be straight (or orthogonal) and spot arrows. You'd sketch a diagram and get a nice vector drawing out at the end. I'd love to have something similar on a tablet today. Oh, the Newton could also recognise some people's handwriting, but not mine (must humans struggle with mine too).
When I was an undergrad (back before most students could afford laptops - I think only two people in my class had them and they were both mature students), I found that taking hand-written notes and then typing them up was best. Taking notes didn't seem to do much, but transcribing them and expanding the short-form notes into something coherent was a good way of lodging the knowledge in my brain.
The version of bash shipped is the last GPLv2 version (plus patches). Apple doesn't ship any GPLv3 code.
I don't get the coffee from a bag, I have a hopper on top of my grinder that stores about a week's worth of beans and grinds one cafetiere's worth at the touch of a button. Ground coffee loses its flavour very quickly. The main problem (ignoring the environmental aspects) with the Keurig machines is that the packaging is very expensive. To keep the costs down, they use cheap beans. For about a quarter of the price of a pod, you can buy an equivalent quantity of far higher quality beans. There's a similar problem with decaf: the process of decaffeination adds a noticeable cost, but companies don't want to sell decaf for more than the cost of normal coffee and so they use much cheaper beans for the decaf.
If you're going to spend a lot of money on coffee, you may as well drink decent coffee. Spending a lot of money on mediocre coffee seems silly.
I'm not from the US, so I can't comment on their instant coffee (it's not something that I've sought out to try when I've visited). Apparently the instant coffee that we get in the UK is pretty good by the standards of instant coffee, but it's still crap in comparison to the real thing.
A modern effects-heavy movie made with thousands of computers doing the rendering? Wow, that's totally unlike every other modern effects-heavy movie. Ever wondered why there are so many Dell computers in action movies with closeups of the Dell logo? Dell either loans, or gives a very steep discount on, the computers for the renderfarms in exchange for product placement.
It's not that bad - it's far better than instant coffee, which would merit that description. It's coffee, it's just not good coffee, though it's better than the over-roast crap you get in places like Starbucks that consider bitterness to be the only flavour that coffee should have.
Neither will be matched by '*.*'. Hidden files, . and .. would be matched by '.*', but a '*.*' glob pattern requires characters before the dot. A '.*' pattern is particularly dangerous for this reason - it looks like it will only match hidden files, but it will also match the current directory and the parent directory.
so gave a sales person root access who promptly ran rm -rf /*.*.
Running rm -rf /*.*. (or even rm -rf /*.*) is unlikely to cause much damage on a *NIX system as it's very unusual to have any files with dots in their names in the root directory and even more unlikely to have any directories that have dots in their names.
The very idea of a cancer cure is idiotic as it is a wide spectrum of different kinds of cellular malfunction
The treatment in TFA (assuming that TFA is about the same treatment that was in the news a couple of days ago) works by allowing the body's natural defences against cancer to work properly. Cells become cancerous all of the time but they're almost always killed immediately by your autoimmune response. Some mutate in a way that prevents the immune system recognising them as broken and can then spread quite quickly. This treatment allows the immune system to recognise the cancer cells, so is likely to work on most, if not all, forms of cancer. The catch is that it requires a specific implementation for each patient.
Swipe left - message. Swipe right - call. Press and hold - for menu. Not hard to learn, very comfortable to use.
How do you learn it? Discoverability is a core concept in UI design. It's one of the reasons that the menu bar, in spite of its other failings, is still a pretty good UI model: everything that you can do is visible in the menu bar and a user can explore the functionality of a new application by browsing the menu. Gestures tend to be terrible for discoverability. Do I need to go and read some documentation to discover that I can swipe those things (or read a pop-up hint that either pops up once and then I forget the gesture, or pops up repeatedly and annoys me)? Or do I just gesture at every widget and eventually find the ones where it works? You can bet, given Google's horrendous track record for HCI, that they aren't going to make swipeable controls visually distinctive in any way.
Blue people make good soldiers, when led by white officers.