Show me the part of the constitution where it says
What happens at schools in the USA that makes so many citizens think 'ethical behaviour' means 'on a bit of paper that a bunch of dead slave-owners wrote down as the minimum set of things that they could agree on a couple of hundred years ago'?
Because it's not a single axis. There is smart vs sloppy, but there's also formal vs informal. If you dress too formally for an occasion, then you risk making other people feel uncomfortable.
It's important to understand the rules, then you can decide whether to break them. If you choose to say 'this rules is stupid, I'm not going to conform to it', then that's very different from saying 'I didn't know that rule exists'. When you're surrounded by other people who have a particular set of social rules, you will insult some of them if you break those rules. If you insult them by accident, then that is practically the definition of social ineptitude. If you insult them because you've decided to, then that's a choice, and you accept the consequences (which may be that you end up with a less uptight set of friends).
Hi, welcome to Slashdot. For future reference, you may be aware of this little thing called 'context'. This means that words written in one post relate to things mentioned in the post that it is a response to. For example, my post was in reply to a post suggesting using more RAM, instead of flash. The context of my comment was that flash [as a cache] won't make a difference to read performance [relative to a disk behind a large RAM cache]. Interpreting it as saying 'an SSD is no faster for reading than a spinning disk' makes you look like you've just failed the Turing Test.
Flash makes little difference to read performance, but can make a huge different to write speeds. RAM, being non-volatile, means that if an application calls fsync, you block until all of the data has been flushed to the disk. With a flash write cache, you can buffer a load of writes and return almost immediately (writes into flash can easily go at 100+MB/s) and then write them out to disk when it is idle or less loaded.
Once you don't compare "cheapest PC" vs. "cheapest Mac", but "the Mac I want" vs. "a PC with the same specs, bought from a reputable company", the Mac hardware will beat most PCs of the same price.
That's true, but it's also not a very good comparison. The real test is to start with the specs you want, and then compare a PC from a reputable supplier and a Mac that most closely approximate them. The Mac you want will almost certainly have some features you don't need, as will the PC that most closely approximates it. Both are likely to lack some features you want.
There's a question of how much targeting they do, and how much users are aware of it. I have a pretty good understanding of the data mining that Google does, but then I have a PhD in computer science and work one floor down from people doing research in this area. How many gmail users regard the ads that they see as being different from the ones that appear on TV to keep those programs free? Of the non-geek ones that I've talked to, very few, and yet the TV ads are broadcast to everyone whereas the gmail ones are selected for you based on a detailed profile that Google has harvested from your activity on a variety of web sites.
In the seL4 case, they first write a formal specification. Then they (oversimplifying slightly) prove an equivalence of their implementation (in a restricted subset of C) and their specification. Then they prove properties (e.g. isolation) of their specification. You can't just take some C code and say 'is this correct' without a spec, and you typically can't take arbitrary C code and say 'do these properties hold for this code'. Even in the seL4 case, there are some issues, for example they correct functioning of the MMU is taken as axiomatic.
In theory, an H-1B is only supposed to be granted if it is not possible to find a qualified person in the native labour pool. If someone has a degree that is so specialised that there are no US citizens with that skill, then why wouldn't they expect to be paid $100K or more?
H-1Bs are different. If a US citizen decides that they are being screwed, they can give notice, quit, and find another job. If an H-1B decides that they're being screwed, then they can't move jobs unless they can find another company that will go through the H-1B sponsorship process, which takes time, before the short grace period expires and they get deported. It gives their new employer a really strong bargaining position if every day that they delay finalising the remuneration agreements puts their potential employee a day closer to being deported.
My watch (nice, but not so nice that I'm terrified of breaking or losing it) was about £100, and it's now six years old. In that time, the battery has been replaced once. My father had a Rolex that he got as a wedding present. It was still working fine when it fell off his wrist into the Atlantic while windsurfing (teaching him to buy a cheap watch for sailing...) about 15 years later. Other people I know have inherited watches. At the opposite end of the spectrum, I know a lot of people who bought watches for about £1, as purely disposable timepieces that they can spill things on and not care when they're destroyed.
If the iWatch is going to be successful it's going to have to be in a different market than any of these. I don't know where Apple is going to position it, but it won't have the longevity of an expensive watch, it will need regular charging, and it will be too expensive to be a cheap disposable device.
It's fundamentally primitive and clunky to try and represent artist, composer, album, CD, song etc by the use of directory and filenames. That's what ID3 tags are for.
Actually, BeOS did it correctly. The ID3 tags were just backup. The information was recorded in filesystem metadata and you could view it in the Tracker (BeOS file manager). You could sort a directory of music by album or artist, and you could create folders whose contents were search results, so you could just create a folder that contained all music, or all music matching some other criteria (year, album name, genre, play count, whatever). And you could do the same thing with other file types. For example, there was no contact manager on BeOS: people were represented as filesystem objects with metadata for things like name, address, and so on. There was an application for viewing and editing a single person, but not for browsing or searching them, because that's what the Tracker was for.
You criticize music as being non-hierarchical but yourself like a two tier hierarchy.
iTunes doesn't expose a hierarchy, it exposes filters. You can select a set of albums, a set of artists, or a set of genres. It also has smart playlists, where you can define a set of predicates that songs must match to be in it. For comparison, in an hierarchical filesystem view, how long does it take you to say 'play all of my rock music from the '60s that isn't by The Beatles'?
Wow. First, the original iPod used FireWire, whereas competing devices used USB 1.1. That meant that you could copy music onto the iPod and be limited by the drive speed, i.e. in about 10 minutes to completely fill the 5GB drive. Filling other devices around at the same time took hours.
Second, are you really arguing that plugging in the device and hitting 'sync' or, in the case that you had a lot of music (remember, desktop hard drives were in the 20-40GB range when the iPod launched, so more than 5GB of music was unusual) selecting a playlist and hitting 'sync' is harder than copying files across? If you've got your music arranged in folders (which, basically, means that either you're a geek or you're using a program that does it for you like... iTunes) then it's not a massive pain to copy it across, but it's still far more time consuming than just pressing the sync button. And you'd better hope that all of your M3U playlists used relative paths (lost of players created them with absolute paths by default) and that you created exactly the same hierarchy on the device, or your playlists wouldn't transfer.
Actually, I think I'm overstating the complexity of the iTunes sync (it's been quite a few years since I've used it). As I recall, you just plugged in the iPod, the sync happened automatically if there was enough space. If not, then you had to select a playlist for syncing.
Two data points don't make a trend. 6 months ago, AAPL was at 700, now it's at 420. But one year ago it was at 533, and at the start of 2012 it was just passing 420, which was the highest it had been ever. Or, to put it another way, if you'd bought Apple stock any time between 1985 and the start of 2012, and you sold it today, you'd make a profit. Their stock is currently at the lowest that it's been for a year, but it dipped to 530 a few times in the last year - the 700 was a big spike just before the iPhone 5 was announced. You can see a similar spike before or after each Apple event, as lots of speculators either buy or short the stock based on rumours and then sell or buy after the announcement.
Google is currently at the highest point in the last year, up at 820, but in October they dipped from 755 to 695 in a single day. In short, both stocks have so many speculators in the market, so their value is largely decoupled from actual performance.
A lot of people do. It amazes me the number of people I see pulling out their phones to check the time. It seems very cumbersome to me, but then I grew up using a watch. A lot of people never owned one and so never got used to the convenience. The first time you wear a watch, it feels strange and it takes a while to get used to having one on. For people in their teens and early 20s, their reaction is often to say 'why do I need this uncomfortable thing, I could just use my phone' and never even spend an hour getting used to the feel.
It also doesn't help that most mens watches seem to be designed for people who feel that it's an affront to their masculinity not to be lifting weights all the time, so are huge blobs of metal. I have a very comfortable Skagen watch that tells the time and is sufficiently thin and light that it's easy to forget that I'm wearing it, but most male watches look like they'd be more useful as a bludgeon than a timepiece.
There used to be PalmOS watches. There are now Android watches. I know someone who has one - it was bought out of a research grant for various HCI and ubicomp projects. They try to solve a problem that doesn't really exist. The main features you want from a watch is never having to think about charging it. Even with an eInk screen, getting reasonable battery life is going to be painful. It's essentially a passive device - something you glance at, but with a UI that's too small for anything active. The old PalmOS ones had a 4-icon screen, and even they were painful to use. Being able to sync your calendar alarms to a watch is somewhat useful, but only for people who are likely to carry a watch in places where they want alarms but don't want to carry a phone or computer. Possibly Apple has found a use-case that everyone else missed, but it seems a bit more like bandwagon jumping than anything else.
This has nothing to do with Oracle. The browser plugin has a long history of security holes going back well over a decade and the bitching has been going on since 1995. The problem is that writing a language implementation that is both fast and 100% correct is really hard. The safety properties of Java (and any other managed language) rely on the implementation being 100% correct. This is relatively easy for something like the Squeak Smalltalk VM, which is a single-threaded bytecode interpreter with a stop-the-world garbage collector, but people insist on the JVM doing all sorts of optimisations, supporting multiple threads and so on. The early complaints about Java were that it was slow. The more recent complaints are that it's not correct. Well, you have three choices:
Have a slow VM.
Have a fast, but incorrect, VM, and be aware that every error is a potential security hole.
Formally verify your VM. Be aware that this will cost at least 30 times as much[1] as the non-verified version.
Relying on software enforcement for security is just asking for trouble.
[1] The factor of 30 comes from seL4 which, to mu knowledge, is the formally verified project that managed the smallest overhead. Other estimates from other projects are 100 or more times the cost.
And it's a good metaphor for DRM. It costs more to make, because this chair has a microprocessor. That cost gets passed on to the consumer, yet makes the product less valuable to them.
I've seen a few cyclists do dangerous things, but rarely things that are likely to endanger anyone other than themselves. In contrast, drivers seem to turn into drooling morons as soon as the see a bicycle. For example, overtaking a bike that's decelerating because there's an obstacle in front (e.g. a parked bus), then pulling in in front of the bike and slamming on the breaks. Or deciding that the place where the road is only just two card widths wide, on a bend, is the best place to overtake a bike - when there's oncoming traffic that then needs to do an emergency stop to avoid it. Or deciding not to bother indicating when stopped at traffic lights directly in front of a junction and then as soon as the lights change pull in front of a bike and turn across its path, then act surprised that the bike has to break hard and swerve to avoid the collision. Or, my personal favourite, start to overtake just before the road narrows, not have enough space to get all the way past, and instead of hanging back and trying again later, cut in front so that the bike has to either brake hard or swerve.
He doesn't say that bicycles produce more CO2 than cars, he says that:
Drivers pay road tax to cover the costs of roads, including bike lanes, why shouldn't bikes pay some of this?
Cycling increases your respiration rate so produces more CO2 than not cycling.
Both of these are true. The only one he is actually using to justify his position (that bikes should pay road tax) is the former, the second point is refuting the point that bikes are environmentally friendly. The second point is debatable: it's a question of what the basic comparison is. Cycling is more polluting than staying at home, less polluting than driving a car.
There are lots of valid reasons to mock Republicans, we don't need to make more up.
The purpose of a good middle manager is to deal with the problems that get in the way of the skilled workers being productive. This includes unreasonable deadlines, poorly specified problems, lack of equipment, poor communication with other teams, and so on. They are also responsible for ensuring that the upper management is aware of their team's capabilities and limitations, so that they are able to set realistic overall objectives for the company. Someone doing this job well can obviously provide a significant benefit to a company. The problem is that a great many people in this role do not do it well.
Slashdot Mirror
a load inside her inside her insides
Recursion porn?
Show me the part of the constitution where it says
What happens at schools in the USA that makes so many citizens think 'ethical behaviour' means 'on a bit of paper that a bunch of dead slave-owners wrote down as the minimum set of things that they could agree on a couple of hundred years ago'?
Because it's not a single axis. There is smart vs sloppy, but there's also formal vs informal. If you dress too formally for an occasion, then you risk making other people feel uncomfortable.
It's important to understand the rules, then you can decide whether to break them. If you choose to say 'this rules is stupid, I'm not going to conform to it', then that's very different from saying 'I didn't know that rule exists'. When you're surrounded by other people who have a particular set of social rules, you will insult some of them if you break those rules. If you insult them by accident, then that is practically the definition of social ineptitude. If you insult them because you've decided to, then that's a choice, and you accept the consequences (which may be that you end up with a less uptight set of friends).
Hi, welcome to Slashdot. For future reference, you may be aware of this little thing called 'context'. This means that words written in one post relate to things mentioned in the post that it is a response to. For example, my post was in reply to a post suggesting using more RAM, instead of flash. The context of my comment was that flash [as a cache] won't make a difference to read performance [relative to a disk behind a large RAM cache]. Interpreting it as saying 'an SSD is no faster for reading than a spinning disk' makes you look like you've just failed the Turing Test.
Flash makes little difference to read performance, but can make a huge different to write speeds. RAM, being non-volatile, means that if an application calls fsync, you block until all of the data has been flushed to the disk. With a flash write cache, you can buffer a load of writes and return almost immediately (writes into flash can easily go at 100+MB/s) and then write them out to disk when it is idle or less loaded.
Once you don't compare "cheapest PC" vs. "cheapest Mac", but "the Mac I want" vs. "a PC with the same specs, bought from a reputable company", the Mac hardware will beat most PCs of the same price.
That's true, but it's also not a very good comparison. The real test is to start with the specs you want, and then compare a PC from a reputable supplier and a Mac that most closely approximate them. The Mac you want will almost certainly have some features you don't need, as will the PC that most closely approximates it. Both are likely to lack some features you want.
I think I'd rather have an incompetent company try to harvest data about me than a competent one...
There's a question of how much targeting they do, and how much users are aware of it. I have a pretty good understanding of the data mining that Google does, but then I have a PhD in computer science and work one floor down from people doing research in this area. How many gmail users regard the ads that they see as being different from the ones that appear on TV to keep those programs free? Of the non-geek ones that I've talked to, very few, and yet the TV ads are broadcast to everyone whereas the gmail ones are selected for you based on a detailed profile that Google has harvested from your activity on a variety of web sites.
In the seL4 case, they first write a formal specification. Then they (oversimplifying slightly) prove an equivalence of their implementation (in a restricted subset of C) and their specification. Then they prove properties (e.g. isolation) of their specification. You can't just take some C code and say 'is this correct' without a spec, and you typically can't take arbitrary C code and say 'do these properties hold for this code'. Even in the seL4 case, there are some issues, for example they correct functioning of the MMU is taken as axiomatic.
$100K for a new grad? Really?
In theory, an H-1B is only supposed to be granted if it is not possible to find a qualified person in the native labour pool. If someone has a degree that is so specialised that there are no US citizens with that skill, then why wouldn't they expect to be paid $100K or more?
H-1Bs are different. If a US citizen decides that they are being screwed, they can give notice, quit, and find another job. If an H-1B decides that they're being screwed, then they can't move jobs unless they can find another company that will go through the H-1B sponsorship process, which takes time, before the short grace period expires and they get deported. It gives their new employer a really strong bargaining position if every day that they delay finalising the remuneration agreements puts their potential employee a day closer to being deported.
If the iWatch is going to be successful it's going to have to be in a different market than any of these. I don't know where Apple is going to position it, but it won't have the longevity of an expensive watch, it will need regular charging, and it will be too expensive to be a cheap disposable device.
It's fundamentally primitive and clunky to try and represent artist, composer, album, CD, song etc by the use of directory and filenames. That's what ID3 tags are for.
Actually, BeOS did it correctly. The ID3 tags were just backup. The information was recorded in filesystem metadata and you could view it in the Tracker (BeOS file manager). You could sort a directory of music by album or artist, and you could create folders whose contents were search results, so you could just create a folder that contained all music, or all music matching some other criteria (year, album name, genre, play count, whatever). And you could do the same thing with other file types. For example, there was no contact manager on BeOS: people were represented as filesystem objects with metadata for things like name, address, and so on. There was an application for viewing and editing a single person, but not for browsing or searching them, because that's what the Tracker was for.
You criticize music as being non-hierarchical but yourself like a two tier hierarchy.
iTunes doesn't expose a hierarchy, it exposes filters. You can select a set of albums, a set of artists, or a set of genres. It also has smart playlists, where you can define a set of predicates that songs must match to be in it. For comparison, in an hierarchical filesystem view, how long does it take you to say 'play all of my rock music from the '60s that isn't by The Beatles'?
Wow. First, the original iPod used FireWire, whereas competing devices used USB 1.1. That meant that you could copy music onto the iPod and be limited by the drive speed, i.e. in about 10 minutes to completely fill the 5GB drive. Filling other devices around at the same time took hours.
Second, are you really arguing that plugging in the device and hitting 'sync' or, in the case that you had a lot of music (remember, desktop hard drives were in the 20-40GB range when the iPod launched, so more than 5GB of music was unusual) selecting a playlist and hitting 'sync' is harder than copying files across? If you've got your music arranged in folders (which, basically, means that either you're a geek or you're using a program that does it for you like... iTunes) then it's not a massive pain to copy it across, but it's still far more time consuming than just pressing the sync button. And you'd better hope that all of your M3U playlists used relative paths (lost of players created them with absolute paths by default) and that you created exactly the same hierarchy on the device, or your playlists wouldn't transfer.
Actually, I think I'm overstating the complexity of the iTunes sync (it's been quite a few years since I've used it). As I recall, you just plugged in the iPod, the sync happened automatically if there was enough space. If not, then you had to select a playlist for syncing.
Two data points don't make a trend. 6 months ago, AAPL was at 700, now it's at 420. But one year ago it was at 533, and at the start of 2012 it was just passing 420, which was the highest it had been ever. Or, to put it another way, if you'd bought Apple stock any time between 1985 and the start of 2012, and you sold it today, you'd make a profit. Their stock is currently at the lowest that it's been for a year, but it dipped to 530 a few times in the last year - the 700 was a big spike just before the iPhone 5 was announced. You can see a similar spike before or after each Apple event, as lots of speculators either buy or short the stock based on rumours and then sell or buy after the announcement.
Google is currently at the highest point in the last year, up at 820, but in October they dipped from 755 to 695 in a single day. In short, both stocks have so many speculators in the market, so their value is largely decoupled from actual performance.
A lot of people do. It amazes me the number of people I see pulling out their phones to check the time. It seems very cumbersome to me, but then I grew up using a watch. A lot of people never owned one and so never got used to the convenience. The first time you wear a watch, it feels strange and it takes a while to get used to having one on. For people in their teens and early 20s, their reaction is often to say 'why do I need this uncomfortable thing, I could just use my phone' and never even spend an hour getting used to the feel.
It also doesn't help that most mens watches seem to be designed for people who feel that it's an affront to their masculinity not to be lifting weights all the time, so are huge blobs of metal. I have a very comfortable Skagen watch that tells the time and is sufficiently thin and light that it's easy to forget that I'm wearing it, but most male watches look like they'd be more useful as a bludgeon than a timepiece.
There used to be PalmOS watches. There are now Android watches. I know someone who has one - it was bought out of a research grant for various HCI and ubicomp projects. They try to solve a problem that doesn't really exist. The main features you want from a watch is never having to think about charging it. Even with an eInk screen, getting reasonable battery life is going to be painful. It's essentially a passive device - something you glance at, but with a UI that's too small for anything active. The old PalmOS ones had a 4-icon screen, and even they were painful to use. Being able to sync your calendar alarms to a watch is somewhat useful, but only for people who are likely to carry a watch in places where they want alarms but don't want to carry a phone or computer. Possibly Apple has found a use-case that everyone else missed, but it seems a bit more like bandwagon jumping than anything else.
Relying on software enforcement for security is just asking for trouble.
[1] The factor of 30 comes from seL4 which, to mu knowledge, is the formally verified project that managed the smallest overhead. Other estimates from other projects are 100 or more times the cost.
And it's a good metaphor for DRM. It costs more to make, because this chair has a microprocessor. That cost gets passed on to the consumer, yet makes the product less valuable to them.
Is it just me, or was the best bit about that the fact that he was described by the subtitle as the chairperson?
I've seen a few cyclists do dangerous things, but rarely things that are likely to endanger anyone other than themselves. In contrast, drivers seem to turn into drooling morons as soon as the see a bicycle. For example, overtaking a bike that's decelerating because there's an obstacle in front (e.g. a parked bus), then pulling in in front of the bike and slamming on the breaks. Or deciding that the place where the road is only just two card widths wide, on a bend, is the best place to overtake a bike - when there's oncoming traffic that then needs to do an emergency stop to avoid it. Or deciding not to bother indicating when stopped at traffic lights directly in front of a junction and then as soon as the lights change pull in front of a bike and turn across its path, then act surprised that the bike has to break hard and swerve to avoid the collision. Or, my personal favourite, start to overtake just before the road narrows, not have enough space to get all the way past, and instead of hanging back and trying again later, cut in front so that the bike has to either brake hard or swerve.
Both of these are true. The only one he is actually using to justify his position (that bikes should pay road tax) is the former, the second point is refuting the point that bikes are environmentally friendly. The second point is debatable: it's a question of what the basic comparison is. Cycling is more polluting than staying at home, less polluting than driving a car.
There are lots of valid reasons to mock Republicans, we don't need to make more up.
The purpose of a good middle manager is to deal with the problems that get in the way of the skilled workers being productive. This includes unreasonable deadlines, poorly specified problems, lack of equipment, poor communication with other teams, and so on. They are also responsible for ensuring that the upper management is aware of their team's capabilities and limitations, so that they are able to set realistic overall objectives for the company. Someone doing this job well can obviously provide a significant benefit to a company. The problem is that a great many people in this role do not do it well.