Indeed - although I'd say that 'bad implementation' that leads to a 'security hole' qualifies as a bug. In fact, probably a priority one bug. But I'm picky about how the apps I ship treat their users' data.
Easily mitigated, too, of course, because there's no rule that says that Skype's installer has to register the same executable as its protocol handler as it uses to handle user input at the command line. Use different executables, and the risk goes away. A small protocol handling exe that validates the input, extracts the information it needs, and uses it to launch the main Skype program is not going to be a huge additional component in an installation.
Register a protocol handler in the Windows registry, and you're taking responsibility for handling any URL you're given that begins with that protocol. It's not that hard to understand. Why people keep on insisting this is somehow a flaw in Windows I don't know...
Well when *I* were a lad, we could tell t' difference between Monty Python and At Last the 1948 Show. Now I'll grant you, half of the performing/writing team that made The Four Yorkshiremen sketch went on to become members of Monty Python, but for goodness sake, the sketch was co-written by Tim Brooke Taylor and Marty Feldman, neither of whom were even involved in Python.
But you try telling that to young Monty Python fans, and they won't believe you. Maybe they'll believe Wikipedia?
The majority of material deleted (Not Only... But Also, Doctor Who, that kind of thing) was entertainment programming - thrown out to make room for the BBC to archive its current affairs output. I don't think the beeb has any worry that when it comes to releasing its complete news archive it won't have every single regional news program from the last 20 years to throw in to the mix. Certainly they will have the vast majority of national news footage that has ever been broadcast, every episode of Panorama, every Question Time...
Firefox (and other Mozilla derivatives) support a preloading link. When they encounter such a link in one page, they begin downloading the content for the linked page, so they have it ready. Google assumes that you're reasonably likely to click on the first link they've sent you for some types of search result (probably where there's a very high search ranking for one particular site for the term you searched for), so sends Mozilla/firefox users a preload warning along with the search result page, with the URL of the first search result page. Firefox does its thing and starts downloading the page content for the first search result before you even click on it - including any cookies.
And of course the computer is right to identify the ambiguity. If I give you the sentences in these two (hastily constructed) contexts, the 'incorrect' interpretations would both be correct:
'Buzzing around the Doctor were millions of tiny insects. "Time flies!" he shouted over the rising buzzing, "we must get out of here." "Doctor! Over here! There's a signpost!" "Keep away from it - it has an arrow on it. Time flies like an arrow. You'll be eaten alive!"'
'Having conducted aerodynamic tests with the trebuchet and a variety of fruit, the scientists concluded that variations based on type of fruit were minor, and the more detailed measurements could be taken by using a single fruit as the test subject and extrapolating. The banana was selected as the median fruit, since the tests had shown that, in a very general sense, all fruit flies like a banana.'
We just have a 'nonsense filter' that evaluates the possible interpretations for the two sentences and chooses the one that makes most sense in the current context.
> We can say, Earlier you educated me. but not Earlier you teached me. Why?
We say 'earlier you taught me' instead. What is your point?
In terms of language evolution, the word 'taught' has the same relationship to 'teach' as 'wrought' has to 'wreak', and similar relationships to 'thought'-'think', 'brought'-'bring' and (less so) 'bought'-'buy'. The pretirite form of each of these verbs is actually formed by a very similar linguistic rule to the one that forms 'educated' from 'educate' - the basic rule in germanic languages being that you stick a dental plosive 't' or 'd' sound on the end of the verb (ignore how the words are spelled, as that's really an irrelevance to the evolution of the words in the first place - we're talking about sounds here). Once this form has been created, however, it can create an awkward sound at the end of the word - 'ct', 'ngd', 'nct', etc. Language users don't like awkward sounds, they change them, preserving the distinctiveness, but losing some of the closeness to the original word. Also bear in mind that 'ch' was not always the sound at the end of the word 'teach' - it was once a much harder sound.
Add to this general rule the tendency in germanic languages for certain verbs ('strong verbs') to change their vowel sound in the past tunse (cf: 'run'-'ran', 'sing'-'sang', etc.), and you can see roughly where 'taught' came from. It's not really an 'exception', just a very old word that's had time to be moulded into a more comfortable shape through usage.
When trying to reduce a living language to a syntax, you miss out on the richness imparted to languages by the conventions that they gather through continual usage. English has simple syntax rules - I can coin a new verb and use it in grammatical sentences without anybody having any doubt about what syntactic role it is playing - look at the rise of 'google' as a verb - nobody had to teach you the words 'googles', 'googled' and 'googling', but you would happily use them. But once words are accepted into the language and used, they move over time, sometimes not in the same direction as their near relatives (as 'teach' and 'taught'). To explain where these words come from you need to look at the syntax rules prevailing at the time the derivative word was coined, and the pressures and modifications the words have been subjected to since. This is exactly what we mean by a 'living language'.
Whoosh... the sound of a point going straight over your head.
The GP was saying that the problem is people have in the past tended to believe whatever they read, because only 'journalists' and 'authorities' and people with 'responsibility' have been able to stick printed words in front of them, in newspapers, magazines, books and the like. In the internet age, the boundary around 'authorities' of this sort has been eroded - we can no longer tell whether someone is just some guy on a power trip, someone in the pay of a loon with an agenda, or a genuine person with a mission to inform.
The interesting thing is, that this has actually always been the case - it's just that the internet draws attention to the fact that anything you read could actually have been written by anybody.
The GP was saying that things are getting better because people are learning to be more cynical about the sources they read in any print medium.
I'm not sure how true that is, but it is an interesting thought...
> take the time in the next couple of days to donate blood
Erm... it's a flood. Lots of water has fallen from the sky and risen from the sea and the river and washed away lots of people, their possessions and livelihoods, and significant amounts of infrastructure. Sewerage has been inundated and potable water and hygienic food resources are scarce. People have no homes - no beds to sleep in; they may have only the clothes they were rescued in. One problem they don't have, though, in the vast majority of cases, is significant blood loss.
While there are doubtless injuries and possibly some diseases resulting from these conditions that will benefit from blood transfusions, I hardly think that is the principal problem the people of the delta area face right now. A million people may have been displaced by these flood waters. A MILLION people.
Giving blood is a good idea; you should do it whenever you can; if you want to help in this specific case, though, maybe the red cross would rather you were donating clean clothes, blankets, bedding... I don't know, but I'm sure your local Red Cross branch can tell you what they need. Blood's probably quite low on the list though.
When you are not a 90th percentile male involved in a typical RTA?
Seatbelts are designed to ensure you slow down almost as quickly as your car does, but quicker than your steering wheel does. They do that because they're tuned to people whose physical dimensions, mass, form, and flexibility fall within certain statistical ranges. If those assumptions are wrong, they might slow you down too quickly, or not quickly enough. They can cause internal injuries, amputations, neck injuries - especially if they're wor wrong, or you aren't in the right position in the seat.
5' tall, female, pregnant? In a low-speed impact, you might well be worse off because you've got your seatbelt on than if you don't.
Fallen asleep in the passenger seat with your head resting on the window, slumped down in the seat? sorry, your seatbelt wasn't designed to protect you in that posture...
But of course, we all should always wear seatbelts because we can't predict what kind of crash we'll be in and in most crashes most of us will be better off with than without. Probability, as the grandparent said.
I mean, the record companies want to sell albums, the record companies pay radio station to air adverts for the albums... but if they pay them to broadcast free samples from the album, that's suddenly wrong?
I never figured out why radio stations had to pay record companies for the right to broadcast advertising material for them. The recording industry's greatest ever scam was reversing the advertising model to such an extent that if they are caught actually paying for their ads to be broadcast, it's seen as wrong...
Who do you think generally has the high ground in the battle for supremacy between bookstores and publishers? time was, back before Borders and Barnes and Noble superstores and - especially - Amazon, not to mention the role of Walmart and the like in book distribution, it was the publisher, no question. Bookstores used to get royally screwed by publishers - look at the net book agreement in the UK, which used to prevent bookstores from ever offering a book for sale at a lower price than the publisher required. Now, though, with bookstores consolidated almost completely into the hands of big retail chains, the publisher isn't in such a strong position.
These bookstore chains have one objective in mind - they want to be certain that when a member of the public decides they want to buy a particular book, and walks into their local branch of the chain, they walk out with a copy of that book. They never want to be faced with the possibility of someone coming in, not finding the book, and going down the road, or online to get it instead.
That means they need publishers to overstock them. They require more copies than they will ever sell so that they can guarantee stock in every store (and what's more, they buy them all on sale or return - the publisher is always left holding the unsold stock, hence the remainders trade). And if publishers want to be present on the shelves of B&N or Borders at all, they need to print and distribute enough to fill the shelves of every store in the country - a logistical challenge, since books are bulky, weather-sensitive, and very heavy.
The agreement by the bookstores not to start selling the book until it's reached every store is simply a result of them agreeing with the publisher not to jump the gun provided nobody else does. They require the publisher to enforce the deal, otherwise they'll breach it too.
Even with a book like Harry Potter, remember, the publisher is nowhere if it's not in B&N, Borders and Amazon. In this respect, the bookstores have the publishers over a barrel.
It's independent bookstores and small or specialist publishers that suffer most, of course.
A worse side effect of police cars on traffic: a typical situation on a British motorway is to hear a siren and see flashing blues in the distance in your rear-view mirror - police on an emergency, trying to get somewhere fast. Obviously, you'll want to pull over and let them pass in the right-hand lane. Except you can't, because everybody sees those blue lights, and slows down to 70mph (they were all doing 80-90 before, of course).
And when everybody's doing 70mph, they form a rolling roadblock preventing the police car from coming past. And preventing you from pulling left out of the overtaking lane UNLESS you do what the police car would prefer you to, which is to accelerate the hell out of their way.
So speed governors sound like a supremely bad idea on motorways.
BBC Radio, some of the finest radio in the world, available via podcast.
This is the same organisation that's been making MP3 recordings of the BBC Philharmonic Orchestra's recent performances of all of Beethoven's symphonies available for download.
Podcast support built into iTunes only makes it more likely that the BBC's trials of this technology will be extended.
Re:Star Wars is Philosophy & Star Trek is Tech
on
The Science of Star Wars
·
· Score: 2, Insightful
Actually, worse than simplistic 'good vs Nazi Evil' is, perhaps, the fact that the overriding moral point made by the six part trilogy is:
People have great potential. They can choose to use it for good, but it's hard work. It's easy to use it for evil. But even if, because you're weak and vulnerable to human emotion, you do use it for evil - huge, Hitler-dwarfing, child-butchering, planet destroying, evil - you can make up for it so long as you eventually realise you were wrong, and ask the forgiveness of your kids.
Dell Latitude D600, ~ 6 months old. 'eraser' point thing and trackpad included.
Great screen resolution, nice formfactor, and apart from the intermittent bluetooth driver issue, very nice laptop indeed. Wouldn't leave home without it.
anyway, point is, I'm sure it isn't the only laptop in the Dell line with a touchpoint whotsit in the middle of the keyboard. And, what's more, unlike the IBM laptops I've used, I don't find I accidentally end up hitting it while typing.
If someone in your company wants to do something illegal, and they need network access to do it, and they choose to steal your password, how are you going to prove, when the audit trail points to it being you who raided the pension funds, that it was done with a stolen password?
Most hacks are inside jobs. Insiders know the systems, they know what checks are in place, they can find loopholes and exploit them. If they can make it look like it's you doing it, so much the better.
Slashdot Mirror
Can starbucks ask you to leave if you walk through an unlocked door that says 'staff only' on it?
Can starbucks clamp your car if you park it in their 'staff only' parking spaces?
what about if you use their 'for patrons only' parking without buying a coffee?
What about if you use their 'for patrons only' wireless network without buying a coffee?
See how this works?
Indeed - although I'd say that 'bad implementation' that leads to a 'security hole' qualifies as a bug. In fact, probably a priority one bug. But I'm picky about how the apps I ship treat their users' data.
Easily mitigated, too, of course, because there's no rule that says that Skype's installer has to register the same executable as its protocol handler as it uses to handle user input at the command line. Use different executables, and the risk goes away. A small protocol handling exe that validates the input, extracts the information it needs, and uses it to launch the main Skype program is not going to be a huge additional component in an installation.
Register a protocol handler in the Windows registry, and you're taking responsibility for handling any URL you're given that begins with that protocol. It's not that hard to understand. Why people keep on insisting this is somehow a flaw in Windows I don't know...
Well when *I* were a lad, we could tell t' difference between Monty Python and At Last the 1948 Show. Now I'll grant you, half of the performing/writing team that made The Four Yorkshiremen sketch went on to become members of Monty Python, but for goodness sake, the sketch was co-written by Tim Brooke Taylor and Marty Feldman, neither of whom were even involved in Python.
But you try telling that to young Monty Python fans, and they won't believe you. Maybe they'll believe Wikipedia?
The majority of material deleted (Not Only ... But Also, Doctor Who, that kind of thing) was entertainment programming - thrown out to make room for the BBC to archive its current affairs output. I don't think the beeb has any worry that when it comes to releasing its complete news archive it won't have every single regional news program from the last 20 years to throw in to the mix. Certainly they will have the vast majority of national news footage that has ever been broadcast, every episode of Panorama, every Question Time...
Google has all the details here: http://www.google.com/help/features.html#prefetch
Sounds like preloading.
Firefox (and other Mozilla derivatives) support a preloading link. When they encounter such a link in one page, they begin downloading the content for the linked page, so they have it ready. Google assumes that you're reasonably likely to click on the first link they've sent you for some types of search result (probably where there's a very high search ranking for one particular site for the term you searched for), so sends Mozilla/firefox users a preload warning along with the search result page, with the URL of the first search result page. Firefox does its thing and starts downloading the page content for the first search result before you even click on it - including any cookies.
And of course the computer is right to identify the ambiguity. If I give you the sentences in these two (hastily constructed) contexts, the 'incorrect' interpretations would both be correct:
'Buzzing around the Doctor were millions of tiny insects. "Time flies!" he shouted over the rising buzzing, "we must get out of here." "Doctor! Over here! There's a signpost!" "Keep away from it - it has an arrow on it. Time flies like an arrow. You'll be eaten alive!"'
'Having conducted aerodynamic tests with the trebuchet and a variety of fruit, the scientists concluded that variations based on type of fruit were minor, and the more detailed measurements could be taken by using a single fruit as the test subject and extrapolating. The banana was selected as the median fruit, since the tests had shown that, in a very general sense, all fruit flies like a banana.'
We just have a 'nonsense filter' that evaluates the possible interpretations for the two sentences and chooses the one that makes most sense in the current context.
> We can say, Earlier you educated me. but not Earlier you teached me. Why?
We say 'earlier you taught me' instead. What is your point?
In terms of language evolution, the word 'taught' has the same relationship to 'teach' as 'wrought' has to 'wreak', and similar relationships to 'thought'-'think', 'brought'-'bring' and (less so) 'bought'-'buy'. The pretirite form of each of these verbs is actually formed by a very similar linguistic rule to the one that forms 'educated' from 'educate' - the basic rule in germanic languages being that you stick a dental plosive 't' or 'd' sound on the end of the verb (ignore how the words are spelled, as that's really an irrelevance to the evolution of the words in the first place - we're talking about sounds here). Once this form has been created, however, it can create an awkward sound at the end of the word - 'ct', 'ngd', 'nct', etc. Language users don't like awkward sounds, they change them, preserving the distinctiveness, but losing some of the closeness to the original word. Also bear in mind that 'ch' was not always the sound at the end of the word 'teach' - it was once a much harder sound.
Add to this general rule the tendency in germanic languages for certain verbs ('strong verbs') to change their vowel sound in the past tunse (cf: 'run'-'ran', 'sing'-'sang', etc.), and you can see roughly where 'taught' came from. It's not really an 'exception', just a very old word that's had time to be moulded into a more comfortable shape through usage.
When trying to reduce a living language to a syntax, you miss out on the richness imparted to languages by the conventions that they gather through continual usage. English has simple syntax rules - I can coin a new verb and use it in grammatical sentences without anybody having any doubt about what syntactic role it is playing - look at the rise of 'google' as a verb - nobody had to teach you the words 'googles', 'googled' and 'googling', but you would happily use them. But once words are accepted into the language and used, they move over time, sometimes not in the same direction as their near relatives (as 'teach' and 'taught'). To explain where these words come from you need to look at the syntax rules prevailing at the time the derivative word was coined, and the pressures and modifications the words have been subjected to since. This is exactly what we mean by a 'living language'.
Whoosh... the sound of a point going straight over your head.
The GP was saying that the problem is people have in the past tended to believe whatever they read, because only 'journalists' and 'authorities' and people with 'responsibility' have been able to stick printed words in front of them, in newspapers, magazines, books and the like. In the internet age, the boundary around 'authorities' of this sort has been eroded - we can no longer tell whether someone is just some guy on a power trip, someone in the pay of a loon with an agenda, or a genuine person with a mission to inform.
The interesting thing is, that this has actually always been the case - it's just that the internet draws attention to the fact that anything you read could actually have been written by anybody.
The GP was saying that things are getting better because people are learning to be more cynical about the sources they read in any print medium.
I'm not sure how true that is, but it is an interesting thought...
> take the time in the next couple of days to donate blood
Erm... it's a flood. Lots of water has fallen from the sky and risen from the sea and the river and washed away lots of people, their possessions and livelihoods, and significant amounts of infrastructure. Sewerage has been inundated and potable water and hygienic food resources are scarce. People have no homes - no beds to sleep in; they may have only the clothes they were rescued in. One problem they don't have, though, in the vast majority of cases, is significant blood loss.
While there are doubtless injuries and possibly some diseases resulting from these conditions that will benefit from blood transfusions, I hardly think that is the principal problem the people of the delta area face right now. A million people may have been displaced by these flood waters. A MILLION people.
Giving blood is a good idea; you should do it whenever you can; if you want to help in this specific case, though, maybe the red cross would rather you were donating clean clothes, blankets, bedding... I don't know, but I'm sure your local Red Cross branch can tell you what they need. Blood's probably quite low on the list though.
> It gets them to think about science, and costs a few grams added to the probe.
Man - how long are your two kids' names?
When you are not a 90th percentile male involved in a typical RTA?
Seatbelts are designed to ensure you slow down almost as quickly as your car does, but quicker than your steering wheel does. They do that because they're tuned to people whose physical dimensions, mass, form, and flexibility fall within certain statistical ranges. If those assumptions are wrong, they might slow you down too quickly, or not quickly enough. They can cause internal injuries, amputations, neck injuries - especially if they're wor wrong, or you aren't in the right position in the seat.
5' tall, female, pregnant? In a low-speed impact, you might well be worse off because you've got your seatbelt on than if you don't.
Fallen asleep in the passenger seat with your head resting on the window, slumped down in the seat? sorry, your seatbelt wasn't designed to protect you in that posture...
But of course, we all should always wear seatbelts because we can't predict what kind of crash we'll be in and in most crashes most of us will be better off with than without. Probability, as the grandparent said.
Isn't crediting it to Britney Spears (which is essentially the brand name of a trading division of XYZ records) essentially saying that?
When they play ads, there's nothing to indicate that they're playing them because they've received money - they might just like the sound of them.
I mean, the record companies want to sell albums, the record companies pay radio station to air adverts for the albums... but if they pay them to broadcast free samples from the album, that's suddenly wrong?
I never figured out why radio stations had to pay record companies for the right to broadcast advertising material for them. The recording industry's greatest ever scam was reversing the advertising model to such an extent that if they are caught actually paying for their ads to be broadcast, it's seen as wrong...
So why aren't they equilateral triangles?
Oh wait, some are.
Who do you think generally has the high ground in the battle for supremacy between bookstores and publishers? time was, back before Borders and Barnes and Noble superstores and - especially - Amazon, not to mention the role of Walmart and the like in book distribution, it was the publisher, no question. Bookstores used to get royally screwed by publishers - look at the net book agreement in the UK, which used to prevent bookstores from ever offering a book for sale at a lower price than the publisher required. Now, though, with bookstores consolidated almost completely into the hands of big retail chains, the publisher isn't in such a strong position.
These bookstore chains have one objective in mind - they want to be certain that when a member of the public decides they want to buy a particular book, and walks into their local branch of the chain, they walk out with a copy of that book. They never want to be faced with the possibility of someone coming in, not finding the book, and going down the road, or online to get it instead.
That means they need publishers to overstock them. They require more copies than they will ever sell so that they can guarantee stock in every store (and what's more, they buy them all on sale or return - the publisher is always left holding the unsold stock, hence the remainders trade). And if publishers want to be present on the shelves of B&N or Borders at all, they need to print and distribute enough to fill the shelves of every store in the country - a logistical challenge, since books are bulky, weather-sensitive, and very heavy.
The agreement by the bookstores not to start selling the book until it's reached every store is simply a result of them agreeing with the publisher not to jump the gun provided nobody else does. They require the publisher to enforce the deal, otherwise they'll breach it too.
Even with a book like Harry Potter, remember, the publisher is nowhere if it's not in B&N, Borders and Amazon. In this respect, the bookstores have the publishers over a barrel.
It's independent bookstores and small or specialist publishers that suffer most, of course.
Roll on electronic paper e-books, I say.
A worse side effect of police cars on traffic: a typical situation on a British motorway is to hear a siren and see flashing blues in the distance in your rear-view mirror - police on an emergency, trying to get somewhere fast. Obviously, you'll want to pull over and let them pass in the right-hand lane. Except you can't, because everybody sees those blue lights, and slows down to 70mph (they were all doing 80-90 before, of course).
And when everybody's doing 70mph, they form a rolling roadblock preventing the police car from coming past. And preventing you from pulling left out of the overtaking lane UNLESS you do what the police car would prefer you to, which is to accelerate the hell out of their way.
So speed governors sound like a supremely bad idea on motorways.
In town, though... maybe they have a point.
http://www.bbc.co.uk/radio4/mp3.shtmlt ml#programmes
http://www.bbc.co.uk/radio/downloadtrial/index.sh
BBC Radio, some of the finest radio in the world, available via podcast.
This is the same organisation that's been making MP3 recordings of the BBC Philharmonic Orchestra's recent performances of all of Beethoven's symphonies available for download.
Podcast support built into iTunes only makes it more likely that the BBC's trials of this technology will be extended.
Actually, worse than simplistic 'good vs Nazi Evil' is, perhaps, the fact that the overriding moral point made by the six part trilogy is:
People have great potential. They can choose to use it for good, but it's hard work. It's easy to use it for evil. But even if, because you're weak and vulnerable to human emotion, you do use it for evil - huge, Hitler-dwarfing, child-butchering, planet destroying, evil - you can make up for it so long as you eventually realise you were wrong, and ask the forgiveness of your kids.
That isn't simplistic - it's insulting.
1) Define 'work well'
2) Explain how electronic voting machines work better
Dell Latitude D600, ~ 6 months old. 'eraser' point thing and trackpad included.
Great screen resolution, nice formfactor, and apart from the intermittent bluetooth driver issue, very nice laptop indeed. Wouldn't leave home without it.
anyway, point is, I'm sure it isn't the only laptop in the Dell line with a touchpoint whotsit in the middle of the keyboard. And, what's more, unlike the IBM laptops I've used, I don't find I accidentally end up hitting it while typing.
My country has had paper and pencil ballots for centuries, and they work fine, too. Why would we want electronic voting machines?
> Have you ever bought a used CD? You can buy the used CD, rip it to MP3's, and be perfectly legal.
Of course, the person you bought it from has to make sure they delete their MP3 rips of the CD, right?
I find it amazing that, of all the places that Mount Diablo Buckwheat should turn up, it'd be on the slopes of Mount Diablo.
It's funny nobody thought of looking there before...
You aren't paranoid enough.
If someone in your company wants to do something illegal, and they need network access to do it, and they choose to steal your password, how are you going to prove, when the audit trail points to it being you who raided the pension funds, that it was done with a stolen password?
Most hacks are inside jobs. Insiders know the systems, they know what checks are in place, they can find loopholes and exploit them. If they can make it look like it's you doing it, so much the better.