Not really. We're not surprised that systems from 40 years ago are still able to do the things that they were designed to do, we're surprised that the requirements haven't morphed beyond all recognition in this time. If you spend three years developing a software system, and at the end of those three years your requirements look even remotely similar to the ones you started with, then these days you consider yourself very lucky. The idea that you could deploy a system and 40 years later your customer would still want the system to be doing the same thing sounds too good to be true.
Even good code starts to struggle after the requirements have changed completely for the 20th time...
I am always amazed how buildings constructed thousands, or even hundreds, of years ago are still standing although often in a state of disrepair due to neglect
That's what's known as survivor bias. The only examples you see of thousand-year-old buildings are the ones that didn't fall down. The ones that collapsed within a decade are long forgotten.
A modern-day castle might survive a century whereas the castles throughout Europe remain or at least remnants of their existence survive to this day
And yet, in the village where I grew up, and near countless other villages in Britain, there was a hill with a raised mound on top, which was the only remaining evidence of the castle that stood there 900 years ago.
North Korea, with its tiny allocation, is not exactly the bastion of well-secured machines. It's entirely plausible that a false flag operation launched some (likely trivial) part of the operation from a compromised machine in North Korea because they knew that as soon as the FBI found a North Korean IP in their traffic they'd stop bothering to look any further.
I thought they'd done that with KitKat. I don't have it on my phone (Moto G), but I don't really see it as a problem. I installed FireFox from F-Droid, disabled Chrome, and everything seems to work fine. I'd much rather have fewer things in the default install, as long as there are good replacements. I'm much more unhappy about the fact that they removed the character count from the SMS app and there's no open source version available.
I'd agree with your second point, but not the first. I really like Windows Phone's UI and structure, but it lacks several important bits of core functionality and (more importantly) lacks third-party apps to fill in the gaps. If Microsoft had managed to get developer mindshare earlier (not helped by breaking all existing Windows mobile apps), they'd be in a much better position.
Does it also come with UI regressions, like the change around 3-4 that turned putting the clock into night mode from a one-tap operation into a 4 tap sequence?
There's a weird trend in US hotels to give free WiFi if you're a member of their loyalty program. One hotel I stayed in gave members of their loyalty program free WiFi, wired Internet in the room, lounge access, cookies, and a few other perks, even if you joined the program when you booked the room and never intended on going to another one. I have a huge collection of 'loyalty' cards for hotel chains now, because I always pick the cheapest one.
+ $30-$75 is what I usually see quoted for the extra legroom seats, OR you can pick a seat near the back that is more likely to have an empty neighbor
Nearer the back means louder, which isn't great for long flights. If you fly regularly though, the extra legroom seats are often free. United gives them out to anyone who got more than 25K miles in the previous year.
The problem is the channels that carry so much advertising (especially shopping channels that are nothing but ads) that they pay to be included in bundles. This lowers the price of the bundle overall, in the hope that someone will watch some ads while channel surfing. No one is going to want to opt in to these and it wouldn't make sense in an a la carte model to allow people to opt into them for a 50 discount (or whatever), because most of the people who would subscribe would be the ones not in the target demographic, who just wanted the discount but would never watch the channel.
My mother recently moved to France and having seen her experience I think a lot of the rest of the world could benefit from adopting the French mode. Estate agents work solely for the buyer. There is no conflict of interest, they are paid to act as the buyer's agent and no one else's. As part of the justification for their fee, they accept a big chunk of liability for anything that is incorrectly represented to the buyer.
In this case, the porn owners are falsely claiming ownership of content on the Github pages.
No they aren't. There are two claims in a DMCA takedown notice. One is that you are authorised to act on behalf of the owners of copyright work A. The second is that work B infringes on work A. Only the first claim is made under penalty of perjury. There is a good reason for this: determining whether a work actually infringes (and is not covered by fair use and so on) requires a court ruling. The problem is that it's overly broad. I can write a DMCA notice that says that I own the copyright on this post (true) and that some other random site infringes on this copyright (false), and that's fine. The protection for this under the DMCA is that you can send a counter-notice and invalidate the original. The problem with this is that it's far easier to spam a load of take-down notices than for everyone to respond with counternotices.
If you want to get this fixed, send DMCA take-down notices to YouTube, Twitter, and Facebook for everything that every current member of Congress has online. Send notices to their web hosts. Do the same for every Republican and Democrat party web site. Knock both parties off the web until they fix their stupid laws.
Not sure about Canada, but in the USA the fines for infringement can be significantly larger for 'wilful infringement', which means that you must prove that the infringer knew that what they were doing was in violation of the relevant laws. If you distribute something that you were under the impression was in the public domain, or licensed in such a way that would permit distribution in the way that you are distributing it, then the penalties are typically lower. If you are an organisation responsible for telling people about copyright law and the stuff that you are infringing needed a subscription to access, then there's no defence against accusations of wilful infringement.
But if you saw one then would you recognise it? If they travel faster than light then you won't see them except when they're feeding. If they don't, then most of their life is likely to be in a dormant state as they spend a few thousand years between stars. Then there's the question of how they eat. If they eat the entire star at once, then you'll notice a star vanishing, but we don't have continuous observation on most stars, so there's a good chance that we'd see something odd in the data but not be able to tell what. If they eat in a more plausible way, then how would we tell it apart from, for example, a star near a superdense non-alive object that is drawing matter away from it?
No it isn't. There's a reason that AMD spun off their fab business. Fabs are insanely expensive and AMD doesn't have the volumes needed to justify a cutting edge fab by themselves. Having a fab running at anything below full production capacity is just throwing money away, but no one wants to buy capacity on an AMD fab because they know that when yields are low AMD chips will get priority. They'll buy capacity in a GlobalFoundaries or TMSC fab, because then they're just one customer in many.
The real question is why Intel won't fab the chips. Their commercial advantage has always been in production techniques, not in chip design. Throughout the lifetime of x86, other people have been able to compete with Intel designs, in spite of being one or two generations behind in process tech.
There are two universities in Brussels. Université Libre de Bruxelles (French for Free University of Brussels) and Vrije Universiteit Brussel (Dutch for Free University of Brussels). Translating the name of either into English makes it impossible to tell which institution he is a member of.
It is because research can never prove what the ROI will be.
That's not really true. You can look at a research lab and measure the ROI retrospectively quite easily and use this to make forward looking decisions, and that's what a lot of companies do. They'll close research labs that haven't produced anything useful in the last 5-10 years, but they'll increase funding to ones that have.
Google is one of the few companies that invests in products that might become useless.
No it isn't, it's just one of the few that labels them products and trumpets them in the press. Apple is about the only tech company that spends less on blue-sky research than Google. Companies like Microsoft and IBM spend vastly more. Aerospace companies would die within a decade if they didn't spend a significant amount on research and the same is true for a lot of other fields.
This is the same argument used for military R&D spending - there are lots of useful civilian results. The problem is that if you throw a big pile of R&D money at anything you'll likely get some useful results. The question is whether you get a good ROI. Compare NASA to, for example, Xerox PARC (Ethernet, the GUI, laser printers, etc.) or Bell Labs (the transistor, access control lists, UNIX, etc.) and see which produced more inventions that benefitted the economy as a whole per dollar spent.
Each shuttle launch cost, on average, $1.5bn. The cost of one launch would fund over ten thousand PhDs, or several hundred DARPA programs. Do you really think that NASA is the best ROI for taxpayers?
The population of the USA is about 320M, so that works out at 160M, or $1.6M. NASA's budget is actually around $17,647M, so you're off by three orders of magnitude. Do you, by any chance, work for Verizon? It's actually about $55 per person in the USA (including children).
That certainly used to be true, but is it still? Even economy on the 787 has power sockets under the seats and enough space to use a 15" laptop. 8 hours of time when no one is interrupting you can be very productive. In business class, there's always power and a lot more space. Or, if you get an overnight flight, it's comfortable enough to sleep and arrive at your destination feeling refreshed. Would you really pay 2-4 times as much for a 7 hour flight in an economy seat than for a 14 hour flight in a first-class seat? Because that's the kind of price difference Concorde had...
Fuel and price are not the same thing. Busses don't cross the ocean very well, but boats do. The problem with the boat is that it takes a week or so to get from Europe to the USA, which means that you're having to pay for food for that trip on top of the ticket and you need a lot more space (no one would put up with an economy airline seat for a week!), which drives up the cost a lot. And that's ignoring the opportunity cost from the lost time.
You can fly from London to New York in about 7-8 hours. How much is it worth to double that speed? It will save you 4 hours, but the total transit time includes waiting at the airport and getting to and from the airport, which adds another 3-4 hours (at least - assuming that the airport is relatively close to your endpoint). So now it's taking around 8 hours instead of 12. Is that worth it? That's assuming that both planes leave at the same time though. Expensive flights tend to leave once a day, so it may not get to your destination any sooner, it just reduces travelling time.
It's definitely worth the money for a lot of people to get there in one day rather than seven, but you rapidly start to hit diminishing returns once the transit time is less than a day. At that point, being comfortable on the way is more important for a lot of people. If you've got money to burn, would you rather have an 8 hour flight in a first class seat which reclines to be completely horizontal, will have decent food and champagne, or be cramped into a Concorde seat that's closer to an economy seat on another airline for 3-4 hours?
If money really isn't an issue, then charter a private plane and you'll get there much sooner because it will fly from a smaller airfield near you, leave as soon as you're ready, and fly to an airfield near where you really want to go.
Bad in this context means (as my post said) a missed optimisation opportunity. Compilers should be doing every optimisation that can be done mechanically, programmers should be doing the ones that require knowledge of the problem domain, data, and so on. If a particular bit of code generates redundant operations, or does not manage to pattern match to the optimal instruction sequence, then we want to know. This is part of the reason that superoptimisers exist. If we don't have a set of test cases then it's hard to measure improvement (and no, SPEC is not even remotely representative of real code, but we use it anyway).
I'm not sure how true this really is. My mother lives in the south of France, I live in the UK. I looked at visiting her by taking the Eurostar to Paris and then the train from Paris to her, or by flying directly. The flight took about as long as the second leg of the train, which would also involve crossing Paris on the Metro (different stations). In terms of both travel time and speed, the flight was a better bet. Even getting from the north of France to the south can be cheaper and faster on a plane. Within the UK, the trains are stupidly expensive, to the point that it is often cheaper to fly than take the train. I did take the train to Edinburgh last year, but it worked out at about the same price as flying and took about an hour longer (including time spent getting to the airport and waiting) - I did it because the train is a bit more comfortable and I could get some work done on a long train trip, whereas a flight with lots of interruptions (train, airport, plane, bus) was much less convenient.
It always becomes a difficult sell once you factor in the travel time. Modern airship designs have an airspeed of up to about 50 knots. That's about a tenth of the speed of a jet and makes a transatlantic flight a week-long endeavour. A week on a large cruise ship isn't so bad, because there's a lot of stuff to do, but an airship by necessity doesn't have as much space. By the time that you've paid for food for a week, you're hard pressed to make it cheaper. For shorter overland hops, rail is likely to be 2-3 times the speed.
The main place where airships could come back is overland cargo flights. Speed doesn't matter too much, and ships can't compete because they have to go the long way around. One amusing idea was to produce Hydrogen from solar electricity in the middle of the Sahara, and use fuel-cell powered airships to transport it to Europe and Asia for feeding hydrogen-burning power plants...
Slashdot Mirror
Not really. We're not surprised that systems from 40 years ago are still able to do the things that they were designed to do, we're surprised that the requirements haven't morphed beyond all recognition in this time. If you spend three years developing a software system, and at the end of those three years your requirements look even remotely similar to the ones you started with, then these days you consider yourself very lucky. The idea that you could deploy a system and 40 years later your customer would still want the system to be doing the same thing sounds too good to be true.
Even good code starts to struggle after the requirements have changed completely for the 20th time...
I am always amazed how buildings constructed thousands, or even hundreds, of years ago are still standing although often in a state of disrepair due to neglect
That's what's known as survivor bias. The only examples you see of thousand-year-old buildings are the ones that didn't fall down. The ones that collapsed within a decade are long forgotten.
A modern-day castle might survive a century whereas the castles throughout Europe remain or at least remnants of their existence survive to this day
And yet, in the village where I grew up, and near countless other villages in Britain, there was a hill with a raised mound on top, which was the only remaining evidence of the castle that stood there 900 years ago.
North Korea, with its tiny allocation, is not exactly the bastion of well-secured machines. It's entirely plausible that a false flag operation launched some (likely trivial) part of the operation from a compromised machine in North Korea because they knew that as soon as the FBI found a North Korean IP in their traffic they'd stop bothering to look any further.
I thought they'd done that with KitKat. I don't have it on my phone (Moto G), but I don't really see it as a problem. I installed FireFox from F-Droid, disabled Chrome, and everything seems to work fine. I'd much rather have fewer things in the default install, as long as there are good replacements. I'm much more unhappy about the fact that they removed the character count from the SMS app and there's no open source version available.
I'd agree with your second point, but not the first. I really like Windows Phone's UI and structure, but it lacks several important bits of core functionality and (more importantly) lacks third-party apps to fill in the gaps. If Microsoft had managed to get developer mindshare earlier (not helped by breaking all existing Windows mobile apps), they'd be in a much better position.
Does it also come with UI regressions, like the change around 3-4 that turned putting the clock into night mode from a one-tap operation into a 4 tap sequence?
There's a weird trend in US hotels to give free WiFi if you're a member of their loyalty program. One hotel I stayed in gave members of their loyalty program free WiFi, wired Internet in the room, lounge access, cookies, and a few other perks, even if you joined the program when you booked the room and never intended on going to another one. I have a huge collection of 'loyalty' cards for hotel chains now, because I always pick the cheapest one.
+ $30-$75 is what I usually see quoted for the extra legroom seats, OR you can pick a seat near the back that is more likely to have an empty neighbor
Nearer the back means louder, which isn't great for long flights. If you fly regularly though, the extra legroom seats are often free. United gives them out to anyone who got more than 25K miles in the previous year.
The problem is the channels that carry so much advertising (especially shopping channels that are nothing but ads) that they pay to be included in bundles. This lowers the price of the bundle overall, in the hope that someone will watch some ads while channel surfing. No one is going to want to opt in to these and it wouldn't make sense in an a la carte model to allow people to opt into them for a 50 discount (or whatever), because most of the people who would subscribe would be the ones not in the target demographic, who just wanted the discount but would never watch the channel.
My mother recently moved to France and having seen her experience I think a lot of the rest of the world could benefit from adopting the French mode. Estate agents work solely for the buyer. There is no conflict of interest, they are paid to act as the buyer's agent and no one else's. As part of the justification for their fee, they accept a big chunk of liability for anything that is incorrectly represented to the buyer.
In this case, the porn owners are falsely claiming ownership of content on the Github pages.
No they aren't. There are two claims in a DMCA takedown notice. One is that you are authorised to act on behalf of the owners of copyright work A. The second is that work B infringes on work A. Only the first claim is made under penalty of perjury. There is a good reason for this: determining whether a work actually infringes (and is not covered by fair use and so on) requires a court ruling. The problem is that it's overly broad. I can write a DMCA notice that says that I own the copyright on this post (true) and that some other random site infringes on this copyright (false), and that's fine. The protection for this under the DMCA is that you can send a counter-notice and invalidate the original. The problem with this is that it's far easier to spam a load of take-down notices than for everyone to respond with counternotices.
If you want to get this fixed, send DMCA take-down notices to YouTube, Twitter, and Facebook for everything that every current member of Congress has online. Send notices to their web hosts. Do the same for every Republican and Democrat party web site. Knock both parties off the web until they fix their stupid laws.
Not sure about Canada, but in the USA the fines for infringement can be significantly larger for 'wilful infringement', which means that you must prove that the infringer knew that what they were doing was in violation of the relevant laws. If you distribute something that you were under the impression was in the public domain, or licensed in such a way that would permit distribution in the way that you are distributing it, then the penalties are typically lower. If you are an organisation responsible for telling people about copyright law and the stuff that you are infringing needed a subscription to access, then there's no defence against accusations of wilful infringement.
For evolution, beings should multiply
For Darwinian evolution, sure, but Lamarckian evolution is still somewhat plausible, even if we haven't seen any examples of it on Earth.
But if you saw one then would you recognise it? If they travel faster than light then you won't see them except when they're feeding. If they don't, then most of their life is likely to be in a dormant state as they spend a few thousand years between stars. Then there's the question of how they eat. If they eat the entire star at once, then you'll notice a star vanishing, but we don't have continuous observation on most stars, so there's a good chance that we'd see something odd in the data but not be able to tell what. If they eat in a more plausible way, then how would we tell it apart from, for example, a star near a superdense non-alive object that is drawing matter away from it?
No it isn't. There's a reason that AMD spun off their fab business. Fabs are insanely expensive and AMD doesn't have the volumes needed to justify a cutting edge fab by themselves. Having a fab running at anything below full production capacity is just throwing money away, but no one wants to buy capacity on an AMD fab because they know that when yields are low AMD chips will get priority. They'll buy capacity in a GlobalFoundaries or TMSC fab, because then they're just one customer in many.
The real question is why Intel won't fab the chips. Their commercial advantage has always been in production techniques, not in chip design. Throughout the lifetime of x86, other people have been able to compete with Intel designs, in spite of being one or two generations behind in process tech.
There are two universities in Brussels. Université Libre de Bruxelles (French for Free University of Brussels) and Vrije Universiteit Brussel (Dutch for Free University of Brussels). Translating the name of either into English makes it impossible to tell which institution he is a member of.
Yes, apparently I'm the one applying for a job at Verizon...
It is because research can never prove what the ROI will be.
That's not really true. You can look at a research lab and measure the ROI retrospectively quite easily and use this to make forward looking decisions, and that's what a lot of companies do. They'll close research labs that haven't produced anything useful in the last 5-10 years, but they'll increase funding to ones that have.
Google is one of the few companies that invests in products that might become useless.
No it isn't, it's just one of the few that labels them products and trumpets them in the press. Apple is about the only tech company that spends less on blue-sky research than Google. Companies like Microsoft and IBM spend vastly more. Aerospace companies would die within a decade if they didn't spend a significant amount on research and the same is true for a lot of other fields.
This is the same argument used for military R&D spending - there are lots of useful civilian results. The problem is that if you throw a big pile of R&D money at anything you'll likely get some useful results. The question is whether you get a good ROI. Compare NASA to, for example, Xerox PARC (Ethernet, the GUI, laser printers, etc.) or Bell Labs (the transistor, access control lists, UNIX, etc.) and see which produced more inventions that benefitted the economy as a whole per dollar spent.
Each shuttle launch cost, on average, $1.5bn. The cost of one launch would fund over ten thousand PhDs, or several hundred DARPA programs. Do you really think that NASA is the best ROI for taxpayers?
The population of the USA is about 320M, so that works out at 160M, or $1.6M. NASA's budget is actually around $17,647M, so you're off by three orders of magnitude. Do you, by any chance, work for Verizon? It's actually about $55 per person in the USA (including children).
Time spent in the air is generally wasted time
That certainly used to be true, but is it still? Even economy on the 787 has power sockets under the seats and enough space to use a 15" laptop. 8 hours of time when no one is interrupting you can be very productive. In business class, there's always power and a lot more space. Or, if you get an overnight flight, it's comfortable enough to sleep and arrive at your destination feeling refreshed. Would you really pay 2-4 times as much for a 7 hour flight in an economy seat than for a 14 hour flight in a first-class seat? Because that's the kind of price difference Concorde had...
Fuel and price are not the same thing. Busses don't cross the ocean very well, but boats do. The problem with the boat is that it takes a week or so to get from Europe to the USA, which means that you're having to pay for food for that trip on top of the ticket and you need a lot more space (no one would put up with an economy airline seat for a week!), which drives up the cost a lot. And that's ignoring the opportunity cost from the lost time.
You can fly from London to New York in about 7-8 hours. How much is it worth to double that speed? It will save you 4 hours, but the total transit time includes waiting at the airport and getting to and from the airport, which adds another 3-4 hours (at least - assuming that the airport is relatively close to your endpoint). So now it's taking around 8 hours instead of 12. Is that worth it? That's assuming that both planes leave at the same time though. Expensive flights tend to leave once a day, so it may not get to your destination any sooner, it just reduces travelling time.
It's definitely worth the money for a lot of people to get there in one day rather than seven, but you rapidly start to hit diminishing returns once the transit time is less than a day. At that point, being comfortable on the way is more important for a lot of people. If you've got money to burn, would you rather have an 8 hour flight in a first class seat which reclines to be completely horizontal, will have decent food and champagne, or be cramped into a Concorde seat that's closer to an economy seat on another airline for 3-4 hours?
If money really isn't an issue, then charter a private plane and you'll get there much sooner because it will fly from a smaller airfield near you, leave as soon as you're ready, and fly to an airfield near where you really want to go.
Bad in this context means (as my post said) a missed optimisation opportunity. Compilers should be doing every optimisation that can be done mechanically, programmers should be doing the ones that require knowledge of the problem domain, data, and so on. If a particular bit of code generates redundant operations, or does not manage to pattern match to the optimal instruction sequence, then we want to know. This is part of the reason that superoptimisers exist. If we don't have a set of test cases then it's hard to measure improvement (and no, SPEC is not even remotely representative of real code, but we use it anyway).
I'm not sure how true this really is. My mother lives in the south of France, I live in the UK. I looked at visiting her by taking the Eurostar to Paris and then the train from Paris to her, or by flying directly. The flight took about as long as the second leg of the train, which would also involve crossing Paris on the Metro (different stations). In terms of both travel time and speed, the flight was a better bet. Even getting from the north of France to the south can be cheaper and faster on a plane. Within the UK, the trains are stupidly expensive, to the point that it is often cheaper to fly than take the train. I did take the train to Edinburgh last year, but it worked out at about the same price as flying and took about an hour longer (including time spent getting to the airport and waiting) - I did it because the train is a bit more comfortable and I could get some work done on a long train trip, whereas a flight with lots of interruptions (train, airport, plane, bus) was much less convenient.
It always becomes a difficult sell once you factor in the travel time. Modern airship designs have an airspeed of up to about 50 knots. That's about a tenth of the speed of a jet and makes a transatlantic flight a week-long endeavour. A week on a large cruise ship isn't so bad, because there's a lot of stuff to do, but an airship by necessity doesn't have as much space. By the time that you've paid for food for a week, you're hard pressed to make it cheaper. For shorter overland hops, rail is likely to be 2-3 times the speed.
The main place where airships could come back is overland cargo flights. Speed doesn't matter too much, and ships can't compete because they have to go the long way around. One amusing idea was to produce Hydrogen from solar electricity in the middle of the Sahara, and use fuel-cell powered airships to transport it to Europe and Asia for feeding hydrogen-burning power plants...