Yeah, but the slowdown in growth will almost certainly reflect in a slowdown in share-price growth (which is normal). So Amazon will have to consider paying a divided which they currently don't.
I disagree. I think the aircraft is quite flyable and can safely be flown if the pilots are properly trained and skilled. All aircraft in this configuration have this pitch up tendency when power is applied, just based on physics.
If you read the analysis of the 737MAX the difference between it and the 737 is that the wings were moved forward and up on the design to accomomdate larger engines. This had the result of making the MAX aerodynamically unstable where the 737 was stable. To compensate for the instability so the MAX would fly like the 737 they added MCAS which adjusted trim quote aggressively based on two AoA sensors, and if the AoA sensors disagreed, it just picked one.
They did it this way to avoid the cost of designing and recertifying a new aircraft in that segment, where Airbus was eating their lunch at the time. To preserve the illusion that it was the same as the 737, but better, the decision was to not document MCAS in the Flight Operations Manual, and to not require retraining.
So there was the engineering design flaw and the systems flaw. Arguably the crew on the LionAir flight should have been able to resolve it - apparently the checklist for this time of issue has the stabilization trim procedure that would have resolved the issue in it - but the plane was sold as flying the same and handling the same as the 737 and that was a lie.
Whereas complicated, toxic, difficult to recycle products like electronic gadgets might see huge price increases. Those huge price increases will motivate producers to build less toxic, longer lasting, easier to recycle, easier to repair products.
This already happens in Europe, with the WEEE (Waste Electrical and Electronic Equipment) directive. Every electronic device has a charge associated with it and built into the price that means it can be dropped off at municipal centers at no cost when it's time to recycle. In Ireland, if a company delivers an electronic device they're obliged to remove the original if requested - useful if you're replacing a large item like a fridge. It works fairly well.
I'm pointing out that this is hardly slam-dunk on the SEC's side. I think it will be very difficult to get a judge to agree that he violated the "material information" clause given the previous investor call. Material information has a specific definition and him saying "350-500K" in the investor call, then causally saying "around 500k" on a tweet might not breach it. It almost certainly won't because the more specific information in the investor call would be more material than the tweet and occurred 2 weeks earlier. We'll see.
That's basically his point. Unless you specifically need that, then blockchain is probably not the technology you need. So (for instance) in the case of supply chain management, blockchain equivalently useful to the USPS "Your thing was here, at this time" tracking. It does not add anything new or better.
In fairness, they're claiming he broke the agreement, and he disagrees.
His counterpoint is that the number he tweeted was already available in the earnings report and thus could not move the stock price. This is reasonable - the SEC cannot expect him to not publicize Tesla's successes, and given this was public information for investors already it seems unlikely to move the stock price.
I see the misunderstanding - the Fn key was moved from its previous position between Ctrl and the Windows/Meta key. This is (as far as I've ever experienced it) the standard location for Fn so no-one would have expected it there until Lenovo moved it, between 2 and 3 years ago. As you point out, it enables extra functions on the top level of keys (depending on the laptop it could be the function keys, number keys, or both) so moving it made that secondary (or tertiary) function easily accesible, sure. Functions like volume, or switch screen, or brightness, as opposed to functions like cut, copy and paste? I find it difficult to believe that the Fn activated functions are used more than the standard windows cut/copy/paste actions.
I had a Lenovo Ideapad for personal use previously, but when switching from a desktop to a laptop in work, I decided I'd prefer the Dell. The keyboard was a large part of the decision.
Rearranging keys like this is suprisingly destructive of your ability to do things without looking at the keyboard, which is a context switch. There's no good reason to move the Fn key, and lots of good reasons to have the primary meta key in one of the easiest to recognise without looking places on the keyboard.
It wasn't just this (and I did exaggerate a bit - I did not return the thinkpad, I refused it in the first place in favour of the precision) but the quality of the Lenovo keyboards dropped noticeably over the last couple of years.
Most albums were 10-15 as I recall. Really popular bands would have fancy artwork and music videos and that was 19.99. Paying 30 for an album meant you were getting the one with the t-shirt and poster.
It's a fair question. In this link 1997 music sales in the US were around 750 million units, in 1996 that was around 780 million and in 1998 it was around 850 million which averages out at less than 3 albums per person. According to this the peak was 953 million units per year. The estimated total sales figures from the first link give approximately $13 per CD, which matches my recollections of those long ago days:) Those figures are for the US, Europe and rest of world was lower (I am not in the US so my recollection will be different).
I don't have data on average CD's-purchased-per-person as that does not seem to be available (it's all units sold and revenue) but if we do a rough exercise and say that the population of the US was 282.2 million in 2000 and that half the population weren't buying music at all, that leaves us with an average of 6 albums per person for around 80 dollars per year, or 6 dollars per month. You were clearly a music aficionado if you were spending way more than 120 per year on CDs!
My point really is that they can either go for quantity or quality. At $5 or $6 a month they will get more people than at $10 or $14. Only they can say if the numbers will add up.
It doesn't matter until the pricing model improves. Before streaming, the majority of people did not spend $120 a year on albums, which would be roughly 12 albums a year. Some of the streaming services are doing ok now, but until the price point drops to somewhere between $5 and $7 a month and until the messy regional licensing stuff gets sorted out, they're all going to stagnate.
They have to choose - lose the extra revenue for the real music buffs or get everyone onboard and earn less per subscriber, but for more people. Maybe charge extra for high fidelity recordings. I was making the point to someone recently that Netflix is competing against cable minus sports, so it looks competitive because it's less than a cable subscription. The music streamers are competing against radio, which is free at the point of consumption, so it looks comparatively less competitive.
hifting the cost of fraud away from them and onto you was one of the âoeselling pointsâ that got US card issuers to finally embrace this âoeupgradeâ.
No, it wasn't. There was no selling point. MasterCard, Amex, Diners and Visa told them that if they did not use chip, they were liable for fraud. Simple as that. So the rule now is, if the card is not chip enabled, the Issuer (your bank) are liable. If the card is chip enabled but the POS does not support chip, the merchant is liable. That was what liability shift meant - it did not shift liability to the consumer.
I'd be interested in the GP's understanding of what actually happened. There's only been one practical breach of EMV enabled card system, and it involved a highly technical operation precisely placing a dummy chip over the original chip in the card that flipped a couple of bits and indicated that the terminal had verified the PIN offline. It was quickly remedied without any hardware change. Any other fraud has been either using the magstripe, or card-not-present (like online).
I am using "OA" as Open Access, where you seem to have misread it as QA.
You are right that it does provide a handy rule of thumb, but the funding authorities will have to figure that out - after all, they have to allocate their funding.
there is little productivity difference between a jr accountant with 3 years experience vs senior accountant with 10 years experience
That's exactly the same as someone telling you there's no difference between a programmer with 3 vs 10 years experience. You build experience over time and a 10 year experienced accountant is likely be to be able to pick up something new in their domain much quicker than a 3 year experienced one.
The problem for me is not Finance, it's management not understanding the value that people bring and looking at everything as a cost. The company's annual budget is (or should be) a give and take between inputs and outputs and it's up to the group leadership to advocate for more of the pie. If they can't, that's their failing (and the failing of executive leadership also if they're letting the rot set in).
If a chemical process works on Mars it will almost certainly work here. It's also possible that this is Mars-focused to avoid the inevitable political wrangling if it was directly aimed at climate change.
I'm not stuck on C/C++ except that when you're learning there is a progress that I personally feel is easier using C++ than any other teaching language.
A language that abstracts away memory management is not a good teaching tool for a computer science student because that is precisely what they should be learning about and experience always beats theory. Reading about memory corruption/underflow/overflow is different from experiencing it. Python, Java, Lua, ECMAScript, Erlang, Prolog, Fortran, BASIC, C, C++, Rust, PERL - they're all tools with strengths and weaknesses. I'm not advocating that one is "better" than the other in general, they all have applications that suit them or not. But for teaching I strongly believe that you have to start with a language that does not abstract too much away because the experience of working with Java gives a different perspective than working with other languages and I believe that C++ allows students to experience a wider range of problems.
They'll end up learning Java, and Rust, and ECMAScript and Python, but for the same reason I would not recommend teaching programming with loosely typed language, I think that memory management abstraction is not a positive. If nothing else, you end up appreciating the convenience later!
You are correct at the moment. But the review is important for one reason - funding. If the funding authorities have made a decision that they require OA, then they will not punish research groups for not publishing in journals that do not conform to their standard.
I expect what they will do is drastically downgrade the ratings of the journals that do not conform, so that if Nature was #1 ranked, but its publishing rules do not support OA, then they will be downgraded to #3 or even be a negative. So funding will not be dependent on publication in those journals and that incentive is removed.
That is actually for many people not easy. It is much simpler to start with objects/classes right away.
Well, yeah, except that OO is not the only way to program. And the goal of CS is not to teach programming, but to teach underlying principles which Java makes more difficult by abstracting away. The most active areas of CS research are basically all about resource management and timing. Java hides all of that.
Elsevier and the other companies are not paying researchers or research assistants. It's the actual funding authorities who are paying those salaries who are clamping down on for-profit publishers.
The mistake people in this thread are making is in assuming that for-profit publishing is providing a financial benefit to researchers. Actually, it provides no financial benefit to researchers, and they have to pay to access the research of others so it's a financial negative. They provide a service in curating, storing and indexing the papers as well as (obviously) publishing them. But their role is not necessarily as important as it was previously because it's cheaper and easier to publish these days.
Java is not a good teaching language. In particular, when learning to program, having a garbage collector is not a great idea because it abstracts away one of the most important underlying details of programming, which is memory management. This is the basic type of resource management and the aim of teaching programming in university should be the application of underlying principles over the syntax of a particular language. Java tries to prevent you doing too many things to make a good teaching language.
Also it enforces an object orientated view of programming which is not the only way to program.
I am disappointed at the move away from C/C++ as a teaching tool. It was easy to start with C for imperative programming, then easy to move on to C++ for object orientation. Java is a useful tool and good to learn as a follow on (the syntax is fairly close and it's in many ways simpler than C++ so it's easier to learn once you have the basics) and having it as a follow on allows greater understanding of the limits it imposes and also why they were imposed - you'll definitely have run into some C++ bugs that you simply can't do in Java when you're learning. That's a great springboard to the wider world of programming languages and their different philosophies, strengths and weaknesses.
He doesn't say anything like that. In fact he says
I don't regret any design decisions, but if I was designing it now I would do everything slightly differently
This is reasonable - the concepts are largely good, but some of the design decisions did not age well and the processor designs we have now make some of those decisions quaint. Other things, like being constantly under attack, were not on anyone's mind when C++ was first released
You mean Israel who still do not acknowledge that they have nuclear weapons are complaining about Iran not being completely clear about their nuclear program before the deal? All that information in Bibi's presentation was old hat - from around 2008 I believe.
And look, building a nuclear bomb is not actually difficult. A chap in the US recreated the Nagasaki bomb perfectly from file photos, in his backyard. The hard part is enriching the uranium and the equipment to do that is not portable. The radiation is not easy to mask, either, and does not dissipate quickly. So military sites can be inspected with a 3 week lead time, but that's not enough to move a centrifuge or for evidence of high radiation levels to disappear.
And Iran do not want a war. Their current adult generation is the one that lived through a time where the oldest class at school were pulled out, given guns, and sent to the front in the Iran-Iraq war, then next month the next oldest class were sent - a war where the US backed the Iraqi invasion and helped the Iraqis deploy chemical weapons. They do not want to do that again and their leadership does not trust the US at all. This is not going to help that situation at all
Frankly, after the Russian annexation of Crimea and meddling in Donbass, no country in the world is ever going to give up nuclear weapons again. Ukraine did it in exchange for security guarantees from the UK, US and Russia - Russia invaded and annexed territory and the other signatories shrugged and said "Oh, it was a memorandum, not a treaty so, you know, tough luck". Now NK have nukes and are being treated like real boys. What message is that sending out?
Actually, 25% of their European workforce is in Ireland, with 4,000 direct employees. Small by US standards, but given the entire population is 5 million-ish, it's nearly 0.1% of the population....
Slashdot Mirror
Yeah, but the slowdown in growth will almost certainly reflect in a slowdown in share-price growth (which is normal). So Amazon will have to consider paying a divided which they currently don't.
I disagree. I think the aircraft is quite flyable and can safely be flown if the pilots are properly trained and skilled. All aircraft in this configuration have this pitch up tendency when power is applied, just based on physics.
If you read the analysis of the 737MAX the difference between it and the 737 is that the wings were moved forward and up on the design to accomomdate larger engines. This had the result of making the MAX aerodynamically unstable where the 737 was stable. To compensate for the instability so the MAX would fly like the 737 they added MCAS which adjusted trim quote aggressively based on two AoA sensors, and if the AoA sensors disagreed, it just picked one.
They did it this way to avoid the cost of designing and recertifying a new aircraft in that segment, where Airbus was eating their lunch at the time. To preserve the illusion that it was the same as the 737, but better, the decision was to not document MCAS in the Flight Operations Manual, and to not require retraining.
So there was the engineering design flaw and the systems flaw. Arguably the crew on the LionAir flight should have been able to resolve it - apparently the checklist for this time of issue has the stabilization trim procedure that would have resolved the issue in it - but the plane was sold as flying the same and handling the same as the 737 and that was a lie.
Whereas complicated, toxic, difficult to recycle products like electronic gadgets might see huge price increases. Those huge price increases will motivate producers to build less toxic, longer lasting, easier to recycle, easier to repair products.
This already happens in Europe, with the WEEE (Waste Electrical and Electronic Equipment) directive. Every electronic device has a charge associated with it and built into the price that means it can be dropped off at municipal centers at no cost when it's time to recycle. In Ireland, if a company delivers an electronic device they're obliged to remove the original if requested - useful if you're replacing a large item like a fridge. It works fairly well.
I'm pointing out that this is hardly slam-dunk on the SEC's side. I think it will be very difficult to get a judge to agree that he violated the "material information" clause given the previous investor call. Material information has a specific definition and him saying "350-500K" in the investor call, then causally saying "around 500k" on a tweet might not breach it. It almost certainly won't because the more specific information in the investor call would be more material than the tweet and occurred 2 weeks earlier. We'll see.
That's basically his point. Unless you specifically need that, then blockchain is probably not the technology you need. So (for instance) in the case of supply chain management, blockchain equivalently useful to the USPS "Your thing was here, at this time" tracking. It does not add anything new or better.
In fairness, they're claiming he broke the agreement, and he disagrees.
His counterpoint is that the number he tweeted was already available in the earnings report and thus could not move the stock price. This is reasonable - the SEC cannot expect him to not publicize Tesla's successes, and given this was public information for investors already it seems unlikely to move the stock price.
I see the misunderstanding - the Fn key was moved from its previous position between Ctrl and the Windows/Meta key. This is (as far as I've ever experienced it) the standard location for Fn so no-one would have expected it there until Lenovo moved it, between 2 and 3 years ago.
As you point out, it enables extra functions on the top level of keys (depending on the laptop it could be the function keys, number keys, or both) so moving it made that secondary (or tertiary) function easily accesible, sure. Functions like volume, or switch screen, or brightness, as opposed to functions like cut, copy and paste? I find it difficult to believe that the Fn activated functions are used more than the standard windows cut/copy/paste actions.
I had a Lenovo Ideapad for personal use previously, but when switching from a desktop to a laptop in work, I decided I'd prefer the Dell. The keyboard was a large part of the decision.
Rearranging keys like this is suprisingly destructive of your ability to do things without looking at the keyboard, which is a context switch. There's no good reason to move the Fn key, and lots of good reasons to have the primary meta key in one of the easiest to recognise without looking places on the keyboard.
It wasn't just this (and I did exaggerate a bit - I did not return the thinkpad, I refused it in the first place in favour of the precision) but the quality of the Lenovo keyboards dropped noticeably over the last couple of years.
Most albums were 10-15 as I recall. Really popular bands would have fancy artwork and music videos and that was 19.99. Paying 30 for an album meant you were getting the one with the t-shirt and poster.
It's a fair question. In this link 1997 music sales in the US were around 750 million units, in 1996 that was around 780 million and in 1998 it was around 850 million which averages out at less than 3 albums per person. According to this the peak was 953 million units per year. The estimated total sales figures from the first link give approximately $13 per CD, which matches my recollections of those long ago days :)
Those figures are for the US, Europe and rest of world was lower (I am not in the US so my recollection will be different).
I don't have data on average CD's-purchased-per-person as that does not seem to be available (it's all units sold and revenue) but if we do a rough exercise and say that the population of the US was 282.2 million in 2000 and that half the population weren't buying music at all, that leaves us with an average of 6 albums per person for around 80 dollars per year, or 6 dollars per month.
You were clearly a music aficionado if you were spending way more than 120 per year on CDs!
My point really is that they can either go for quantity or quality. At $5 or $6 a month they will get more people than at $10 or $14. Only they can say if the numbers will add up.
It doesn't matter until the pricing model improves. Before streaming, the majority of people did not spend $120 a year on albums, which would be roughly 12 albums a year. Some of the streaming services are doing ok now, but until the price point drops to somewhere between $5 and $7 a month and until the messy regional licensing stuff gets sorted out, they're all going to stagnate.
They have to choose - lose the extra revenue for the real music buffs or get everyone onboard and earn less per subscriber, but for more people. Maybe charge extra for high fidelity recordings. I was making the point to someone recently that Netflix is competing against cable minus sports, so it looks competitive because it's less than a cable subscription. The music streamers are competing against radio, which is free at the point of consumption, so it looks comparatively less competitive.
On the Thinkpad they moved Fn to the outside on the left, where Ctrl used to be.
I returned the Thinkpad to IT and got a Dell Precision instead.
hifting the cost of fraud away from them and onto you was one of the âoeselling pointsâ that got US card issuers to finally embrace this âoeupgradeâ.
No, it wasn't. There was no selling point. MasterCard, Amex, Diners and Visa told them that if they did not use chip, they were liable for fraud. Simple as that. So the rule now is, if the card is not chip enabled, the Issuer (your bank) are liable. If the card is chip enabled but the POS does not support chip, the merchant is liable. That was what liability shift meant - it did not shift liability to the consumer.
I'd be interested in the GP's understanding of what actually happened. There's only been one practical breach of EMV enabled card system, and it involved a highly technical operation precisely placing a dummy chip over the original chip in the card that flipped a couple of bits and indicated that the terminal had verified the PIN offline. It was quickly remedied without any hardware change. Any other fraud has been either using the magstripe, or card-not-present (like online).
I am using "OA" as Open Access, where you seem to have misread it as QA.
You are right that it does provide a handy rule of thumb, but the funding authorities will have to figure that out - after all, they have to allocate their funding.
there is little productivity difference between a jr accountant with 3 years experience vs senior accountant with 10 years experience
That's exactly the same as someone telling you there's no difference between a programmer with 3 vs 10 years experience. You build experience over time and a 10 year experienced accountant is likely be to be able to pick up something new in their domain much quicker than a 3 year experienced one.
The problem for me is not Finance, it's management not understanding the value that people bring and looking at everything as a cost. The company's annual budget is (or should be) a give and take between inputs and outputs and it's up to the group leadership to advocate for more of the pie. If they can't, that's their failing (and the failing of executive leadership also if they're letting the rot set in).
If a chemical process works on Mars it will almost certainly work here. It's also possible that this is Mars-focused to avoid the inevitable political wrangling if it was directly aimed at climate change.
I'm not stuck on C/C++ except that when you're learning there is a progress that I personally feel is easier using C++ than any other teaching language.
A language that abstracts away memory management is not a good teaching tool for a computer science student because that is precisely what they should be learning about and experience always beats theory. Reading about memory corruption/underflow/overflow is different from experiencing it. Python, Java, Lua, ECMAScript, Erlang, Prolog, Fortran, BASIC, C, C++, Rust, PERL - they're all tools with strengths and weaknesses. I'm not advocating that one is "better" than the other in general, they all have applications that suit them or not. But for teaching I strongly believe that you have to start with a language that does not abstract too much away because the experience of working with Java gives a different perspective than working with other languages and I believe that C++ allows students to experience a wider range of problems.
They'll end up learning Java, and Rust, and ECMAScript and Python, but for the same reason I would not recommend teaching programming with loosely typed language, I think that memory management abstraction is not a positive. If nothing else, you end up appreciating the convenience later!
You are correct at the moment. But the review is important for one reason - funding. If the funding authorities have made a decision that they require OA, then they will not punish research groups for not publishing in journals that do not conform to their standard.
I expect what they will do is drastically downgrade the ratings of the journals that do not conform, so that if Nature was #1 ranked, but its publishing rules do not support OA, then they will be downgraded to #3 or even be a negative. So funding will not be dependent on publication in those journals and that incentive is removed.
That is actually for many people not easy. It is much simpler to start with objects/classes right away.
Well, yeah, except that OO is not the only way to program. And the goal of CS is not to teach programming, but to teach underlying principles which Java makes more difficult by abstracting away. The most active areas of CS research are basically all about resource management and timing. Java hides all of that.
Elsevier and the other companies are not paying researchers or research assistants. It's the actual funding authorities who are paying those salaries who are clamping down on for-profit publishers.
The mistake people in this thread are making is in assuming that for-profit publishing is providing a financial benefit to researchers. Actually, it provides no financial benefit to researchers, and they have to pay to access the research of others so it's a financial negative. They provide a service in curating, storing and indexing the papers as well as (obviously) publishing them. But their role is not necessarily as important as it was previously because it's cheaper and easier to publish these days.
Java is not a good teaching language. In particular, when learning to program, having a garbage collector is not a great idea because it abstracts away one of the most important underlying details of programming, which is memory management. This is the basic type of resource management and the aim of teaching programming in university should be the application of underlying principles over the syntax of a particular language. Java tries to prevent you doing too many things to make a good teaching language.
Also it enforces an object orientated view of programming which is not the only way to program.
I am disappointed at the move away from C/C++ as a teaching tool. It was easy to start with C for imperative programming, then easy to move on to C++ for object orientation. Java is a useful tool and good to learn as a follow on (the syntax is fairly close and it's in many ways simpler than C++ so it's easier to learn once you have the basics) and having it as a follow on allows greater understanding of the limits it imposes and also why they were imposed - you'll definitely have run into some C++ bugs that you simply can't do in Java when you're learning. That's a great springboard to the wider world of programming languages and their different philosophies, strengths and weaknesses.
He doesn't say anything like that. In fact he says
I don't regret any design decisions, but if I was designing it now I would do everything slightly differently
This is reasonable - the concepts are largely good, but some of the design decisions did not age well and the processor designs we have now make some of those decisions quaint. Other things, like being constantly under attack, were not on anyone's mind when C++ was first released
Sounds more like something President Trump would say, doesn't it?!
You mean Israel who still do not acknowledge that they have nuclear weapons are complaining about Iran not being completely clear about their nuclear program before the deal? All that information in Bibi's presentation was old hat - from around 2008 I believe.
And look, building a nuclear bomb is not actually difficult. A chap in the US recreated the Nagasaki bomb perfectly from file photos, in his backyard. The hard part is enriching the uranium and the equipment to do that is not portable. The radiation is not easy to mask, either, and does not dissipate quickly. So military sites can be inspected with a 3 week lead time, but that's not enough to move a centrifuge or for evidence of high radiation levels to disappear.
And Iran do not want a war. Their current adult generation is the one that lived through a time where the oldest class at school were pulled out, given guns, and sent to the front in the Iran-Iraq war, then next month the next oldest class were sent - a war where the US backed the Iraqi invasion and helped the Iraqis deploy chemical weapons. They do not want to do that again and their leadership does not trust the US at all. This is not going to help that situation at all
Frankly, after the Russian annexation of Crimea and meddling in Donbass, no country in the world is ever going to give up nuclear weapons again. Ukraine did it in exchange for security guarantees from the UK, US and Russia - Russia invaded and annexed territory and the other signatories shrugged and said "Oh, it was a memorandum, not a treaty so, you know, tough luck". Now NK have nukes and are being treated like real boys. What message is that sending out?
Actually, 25% of their European workforce is in Ireland, with 4,000 direct employees. Small by US standards, but given the entire population is 5 million-ish, it's nearly 0.1% of the population....