Not your government. But right at the time when the Americans were bailing out the car companies Ford went to the Australian government for a handout threatening the closure of it's last manufacturing plant. It got it too to the tune of several billion AUD.... And then proceeded to close the factory anyway.
That is a drop in the bucket to what it would take to make them not annoyingly inconvenient to find.
Why would you be trying to find them? Your car will be fully charged at home. The only reason you would use one of these 400 stations is so you can drive internationally across Europe in an EV. You won't need to find them, they will all be highway stop locations.
So, 400 charging stations for all of Europe is supposed to impress me?
At the end of this you will be able to drive your EV from South of Spain to the north of Finland without range anxiety. If you're not impressed then maybe up your dose on the anti-depressants.
Lots of folks in Europe live, like I do, in an inner city. I am now privileged, and have an apartment with a garage, but for most folks, they just park their cars on the street, and have no method of charging there.
I live in Europe in the inner city. There are 3 charging points in my street alone and the number keep increasing every year.
As an ideal . . . I would like to see EV charging stations to be like gas stations are now: You can stop anywhere and tank up, and just pick a station that has a reasonable price. No proprietary connectors.
NO! That is a horrible waste of time idea that serves only to fit EVs to an outdated model of having to "fill up" a car while at the same time creating incredibly difficult to solve problems for the energy grid. Cars should be fully ready to go when you get in them. The only reason you should have to fill up on the go is for an international road trip.
These companies know this as well. The 400 charging stations won't be in any city but rather on major interstate and international highway routes.
I walk or take public transportation most of the time, so it is sort of a moot point for me.
My brother does too, he doesn't own a car but subscribes to a car sharing service. All the more reason that cars on the side of the road should be ready to go with a full tank at all times.
Gold is a noble metal, diamond is one of the hardest substances on the planet. Both have incredibly vast uses in high quantities.
I think what you meant was in large single sizes. There's little use for a single large diamond, but the same weight in diamond dust you'd think would be worth it's weight in... diamond. Whereas in fact it costs so little that we use it expendebly. Gold likewise finds its greatest utility in its smallest thickness rather than its biggest chunk.
Yes the report does identify that the horns may have helped avoid the collision, but that doesn't fundamentally change that human decisions and misunderstanding of the control system lead people to manually accidentally steer the ship into the collision. It *may* have mitigated the problem, but another supervisory control system when all others were disabled and actively fought against would unlikely have helped maintain control.
Of course... most of the internet traffic is going outside Australia
No it doesn't. The vast majority of the traffic is delivered by local data-centres and CDNs. In Australia it's exclusively the last mile which is utter garbage. 50 year old copper telephone cabling in complete disrepair, where it has been repaired it was done so by connectors which have been discovered to be corrosive, and long runs between nodes and houses such that even some apartments with 4km of the centre of the city are stuck with internet that can at best be described as third world.
Have you seen the state of Australian broadband? Your proposal will simply cause a new headline to run: "Australians taken shotgun to shitty Billion-Dollar Broadband"
What version of an OS was actually wanted? These aren't iPhones, or Androids where new version brings some fundamental functionality. Windows ultimately even with major version jumps are incredibly minor feature improvements over their predecessor on a platform open enough that any problem you have was likely already resolved by a 3rd party program you installed.
People don't get excited about basic UI, or some incremental changes unless there's something fundamentally missing in the previous system. Mobile phones are getting to that stage now too. I used to get excited about an Android release, or an iOS release. "OMG FINALLY it can copy and paste!" or similar such fundamental improvements. Nowadays,... meh.
Windows has been the same for a long time. If you're security conscious you'll upgrade when long term support expires. If not, you'll update with your new computer.
After reading many, many accident investigation reports, I would say that the FAA strongly disagrees with you. Nearly every report includes the phrase "pilot error". As in, "after the left wing fell off, the pilot failed to maintain coordinated flight..."
That's called a contributing factor, not a root cause. And just because something is done in the FAA doesn't make it necessarily right, especially when it comes to something like the human error of not understanding the display that was presented to them.
It's an interesting dichotomy given the FAA's long history of blaming the pilot while also having an even longer history of developing exceptionally good human interaction models that other industries look to keenly for guidance.
Read the report, there's nothing of the sort. The entire issue could have been resolved by one person knowing that he moved his steering command to another panel, and a follow up with another person knowing that his recently enabled steering and propulsion system had it's propulsion setup differently than expected.
The report itself chronicals if anything that the cooks and chefs all actually worked quite well together. From the moment steering was incorrectly assumed lost, to 45 seconds later the bridge being reconfigured to have steering independently controlled by someone elsewhere. There were absolutely no unreasonable delays between orders and execution at any stage.
There was however a great lack of situational awareness by 2 key people caused by their control system giving them garbage information.
What's the point? One less body on the bridge, saving a few bucks?
Actually this is more expensive. But the body on the bridge thing is interesting especially in a war type situation. A ship that can be crewed at a minimum is critical. But ultimately it wasn't even a case of the number of crew or redundant personnel, but rather redundant systems.
The entire control fuckup seems to be based on the idea that you could transfer control between two stations. This has been fairly standard in everything from large ship navigation to chemical plants, or even basic things like breweries since the 80s.
Why is this relevant? It wasn't a charting issue in the slightest. It was a control issue. The plan was just fine.
I'm more interested in who would implement a controller that: a) doesn't show information about if it is currently in control b) doesn't show information about who currently has control c) doesn't bumplessly transfer to another station but instead assumes you wanted to go straight ahead d) doesn't display if engine speed is ganged or not. e) assumes the default is that you want to control engine speed independently screwing up your steering instead of defaulting to ganged.
It wasn't the design of the controls which caused the problem, but rather if they were active on the station. If your follow up question is wouldn't it be better to only have controls in one place, let me answer with a resounding FUCK NO!
Yes but if you had more than one control, and one stopped working, wouldn't your first thought be to try the other one? I can't accept everybody just ran about waving their hands in the air shouting "we've lost steering control".
Yes and no. If you have 2 controls but they run through a common system, and your "we've lost steering control" strategy is to use an independent 3rd system then it is perfectly reasonable to assume that on loss of one control you jump straight to the air shouting strategy. Mind you that strategy is quite solid:
Look at the timeline: 0521 The Helm reported loss of steering to the Officer of the Deck. 0521:13 Steering units on the port rudder were shifted as ordered. 0521:15 Steering units on the starboard rudder were shifted as ordered 0521:55 The first watchstander reported to After Steering.
Remember that a shitty UI brought them into confusion in the first place. Their backup strategy was in place within 45 seconds. How long do you dedicate time to debugging your control loop on a shitty UI?
The problem wasn't the speed of switching to a backup, it was diagnosing the drift in the first place, compounded by the followup action that the engines weren't ganged. Reading through the report I get the opinion that the collision may have been avoided if the order wasn't given to reduce propulsion, or if that part of the order was executed correctly because only 10 seconds after steering control was taken by the offline units the course had already been largely corrected.
This! I have the option to work from home. I don't because of the inefficiencies that result. The number of problems that I've solved just at the coffee machine at work. Oh and the coffee is free there, that's the other reason I commute.
That is why a VCO is not used; instead, a VCXO is used. Then the benefit of high PLL bandwidth which would otherwise lower output jitter within the PLL bandwidth from the source clock is not required which is just as well because in this case the source clock has high jitter. So instead a low PLL bandwidth is used to control an oscillator which already has low jitter.
I used the term interchangeably. There is *NO* VCXO on the market with total jitter lower than the result of ASRC + a far cheaper fixed oscillator, and the floor performance of your PLL is decided by the bandwidth and the oscillator its referencing.
The problem with ASRC is that it is only suitable when integrated which in practice means already part of a device side ASIC. If implemented on the CPU, it requires too much power.
Not sure why that's relevant. It's basically always implemented int he former way.
There is a different reason a phased locked VCXO is not used. It limits the number of sample rates (although most could be covered with two crystals), cannot be integrated, takes up too much space and perhaps power in a small portable device, and most source material is so poor that it hardly matters.
It isn't any more limiting than the traditional approach (I didn't realise we were talking portable battery here since I haven't seen a battery powered S/PDIF device since the late 90s). DACs are definitely not short on space, and typical DACs already limit sample rates to 2x multiples of 24.567MHz clocks, though ultimately many implement clock switching between that and a secondary 22.5792MHz clock. Most DIR's provide a configurable digital output to show which clock it uses for this reason. Between those two you cover all audio sample rates except for the rarely used 132.3KHz and 144KHz which most DACs on the market don't support for precisely this reason.
but the narrative that they are wildly successful and overshadowing Apple's products seems a bit overstated.
I didn't say it was successful and overshadowing Apple. I said it was wildly successful to the point where everyone has adopted the form factor including a company which was dead set against it from the onset.
Surface revenues seem
Surface revenues are entirely irrelevant in the success of what the Surface has achieved: Moving the PC paradigm to successfully incorporate pen and touch across the board, incorporating new use cases for the Windows platform as they go. These efforts date back to Windows XP, and the Surface has now finally done what MS has been attempting to do for the best part of 20 years. Comparing them to hardware sales from Apple is utterly daft, especially since MS hasn't even given guarantees that it would stay in this hardware market once it was adopted by OEMs.
Maybe more recent quarters are substantially different, but it doesnt look like Apple should be panicing.
They aren't. Apple aren't panicking from MS. They were "panicking" from an industry widely adopting pen computing. And I use panic as a term quite loosely given their actual response: Release a "Pro" version of a tablet. Include the one device they swore would never be used on a tablet. Compare the tablet to a full blown computer in the marketing materials. Directly advertise the weaknesses of competitors products in the one feature they didn't want to include in the first place: the pen. They are simply putting in a half hearted: "Hey slate devices are dumb, but if you do want one we have something similar with an Apple logo"
Again how many times will MS get the complaint is the issue.
Yeah that's kind of what I wrote.
Again you are making multiple assumptions: 1) every device is still under warranty 2) that every vendor is available 24/7 for tech support (they are not) and 3) every person has unlimited time/opportunity/willingness to go the vendor.
No actually I'm not making a single one of those assumptions.
Please if you're going to "counter" my argument it helps if you at the very least read the text you're quoting at me if not the whole post.
Insufficient training(What Ars Technica neglects(possibly deliberately)
I would deliberately not mention that too. If you require specific training for something as simple as deciding which direction you're pointing and how fast you're getting there then there's something very VERY wrong with the interface.
The rest of what you talk about is a critical disaster management strategy. The fact that they were able to get to that position was the disaster initiator. You can't eliminate human error, so you don't consider it a true root cause of failure.
Everyone. The dumber question: who would know what good design looks like? That ultimately is the key problem here. Everyone is aware of the importance of design but there petty few that are actually aware of what "good" looks like.
Slashdot Mirror
Not your government. But right at the time when the Americans were bailing out the car companies Ford went to the Australian government for a handout threatening the closure of it's last manufacturing plant. It got it too to the tune of several billion AUD. ... And then proceeded to close the factory anyway.
Fuck them along with the other leeches.
That is a drop in the bucket to what it would take to make them not annoyingly inconvenient to find.
Why would you be trying to find them? Your car will be fully charged at home. The only reason you would use one of these 400 stations is so you can drive internationally across Europe in an EV. You won't need to find them, they will all be highway stop locations.
So, 400 charging stations for all of Europe is supposed to impress me?
At the end of this you will be able to drive your EV from South of Spain to the north of Finland without range anxiety. If you're not impressed then maybe up your dose on the anti-depressants.
Lots of folks in Europe live, like I do, in an inner city. I am now privileged, and have an apartment with a garage, but for most folks, they just park their cars on the street, and have no method of charging there.
I live in Europe in the inner city. There are 3 charging points in my street alone and the number keep increasing every year.
As an ideal . . . I would like to see EV charging stations to be like gas stations are now: You can stop anywhere and tank up, and just pick a station that has a reasonable price. No proprietary connectors.
NO! That is a horrible waste of time idea that serves only to fit EVs to an outdated model of having to "fill up" a car while at the same time creating incredibly difficult to solve problems for the energy grid. Cars should be fully ready to go when you get in them. The only reason you should have to fill up on the go is for an international road trip.
These companies know this as well. The 400 charging stations won't be in any city but rather on major interstate and international highway routes.
I walk or take public transportation most of the time, so it is sort of a moot point for me.
My brother does too, he doesn't own a car but subscribes to a car sharing service. All the more reason that cars on the side of the road should be ready to go with a full tank at all times.
which have very limited usage in large quantities
Gold is a noble metal, diamond is one of the hardest substances on the planet. Both have incredibly vast uses in high quantities.
I think what you meant was in large single sizes. There's little use for a single large diamond, but the same weight in diamond dust you'd think would be worth it's weight in ... diamond. Whereas in fact it costs so little that we use it expendebly. Gold likewise finds its greatest utility in its smallest thickness rather than its biggest chunk.
This is very, very impressive.
I have to agree. I happily shit on Apple every chance I get, but the main PCB sandwich gave me a huge metaphorical (and partially literal) stiffy.
hours waits for activation
I've never had hours waits for activation. I've never had minutes, seconds, milliseconds, or microseconds waits either.
Yes the report does identify that the horns may have helped avoid the collision, but that doesn't fundamentally change that human decisions and misunderstanding of the control system lead people to manually accidentally steer the ship into the collision. It *may* have mitigated the problem, but another supervisory control system when all others were disabled and actively fought against would unlikely have helped maintain control.
Of course... most of the internet traffic is going outside Australia
No it doesn't. The vast majority of the traffic is delivered by local data-centres and CDNs. In Australia it's exclusively the last mile which is utter garbage. 50 year old copper telephone cabling in complete disrepair, where it has been repaired it was done so by connectors which have been discovered to be corrosive, and long runs between nodes and houses such that even some apartments with 4km of the centre of the city are stuck with internet that can at best be described as third world.
Have you seen the state of Australian broadband? Your proposal will simply cause a new headline to run: "Australians taken shotgun to shitty Billion-Dollar Broadband"
I'd believe it only if you include the sum total of dodgy chinese ebay sellers in the stats. That obviously makes the entire assertion worthless.
and not because Windows 10 is actually wanted
What version of an OS was actually wanted? These aren't iPhones, or Androids where new version brings some fundamental functionality. Windows ultimately even with major version jumps are incredibly minor feature improvements over their predecessor on a platform open enough that any problem you have was likely already resolved by a 3rd party program you installed.
People don't get excited about basic UI, or some incremental changes unless there's something fundamentally missing in the previous system. Mobile phones are getting to that stage now too. I used to get excited about an Android release, or an iOS release. "OMG FINALLY it can copy and paste!" or similar such fundamental improvements. Nowadays, ... meh.
Windows has been the same for a long time. If you're security conscious you'll upgrade when long term support expires. If not, you'll update with your new computer.
After reading many, many accident investigation reports, I would say that the FAA strongly disagrees with you. Nearly every report includes the phrase "pilot error". As in, "after the left wing fell off, the pilot failed to maintain coordinated flight..."
That's called a contributing factor, not a root cause. And just because something is done in the FAA doesn't make it necessarily right, especially when it comes to something like the human error of not understanding the display that was presented to them.
It's an interesting dichotomy given the FAA's long history of blaming the pilot while also having an even longer history of developing exceptionally good human interaction models that other industries look to keenly for guidance.
Read the report, there's nothing of the sort. The entire issue could have been resolved by one person knowing that he moved his steering command to another panel, and a follow up with another person knowing that his recently enabled steering and propulsion system had it's propulsion setup differently than expected.
The report itself chronicals if anything that the cooks and chefs all actually worked quite well together. From the moment steering was incorrectly assumed lost, to 45 seconds later the bridge being reconfigured to have steering independently controlled by someone elsewhere. There were absolutely no unreasonable delays between orders and execution at any stage.
There was however a great lack of situational awareness by 2 key people caused by their control system giving them garbage information.
What's the point? One less body on the bridge, saving a few bucks?
Actually this is more expensive. But the body on the bridge thing is interesting especially in a war type situation. A ship that can be crewed at a minimum is critical. But ultimately it wasn't even a case of the number of crew or redundant personnel, but rather redundant systems.
The entire control fuckup seems to be based on the idea that you could transfer control between two stations. This has been fairly standard in everything from large ship navigation to chemical plants, or even basic things like breweries since the 80s.
Why is this relevant? It wasn't a charting issue in the slightest. It was a control issue. The plan was just fine.
I'm more interested in who would implement a controller that:
a) doesn't show information about if it is currently in control
b) doesn't show information about who currently has control
c) doesn't bumplessly transfer to another station but instead assumes you wanted to go straight ahead
d) doesn't display if engine speed is ganged or not.
e) assumes the default is that you want to control engine speed independently screwing up your steering instead of defaulting to ganged.
It wasn't the design of the controls which caused the problem, but rather if they were active on the station.
If your follow up question is wouldn't it be better to only have controls in one place, let me answer with a resounding FUCK NO!
Makes a pretty good UI.
Makes for a horrible UI in a computer aided control environment.
Yes but if you had more than one control, and one stopped working, wouldn't your first thought be to try the other one?
I can't accept everybody just ran about waving their hands in the air shouting "we've lost steering control".
Yes and no. If you have 2 controls but they run through a common system, and your "we've lost steering control" strategy is to use an independent 3rd system then it is perfectly reasonable to assume that on loss of one control you jump straight to the air shouting strategy. Mind you that strategy is quite solid:
Look at the timeline:
0521 The Helm reported loss of steering to the Officer of the Deck.
0521:13 Steering units on the port rudder were shifted as ordered.
0521:15 Steering units on the starboard rudder were shifted as ordered
0521:55 The first watchstander reported to After Steering.
Remember that a shitty UI brought them into confusion in the first place. Their backup strategy was in place within 45 seconds. How long do you dedicate time to debugging your control loop on a shitty UI?
The problem wasn't the speed of switching to a backup, it was diagnosing the drift in the first place, compounded by the followup action that the engines weren't ganged. Reading through the report I get the opinion that the collision may have been avoided if the order wasn't given to reduce propulsion, or if that part of the order was executed correctly because only 10 seconds after steering control was taken by the offline units the course had already been largely corrected.
This! I have the option to work from home. I don't because of the inefficiencies that result. The number of problems that I've solved just at the coffee machine at work. Oh and the coffee is free there, that's the other reason I commute.
This seems like a bad precedent to set
No this seems like a natural consequence. The bad precedent was set when education became unaffordable in the first place.
That is why a VCO is not used; instead, a VCXO is used. Then the benefit of high PLL bandwidth which would otherwise lower output jitter within the PLL bandwidth from the source clock is not required which is just as well because in this case the source clock has high jitter. So instead a low PLL bandwidth is used to control an oscillator which already has low jitter.
I used the term interchangeably. There is *NO* VCXO on the market with total jitter lower than the result of ASRC + a far cheaper fixed oscillator, and the floor performance of your PLL is decided by the bandwidth and the oscillator its referencing.
The problem with ASRC is that it is only suitable when integrated which in practice means already part of a device side ASIC. If implemented on the CPU, it requires too much power.
Not sure why that's relevant. It's basically always implemented int he former way.
There is a different reason a phased locked VCXO is not used. It limits the number of sample rates (although most could be covered with two crystals), cannot be integrated, takes up too much space and perhaps power in a small portable device, and most source material is so poor that it hardly matters.
It isn't any more limiting than the traditional approach (I didn't realise we were talking portable battery here since I haven't seen a battery powered S/PDIF device since the late 90s). DACs are definitely not short on space, and typical DACs already limit sample rates to 2x multiples of 24.567MHz clocks, though ultimately many implement clock switching between that and a secondary 22.5792MHz clock. Most DIR's provide a configurable digital output to show which clock it uses for this reason. Between those two you cover all audio sample rates except for the rarely used 132.3KHz and 144KHz which most DACs on the market don't support for precisely this reason.
but the narrative that they are wildly successful and overshadowing Apple's products seems a bit overstated.
I didn't say it was successful and overshadowing Apple. I said it was wildly successful to the point where everyone has adopted the form factor including a company which was dead set against it from the onset.
Surface revenues seem
Surface revenues are entirely irrelevant in the success of what the Surface has achieved: Moving the PC paradigm to successfully incorporate pen and touch across the board, incorporating new use cases for the Windows platform as they go. These efforts date back to Windows XP, and the Surface has now finally done what MS has been attempting to do for the best part of 20 years. Comparing them to hardware sales from Apple is utterly daft, especially since MS hasn't even given guarantees that it would stay in this hardware market once it was adopted by OEMs.
Maybe more recent quarters are substantially different, but it doesnt look like Apple should be panicing.
They aren't. Apple aren't panicking from MS. They were "panicking" from an industry widely adopting pen computing. And I use panic as a term quite loosely given their actual response: Release a "Pro" version of a tablet. Include the one device they swore would never be used on a tablet. Compare the tablet to a full blown computer in the marketing materials. Directly advertise the weaknesses of competitors products in the one feature they didn't want to include in the first place: the pen. They are simply putting in a half hearted: "Hey slate devices are dumb, but if you do want one we have something similar with an Apple logo"
Again you are assuming.
Yeah that's kind of what I wrote.
Again how many times will MS get the complaint is the issue.
Yeah that's kind of what I wrote.
Again you are making multiple assumptions: 1) every device is still under warranty 2) that every vendor is available 24/7 for tech support (they are not) and 3) every person has unlimited time/opportunity/willingness to go the vendor.
No actually I'm not making a single one of those assumptions.
Please if you're going to "counter" my argument it helps if you at the very least read the text you're quoting at me if not the whole post.
Insufficient training(What Ars Technica neglects(possibly deliberately)
I would deliberately not mention that too. If you require specific training for something as simple as deciding which direction you're pointing and how fast you're getting there then there's something very VERY wrong with the interface.
The rest of what you talk about is a critical disaster management strategy. The fact that they were able to get to that position was the disaster initiator. You can't eliminate human error, so you don't consider it a true root cause of failure.
Who would have expected not having a good design
Everyone. The dumber question: who would know what good design looks like? That ultimately is the key problem here. Everyone is aware of the importance of design but there petty few that are actually aware of what "good" looks like.