Not sure about Switzerland (which is not in EU) but in Poland (and from what I heard, most other countries) you have a range of plans for very moderate prices.
Yes, all-inclusive unlimited is expensive like hell, but you'd better need to spend 12h a day on the phone if you really need it. This is not a plan for anyone who wants something above the minimum, like it is done in the US. This is for heavy abusers of the phone. There are extra services of unlimited calls to one number within the same network (or landline), but universal unlimited is just a disused option for rich lunatics.
Actual plans range from $6/mo for 40 minutes of calls included in the price (and about $0.15/minute above that limit) to some $150/mo with 15 hours of calls free, and $0.08 above the limit. This is for individual offers. Companies may get much better offers if they sign a contract for a bulk number of phones.
The plans often differ in "flat rate for all days, all hours" vs "cheap evenings, expensive business hours", "flat rate to all operators" vs varied pricing depending on which operator you call, extra services like data transfer in packets starting with 50MB and ending with 50GB of transfer for various prices (and either throttling or entering default no-premium-service pricing upon exhausting the packet), varied extra SMS packages and so on. Still, you never pay for incoming calls.
Prepaid is more expensive and you don't get many of the premium services. But interestingly, in some prepaid plans YOU GET PAID for incoming calls. Not with money but with outgoing calls credit - like, for 3 minutes of incoming call you get one minute of outgoing calls credit. Simply, the companies have deals signed and the caller's company pays the receiver's company a share from the caller's fees. And since it's still a valid source of income, you don't even have to pay for maintaining your phone number as long as you receive enough calls - your callers will pay for it.
Currently, I just use the blacklist feature of my Android phone. It works well - the spammers usually don't have too many numbers to block, and occasionally receiving an undesired call is not a tragedy.
Now I wonder if it would be viable to get an "automatic service" tarpit app for whoever calls. "If you are calling concerning business, press 1. If you're calling concerning family matters, press 2 and state your business. If you are a friend or colleague, press # and enter your access code."
Robots would quickly get disconnected for talking to the hand (actually, continuous talking while a message is playing could be taken as a hint to disconnect and blacklist). And live telemarketers would sink in the labyrinth of options to press, voice-selections, dead ends, PIN codes and so on. Ah, and they would need to agree to a soul-stealing license they never got to see.
Anyway, a simple voice captcha ("To connect call, press the number three") would get rid of most robot calls. Of course this would still rather be in a plan that doesn't force fees for incoming calls.
If you are unemployed and awaiting a call from your potential employee? If you partake in a lottery and await a call for the winner? If an emergency happened to a member of your family and the hospital calls? Or they take their one call from arrest?
There's a hundred valid reasons to receive calls from unknown numbers. And the spammers make people deny calls when the real need arises. You should not be subjected to extra costs if you accept a call before you know the name and business of whoever is calling you.
If you merely press ctrl+D, they blow up without leaving any time to heal, not even a corpse. OTOH, if you blow up in a car... you can even blow up a few times in sequence, and then heal.
At least that's how Amiga version behaved.
Still, even better is that persuaded crowd member has unlimited ammo to whatever weapon they picked. So bring lots of gauss guns, drop them and drive the crowd over them. One shot from minigun results in ~30 shots from gauss gun. Remember Gauss Gun? The big explosion? Now imagine 30 of them shot at once. With no ammo limit.
I preferred to take a tactical approach even if it wasn't necessary. I would plan ahead. There were missions where I would only take one agent and sprint through the entire thing. There were ones where I would take two, one just to carry lots of heavy guns. Interestingly, there were even missions where I deemed exactly three agents necessary.
I don't know, maybe it was the PC version, but the Amiga AI wasn't all that bad. Some enemies would be triggered by events, some by proximity, some would attack after certain time. They had decent pathfinding algorithms, and would attack according to selected weapon ranges. Sure it wasn't Alien Breed 3D 2 (there are few games to this day with better AI than it) but it was quite challenging at times.
Syndicate would play out pretty great in 3D as RTS where you control disembodied camera to drive the guys, with option to switch to FPS controlling one of them.
But if you do away with "control a tactical team" and change it into "control one guy" you lose the main point of the game. It's not isometric graphics that defined Syndicate, it was the tactical aspect of splitting the team.
Sacrifice half the team to draw enemies away from the target, then assault it with the remaining agent. Send one by one each of the 4 agents to tear further through lethal defenses. Split the team and ambush the enemy. Send away one team member for a vehicle, then drive it like a tank. Send one agent on a suicide mission, self-destruct at the target. Complete the mission using remaining agents. Split the team to kill four hits at four locations before the enemy can react. Set one agent with a flamethrower behind a corner creating unpassable wall of fire while others perform the mission tasks.
And of course "Persuade" half the town, arm them and run the mob at the target.
Actually, in the original, the city areas were varied, and some were very tidy and pretty, with neatly trimmed grass, perfect lines of trees and nicely angled hills. To this day I get "Syndicate vibes" from certain gardens, squares, parks that are all too regularly trimmed.
Certainly migrating any common daily activity has been done plenty of times before, using talking to people, signing papers, talking over the phone, pressing a button in standalone vending machines, or just getting an opinion through word-on-mouth or driving there and looking at the thing yourself.
Yet it didn't stop the patent office from issuing thousands of patents on common daily activity using the Internet.
As soon as you use a buzzword: "On the Internet", "Nuclear", "In Space", "Quantum" - often even "Portable" or "Ecological", the part about obviousness of a patent goes out through the window and the patent office grants the patents for simplest of things.
Look at multitouch gesture patents which became viable (and obvious) once multitouch screens became viable. And look at the shitstorm Apple started about them.
Oh, but before I build a prototype, I have a solid belief I know all the steps necessary. Of course life will invalidate my assumption and the final device will appear vastly more complex. But at the project stage, I can just make up solutions to all the problems I can imagine and write them down.
Yep, that would invalidate the warp drive patent. But nuclear fusion water desalination plant? Best to my knowledge, the water won't be irradiated. The salt depositing won't be a problem if we use plastic for the separation unit. The plastic won't melt if we use plasma temperature low enough. And we can control the temperature by regulating the water flow with a generic off-the-shelf PLC device. Of course we reserve rights to other means of salt-clogging protection, other means of temperature control and the actual fusion reactor is yet to be built - its construction is currently on hold due to budget cuts. But who are you to say this won't work using the basic components we listed? And any work of actual desalination plant will have to build upon our patent - extend and correct it, but the key parts will remain the same.
Yes, but this is moot. I don't care about building a nuclear reactor. I only care about reaping profits off those who build it. They can patent the pumps and the fountain, I don't care. I don't need the pumps or their patents. But they need MY patent to run the reactor, even though they understand it thousandfold better than I ever did.
Yes, the other guy was right - patenting warp drive... warp space, create hole between points, transport vehicle through hole. Easy-peasy. I don't care about your singularity machine, space warper, or long distance space ship design. You won't get it running without implementing a wrap drive, and I have a patent on that.
Actually, "ironing the details out" may take a long time and a race to finish a prototype is not a nice thing. But I guess a combined system, when you get, say, 3 years for an idea patent and then prolonging it by quite a bit for a working prototype would be better. Primarily stop the submarine patents to lie in wait for 15 years and more before something becomes mainstream.
Or alternatively, Trademark-like system of "if you don't defend it, you lose it". Patents that become industry standards cease to be patents.
The worst part is the patent lists the basic fundamental idea which you know is sound, simple and will work in the end once the obstacles are out of the way. And 98% of the work (and cost) is removing these obstacles, solving all the little caveats, to get it working.
Nuclear reactor? Trivial. Stack some radioactives, run water through them, blow the resulting steam at turbines. Easy-peasy. I can draw the schematics in 5 minutes and submit the patent application tomorrow. Now for details like stopping the core from overheating, dealing with pressures of thousands atmospheres, cooling tons of water per second before it returns into the system, stopping the radioactivity from leaking... Let someone take care of that and I'll just reap profits from my patent application.
I think if I had a lot of carbon in form of diamond, it would have effect on global warming. Just think of environmental cost of production of all the stuff I would buy...
While Communism has been proven to be a wrong answer to all the evils of capitalism, it doesn't make these evils any less true or evil. Marx spotted the problems right on, it's just the solutions he came up with weren't realistic, assuming humans aren't inherently lazy, dishonest and close-minded.
I'd say this is the main problem with the Zeitgeist series: the problems it points out are spot on, it's the solutions that are unrealistic.
People somehow feel "if I take something as big as Linux, it would be only fair if I contributed something of similar scale. And since I can't, too bad..."
No. If a set of pieces of software is used by 100,000 people, and each contributes equivalent of only 0.001% of the volume/workload/cost of the set, then the set will double in size.
Grab as much as you wish, with both hands, freely, then slightly nudge a single small thing ahead in exchange, and everything is fine and fair. If everyone gives such a small nudge, the progress will be rapid and fluent, no matter how much they take from the common pool. You don't have to
Green wave doesn't work for such distances. The maximum allowed cycle length is 3 minutes [tests showed most people get impatient over that wait time and tend to ignore the lights, regulations followed the tests.] Longer time between traffic lights, bigger differences at minor speed differences... and infrastructure to connect them all. This works best for a line of 3-5 traffic lights within one mile. And forces unoptimal traffic for intersecting directions. It's quite likely only one or two major crossings in a line are synchronized. Next major crossing may be quite independent, or start a new line.
Your impression of "synchronized for 90MPH"? Most likely not synchronized at all. Each simply running their own cycle, and if the cycles are of the same length (likely - same traffic, similar requirements, same author with personal preferences), any apparent synchronization is dependent on dumb luck of the time of last reset of the controller.
Besides, due to unequal distances, green wave can be guaranteed only in one direction. The opposite direction can be lucky to get in it, or miss it on some crossings. The real life application gives both directions roughly equal share of "goodness" so the number of stops will be reduced but hardly ever zeroed.
...and THAT'S where videoradar really helps. One that is clearly visible, advertised by big signs... and possibly even just a dummy, an empty box. Of course depends on road conditions. If the road is safe to travel 70mph, just set the limit so...
Poland here. Timing lights above speed limit is simply in violation of regulations. There is a simple, clear process of challenging any setting of traffic lights. If they are not to code, they must be fixed. (of course few "citizens" ever bother to research the process and prefer to bicker about these pesky traffic lights over a drink instead)
The guy "up there" who wrote the regulations is a dread to all of the traffic business people. The regulations are extremely precise, very highly demanding, and the guy is a total no-life, no-compromise ivory-tower academic. Some of the requirements make the project makers, the firmware developers, the managers tear their hair from their heads (multiple redundancy, no-compromise geometry requirements that are often impossible in real life etc) but while some are way over the top, there is very little room left for gaming/abusing the system for profit.
For example, the US "red light" cameras, which shoot a car that passed on "late yellow/early red"? No room for such thing here. While yes, you are not legally allowed to enter the crossroads on red light, the project must account for drivers that do, and on a high-traffic road assume first 2-3 seconds of red as pretty much equivalent to yellow, and delay other directions accordingly. So yes, red-light cameras are a go, but they can start shooting no earlier than 5s after the red light was lit. And yellow light must last at least 3s. More is allowed if circumstances require. Less is illegal.
Imagine a modding engine where to get each character you use in a script of the mod you need to enter an ancient dungeon, fight skeletons and then mine the walls for coding commands. Typing "if(x=0)" feels different than usual if obtaining the "if" statement was forged from two rare ores found on the bottom of a pit filled with scorpions, the "x" was guarded by goblins, the "0" was found floating in shark-infested waters on the seaside and you needed to assemble the "==" from two trees you chopped down.
Programming using a nail and a box of punch cards seems leisurely by comparison.
Oh, one more problem with your approach: Low traffic on a crossing with adaptative lights.
In case of two roads of similar level of traffic, the default program type is "AllRed". Meaning, by default if there is no traffic, all directions get red signal, so a car approaching from any direction will get green before they reach the crossing, without waiting at all.
First detectors are located 50 and more meters from the crossing, and as soon as a car is detected, the green signal sequence for that direction starts. So by the time you reach the crossing, you have green light.
Now you saw the red light from 400 meters away. You slow down and roll slowly, waiting for it to change. As you enter the detector zone, you get green signal "at long last, why did it take so long?". It wouldn't take so long if you got there faster. And due to you driving so slowly, you will keep your green signal longer, meaning cars at the opposite lanes may need to stop and wait for green, instead of just dashing through after your minimal green ended.
Slashdot Mirror
Not sure about Switzerland (which is not in EU) but in Poland (and from what I heard, most other countries) you have a range of plans for very moderate prices.
Yes, all-inclusive unlimited is expensive like hell, but you'd better need to spend 12h a day on the phone if you really need it. This is not a plan for anyone who wants something above the minimum, like it is done in the US. This is for heavy abusers of the phone. There are extra services of unlimited calls to one number within the same network (or landline), but universal unlimited is just a disused option for rich lunatics.
Actual plans range from $6/mo for 40 minutes of calls included in the price (and about $0.15/minute above that limit) to some $150/mo with 15 hours of calls free, and $0.08 above the limit. This is for individual offers. Companies may get much better offers if they sign a contract for a bulk number of phones.
The plans often differ in "flat rate for all days, all hours" vs "cheap evenings, expensive business hours", "flat rate to all operators" vs varied pricing depending on which operator you call, extra services like data transfer in packets starting with 50MB and ending with 50GB of transfer for various prices (and either throttling or entering default no-premium-service pricing upon exhausting the packet), varied extra SMS packages and so on. Still, you never pay for incoming calls.
Prepaid is more expensive and you don't get many of the premium services. But interestingly, in some prepaid plans YOU GET PAID for incoming calls. Not with money but with outgoing calls credit - like, for 3 minutes of incoming call you get one minute of outgoing calls credit. Simply, the companies have deals signed and the caller's company pays the receiver's company a share from the caller's fees. And since it's still a valid source of income, you don't even have to pay for maintaining your phone number as long as you receive enough calls - your callers will pay for it.
Currently, I just use the blacklist feature of my Android phone. It works well - the spammers usually don't have too many numbers to block, and occasionally receiving an undesired call is not a tragedy.
Now I wonder if it would be viable to get an "automatic service" tarpit app for whoever calls.
"If you are calling concerning business, press 1. If you're calling concerning family matters, press 2 and state your business. If you are a friend or colleague, press # and enter your access code."
Robots would quickly get disconnected for talking to the hand (actually, continuous talking while a message is playing could be taken as a hint to disconnect and blacklist). And live telemarketers would sink in the labyrinth of options to press, voice-selections, dead ends, PIN codes and so on. Ah, and they would need to agree to a soul-stealing license they never got to see.
Anyway, a simple voice captcha ("To connect call, press the number three") would get rid of most robot calls. Of course this would still rather be in a plan that doesn't force fees for incoming calls.
If you are unemployed and awaiting a call from your potential employee?
If you partake in a lottery and await a call for the winner?
If an emergency happened to a member of your family and the hospital calls? Or they take their one call from arrest?
There's a hundred valid reasons to receive calls from unknown numbers. And the spammers make people deny calls when the real need arises. You should not be subjected to extra costs if you accept a call before you know the name and business of whoever is calling you.
Still, petty crime - like pickpocketing or shoplifting, vs serious crime like exortion or murder...?
If you merely press ctrl+D, they blow up without leaving any time to heal, not even a corpse.
OTOH, if you blow up in a car... you can even blow up a few times in sequence, and then heal.
At least that's how Amiga version behaved.
Still, even better is that persuaded crowd member has unlimited ammo to whatever weapon they picked. So bring lots of gauss guns, drop them and drive the crowd over them. One shot from minigun results in ~30 shots from gauss gun. Remember Gauss Gun? The big explosion? Now imagine 30 of them shot at once. With no ammo limit.
Let me guess, in Hitman - Silent Assassin your preferred weapon was a minigun?
I preferred to take a tactical approach even if it wasn't necessary. I would plan ahead. There were missions where I would only take one agent and sprint through the entire thing. There were ones where I would take two, one just to carry lots of heavy guns. Interestingly, there were even missions where I deemed exactly three agents necessary.
I don't know, maybe it was the PC version, but the Amiga AI wasn't all that bad. Some enemies would be triggered by events, some by proximity, some would attack after certain time. They had decent pathfinding algorithms, and would attack according to selected weapon ranges. Sure it wasn't Alien Breed 3D 2 (there are few games to this day with better AI than it) but it was quite challenging at times.
Syndicate would play out pretty great in 3D as RTS where you control disembodied camera to drive the guys, with option to switch to FPS controlling one of them.
But if you do away with "control a tactical team" and change it into "control one guy" you lose the main point of the game. It's not isometric graphics that defined Syndicate, it was the tactical aspect of splitting the team.
Sacrifice half the team to draw enemies away from the target, then assault it with the remaining agent. Send one by one each of the 4 agents to tear further through lethal defenses. Split the team and ambush the enemy. Send away one team member for a vehicle, then drive it like a tank. Send one agent on a suicide mission, self-destruct at the target. Complete the mission using remaining agents. Split the team to kill four hits at four locations before the enemy can react. Set one agent with a flamethrower behind a corner creating unpassable wall of fire while others perform the mission tasks.
And of course "Persuade" half the town, arm them and run the mob at the target.
Actually, in the original, the city areas were varied, and some were very tidy and pretty, with neatly trimmed grass, perfect lines of trees and nicely angled hills. To this day I get "Syndicate vibes" from certain gardens, squares, parks that are all too regularly trimmed.
Certainly migrating any common daily activity has been done plenty of times before, using talking to people, signing papers, talking over the phone, pressing a button in standalone vending machines, or just getting an opinion through word-on-mouth or driving there and looking at the thing yourself.
Yet it didn't stop the patent office from issuing thousands of patents on common daily activity using the Internet.
As soon as you use a buzzword: "On the Internet", "Nuclear", "In Space", "Quantum" - often even "Portable" or "Ecological", the part about obviousness of a patent goes out through the window and the patent office grants the patents for simplest of things.
Look at multitouch gesture patents which became viable (and obvious) once multitouch screens became viable.
And look at the shitstorm Apple started about them.
Oh, but before I build a prototype, I have a solid belief I know all the steps necessary. Of course life will invalidate my assumption and the final device will appear vastly more complex. But at the project stage, I can just make up solutions to all the problems I can imagine and write them down.
Yep, that would invalidate the warp drive patent. But nuclear fusion water desalination plant? Best to my knowledge, the water won't be irradiated. The salt depositing won't be a problem if we use plastic for the separation unit. The plastic won't melt if we use plasma temperature low enough. And we can control the temperature by regulating the water flow with a generic off-the-shelf PLC device. Of course we reserve rights to other means of salt-clogging protection, other means of temperature control and the actual fusion reactor is yet to be built - its construction is currently on hold due to budget cuts. But who are you to say this won't work using the basic components we listed? And any work of actual desalination plant will have to build upon our patent - extend and correct it, but the key parts will remain the same.
Yes, but this is moot. I don't care about building a nuclear reactor. I only care about reaping profits off those who build it. They can patent the pumps and the fountain, I don't care. I don't need the pumps or their patents. But they need MY patent to run the reactor, even though they understand it thousandfold better than I ever did.
Yes, the other guy was right - patenting warp drive... warp space, create hole between points, transport vehicle through hole. Easy-peasy. I don't care about your singularity machine, space warper, or long distance space ship design. You won't get it running without implementing a wrap drive, and I have a patent on that.
Actually, "ironing the details out" may take a long time and a race to finish a prototype is not a nice thing. But I guess a combined system, when you get, say, 3 years for an idea patent and then prolonging it by quite a bit for a working prototype would be better. Primarily stop the submarine patents to lie in wait for 15 years and more before something becomes mainstream.
Or alternatively, Trademark-like system of "if you don't defend it, you lose it". Patents that become industry standards cease to be patents.
The worst part is the patent lists the basic fundamental idea which you know is sound, simple and will work in the end once the obstacles are out of the way. And 98% of the work (and cost) is removing these obstacles, solving all the little caveats, to get it working.
Nuclear reactor? Trivial. Stack some radioactives, run water through them, blow the resulting steam at turbines. Easy-peasy. I can draw the schematics in 5 minutes and submit the patent application tomorrow. Now for details like stopping the core from overheating, dealing with pressures of thousands atmospheres, cooling tons of water per second before it returns into the system, stopping the radioactivity from leaking... Let someone take care of that and I'll just reap profits from my patent application.
[insert Pinkie Pie reaction image here]
I think if I had a lot of carbon in form of diamond, it would have effect on global warming. Just think of environmental cost of production of all the stuff I would buy...
While Communism has been proven to be a wrong answer to all the evils of capitalism, it doesn't make these evils any less true or evil. Marx spotted the problems right on, it's just the solutions he came up with weren't realistic, assuming humans aren't inherently lazy, dishonest and close-minded.
I'd say this is the main problem with the Zeitgeist series: the problems it points out are spot on, it's the solutions that are unrealistic.
That's no star. It's artificial sun.
People somehow feel "if I take something as big as Linux, it would be only fair if I contributed something of similar scale. And since I can't, too bad..."
No. If a set of pieces of software is used by 100,000 people, and each contributes equivalent of only 0.001% of the volume/workload/cost of the set, then the set will double in size.
Grab as much as you wish, with both hands, freely, then slightly nudge a single small thing ahead in exchange, and everything is fine and fair. If everyone gives such a small nudge, the progress will be rapid and fluent, no matter how much they take from the common pool. You don't have to
Green wave doesn't work for such distances. The maximum allowed cycle length is 3 minutes [tests showed most people get impatient over that wait time and tend to ignore the lights, regulations followed the tests.] Longer time between traffic lights, bigger differences at minor speed differences... and infrastructure to connect them all. This works best for a line of 3-5 traffic lights within one mile. And forces unoptimal traffic for intersecting directions. It's quite likely only one or two major crossings in a line are synchronized. Next major crossing may be quite independent, or start a new line.
Your impression of "synchronized for 90MPH"? Most likely not synchronized at all. Each simply running their own cycle, and if the cycles are of the same length (likely - same traffic, similar requirements, same author with personal preferences), any apparent synchronization is dependent on dumb luck of the time of last reset of the controller.
Besides, due to unequal distances, green wave can be guaranteed only in one direction. The opposite direction can be lucky to get in it, or miss it on some crossings. The real life application gives both directions roughly equal share of "goodness" so the number of stops will be reduced but hardly ever zeroed.
...and THAT'S where videoradar really helps.
One that is clearly visible, advertised by big signs... and possibly even just a dummy, an empty box.
Of course depends on road conditions. If the road is safe to travel 70mph, just set the limit so...
Poland here.
Timing lights above speed limit is simply in violation of regulations. There is a simple, clear process of challenging any setting of traffic lights. If they are not to code, they must be fixed. (of course few "citizens" ever bother to research the process and prefer to bicker about these pesky traffic lights over a drink instead)
The guy "up there" who wrote the regulations is a dread to all of the traffic business people. The regulations are extremely precise, very highly demanding, and the guy is a total no-life, no-compromise ivory-tower academic. Some of the requirements make the project makers, the firmware developers, the managers tear their hair from their heads (multiple redundancy, no-compromise geometry requirements that are often impossible in real life etc) but while some are way over the top, there is very little room left for gaming/abusing the system for profit.
For example, the US "red light" cameras, which shoot a car that passed on "late yellow/early red"? No room for such thing here. While yes, you are not legally allowed to enter the crossroads on red light, the project must account for drivers that do, and on a high-traffic road assume first 2-3 seconds of red as pretty much equivalent to yellow, and delay other directions accordingly. So yes, red-light cameras are a go, but they can start shooting no earlier than 5s after the red light was lit. And yellow light must last at least 3s. More is allowed if circumstances require. Less is illegal.
Imagine a modding engine where to get each character you use in a script of the mod you need to enter an ancient dungeon, fight skeletons and then mine the walls for coding commands. Typing "if(x=0)" feels different than usual if obtaining the "if" statement was forged from two rare ores found on the bottom of a pit filled with scorpions, the "x" was guarded by goblins, the "0" was found floating in shark-infested waters on the seaside and you needed to assemble the "==" from two trees you chopped down.
Programming using a nail and a box of punch cards seems leisurely by comparison.
Oh, one more problem with your approach: Low traffic on a crossing with adaptative lights.
In case of two roads of similar level of traffic, the default program type is "AllRed". Meaning, by default if there is no traffic, all directions get red signal, so a car approaching from any direction will get green before they reach the crossing, without waiting at all.
First detectors are located 50 and more meters from the crossing, and as soon as a car is detected, the green signal sequence for that direction starts. So by the time you reach the crossing, you have green light.
Now you saw the red light from 400 meters away. You slow down and roll slowly, waiting for it to change. As you enter the detector zone, you get green signal "at long last, why did it take so long?". It wouldn't take so long if you got there faster. And due to you driving so slowly, you will keep your green signal longer, meaning cars at the opposite lanes may need to stop and wait for green, instead of just dashing through after your minimal green ended.