I think you're overlooking the fact that a coder who wins the contest gets something far more valuable: a demonstrable proof of one's mettle and a fairly admirable accomplishment that can only pay dividends for years to come when they're hired by a company who pays them what they're really worth.
true, but in the short-term, bragging rights and resume' bullet points don't pay the bills.
It's too bad the winning entry, at 970x the speed of the algo it replaced, only received $6k. Surely this was worth a lot more to the eggheads than that? You'd have difficulty contracting even simple, low grade code for that amount?
when I've called my ISP to complain about low speed, they usually start out by telling me to go to a specific site to check my speed. (they do the same thing when they send out a tech)
Thanks, no. I'll go to a different site. Anywhere besides the one you just suggested to me. Using what they recommend is like the used car dealer recommending you get a second opinion from his brother Jim.
oh I think this is at least slightly interesting. I remember the "ping of death" (and pissing off a few windows heads in my sights) back in 'th day.
This is basically a DoS attack on hardware. The fact that it can get through someone's firewall makes it a bit more effective. Having your ethernet port check out every five minutes (requiring a reboot to fix) just because someone down the hall (or in Bulgaria) wants to be an ass is definitely annoying and something I'd like to know is a possibility when troubleshooting screwy network problems.
I just got done swapping out a gigabit switch that was being wonky and slow for no obvious reason. I don't mind so much when hardware keels over and dies, but when it throws symptoms that don't immediately suggest where the problem is, those are the real time wasters. And we've come to rely on hardware generally being more reliable than software. So if my ethernet was going out when I VOIP'ed, I might have spent (wasted) a lot of my time troubleshooting the VOIP software.
MS has a game console and a phone. Maybe now they are going to have their own actual consumer computer? Bet the other hardware retailers are gonna just love that...
I was thinking more just ORBIT. You first have to ask yourself what is the height required for the mass to orbit? (this is assuming you can start the plane out at any given direction, and you know the mass of the plane)
After that, anything starting slower, lower, or weighing more will need to do some sort of powered flight to stay up.
If you start a little bit below orbital requirements, the demand will be very minor. The farther down you go, the slower you start, or the heavier you are, the more demand there will be. It's not a yes/no thing. It could be a system just off equilibrium.
Unless you have the scenario they described with the sun, where you don't hit atmosphere until you are well inside orbit (unless you have some impressive speed)
So I guess what I'm saying is that the starting conditions (speed, heading, elevation) are at least as important as the planet you are trying to fly at. Since they somewhat arbitrarily picked them, the resulting comparisons are equally arbitrary. And not by an insignificant amount.
That being said, revisit the starting conditions. If we say you must start with your speed, heading, and elevation, so that you are say, 10% below orbital speed, it becomes a question of atmosphere - "can you sustain flight?", rather than "can you pull out in time?". Can you generate enough lift to achieve... not sure what to call it... stable sub-orbital trajectory? I suppose that's the best definition of "powered flight". aka "straight and level", with an "at stable speed" thrown in for good measure.
Numbers become more critical as your ability to generate lift is lowered, or as that 10% below orbit is raised. At some point it will become a question of whether or not you have time to pull out of the dive. (or whether it's even possible, assuming there's no ground) And then we get into what your "ceiling" would be - the highest altitude you can climb to, where you finally level out while trying to climb, equilibrium of climb. It's interesting to ponder that most craft have TWO ceilings... one is their orbit, and the other is somewhere below that. So what we may really be asking here is, "does the craft have a ceiling other than orbit?"
Although that article does dig a bit deeper than xkcd did, it's still quite a long way from the goal.
So if you apply this reasoning to slide-to-unlock, you can patent a specific slider, but neither the sliding nor the unlocking?
If you word your "slide to unlock" by saying you're "using a touchscreen that tracks finger movement, over an lcd display screen. You have an icon displayed on the screen under the touch surface, that moves in response to touch and drag movement on the finger. And the device becomes active when the icon is moved from one designated point to another", you have described a process. And that is patentable.
Android uses their "draw between the dots" to authenticate an unlock, but it adds the additional step of moving between more than two points, in a required order, and thus the basic slide-to-unlock patented method would not apply to this method. It's a great counterpoint to show that the "slide to unlock" patent isn't overly broad.
And "slide to unlock" isn't an obvious feature. It was probably a "wow, that's slick... that's very slick... that's intuitive and may be the best way to do it", but that doesn't make it unpatentable. Coming up with the easiest, most intuitive method should be highly worthy of patent. Others complaining is just sour grapes. You know in their heart they're not upset that the process was patentable... they're upset that they aren't the ones that got the patent. Just think about it - if you were to ask them, which would you rather have.... it's not patentable, or you have the patent for it? You know what their honest answer would be.
The problem is that there is no clear line on what is hardware and software, and both can implement the same thing.
I would disagree. Hardware that executes instructions is a physical design, and should be obvious to be patentable.
Hardware CAN have firmware. For the purpose of comparison, the firmware should be classified as software.
With that settled.... software is a list of instructions. It's a process, in written form. Processes can be patented.
The question then becomes "how far do you need to deviate from a patented process or hardware design to be considered non infringing? THAT is the big problem here. "broad patents" try to define the breadth of their coverage, and frequently come down to "anything that's even remotely similar to this is ours".
For both hardware and software, the exact implementation is obviously protected. On the other end of the spectrum, you clearly don't have a patent over the result. (unless the result is another physical, separately patentable object, such as chemicals/pharmaceuticals) You can patent a car but not transportation. Therefore you can't simply claim rights because someone else has arrived at the same result as your process. And that's what they're trying to do. It'd be like inventing a calculator and then trying to patent anything that adds. You should be able to patent HOW it's done, but not WHAT it does. Look at the wording at the start of most patents of this type.... "This defines a method and process for...." That's how they have to word it because the law doesn't explicitly protect results. But then they always go on to explain the results in the patent. ("and by doing A, B, and C, it results in X and Y") and then they try to enforce someone else arriving at the X and the Y by a different method or process.
But the current state of patent interpretation and enforcement is allowing results of processes to be enforced. And there's the problem.
you lose a lot of your ability to catch certain elements of "crime" when you introduce any form of anonymity. Anonymity will be used by people that want privacy, people that are paranoid, people that want to whistleblow, harass, defraud, threaten, etc. You will get the good with the bad.
Unfortunately, we seem to be headed in the direction that NO bad is allowed, so we are supposed to be OK with giving up all the good.
As a basic example, you can't punish or prevent all slander while allowing whistleblowing. The concept of allowing truly anonymous good behavior while preventing anonymous bad behavior is an impossible goal.
Normally one would expect "innocent until proven guilty" to take precedence and allow anonymous behavior while tolerating difficulty in catching criminals, but there are currently just too many well-funded groups with a stake that push the legislation in the other direction.
Since when is a legitimate patent holder a 'patent troll'?
A "patent troll" is someone that takes advantage of patent law for monetary gain based on the innovation of others. Patent trolls aren't trying to claim reward for what is theirs. They simply game the system and out-maneuver the innovators such that, by legal definition (but not common sense) they are entitled to reward.
As such, it's mostly counterproductive for the purpose that the method of patenting was intended to serve. (encouraging and rewarding innovation)
Patent trolls siphon off some of the rewards of innovation through litigation and through the licensing of innovations where they themselves were not the innovator.
Fortunately, a lot of courts (at least those that'd don't directly benefit from the litigations, such as texas east district) have identified these people as taking advantage of the legal system and costing it money in exchange not for the support of innovation, but for the enrichment of the trolls and stifling of innovation, and are starting to push for change.
this level of unemployment cannot be fixed by manufacturers moving jobs around. stopping the populace from breeding like rabbits is about the only useful solution. There's a very good reason china is pushing for "one child per family". Their bean counters know how to do math, and see what's coming. It's a problem that takes a generation to manifest. They're already trying to head it off, yet they're already getting their feet wet in the problem. I'd say they took action right on time.
I also find it ironic that the thread has turned in this direction. Employ underage? you child-slaving monster! Don't employ underage? throwing them to the gutters to starve! damned if you do, damned if you don't.
It's not the employers that created this problem, and it's not within their power to fix. All they can do is move the shells around and make it look a little different from a static viewpoint.
interesting to see how a joe average gets smacked down like a gnat with a buick on youtube, but then we see the exact opposite here? Or didn't they file a takedown notice?
I think what concerns the educated public the most is RIM's lack of intelligence
Just because an industry is based on your home turf is no excuse to allow embarrassing performance to continue to drag itself out. If Lenovo thinks they can "fix" RIM, I say let them try. They can't really do much worse at this point. Any business that's not totally dominate by a PHB should have by now at least started diversifying critical resources away from reliance on RIM.
I'm surprised this issue hasn't been tossed out onto a We the People petition?
Although recently we've seen a few of those used for stupid things (death star) as well as being flat out trampled on a few times with responses that basically said "we don't feel like telling you that", it would still be nice to see it out there.
Just make sure that you don't leave the luggage for more than a few minutes, and make nice with the people sitting with you.
Friend of mine got his laptop bag stolen at an airport. He set it on the conveyor to run through the xray and there was someone already on the other end of the scanner waiting for it. He grabbed it, and just plain ran off with it. The wonderful security guards are not allowed to leave their post to give chase, and merely radio'd in the thief. Who got away. They had the regular exits covered, but he ran straight for and out a fire door (setting off the alarm, which obviously he did not care about) to a waiting car.
You don't even have to leave your luggage for a few minutes. Five seconds is enough. You're not good, you're lucky. And one day you won't be lucky.
google for "purse alarm". there are a wide variety of options for things that make noise if disconnected.
for proximity, goole all the options with 'child proximity alarm", this is probably more up your alley. Some only sound an alarm on the kid, some only on the parent with the remote, and some do both. that's up to you what you want.
final option for gps is a smartphone with a "find my phone" option. Like with the iphone where you can see where it's at from a computer. that would be useful if the bag disappeared without setting off whatever else you had watching it like a proximity alarm, or if they managed to outrun you and hop in a car etc with your bag,
I see more to this "milestone" thing than a single hurdle. It's probably better to look at it as a "product development cycle" where there are several tiers a product has to pass through, with similar but higher requirements at each step.
1. throw some money at it 2. add featuers 3. remove / combine features / refine 4. in-house and user testing 5. decide if it's worth continuing
and then it repeats, with more money, fewer new features, more careful and thorough refining, and more thorough testing at each next stage. Go through that three or more times and you will probably have a few winners out of the hundreds you went in with at the start.
You could call passing each iteration's part 5 a "milestone" I suppose. But most of the discussion above acts like it only hits this point once during development. And I suppose for some, that's true. But that's a bad way to do it. You can't consistently produce a lot of winners and very few bombs if your development tree has no depth to it.
I like how a previous comment discussed Phillips and their "throw a bunch of crap at the wall and see what sticks." That really suggests a 1 or 2 iteration development cycle, and it's going to perform poorly over the long run.
I don't see any idealogical difference between each of these "milestones". Either it's worth continuing on, or it's not. If your development wing is only supposed to bring one product to market, then it's a simple game of "survival". If you start with 250 ideas you can just say you're going to go to the next stage after removing 75% of the ideas, then it's just a simple comparison, and you still have enough "keepers" to carry forward that things that are still "a little iffy" can make the cut, at least once. Do that four times and you are down to ONE idea. No need to change milestone strategies at each stage.
If you've got low depth, say 2 levels, think of how you'd have to do it. Split it evenly, and you have to drop over 90% of your ideas both times. The first cut probably won't be hard, but then having to compare 16 items to come up with the one best, you could easily kill what would have been the best idea. Do a hard cut on the first round and you are down to having to pick so few to keep from amongst so many you may as well pick them by lotto. A deep cut on round 2 I hope everyone sees would be suicidal. Milestone depth is the only solution to this problem.
A substantial part of their score was for things that very specifically were not actively being exploited. They were testing the heuristics to see if it could identify "virus and malware-like behavior". You can't rely on software updates and AV definition updates to protect you from zero-day's, that's 100% on the head of your AV software to keep you safe from.
And MS fails miserably at protecting users from zero-days. They flunked, and they deserved to flunk. There's just too many new viruses and malware being developed every day to try to function on a blacklist-only basis. Some behaviors need to be whitelist only, and quite a few need to be greylisted with heuristics so that your system survives long enough for the new exploit to get added to tomorrow's definitions file.
Now the best non-linear game play has to be Deus Ex
I'd have to agree with that. It had good replayability also because you had so many binary upgrade decisions to make. And most levels could be solved with at least two distinct play styles. (stealth and power)
It was also refreshing to have three factions fighting for your loyalty instead of the usual two found in other games that play that way.
The most simplified counter-example is "lets say by random chance your bt client decides to download the first piece of the video last". There's nothing to prevent it from downloading one of the first pieces last. And in that case, regardless of how fast your connection is, you WILL be waiting until the download finishes before starting to watch it.
Not sure if you're trolling or just uninformed. So maybe I'm feeding.
Torrent files don't download sequentially, from start to finish. Clients that follow spec will randomly select a piece to download, possibly influenced by availability or speed of peers with pieces. The larger the file gets, the lower the odds are that you're not missing an early piece. For a movie that has 1000 pieces, the odds of having ALL of the first 250 pieces at any point before you hit the 95% downloaded point are astronomically low.
(and i haven't seen a bt client yet that you can prefer sequential download with)
It's easy to see why they wrote it that way. If one seed starts with 10 peers, and all of them download in order, then the peers won't have many pieces available from each other. (they'll all have piece 1, then all have piece 2, etc) OTOH if the seed hands out a different piece to all 10 peers, when those complete, now each peer has a choice of 9 other pieces to get from other peers, in addition to the seed, and data starts flowing around between peers instead of just between peers and the seed. And that's what makes bt fast.
Slashdot Mirror
true, but in the short-term, bragging rights and resume' bullet points don't pay the bills.
It's too bad the winning entry, at 970x the speed of the algo it replaced, only received $6k. Surely this was worth a lot more to the eggheads than that? You'd have difficulty contracting even simple, low grade code for that amount?
looks to be paywalled, @ $32 for a single article?
when I've called my ISP to complain about low speed, they usually start out by telling me to go to a specific site to check my speed. (they do the same thing when they send out a tech)
Thanks, no. I'll go to a different site. Anywhere besides the one you just suggested to me. Using what they recommend is like the used car dealer recommending you get a second opinion from his brother Jim.
oh I think this is at least slightly interesting. I remember the "ping of death" (and pissing off a few windows heads in my sights) back in 'th day.
This is basically a DoS attack on hardware. The fact that it can get through someone's firewall makes it a bit more effective. Having your ethernet port check out every five minutes (requiring a reboot to fix) just because someone down the hall (or in Bulgaria) wants to be an ass is definitely annoying and something I'd like to know is a possibility when troubleshooting screwy network problems.
I just got done swapping out a gigabit switch that was being wonky and slow for no obvious reason. I don't mind so much when hardware keels over and dies, but when it throws symptoms that don't immediately suggest where the problem is, those are the real time wasters. And we've come to rely on hardware generally being more reliable than software. So if my ethernet was going out when I VOIP'ed, I might have spent (wasted) a lot of my time troubleshooting the VOIP software.
MS has a game console and a phone. Maybe now they are going to have their own actual consumer computer? Bet the other hardware retailers are gonna just love that...
I was thinking more just ORBIT. You first have to ask yourself what is the height required for the mass to orbit? (this is assuming you can start the plane out at any given direction, and you know the mass of the plane)
After that, anything starting slower, lower, or weighing more will need to do some sort of powered flight to stay up.
If you start a little bit below orbital requirements, the demand will be very minor. The farther down you go, the slower you start, or the heavier you are, the more demand there will be. It's not a yes/no thing. It could be a system just off equilibrium.
Unless you have the scenario they described with the sun, where you don't hit atmosphere until you are well inside orbit (unless you have some impressive speed)
So I guess what I'm saying is that the starting conditions (speed, heading, elevation) are at least as important as the planet you are trying to fly at. Since they somewhat arbitrarily picked them, the resulting comparisons are equally arbitrary. And not by an insignificant amount.
That being said, revisit the starting conditions. If we say you must start with your speed, heading, and elevation, so that you are say, 10% below orbital speed, it becomes a question of atmosphere - "can you sustain flight?", rather than "can you pull out in time?". Can you generate enough lift to achieve... not sure what to call it... stable sub-orbital trajectory? I suppose that's the best definition of "powered flight". aka "straight and level", with an "at stable speed" thrown in for good measure.
Numbers become more critical as your ability to generate lift is lowered, or as that 10% below orbit is raised. At some point it will become a question of whether or not you have time to pull out of the dive. (or whether it's even possible, assuming there's no ground) And then we get into what your "ceiling" would be - the highest altitude you can climb to, where you finally level out while trying to climb, equilibrium of climb. It's interesting to ponder that most craft have TWO ceilings... one is their orbit, and the other is somewhere below that. So what we may really be asking here is, "does the craft have a ceiling other than orbit?"
Although that article does dig a bit deeper than xkcd did, it's still quite a long way from the goal.
Rovers still have the best hardware
This one's even got a laser tough enough to blast rocks. It's gonna be awhile before we see a cell phone with those kind of specs.
If you word your "slide to unlock" by saying you're "using a touchscreen that tracks finger movement, over an lcd display screen. You have an icon displayed on the screen under the touch surface, that moves in response to touch and drag movement on the finger. And the device becomes active when the icon is moved from one designated point to another", you have described a process. And that is patentable.
Android uses their "draw between the dots" to authenticate an unlock, but it adds the additional step of moving between more than two points, in a required order, and thus the basic slide-to-unlock patented method would not apply to this method. It's a great counterpoint to show that the "slide to unlock" patent isn't overly broad.
And "slide to unlock" isn't an obvious feature. It was probably a "wow, that's slick... that's very slick... that's intuitive and may be the best way to do it", but that doesn't make it unpatentable. Coming up with the easiest, most intuitive method should be highly worthy of patent. Others complaining is just sour grapes. You know in their heart they're not upset that the process was patentable... they're upset that they aren't the ones that got the patent. Just think about it - if you were to ask them, which would you rather have.... it's not patentable, or you have the patent for it? You know what their honest answer would be.
I would disagree. Hardware that executes instructions is a physical design, and should be obvious to be patentable.
Hardware CAN have firmware. For the purpose of comparison, the firmware should be classified as software.
With that settled.... software is a list of instructions. It's a process, in written form. Processes can be patented.
The question then becomes "how far do you need to deviate from a patented process or hardware design to be considered non infringing? THAT is the big problem here. "broad patents" try to define the breadth of their coverage, and frequently come down to "anything that's even remotely similar to this is ours".
For both hardware and software, the exact implementation is obviously protected. On the other end of the spectrum, you clearly don't have a patent over the result. (unless the result is another physical, separately patentable object, such as chemicals/pharmaceuticals) You can patent a car but not transportation. Therefore you can't simply claim rights because someone else has arrived at the same result as your process. And that's what they're trying to do. It'd be like inventing a calculator and then trying to patent anything that adds. You should be able to patent HOW it's done, but not WHAT it does. Look at the wording at the start of most patents of this type.... "This defines a method and process for ...." That's how they have to word it because the law doesn't explicitly protect results. But then they always go on to explain the results in the patent. ("and by doing A, B, and C, it results in X and Y") and then they try to enforce someone else arriving at the X and the Y by a different method or process.
But the current state of patent interpretation and enforcement is allowing results of processes to be enforced. And there's the problem.
you lose a lot of your ability to catch certain elements of "crime" when you introduce any form of anonymity. Anonymity will be used by people that want privacy, people that are paranoid, people that want to whistleblow, harass, defraud, threaten, etc. You will get the good with the bad.
Unfortunately, we seem to be headed in the direction that NO bad is allowed, so we are supposed to be OK with giving up all the good.
As a basic example, you can't punish or prevent all slander while allowing whistleblowing. The concept of allowing truly anonymous good behavior while preventing anonymous bad behavior is an impossible goal.
Normally one would expect "innocent until proven guilty" to take precedence and allow anonymous behavior while tolerating difficulty in catching criminals, but there are currently just too many well-funded groups with a stake that push the legislation in the other direction.
A "patent troll" is someone that takes advantage of patent law for monetary gain based on the innovation of others. Patent trolls aren't trying to claim reward for what is theirs. They simply game the system and out-maneuver the innovators such that, by legal definition (but not common sense) they are entitled to reward.
As such, it's mostly counterproductive for the purpose that the method of patenting was intended to serve. (encouraging and rewarding innovation)
Patent trolls siphon off some of the rewards of innovation through litigation and through the licensing of innovations where they themselves were not the innovator.
Fortunately, a lot of courts (at least those that'd don't directly benefit from the litigations, such as texas east district) have identified these people as taking advantage of the legal system and costing it money in exchange not for the support of innovation, but for the enrichment of the trolls and stifling of innovation, and are starting to push for change.
Here's the droid you're looking for:
http://4.bp.blogspot.com/-bg_P3VNyJJ0/TnTMgl3jtSI/AAAAAAAAA68/FcUfXOeEeQw/s1600/hell%2527s%2Bsuggestion%2Bbox%2B-%2Bfar%2Bside.jpg
this level of unemployment cannot be fixed by manufacturers moving jobs around. stopping the populace from breeding like rabbits is about the only useful solution. There's a very good reason china is pushing for "one child per family". Their bean counters know how to do math, and see what's coming. It's a problem that takes a generation to manifest. They're already trying to head it off, yet they're already getting their feet wet in the problem. I'd say they took action right on time.
I also find it ironic that the thread has turned in this direction. Employ underage? you child-slaving monster! Don't employ underage? throwing them to the gutters to starve! damned if you do, damned if you don't.
It's not the employers that created this problem, and it's not within their power to fix. All they can do is move the shells around and make it look a little different from a static viewpoint.
interesting to see how a joe average gets smacked down like a gnat with a buick on youtube, but then we see the exact opposite here? Or didn't they file a takedown notice?
I think what concerns the educated public the most is RIM's lack of intelligence
Just because an industry is based on your home turf is no excuse to allow embarrassing performance to continue to drag itself out. If Lenovo thinks they can "fix" RIM, I say let them try. They can't really do much worse at this point. Any business that's not totally dominate by a PHB should have by now at least started diversifying critical resources away from reliance on RIM.
I'm surprised this issue hasn't been tossed out onto a We the People petition?
Although recently we've seen a few of those used for stupid things (death star) as well as being flat out trampled on a few times with responses that basically said "we don't feel like telling you that", it would still be nice to see it out there.
Wasn't it inksys that bought cisco awhile ago? and now the hybrid is selling linksys?
either I was confused, or I am confused. Or maybe a bit of both.
Friend of mine got his laptop bag stolen at an airport. He set it on the conveyor to run through the xray and there was someone already on the other end of the scanner waiting for it. He grabbed it, and just plain ran off with it. The wonderful security guards are not allowed to leave their post to give chase, and merely radio'd in the thief. Who got away. They had the regular exits covered, but he ran straight for and out a fire door (setting off the alarm, which obviously he did not care about) to a waiting car.
You don't even have to leave your luggage for a few minutes. Five seconds is enough. You're not good, you're lucky. And one day you won't be lucky.
google for "purse alarm". there are a wide variety of options for things that make noise if disconnected.
for proximity, goole all the options with 'child proximity alarm", this is probably more up your alley. Some only sound an alarm on the kid, some only on the parent with the remote, and some do both. that's up to you what you want.
final option for gps is a smartphone with a "find my phone" option. Like with the iphone where you can see where it's at from a computer. that would be useful if the bag disappeared without setting off whatever else you had watching it like a proximity alarm, or if they managed to outrun you and hop in a car etc with your bag,
I see more to this "milestone" thing than a single hurdle. It's probably better to look at it as a "product development cycle" where there are several tiers a product has to pass through, with similar but higher requirements at each step.
1. throw some money at it
2. add featuers
3. remove / combine features / refine
4. in-house and user testing
5. decide if it's worth continuing
and then it repeats, with more money, fewer new features, more careful and thorough refining, and more thorough testing at each next stage. Go through that three or more times and you will probably have a few winners out of the hundreds you went in with at the start.
You could call passing each iteration's part 5 a "milestone" I suppose. But most of the discussion above acts like it only hits this point once during development. And I suppose for some, that's true. But that's a bad way to do it. You can't consistently produce a lot of winners and very few bombs if your development tree has no depth to it.
I like how a previous comment discussed Phillips and their "throw a bunch of crap at the wall and see what sticks." That really suggests a 1 or 2 iteration development cycle, and it's going to perform poorly over the long run.
I don't see any idealogical difference between each of these "milestones". Either it's worth continuing on, or it's not. If your development wing is only supposed to bring one product to market, then it's a simple game of "survival". If you start with 250 ideas you can just say you're going to go to the next stage after removing 75% of the ideas, then it's just a simple comparison, and you still have enough "keepers" to carry forward that things that are still "a little iffy" can make the cut, at least once. Do that four times and you are down to ONE idea. No need to change milestone strategies at each stage.
If you've got low depth, say 2 levels, think of how you'd have to do it. Split it evenly, and you have to drop over 90% of your ideas both times. The first cut probably won't be hard, but then having to compare 16 items to come up with the one best, you could easily kill what would have been the best idea. Do a hard cut on the first round and you are down to having to pick so few to keep from amongst so many you may as well pick them by lotto. A deep cut on round 2 I hope everyone sees would be suicidal. Milestone depth is the only solution to this problem.
A substantial part of their score was for things that very specifically were not actively being exploited. They were testing the heuristics to see if it could identify "virus and malware-like behavior". You can't rely on software updates and AV definition updates to protect you from zero-day's, that's 100% on the head of your AV software to keep you safe from.
And MS fails miserably at protecting users from zero-days. They flunked, and they deserved to flunk. There's just too many new viruses and malware being developed every day to try to function on a blacklist-only basis. Some behaviors need to be whitelist only, and quite a few need to be greylisted with heuristics so that your system survives long enough for the new exploit to get added to tomorrow's definitions file.
I'd have to agree with that. It had good replayability also because you had so many binary upgrade decisions to make. And most levels could be solved with at least two distinct play styles. (stealth and power)
It was also refreshing to have three factions fighting for your loyalty instead of the usual two found in other games that play that way.
The most simplified counter-example is "lets say by random chance your bt client decides to download the first piece of the video last". There's nothing to prevent it from downloading one of the first pieces last. And in that case, regardless of how fast your connection is, you WILL be waiting until the download finishes before starting to watch it.
Not sure if you're trolling or just uninformed. So maybe I'm feeding.
Torrent files don't download sequentially, from start to finish. Clients that follow spec will randomly select a piece to download, possibly influenced by availability or speed of peers with pieces. The larger the file gets, the lower the odds are that you're not missing an early piece. For a movie that has 1000 pieces, the odds of having ALL of the first 250 pieces at any point before you hit the 95% downloaded point are astronomically low.
(and i haven't seen a bt client yet that you can prefer sequential download with)
It's easy to see why they wrote it that way. If one seed starts with 10 peers, and all of them download in order, then the peers won't have many pieces available from each other. (they'll all have piece 1, then all have piece 2, etc) OTOH if the seed hands out a different piece to all 10 peers, when those complete, now each peer has a choice of 9 other pieces to get from other peers, in addition to the seed, and data starts flowing around between peers instead of just between peers and the seed. And that's what makes bt fast.