Google Maps routing is pretty poor quality for the number of years we are from dashboard GPS and desktop routing software. Streets and Trips 2002 did a better job at handling multiple destinations by a huge margin.
Google Maps puts inappropriate weight towards making a route more complicated with short freeway hops (hop on for the next immediate exit). It also looks like they are taking some kind of payola from toll authorities, as it frequently tries to give toll routes when a non-toll route has an ETA within 1-2 minutes.
Lastly, their lack of intelligence on street construction (major multi-month/year affairs) is pathetic when the exact advantage of an online based mapping service should be current, dynamic updates. The only area they are leveraging there is current traffic knowledge, yet they fail to notice when a major street abruptly hits 0 traffic, indicating a closure.
The driving preferences should be vastly more granular and more than on/off check boxes for weighted route offerings. It would be nice if they would bring back future predicted ETAs to the web product and add it to the smartphone apps.
Keep in mind they still call it "Beta" software.
I like Google, I am fine exchanging a controllable portion of anonymized habits for free services, but just keep in mind that they are having trouble serving two masters.
That is a terrible broken design. It fails to accommodate a nice portion of America males. Not because we're fat necessarily, but because it doesn't have space for person over 6'1". Fixed pods are still an issue for the overweight though who if extreme enough can purchase a second seat due the intelligently modular design of armrests. Lastly, while so much additional structure for pods would increase structural rigidity and safety of airframes, the additional weight would cost more in gas than the increased passenger density. The exception would be carbon fiber, but the design is likely too complex to make out of carbon fiber affordably.
The random switches between metric and imperial measurements sounds like a designer / designer's marketing idea. Everyone involved in that design should burn in hell for even releasing publicly an idea with so very many obvious flaws. They are math / physics retarded mechanical engineers, of which there is a shocking population allowed to live.
The next logical step is to remove overhead bins, mostly by using Spirit's model for baggage fees, and then have the passengers stand against partitions. The partitions will have the necessary safety restraints for take off and landing.
Following that, we can remove the partitions and just sell standing space based on a weighted priority boarding system, and let as many board as possible, while being able to hold onto horizontal and vertical railing systems like a subway.
It's incredibly taxing to see the number of senior programmers/engineers/researchers/etc. get moved into management who completely lack the appropriate skill set. Definitely agree though that food service and retail management experiences generally give people the correct skill set. For a variety of reasons, those industries better train and prepare management as well as filter the crap. White collar offices tend to lack effective management training and let's not even get into the whole university MBA factories where management vocabulary trumps actual management ability.
I was thinking more or less this. The reality is that some people are just one trick drones. Many of the office dweller jobs have become ridiculously narrow and monotonous in scope, just to prevent anyone from being able to fail short of blatant negligence or malice. It's beyond the point of specialization to the point that college degree requirements need to be scrapped and replaced with a "you have no experience, so for the first 90 days you work at reduced salary and subject to abrupt termination for poor performance" pseudo apprenticeship system.
This is actually a problem caused exclusively by the logical incorrectness of 0th based numbering to which so many programmers incorrectly subscribe.
The rule anything divided by itself is one does not apply because the concept of zero is that it is NOT in fact anything. It is the definition of nothing, of lacking substance.
Each time I run into equipment or object enumeration which starts with the first object numbered as "0" I shake my head at the fact that some engineer or programmer has missed the point of natural numbers: to delineate quantities of discreet objects. There is no such thing as "port 0" on a switch or "disk 0" in a SAN array, but I see this logic all the time because it makes writing loops and such tidier in software. The first port is "port 1", that is it is 1 whole, discreet object of the kind "port".
This is an area of abstract thinking which is disconnected from the physical world and thus not understanding the intrinsic relationships of physics and mathematics. Zero is a special number used for defining nothing, not something.
I've talked to a couple of guys trained to fly military drones. There isn't a big stress issue going on. The real challenge is that with their training there are very high paying private sector and DoD consultancy jobs.
Not being boots on the ground and such is leaving these guys with less sense of camaraderie than other soldiers. They don't feel compelled to re-up to fight alongside their brethren the way grunts and conventional pilots do.
The stress discussion is just a smokescreen for the fact that they can't keep people in the jobs at the moment.
Aircraft inspection already includes X-ray checks of key areas at various service intervals. It's the best nondestructive way to know the fatigue state of aluminum, which just generally sucks as a material always fatiguing.
Except that alchemy works if you bump over from chemistry to nuclear physics. It's trivial to turn lead into gold, but energetically unfavorable (what you are saying with scalability) and lots of nasty radiation.
It's clearly not an OS designed for millions of users. It's meant for tinkerers who want to run one application at a time because they know they are more productive that way. George R. R. Martin uses an old DOS box with Wordstar for exactly this reason.
Multitasking may let a person do more stuff, but for the majority of people, it results in less total stuff getting done.
More interesting is that Core architecture is based on pre-Netburst P2/3 microprocessor designs. Netburst was a mistake that Intel quickly (in development cycle terms) admitted.
There's also the aggregate bandwidth problem. Also the fact that Google and Facebook have been pushing HTTPS everywhere and customized content so hard that most forms of caching break.
I don't think it is pro-bono if you're providing it to paying customers. If anything this should become the norm, similar to the way an insurance company has lawyers to aid in handling automobile accidents.
There are two simple things which would make the process functional for both sides.
1) Put cameras on the officers
2) Give the magistrate court and a initial public defender / citizen advocate the right to view officer body cam footage and make a provisional judgement to hold or release a person on the likelihood of a crime based on the officer's body cam footage. Our magistrate courts are way under-powered here. All one needs to look at is the Innocence Project case for Michael Hanline in which the magistrate court report was ignored by the District Court. That district court judge belongs in jail and it is as simple as that (http://californiainnocenceproject.org/read-their-stories/michael-hanline/)
If a cop is making an iffy call about obstructing because a person is filming 20 feet away, release the person (no overnight/multi-day stay bullshit) pending the officer and attorney general continuing to press charges. If somebody filming is trying to stand over a body in the street after the cop tells them to back away to an appropriate distance and they refuse, hold them until bail is made.
There is a third stage which neither the police nor much of the unwashed masses are ready for in which one can lawfully refuse an officer and defend themselves as they would against any other violent, armed attacker. The judgement of the average person just isn't there yet, as a function primarily of US culture. There are plenty of other places in the world where that works just fine, but it's not for the US just yet. We have a cultural / social maturity problem and it is what is feeding the us vs them attitude that the average police department has. It's a bit hard to argue with the Baltimore police doctrine, proven repeatedly over the years, that many of their officers believe the value of a black man's life is more than a bullet but less than a taser cartridge when protests turn into looting and vandalism with selfish rather and communal interests as the driving force.
Not to mention the potential ageism in the search criteria. C devs are naturally going to have an age and experience bell curve shifted years older than PHP or Java. This could mean the phrase "ugly hack" is more in keeping with the age range's vernacular. I'm sure the posers churning out PHP are more likely to comment/* omgwtfbbq */ and variations than the C group.
Secondly, more experienced programmers and those working with more complex code, both things more likely with C than the other languages, are likely better able to identify what an "ugly hack" looks like. Lastly, C specifically lends itself to a lot of magic tricks for optimization that could carry "ugly hack" (e.g. the famous Quake Fast Inverse Square Root) which simply aren't worth implementing in other languages because they are too slow to matter.
Security "experts" who do little more than run a pre-built set of scripts against websites. It's five minutes of work for them to roll the dice that they might get a payout. Even if the security firm invested serious time in developing the tools and scripts, their spamming their tests out and wanting a payday not commensurate with the scalability of their approach.
This. And it was stupid of the asker to even pull the "I don't trust cloud providers" with Google and Microsoft not going anywhere for awhile. Microsoft in particular respects privacy and permissions is the better bet. Worried about hackers, as you said, encrypt locally and upload.
Yep, this is exactly how I want computing laws to work, based on the 70-80's standards. It's under those same standards that most forms of internet advertising are illegal forms of hacking and computer time use.
When you roll forward to the concept of smart phones with battery restrictions and "just enough" computing power as a key concept, suddenly the incredible usage loss by the owner whenever inundated by advertisements looks janky. It makes sense as to why no one but the owner / administrator may authorize code to execute as far as the legal wordings go from way back in time.
I make it a rule to never pay for an app to provide what should be basic OS functionality. Whole device encryption is pretty well handled in Android. There are lots of areas where I make do with available tools, like TRWP for point in time / image backups. If I find a specific, basic feature missing, I'm not terribly above obtaining decent software through alternate channels (root / administrative access browsers, configuration changing applications).
It's silly that Windows XP from 2001 still has a better list of basic OS functionality, bundled applications, and diagnostic tools than either Android or iOS can claim today. I'd gladly pay for a phone OS separate from my phone if I could get solid functionality covered.
Android also falls way short of iOS and Windows Phone for device to device migration. Settings and preferences from apps only migrate if they are Google first party or if the app developer stores your settings on their server and uses a Google (or similar) sign in. Setting up a new device, or even ROM is nowadays a longer process (~4 hours) than on a PC to get everything operational, and that is with using tools like Titanium Backup and similar.
If they are abandoning the teaching of core, pure subjects the way it sounds, it is certainly a bad move.
The correct method would be to have core classes in key areas, such as math, various sciences, literature and rhetoric, history and social studies AND THEN have derivative classes which fused concepts in practical and vocational settings. The chief problem in most educational settings is the student's lack of will to connect the dots. When I took a high school job at a restaurant, I could immediately see the applications from core Biology of sterile technique and protein / carbohydrate denaturing, most people fail to see those connections. A topical class like "Cafeteria Services" should teach students where to draw from key knowledge in a derivative and synthetic sense: math for accounting, inventory, and projecting trends, psychology and rhetoric for synthesizing menus and advertisements, biology for cooking and sanitation, and so forth.
This approach also better allows students to learn how to leverage their general "core subject" knowledge into changes in vocation because they have already learned the methods to apply what seems to some like dry, lifeless facts and calculations. Further, using such a system in the secondary levels (middle school / jr high / high school) would help students make better choices in the University system. The lack of topics usefully linking back to art history and anthropology would certainly help students to understand the lack of career value, except unto themselves in those fields. At the same time, it should increase the desire to get at least a basic understanding of such subjects as one sees the limited, but interesting ways to apply a general knowledge base.
The problem is that police would respond with that level of force based upon an anonymous tip.
The problem is more the police than the swatters. The swatters are malicious actors. The police are failing to perform as good actors by following through the least bit of due diligence in these situations. Before breaking down the door, they should at least have a seasoned, senior officer knock to see if anything seems odd first.
The problem is a police force filled with the same adrenaline junky types that call in the swatting. They see an opportunity to break a door down and going running around in full CQC gear and they lose their composure. If they were actually interested in public safety, that wouldn't be their first impulse reaction to a potential emergency situation, negotiation and diffusal would be.
Testing is for chumps who believe in waterfall development and all that jazz. The modern edict of Agile, the end users will quickly pass any issues up through the proper channels and developers will prioritize and fix as them deem appropriate. The customer isn't the programmer's boss.
I'm glad somebody mentioned Checkpoint, though it sounds like you aren't a fan.
Having worked plenty with Cisco, Juniper, Sonicwall and Checkpoint gear, the Checkpoint stuff is my favorite to build out and easiest to administer. Also the easiest by a good amount to take a quick glance at the configuration or log and know exactly what is going on. It does take a lot of overhead in the way of a dedicated configuration utility which only runs on Windows.
Juniper is a close second, and they definitely have superiority in the CLI department. Their C style nested configuration file is the best there is with beautiful logic and organization.
Cisco's configuration text files are a joke. They look like the result of letting every architect have a hand in the action without anyone dictating a direction. ASDM is even worse, featuring every bad design and logic decision possible and just feeling like a senior project at a university with a mediocre CS department.
Getting past the UI side, it's nice dealing with equipment that logically validates the configuration directives (Checkpoint / Juniper) instead of letting anything go even if it won't work.
As the oldest of 5 children, one thought came to mind. How well does it stand up to crayons? I would assume from the hydrophobic nature it will have no problems with water based markers, though I am curious on how well it handles Sharpies and other solvent based ones.
They might get head tracking latency, accuracy, and precision down in this generation of VR. It will be another 20 years at least before they conquer depth of field / focus / light field projection in a meaningful way that works with the biology of the eye. Entirely new, non-planar display technologies will be required.
Slashdot Mirror
Google Maps routing is pretty poor quality for the number of years we are from dashboard GPS and desktop routing software. Streets and Trips 2002 did a better job at handling multiple destinations by a huge margin.
Google Maps puts inappropriate weight towards making a route more complicated with short freeway hops (hop on for the next immediate exit). It also looks like they are taking some kind of payola from toll authorities, as it frequently tries to give toll routes when a non-toll route has an ETA within 1-2 minutes.
Lastly, their lack of intelligence on street construction (major multi-month/year affairs) is pathetic when the exact advantage of an online based mapping service should be current, dynamic updates. The only area they are leveraging there is current traffic knowledge, yet they fail to notice when a major street abruptly hits 0 traffic, indicating a closure.
The driving preferences should be vastly more granular and more than on/off check boxes for weighted route offerings. It would be nice if they would bring back future predicted ETAs to the web product and add it to the smartphone apps.
Keep in mind they still call it "Beta" software.
I like Google, I am fine exchanging a controllable portion of anonymized habits for free services, but just keep in mind that they are having trouble serving two masters.
That is a terrible broken design. It fails to accommodate a nice portion of America males. Not because we're fat necessarily, but because it doesn't have space for person over 6'1". Fixed pods are still an issue for the overweight though who if extreme enough can purchase a second seat due the intelligently modular design of armrests. Lastly, while so much additional structure for pods would increase structural rigidity and safety of airframes, the additional weight would cost more in gas than the increased passenger density. The exception would be carbon fiber, but the design is likely too complex to make out of carbon fiber affordably.
The random switches between metric and imperial measurements sounds like a designer / designer's marketing idea. Everyone involved in that design should burn in hell for even releasing publicly an idea with so very many obvious flaws. They are math / physics retarded mechanical engineers, of which there is a shocking population allowed to live.
The next logical step is to remove overhead bins, mostly by using Spirit's model for baggage fees, and then have the passengers stand against partitions. The partitions will have the necessary safety restraints for take off and landing.
Following that, we can remove the partitions and just sell standing space based on a weighted priority boarding system, and let as many board as possible, while being able to hold onto horizontal and vertical railing systems like a subway.
It's incredibly taxing to see the number of senior programmers/engineers/researchers/etc. get moved into management who completely lack the appropriate skill set. Definitely agree though that food service and retail management experiences generally give people the correct skill set. For a variety of reasons, those industries better train and prepare management as well as filter the crap. White collar offices tend to lack effective management training and let's not even get into the whole university MBA factories where management vocabulary trumps actual management ability.
I was thinking more or less this. The reality is that some people are just one trick drones. Many of the office dweller jobs have become ridiculously narrow and monotonous in scope, just to prevent anyone from being able to fail short of blatant negligence or malice. It's beyond the point of specialization to the point that college degree requirements need to be scrapped and replaced with a "you have no experience, so for the first 90 days you work at reduced salary and subject to abrupt termination for poor performance" pseudo apprenticeship system.
I hear Bernadette's shrill voice muttering about the one on amphetamines that ripped off it's own tail to get free before disemboweling the others.
This is actually a problem caused exclusively by the logical incorrectness of 0th based numbering to which so many programmers incorrectly subscribe.
The rule anything divided by itself is one does not apply because the concept of zero is that it is NOT in fact anything. It is the definition of nothing, of lacking substance.
Each time I run into equipment or object enumeration which starts with the first object numbered as "0" I shake my head at the fact that some engineer or programmer has missed the point of natural numbers: to delineate quantities of discreet objects. There is no such thing as "port 0" on a switch or "disk 0" in a SAN array, but I see this logic all the time because it makes writing loops and such tidier in software. The first port is "port 1", that is it is 1 whole, discreet object of the kind "port".
This is an area of abstract thinking which is disconnected from the physical world and thus not understanding the intrinsic relationships of physics and mathematics. Zero is a special number used for defining nothing, not something.
I've talked to a couple of guys trained to fly military drones. There isn't a big stress issue going on. The real challenge is that with their training there are very high paying private sector and DoD consultancy jobs.
Not being boots on the ground and such is leaving these guys with less sense of camaraderie than other soldiers. They don't feel compelled to re-up to fight alongside their brethren the way grunts and conventional pilots do.
The stress discussion is just a smokescreen for the fact that they can't keep people in the jobs at the moment.
Aircraft inspection already includes X-ray checks of key areas at various service intervals. It's the best nondestructive way to know the fatigue state of aluminum, which just generally sucks as a material always fatiguing.
Except that alchemy works if you bump over from chemistry to nuclear physics. It's trivial to turn lead into gold, but energetically unfavorable (what you are saying with scalability) and lots of nasty radiation.
It's clearly not an OS designed for millions of users. It's meant for tinkerers who want to run one application at a time because they know they are more productive that way. George R. R. Martin uses an old DOS box with Wordstar for exactly this reason.
Multitasking may let a person do more stuff, but for the majority of people, it results in less total stuff getting done.
More interesting is that Core architecture is based on pre-Netburst P2/3 microprocessor designs. Netburst was a mistake that Intel quickly (in development cycle terms) admitted.
There's also the aggregate bandwidth problem. Also the fact that Google and Facebook have been pushing HTTPS everywhere and customized content so hard that most forms of caching break.
I don't think it is pro-bono if you're providing it to paying customers. If anything this should become the norm, similar to the way an insurance company has lawyers to aid in handling automobile accidents.
There are two simple things which would make the process functional for both sides.
1) Put cameras on the officers
2) Give the magistrate court and a initial public defender / citizen advocate the right to view officer body cam footage and make a provisional judgement to hold or release a person on the likelihood of a crime based on the officer's body cam footage. Our magistrate courts are way under-powered here. All one needs to look at is the Innocence Project case for Michael Hanline in which the magistrate court report was ignored by the District Court. That district court judge belongs in jail and it is as simple as that (http://californiainnocenceproject.org/read-their-stories/michael-hanline/)
If a cop is making an iffy call about obstructing because a person is filming 20 feet away, release the person (no overnight/multi-day stay bullshit) pending the officer and attorney general continuing to press charges. If somebody filming is trying to stand over a body in the street after the cop tells them to back away to an appropriate distance and they refuse, hold them until bail is made.
There is a third stage which neither the police nor much of the unwashed masses are ready for in which one can lawfully refuse an officer and defend themselves as they would against any other violent, armed attacker. The judgement of the average person just isn't there yet, as a function primarily of US culture. There are plenty of other places in the world where that works just fine, but it's not for the US just yet. We have a cultural / social maturity problem and it is what is feeding the us vs them attitude that the average police department has. It's a bit hard to argue with the Baltimore police doctrine, proven repeatedly over the years, that many of their officers believe the value of a black man's life is more than a bullet but less than a taser cartridge when protests turn into looting and vandalism with selfish rather and communal interests as the driving force.
Not to mention the potential ageism in the search criteria. C devs are naturally going to have an age and experience bell curve shifted years older than PHP or Java. This could mean the phrase "ugly hack" is more in keeping with the age range's vernacular. I'm sure the posers churning out PHP are more likely to comment /* omgwtfbbq */ and variations than the C group.
Secondly, more experienced programmers and those working with more complex code, both things more likely with C than the other languages, are likely better able to identify what an "ugly hack" looks like. Lastly, C specifically lends itself to a lot of magic tricks for optimization that could carry "ugly hack" (e.g. the famous Quake Fast Inverse Square Root) which simply aren't worth implementing in other languages because they are too slow to matter.
Security "experts" who do little more than run a pre-built set of scripts against websites. It's five minutes of work for them to roll the dice that they might get a payout. Even if the security firm invested serious time in developing the tools and scripts, their spamming their tests out and wanting a payday not commensurate with the scalability of their approach.
This. And it was stupid of the asker to even pull the "I don't trust cloud providers" with Google and Microsoft not going anywhere for awhile. Microsoft in particular respects privacy and permissions is the better bet. Worried about hackers, as you said, encrypt locally and upload.
Yep, this is exactly how I want computing laws to work, based on the 70-80's standards. It's under those same standards that most forms of internet advertising are illegal forms of hacking and computer time use.
When you roll forward to the concept of smart phones with battery restrictions and "just enough" computing power as a key concept, suddenly the incredible usage loss by the owner whenever inundated by advertisements looks janky. It makes sense as to why no one but the owner / administrator may authorize code to execute as far as the legal wordings go from way back in time.
I make it a rule to never pay for an app to provide what should be basic OS functionality. Whole device encryption is pretty well handled in Android. There are lots of areas where I make do with available tools, like TRWP for point in time / image backups. If I find a specific, basic feature missing, I'm not terribly above obtaining decent software through alternate channels (root / administrative access browsers, configuration changing applications).
It's silly that Windows XP from 2001 still has a better list of basic OS functionality, bundled applications, and diagnostic tools than either Android or iOS can claim today. I'd gladly pay for a phone OS separate from my phone if I could get solid functionality covered.
Android also falls way short of iOS and Windows Phone for device to device migration. Settings and preferences from apps only migrate if they are Google first party or if the app developer stores your settings on their server and uses a Google (or similar) sign in. Setting up a new device, or even ROM is nowadays a longer process (~4 hours) than on a PC to get everything operational, and that is with using tools like Titanium Backup and similar.
If they are abandoning the teaching of core, pure subjects the way it sounds, it is certainly a bad move.
The correct method would be to have core classes in key areas, such as math, various sciences, literature and rhetoric, history and social studies AND THEN have derivative classes which fused concepts in practical and vocational settings. The chief problem in most educational settings is the student's lack of will to connect the dots. When I took a high school job at a restaurant, I could immediately see the applications from core Biology of sterile technique and protein / carbohydrate denaturing, most people fail to see those connections. A topical class like "Cafeteria Services" should teach students where to draw from key knowledge in a derivative and synthetic sense: math for accounting, inventory, and projecting trends, psychology and rhetoric for synthesizing menus and advertisements, biology for cooking and sanitation, and so forth.
This approach also better allows students to learn how to leverage their general "core subject" knowledge into changes in vocation because they have already learned the methods to apply what seems to some like dry, lifeless facts and calculations. Further, using such a system in the secondary levels (middle school / jr high / high school) would help students make better choices in the University system. The lack of topics usefully linking back to art history and anthropology would certainly help students to understand the lack of career value, except unto themselves in those fields. At the same time, it should increase the desire to get at least a basic understanding of such subjects as one sees the limited, but interesting ways to apply a general knowledge base.
The problem is that police would respond with that level of force based upon an anonymous tip.
The problem is more the police than the swatters. The swatters are malicious actors. The police are failing to perform as good actors by following through the least bit of due diligence in these situations. Before breaking down the door, they should at least have a seasoned, senior officer knock to see if anything seems odd first.
The problem is a police force filled with the same adrenaline junky types that call in the swatting. They see an opportunity to break a door down and going running around in full CQC gear and they lose their composure. If they were actually interested in public safety, that wouldn't be their first impulse reaction to a potential emergency situation, negotiation and diffusal would be.
Testing is for chumps who believe in waterfall development and all that jazz. The modern edict of Agile, the end users will quickly pass any issues up through the proper channels and developers will prioritize and fix as them deem appropriate. The customer isn't the programmer's boss.
I'm glad somebody mentioned Checkpoint, though it sounds like you aren't a fan.
Having worked plenty with Cisco, Juniper, Sonicwall and Checkpoint gear, the Checkpoint stuff is my favorite to build out and easiest to administer. Also the easiest by a good amount to take a quick glance at the configuration or log and know exactly what is going on. It does take a lot of overhead in the way of a dedicated configuration utility which only runs on Windows.
Juniper is a close second, and they definitely have superiority in the CLI department. Their C style nested configuration file is the best there is with beautiful logic and organization.
Cisco's configuration text files are a joke. They look like the result of letting every architect have a hand in the action without anyone dictating a direction. ASDM is even worse, featuring every bad design and logic decision possible and just feeling like a senior project at a university with a mediocre CS department.
Getting past the UI side, it's nice dealing with equipment that logically validates the configuration directives (Checkpoint / Juniper) instead of letting anything go even if it won't work.
As the oldest of 5 children, one thought came to mind. How well does it stand up to crayons? I would assume from the hydrophobic nature it will have no problems with water based markers, though I am curious on how well it handles Sharpies and other solvent based ones.
They might get head tracking latency, accuracy, and precision down in this generation of VR. It will be another 20 years at least before they conquer depth of field / focus / light field projection in a meaningful way that works with the biology of the eye. Entirely new, non-planar display technologies will be required.