Haven't really followed the One or S, since I don't own them. The issue is that not many of the modders actually use them now - they tend to get so many donations that they run the latest and greatest, and so they don't support the older phones the way they used to.
Back when everybody running Android had an ADP or G1 (identical hardware) 3rd-party updates were very well supported, since that's what EVERYBODY used. However, once the newer phones started coming out, and once donations to the bigger mods ramped up, the leading devs tended to move on and buy new phones all the time.
Don't get me wrong - I'm grateful for their efforts and I get a lot more than I pay for (nothing). However, fragmentation has not really helped the driver hacking scene much - to few devs working on too many phones to really keep most of them up-to-date.
I think the main problem is application developers not understanding anything about database theory. The vast majority of databases I encounter are not normalized at all, and it's almost always because they were designed by a developer with no database background.
Unfortunately I've seen the opposite extreme as well - databases that are SO normalized that to find anything at all you need about 35 joins. Sure, everything is arbitrarily expandable, but basically there is no practical way to work with it without writing code. We have a database at work that anytime I have to actually look inside and see what is going on I end up copying and pasting from my personal library of 1.5 page long SQL SELECTs. People don't take much advantage of all the expandability since doing anything to the system costs a ton of money.
This whole idea of throwing random strings of structured text into a database column, and then relying entirely on the program code to parse and use it... well, why the hell even use a relational database, then?
And that is the opposite extreme. I've also seen a database at work where half the fields in the databases are CLOBs full of XML. That is even more painful to deal with than the 1.5 page SQL since you end up replicating half the database over the network to do anything that wasn't fielded at the DB-level - like a NoSQL from hell.
If someone died, he or she has a row in a Deaths table, which joins to the People table; otherwise, not; no NULLS necessary.
The problem with this sort of logic is that it gets really cumbersome to work with.
I want to store an optional address for each record. So, that means a place table and a many:many relationship between people and places. Oh, but a place might or might not have an address line 2 or zip code, so that means a sub-street and zip table too, maybe 1:0-1 on the first and many:many on the second. You'd end up with 30,000 views in such a database, and those views basically look just like the tables you've gotten rid of, and those are what everybody would use, and if you wanted to take advantage of your highly-normalized design and assign two places to a single person your code would probably break since that now leads to duplicate rows in the view that the code wasn't designed to handle, since we don't value competent programmers in most enterprises that need these big complex databases anyway.
I can get how mathematically it is more elegant to get rid of the nulls, but when you have people with relatively minimal skill programming and manipulating the database it isn't very practical.
I thought much of the Android code is GPL'ed. If they distribute early versions of Android to selected developers, wouldn't they also need to give away the Android source code to anyone else who demands it at that time?
No, you only have to offer the source to the GPL components, and only to those you offer the binaries to. You can't stop them from redistributing it, but they likely have various incentives not to anyway (if they got in on some special deal that gives them a leg up, why would they go and give away their advantage?). For non-GPL components there is no obligation to give anything away.
The official google phone is the ONLY one to buy IMHO.
You sure about that?
The 3.5 year old old official phone (ADP) only received major OS updates for about a year (Donut).
The two year old official phone (Nexus One) only received major OS updates for about a year (Gingerbread came out less than a year after its release, and that is all it runs).
The 1.5 year old official phone (Nexus S) is just only getting Ice Cream Sandwich now - so you could say that it had updates for 1.5 years after its release date, though only to a version that was released less than a year after it came out (so you could argue it is a 1 year update six months late).
Google has yet to roll out any update to one of its phones that was developed more than a year after the phone was FIRST sold. Anybody who bought the phone just before it was discontinued might have gotten an update at most a month or two later, if they got any updates at all.
So, if you buy a non-Nexus phone you'll be lucky to get any updates at all (it is hit and miss), and if you buy a Nexus phone you may get a single update, if you buy it pretty early. I love Android, but none of their devices really support upgrades much at all.
Interesting that he was not permitted a trial by jury... In fact, I don't see anything about a trial at all - just a resolution passed by a legislative body.
I think that's the whole point here. I imagine most of the folks in Gitmo have had some kind of hearing. The issue is that they have not had public criminal trials before a jury of their "peers."
Uh, care to cite any examples of spies caught on a battlefield in that era who were given trials with due process, a jury of peers, and the rest of the bill of rights?
Cases where people have been found innocent after a long period of expense and the destruction of their lives? There are tons of cases of this. Sure, few if any involve people protesting security theater, but lots of people get trampled on in the justice system. I sat on jury duty for one of them - some poor schlep who was held for 30 days in prison after the police botched up a lineup until the DA realized they had the wrong guy.
THAT document says they have rights and demands that no part of the U.S. government shall violate those rights.
Where, exactly?
The last time I checked soldiers don't attempt to arrest enemy LAWFUL combatants (let alone unlawful ones) and hold trials for them - they shoot them. Armies run by the people who wrote the constitution operated no differently, so somehow I doubt they felt that the kinds of absolute rights you are getting at existed.
Also, the formal distinctions between lawful and unlawful combatants date back to the Geneva conventions which are more recent than the constitution. Granted, they're based on practices which are much older.
My suggestion wasn't that actual evidence of imminent harm should be required in order to make ANY arrest. My point was that this should be required if the police want to arrest somebody before the investigation is completed. If a thorough investigation suggests that somebody is guilty of a crime, then get an arrest warrant from a grand jury and then take them to court.
This was in reply to the suggestion that the police generally arrest first and investigate later.
I realize that you're describing how the system often works, and I don't dispute that. However, the result of this policy is lots of innocent people having their lives ruined merely for being suspected of a crime.
If I recall correctly, the Phoenix was never responsible for any air-to-air kills during its service.
Keep in mind that the scenario it was designed to handle never came up either, thankfully (that would have basically been a WW3 scenario). I think the original use case was a group of dozens of bombers headed for the carrier, and the carrier had to scramble and intercept all of them. Since the bombers could fire at long range, the interceptor had to hit them at long range. Bombers would have a difficult time defeating a missile I would think - they weren't really designed to hit modern fighters.
Much of the stuff that was built in the cold war was designed for a scenario that never happened - a full-out war with the USSR.
Arresting someone who is innocent or is later found not guilty - no.
The problem is that in our system simply being arrested results in a huge grind that is effectively punishment regardless of the final outcome. People spend months to years in jail. Their reputations are tarnished immediately and regardless of outcome, in a society where that reputation follows you even internationally (you can't just move to the next town over and start a new life - arrests show up on background checks). People don't own their means of production like farmers of yore did - if you lose your job you have no income, and it is difficult to get a new job. A speedy trial doesn't mean next week either.
Everyone has human rights. That's why they are called human rights, all humans have them.
True, but I've never seen the US agree that people engaging in combat against soldiers have a right to not be shot by those soldiers.
You can call them "unlawful combatants" but they are either civilian criminals or POWs. Not giving them any rights, torturing and treating them as sub human just shames yourselves.
The US military deployed abroad is generally not subject to any non-US law beyond treaties to which the US is a party. That's why we call it the military - their job is to generally make a mess out of other countries (and I'm a big fan of using it a lot less often as a result). If people interfere with them then the US is under treaty obligation to afford those people certain rights if they are wearing a uniform. If they aren't, then the US has no legal obligation to afford them any rights at all. Non-combatants are protected where practical under both treaty obligations and general policy.
If that makes people less willing to host the US army so be it - people shouldn't be all that happy about having foreign armies present, and I'm not a big fan of having the US army deployed in places where they aren't actually busy shooting people (and hence the local opinion of their presence isn't of much worth anyway), and I'm also not a fan of having the army shoot people who haven't been shooting at us either. War is hell, and trying to make it less so ultimately just makes it more so.
Now, a lot of people in Gitmo aren't combatants captured on a battlefield at all, and I completely disagree with keeping these sorts of people there. They are criminals and should be charged as such.
For those who are civilians who did not engage in direct combat against military forces abroad I'd tend to agree with you - they should be arrested and treated as criminals.
For anybody who actually engaged in direct combat/sabotage against the US military, they should be classified as spies. I don't favor indefinite incarceration - probably better to either consider the problems ancient history and cut them loose, or just execute them. Historically execution was the method of choice, but it was typically administered on the spot (catch some guy planting a bomb near your camp not wearing a uniform, you bring them to the officer in charge, they get what intelligence they can out of the guy, and then shoot them (in more ancient times the head might go on a pike).
This status should not be given to anybody arrested in a non-combat situation, or to anybody arrested in a nation the US is not at war with, including anybody captured on US soil. These are criminals, and should be handled under criminal law no matter how inconvenient.
Unlawful combatants (aka civilians with guns) who are captured are not entitled to the rights of a POW, but ARE entitled to the rights of a civilian.
Citation? That is, a citation to some treaty to which the US is a party, or some law binding on the US military?
Foreign civilians on foreign soil have no rights under US law, at least not as far as I can tell. Principles of international law do of course encourage armies to avoid unnecessary harm to non-combatants, but we aren't talking about non-combatants here.
I'm a proponent of closing Gitmo for sure, but the whole reason the situation we're in actually exists is because the US isn't doing what armies have historically done - execute non-uniformed combatants on the spot.
You do realize that an arrest generally occurs before the full investigation, which precedes the decision of whether or not to file criminal charges... right?
Unless there is actual evidence of imminent harm, this should never happen. There's this thing called probable cause...
That the primary problem with people becoming obese being that they consume too much food is hardly news, nor is it news that food is much more abundant these days than it was in the past.
I think the real issue is what the underlying root cause is. Obviously if everybody ate less there would be less obesity. The important question is WHY people don't just do it.
Interesting. I think another factor that people ignore is epigenetics. It was found that the incidence rate of diabetes in some town in Europe that had good records seemed to be related to what people's grandparents ate.
I have no idea whether it is true, but if there are epigenetic factors at work, then the best we can hope to do is prevent obesity in kids who have not been born yet, or find some way to manipulate our own epigenetic programming.
My prediction is that the obesity epidemic is hopeless - people will gain weight no matter what you do. Oh, and best not to live near a graveyard - one of these days one of those old people buried there will collapse into a singularity.
Now let's have a round of useless speculation about what my results mean, without calling into question the fact that a model is not reality.
You should have played it before you bought it then, or choose your friends more carefully.
Sure, buyer's remorse will never be completely eliminated, but you have a LOT more ability to do due diligence when buying something that already exists than when buying something that hasn't been built yet.
What you describe is basically the intentional design of Bitcoin. It was designed so that the value of coins became about equal to the marginal cost of computing power. Of course, in the beginning there was a shortage.
As far as I can tell Bitcoin has some advantages and disadvantages compared to other forms of digital currency: 1. A big disadvantage is that transactions aren't anonymous - so why it would appeal to illegal activity is a mystery to me. It seems even less anonymous than conventional cash (which has serial numbers, but less history tracking in general).
2. The big advantage is that it doesn't need a trusted authority to run. Its value is self-regulating, as anybody can produce it and will do so anytime demand for currency results in it being overvalued. It does lack the ability to deflate the currency should there be an oversupply.
3. The currency also prevents double-spending, which other digital cash systems usually require a central bank to prevent. To some extent this mitigates the anonymity of other systems, since those receiving cash need to deposit it very quickly to confirm that it has value, which makes it harder to make the use of the currency untraceable.
Buying retail requires trust, it's just that the trust has been established.
The kind of trust we're talking about is very different.
In the case of a store I only have to trust that the thing in the box is SuperGame 3. In the case of Kickstarter I have to trust that somebody makes SuperGame 3 and that it is a fun game. The quality of the product itself is not in doubt with the retail model, and I can even buy straight from the manufacturer if I want to pay a little more and be assured of some level of quality.
I don't buy pre-releases - I want until I know that I want to play a game before forking out cash. That is what you lose with the shared risk model. Since you're an investor rather than a consumer you take on more risk. Potentially you also pay less, which is basically how risk works.
My main issue with KeepassX is that it isn't capable of running solely with an extension or bookmarklet, which means that it won't work on every OS I have. I use Lastpass as a result, though I'd prefer something equivalent that is open-source...
Keep in mind that to some degree the problems in Vietnam were more politcal than anything else. We built aircraft equipped for long-range air-to-air contact, and then told the pilots to not shoot until they could see the whites of their eyes, thus negating any advantages they had.
Vietnam was a textbook case of how not to fight a war...
Slashdot Mirror
Haven't really followed the One or S, since I don't own them. The issue is that not many of the modders actually use them now - they tend to get so many donations that they run the latest and greatest, and so they don't support the older phones the way they used to.
Back when everybody running Android had an ADP or G1 (identical hardware) 3rd-party updates were very well supported, since that's what EVERYBODY used. However, once the newer phones started coming out, and once donations to the bigger mods ramped up, the leading devs tended to move on and buy new phones all the time.
Don't get me wrong - I'm grateful for their efforts and I get a lot more than I pay for (nothing). However, fragmentation has not really helped the driver hacking scene much - to few devs working on too many phones to really keep most of them up-to-date.
I think the main problem is application developers not understanding anything about database theory. The vast majority of databases I encounter are not normalized at all, and it's almost always because they were designed by a developer with no database background.
Unfortunately I've seen the opposite extreme as well - databases that are SO normalized that to find anything at all you need about 35 joins. Sure, everything is arbitrarily expandable, but basically there is no practical way to work with it without writing code. We have a database at work that anytime I have to actually look inside and see what is going on I end up copying and pasting from my personal library of 1.5 page long SQL SELECTs. People don't take much advantage of all the expandability since doing anything to the system costs a ton of money.
This whole idea of throwing random strings of structured text into a database column, and then relying entirely on the program code to parse and use it... well, why the hell even use a relational database, then?
And that is the opposite extreme. I've also seen a database at work where half the fields in the databases are CLOBs full of XML. That is even more painful to deal with than the 1.5 page SQL since you end up replicating half the database over the network to do anything that wasn't fielded at the DB-level - like a NoSQL from hell.
Oh, and those two systems talk to each other. :)
If someone died, he or she has a row in a Deaths table, which joins to the People table; otherwise, not; no NULLS necessary.
The problem with this sort of logic is that it gets really cumbersome to work with.
I want to store an optional address for each record. So, that means a place table and a many:many relationship between people and places. Oh, but a place might or might not have an address line 2 or zip code, so that means a sub-street and zip table too, maybe 1:0-1 on the first and many:many on the second. You'd end up with 30,000 views in such a database, and those views basically look just like the tables you've gotten rid of, and those are what everybody would use, and if you wanted to take advantage of your highly-normalized design and assign two places to a single person your code would probably break since that now leads to duplicate rows in the view that the code wasn't designed to handle, since we don't value competent programmers in most enterprises that need these big complex databases anyway.
I can get how mathematically it is more elegant to get rid of the nulls, but when you have people with relatively minimal skill programming and manipulating the database it isn't very practical.
I thought much of the Android code is GPL'ed. If they distribute early versions of Android to selected developers, wouldn't they also need to give away the Android source code to anyone else who demands it at that time?
No, you only have to offer the source to the GPL components, and only to those you offer the binaries to. You can't stop them from redistributing it, but they likely have various incentives not to anyway (if they got in on some special deal that gives them a leg up, why would they go and give away their advantage?). For non-GPL components there is no obligation to give anything away.
The official google phone is the ONLY one to buy IMHO.
You sure about that?
The 3.5 year old old official phone (ADP) only received major OS updates for about a year (Donut).
The two year old official phone (Nexus One) only received major OS updates for about a year (Gingerbread came out less than a year after its release, and that is all it runs).
The 1.5 year old official phone (Nexus S) is just only getting Ice Cream Sandwich now - so you could say that it had updates for 1.5 years after its release date, though only to a version that was released less than a year after it came out (so you could argue it is a 1 year update six months late).
Google has yet to roll out any update to one of its phones that was developed more than a year after the phone was FIRST sold. Anybody who bought the phone just before it was discontinued might have gotten an update at most a month or two later, if they got any updates at all.
So, if you buy a non-Nexus phone you'll be lucky to get any updates at all (it is hit and miss), and if you buy a Nexus phone you may get a single update, if you buy it pretty early. I love Android, but none of their devices really support upgrades much at all.
Interesting that he was not permitted a trial by jury... In fact, I don't see anything about a trial at all - just a resolution passed by a legislative body.
I think that's the whole point here. I imagine most of the folks in Gitmo have had some kind of hearing. The issue is that they have not had public criminal trials before a jury of their "peers."
Go google "man falsely accused" or something.
Uh, care to cite any examples of spies caught on a battlefield in that era who were given trials with due process, a jury of peers, and the rest of the bill of rights?
Cases where people have been found innocent after a long period of expense and the destruction of their lives? There are tons of cases of this. Sure, few if any involve people protesting security theater, but lots of people get trampled on in the justice system. I sat on jury duty for one of them - some poor schlep who was held for 30 days in prison after the police botched up a lineup until the DA realized they had the wrong guy.
THAT document says they have rights and demands that no part of the U.S. government shall violate those rights.
Where, exactly?
The last time I checked soldiers don't attempt to arrest enemy LAWFUL combatants (let alone unlawful ones) and hold trials for them - they shoot them. Armies run by the people who wrote the constitution operated no differently, so somehow I doubt they felt that the kinds of absolute rights you are getting at existed.
Also, the formal distinctions between lawful and unlawful combatants date back to the Geneva conventions which are more recent than the constitution. Granted, they're based on practices which are much older.
My suggestion wasn't that actual evidence of imminent harm should be required in order to make ANY arrest. My point was that this should be required if the police want to arrest somebody before the investigation is completed. If a thorough investigation suggests that somebody is guilty of a crime, then get an arrest warrant from a grand jury and then take them to court.
This was in reply to the suggestion that the police generally arrest first and investigate later.
I realize that you're describing how the system often works, and I don't dispute that. However, the result of this policy is lots of innocent people having their lives ruined merely for being suspected of a crime.
If I recall correctly, the Phoenix was never responsible for any air-to-air kills during its service.
Keep in mind that the scenario it was designed to handle never came up either, thankfully (that would have basically been a WW3 scenario). I think the original use case was a group of dozens of bombers headed for the carrier, and the carrier had to scramble and intercept all of them. Since the bombers could fire at long range, the interceptor had to hit them at long range. Bombers would have a difficult time defeating a missile I would think - they weren't really designed to hit modern fighters.
Much of the stuff that was built in the cold war was designed for a scenario that never happened - a full-out war with the USSR.
Arresting someone who is innocent or is later found not guilty - no.
The problem is that in our system simply being arrested results in a huge grind that is effectively punishment regardless of the final outcome. People spend months to years in jail. Their reputations are tarnished immediately and regardless of outcome, in a society where that reputation follows you even internationally (you can't just move to the next town over and start a new life - arrests show up on background checks). People don't own their means of production like farmers of yore did - if you lose your job you have no income, and it is difficult to get a new job. A speedy trial doesn't mean next week either.
The whole system is in massive need of reform.
Everyone has human rights. That's why they are called human rights, all humans have them.
True, but I've never seen the US agree that people engaging in combat against soldiers have a right to not be shot by those soldiers.
You can call them "unlawful combatants" but they are either civilian criminals or POWs. Not giving them any rights, torturing and treating them as sub human just shames yourselves.
The US military deployed abroad is generally not subject to any non-US law beyond treaties to which the US is a party. That's why we call it the military - their job is to generally make a mess out of other countries (and I'm a big fan of using it a lot less often as a result). If people interfere with them then the US is under treaty obligation to afford those people certain rights if they are wearing a uniform. If they aren't, then the US has no legal obligation to afford them any rights at all. Non-combatants are protected where practical under both treaty obligations and general policy.
If that makes people less willing to host the US army so be it - people shouldn't be all that happy about having foreign armies present, and I'm not a big fan of having the US army deployed in places where they aren't actually busy shooting people (and hence the local opinion of their presence isn't of much worth anyway), and I'm also not a fan of having the army shoot people who haven't been shooting at us either. War is hell, and trying to make it less so ultimately just makes it more so.
Now, a lot of people in Gitmo aren't combatants captured on a battlefield at all, and I completely disagree with keeping these sorts of people there. They are criminals and should be charged as such.
For those who are civilians who did not engage in direct combat against military forces abroad I'd tend to agree with you - they should be arrested and treated as criminals.
For anybody who actually engaged in direct combat/sabotage against the US military, they should be classified as spies. I don't favor indefinite incarceration - probably better to either consider the problems ancient history and cut them loose, or just execute them. Historically execution was the method of choice, but it was typically administered on the spot (catch some guy planting a bomb near your camp not wearing a uniform, you bring them to the officer in charge, they get what intelligence they can out of the guy, and then shoot them (in more ancient times the head might go on a pike).
This status should not be given to anybody arrested in a non-combat situation, or to anybody arrested in a nation the US is not at war with, including anybody captured on US soil. These are criminals, and should be handled under criminal law no matter how inconvenient.
Unlawful combatants (aka civilians with guns) who are captured are not entitled to the rights of a POW, but ARE entitled to the rights of a civilian.
Citation? That is, a citation to some treaty to which the US is a party, or some law binding on the US military?
Foreign civilians on foreign soil have no rights under US law, at least not as far as I can tell. Principles of international law do of course encourage armies to avoid unnecessary harm to non-combatants, but we aren't talking about non-combatants here.
I'm a proponent of closing Gitmo for sure, but the whole reason the situation we're in actually exists is because the US isn't doing what armies have historically done - execute non-uniformed combatants on the spot.
You do realize that an arrest generally occurs before the full investigation, which precedes the decision of whether or not to file criminal charges... right?
Unless there is actual evidence of imminent harm, this should never happen. There's this thing called probable cause...
That the primary problem with people becoming obese being that they consume too much food is hardly news, nor is it news that food is much more abundant these days than it was in the past.
I think the real issue is what the underlying root cause is. Obviously if everybody ate less there would be less obesity. The important question is WHY people don't just do it.
Interesting. I think another factor that people ignore is epigenetics. It was found that the incidence rate of diabetes in some town in Europe that had good records seemed to be related to what people's grandparents ate.
I have no idea whether it is true, but if there are epigenetic factors at work, then the best we can hope to do is prevent obesity in kids who have not been born yet, or find some way to manipulate our own epigenetic programming.
Yup, here's a mathematical model:
masstoday=massyesterday+0.05kg.
My prediction is that the obesity epidemic is hopeless - people will gain weight no matter what you do. Oh, and best not to live near a graveyard - one of these days one of those old people buried there will collapse into a singularity.
Now let's have a round of useless speculation about what my results mean, without calling into question the fact that a model is not reality.
You should have played it before you bought it then, or choose your friends more carefully.
Sure, buyer's remorse will never be completely eliminated, but you have a LOT more ability to do due diligence when buying something that already exists than when buying something that hasn't been built yet.
What you describe is basically the intentional design of Bitcoin. It was designed so that the value of coins became about equal to the marginal cost of computing power. Of course, in the beginning there was a shortage.
As far as I can tell Bitcoin has some advantages and disadvantages compared to other forms of digital currency:
1. A big disadvantage is that transactions aren't anonymous - so why it would appeal to illegal activity is a mystery to me. It seems even less anonymous than conventional cash (which has serial numbers, but less history tracking in general).
2. The big advantage is that it doesn't need a trusted authority to run. Its value is self-regulating, as anybody can produce it and will do so anytime demand for currency results in it being overvalued. It does lack the ability to deflate the currency should there be an oversupply.
3. The currency also prevents double-spending, which other digital cash systems usually require a central bank to prevent. To some extent this mitigates the anonymity of other systems, since those receiving cash need to deposit it very quickly to confirm that it has value, which makes it harder to make the use of the currency untraceable.
Buying retail requires trust, it's just that the trust has been established.
The kind of trust we're talking about is very different.
In the case of a store I only have to trust that the thing in the box is SuperGame 3. In the case of Kickstarter I have to trust that somebody makes SuperGame 3 and that it is a fun game. The quality of the product itself is not in doubt with the retail model, and I can even buy straight from the manufacturer if I want to pay a little more and be assured of some level of quality.
I don't buy pre-releases - I want until I know that I want to play a game before forking out cash. That is what you lose with the shared risk model. Since you're an investor rather than a consumer you take on more risk. Potentially you also pay less, which is basically how risk works.
My main issue with KeepassX is that it isn't capable of running solely with an extension or bookmarklet, which means that it won't work on every OS I have. I use Lastpass as a result, though I'd prefer something equivalent that is open-source...
Keep in mind that to some degree the problems in Vietnam were more politcal than anything else. We built aircraft equipped for long-range air-to-air contact, and then told the pilots to not shoot until they could see the whites of their eyes, thus negating any advantages they had.
Vietnam was a textbook case of how not to fight a war...