I want to know why Obama hasn't closed the damn place yet. One of the major reasons I voted democratic in the last presidential election was to put an end to this sort of thing.
Because Obama is not the dictator of the United States but must faithfully execute[1] the laws passed by the Congress when they are within the power of Congress to regulate. As it happens, Congress has the explicit power to determine what happens to captures[2] during a time of war. So blaming Obama here is somewhat ridiculous as he is simply not in an office charged with
So far, Congress has forbidden the Executive from moving detainees from Guantanamo[3,4] by huge supermajority votes (90-6 in the Senate, for instance). The actual statutory language[5] is quite clear (quoted below). So if you want Obama to close Gitmo then you are essentially asking him to ask in open defiance of the law.
SEC. 1032. PROHIBITION ON THE USE OF FUNDS FOR THE TRANSFER OR RELEASE OF INDIVIDUALS DETAINED AT UNITED STATES NAVAL STATION, GUANTANAMO BAY, CUBA.
None of the funds authorized to be appropriated by this Act for fiscal year 2011 may be used to transfer, release, or assist in the transfer or release to or within the United States, its territories, or possessions of Khalid Sheikh Mohammed or any other detainee who-- (1) is not a United States citizen or a member of the Armed Forces of the United States; and (2) is or was held on or after January 20, 2009, at United States Naval Station, Guantanamo Bay, Cuba, by the Department of Defense.
Worse, thanks to a recent California Supreme Court ruling, police officers may be able to search your smartphone for hours without a warrant if you're arrested for any reason.
First, not all Americans live in California. Other States can (and have) interpreted their 4A equivalents to provide more or less protection than the Federal one.
More to the point, it's probably not true that they can search your cellphone if you are arrested for any reason. Rather, the US Supreme Court explained recently in Gant[1], the idea is that the police can search for things "reasonably believed to contain evidence of the offense of arrest". So searching the cell phone of the CA drug dealer might come out differently than searching the cell phone of (say) a parole violator or a drunk driver.
To be fair, Gant was an automobile search and the court might distinguish a cellphone from a car in some important sense. Nevertheless, the blanket statement in the summary is not likely to hold up if the police do not have some nexus between the arresting crime and the cellphone.
And of course, Gant might be wrong as a matter of policy, although Orin Kerr has a very good writeup[2] of the extensive history of search incident to arrest in Anglo-Saxon law that's worth reading for some historical context.
Your proposals are all technologically sound but, with all due respect, socially retarded. This is what I tried (and failed, apparently) to emphasize in my original reply: security protocols must respect the limits of user compliance and attempt to balance technological requirement with user tolerance for extra procedures. You appear to be evaluating 'best practices' from the point of view of assured user compliance instead of trying to find the set of practices that will actually result in the greatest security given the probable behavior of users when confronted with such a system.
Certainly requiring users to input a folder for what is essentially a disposable updater executable, then manually hunt that executable down, then run it with the appropriate privileges is going to reduce user compliance by a huge factor. Requiring users to copy newly downloaded applications from $USER/downloads to some other arbitrary directory before executing them is far more likely to yield Google searches on how to disable this draconian restriction than to enhance security in any meaningful way (for animated bunny screensaver, copy this to $USER/safe and double click!).
Programs can be executed from the default allowed directories. %programfiles% , %systemroot%\system32, etc, and some designated paths for placing executables in manually, in order to install them.
When Chrome closes it copies over the %ProgramFiles% version if the user have sufficient privileges to do so. That's the best place for it, but given that NTFS does not allow unlinking an exectuable when it is running, having it in %AppData% for the time being is the next best option.
User profile directories including appdata are specifically excluded, because this is best common practice. Programs/executables don't belong in any user's profile or appdata folder (Especially not in any folder used as a default download directory for saving files or temporary directory used by a mail application for opening attachments in a viewer). Contents of appdata is a data folder, and all of a user's profile are data folders, not program folders.
Wait, so if I instruct chrome to download an application, it shouldn't go in $USER/Downloads because executables aren't suppose to be in data folders? To where should setup.exe be downloaded then? In fact, how the heck is any updater supposed to work in this case? Even Firefox downloads an executable to %appdata%\Temp\ and then launches the process.
What you've described isn't best common practice, it's slavish attention to distinctions that are made for the sake of convenience -- allowing the a particular form to entirely straightjacket the function of software that keeps itself updated.
What's more, given that placing roadblocks to updating causes a huge decrease in user compliance, it's not even clear that such draconian measures even improve security. Having those 16 browser vulnerabilities patched as promptly as possible is far more important than adhering to whatever practices seem best in the abstract.
TL;DR: I'm very happy that Google does not adhere to 'best practices' that would result in more people using software with known vulnerabilities for longer.
As for your Firefox issue, go to Tools > Options > Advanced > Update and untick automatically update for Add-ons (and probably search engines). There, job done. Yes it isn't the best user interaction decision to update at startup and block the main UI from loading, but it doesn't mean you have to live with it when it clearly ticks you off so much.
So now I have to manually check for updates? And this is your idea of fixing things?
Is that updates take place silently and promptly without any user intervention even on systems with UAC activated (a copy is installed to %appdata%). Why can't other applications just keep themselves up to date automatically in that way? It's obviously not technologically impossible, we've seen it happen. Even Windows Update is vaguely alright in this respect once you disable the restart-nagging. Debian systems do fine after a simple 'apt-get update && apt-get upgrade -y' in the root crontab although the GUI will occasionally pester you.
Firefox has to be the worst offender in this respect, both in terms of actual software upgrades that block the UI and then add-ons that also block the main UI and then spawn a silly splash to inform you of the amazing upgrade rfom 2.1.6 to 2.1.6(b). Unless it requires a change in the terms of the license or more permissions (Android does this nicely), I don't care and I definitely don't need to be interrupted to see it.
Another free tip for the Mozilla team -- when I open an application is not the time to install any updates. In fact, that is the only time you can be nearly guaranteed that I want to use the application right this second. Schedule updates for when I close the app because it's pretty damn likely I don't need to use it for a few minutes.
Apple could learn the same thing about their infernal updates too, plus an extra special place in hell for pimping their other software at the same time. I still get calls from my parents "Do I need Safari?", hmm, no just upgrade iTunes when it asks you to. "What about quicktime?". Gah.
There's legitimate reasons why employees at a startup would need to put in tons of hours until things get up to speed. The flipside is that the potential for a large payday is significantly greater for the startup employer than for an established firm.
It seems therefore logical that the proper arrangement is to offer the employees a chunk of the profit in exchange for getting the push to release done on-time and with all the features. If your employer doesn't want to pay you like a startup, he has no right to ask for startup-esque sacrifices. Conversely, if the employees are not willing to push hard for release in exchange for such a bonus, they should find employment at a more well-established firm.
We're talking monitoring your supposedly private information behind the scenes
Well, here's the thing about US law (for better or worse, I'm just explaining it as I understand how it actually operates) is that there is no constitutional reasonable expectation of privacy in Facebook stuff, since my assumption you have already shared it with others (if only Facebook Inc). This is called "the third party doctrine", since it covers only information that an individual has voluntarily disclosed some third (non-government) entity. See, e.g. United States v. Miller (1976):
The Fourth Amendment does not prohibit the obtaining of information revealed to a third party and conveyed by him to Government authorities, even if the information is revealed on the assumption that it will be used only for a limited purpose and the confidence placed in the third party will not be betrayed.
The long and short of this is that the act of transmitting to Facebook establishes that you have no REP in whatever you transmit. A lot of ink has been spilled in debating the doctrine, both legally and normatively but that's past the scope of this post so I'll just point you to an article criticizing the doctrine and one defending it. Both contain excellent overviews of the law and the surrounding doctrinal argument.
More interestingly, however, Congress stepped in to provide even more protection than the Court when it passed the Stored Communications Act that provides an intermediate level of scrutiny past the normal scrutiny that attaches to any criminal subpoena[1]. In the SCA, Congress requires the government to prove "specific and articulable facts" that the information is relevant and material to a criminal investigation. That would be the standard applicable to a subpoena to Facebook.
Of course, if Facebook wanted to disclose information voluntarily, that would be well covered by the Third Party Doctrine (as it exists) except to the extent prohibited by the Facebook TOS.
[1] That would be, approximately, 'reasonable possibility that the materials sought will produce information relevant to the investigation'. See, e.g. United States v. R. Enterprises (1991) and FRCP 17.
Maybe I'm missing something, but why does PHP have its own version of strtod()? It's a standard C99 function, so you'll find it in libc or equivalent in any C99-compliant platform (including Windows) and more effort has probably gone into optimising that version than the PHP version
Sometimes the overhead required to take values between languages actually makes doing it in the lower performance language slower, even if it's more highly optimized. This obviously manifests itself much more readily in quicker functions like strtod() where the overhead is likely to be large relative to the actual execution time. .
Just for a taste, in Python (I'm no php expert) in order to call a C function you must context switch (flush registers), unbox the value into the native type, perform the operation, rebox the value into the Python type, then context switch back to the interpreter. Here's the strdod() example in python (I think) with the error handling
int Py_strtod(PyObject const *string, PyObject *target) {
target = 0;
if ( !PyString_check(string) )
return -1;/* Should we set something so the caller knows why this failed? */
const char* const cstr = PyString_AsString(string);
const double cdoub = strtod(cstr,NULL);
if ( 0 != errno )/* something something */
target = PyFloat_FromDouble(cdoub);
if ( PyError_Occured() )/* something something */
return 0;// Success }
So in the process of using the faster C function, we've changed context a twice and had to check for a number of errors that are impossible to throw otherwise. Is it worth it for strtod()? Maybe, but not surely.
While it's one thing to charge people more to discourage excessive data use and maintain your network performance and the like, it's quite another thing to make it part of your business plan to charge unsuspecting users hundreds of dollars when they exceed that cap without realizing it. That's just exploiting people.
This is why AT&T send you a text message at 80%, 90% and 95% of your quota and gives you the option of cutting yourself off instead of getting overages. So they are proactively trying to make sure that you realize what's going on. And at $10/GB for those on the $25 plan, it would be seriously difficult to rack up hundreds of dollars of charges unless you went 500% over your quota.
In other words, what you are describing is exploitative but bears little relation to actual practice. Yes, they did this bullshit in 2008 but they don't any more AFAICT.
The article implies that the judge will just issue the warrant. Not that he will consider it, or hear arguments, or look at evidence: "issues a warrant."
Did it occur to you that the article might be simplifying things and that, in practice, the police would in fact make an affidavit and all that jazz? I mean, judges are not even allowed to issue a warrant without the police applying for one.
I concede without reservation that if the judge automatically issues the warrant without considering the facts sworn by the affiant then you are absolutely correct. Of course, he could that in his office or on the beach just as well as the roadblock. A corrupt judge is corrupt wherever he goes.
I think my point is that an honest (or at least well-meaning within the rules and constrains of the higher courts) judge is honest either in his chambers or on the road just as much.
The U.S. Constitution defines three distinct roles of the government, providing checks and balances. When two branches collude like this, it breaks down those checks. The judiciary (the magistrate) and the executive (the police) should not, in my opinion, be working together like this.
How is spatial proximity the same as collusion? Collusion requires an agreement towards nefarious end, not merely two people that are in the same place at the same time. There is no evidence in the record indicating that the magistrates have agreed to do anything nefarious by sitting at the roadblock and judging warrants by the exact same standard they always do.
Now, maybe you can argue that the judges at the roadblock are not applying the same standard they normally do in the office. But that's not a criticism of the program, it's a criticism of the performance of the judge that (allegedly, as no evidence of this claim is put forth) loosened his standards.
Of course, the argument remains that the standards they always apply are too weak or too deferential or whatever other complaint. But that complaint applies just as well to warrants issued while sitting in chambers as it does while sitting at a roadblock.
In other words, it's not clear exactly what substantive element of the process you think is broken, other than a vague insistence that some incidental elements are themselves evidence of malfeasance.
But in this case, there is not even an illusion of separation of powers.
I don't get it. The police gather evidence, they swear it to a judge, the judge decides. How is that not the same separation of powers that goes on all the time merely because of the spatial and temporal proximity?
Except that you'd be hard pressed to get a warrant to search every house in town as a police officer, but searching all traffic passing through a given intersection is entirely different.
Apples to oranges. Not every driver is being breathalyzed, only the ones for whom there are other indicia of DUI. The way you phrased this, you'd think the judges were on hand to issue a blood-draw warrant for every single car on the road, by analogy to 'every house in town'. Instead, each warrant application covers only a single individual, upon a sworn affidavit establishing probably cause to believe that evidence of a crime will be found in that individual's blood.
By your house analogy, it's like walking down the street looking at all the houses and then applying for a warrant for the ones that smell like s**t from the road on the grounds that they are likely to be violating the septic code. Some may, on closer inspection, be found to be violating the code, others might not. But no one will accuse the city of "searching every house in town".
That and there's a significant number of issues with the breathalyzer machines themselves that can lead to false positives. Not to mention that ni some states it is explicitly forbidden to mention the faults with a breathalyzer as part of your defense, and in at least one state receiving a 0.08 on a breathalyzer machine regardless of it's condition, calibration, any complicating medical issues, or the results of any following blood test makes you guilty of a DUI in and of itself, as the DUI law makes getting such a result on a breathalyzer a crime in and of itself.
The warrants here are for blood tests, not breath tests, which are a fair bit more reliable. In most States (incl. FL) the police cannot compel a blood test except in the case of an accident with serious injury or a search warrant, which is a major reason they are not usually deployed. Courts and juries tend to look more favorably on blood than breath, perhaps for the reasons you cited regarding false positives, so it makes sense that if you are going to get a warrant you should seize the most reliable possible evidence.
Having the judiciary working hand in hand with the police and rubberstamping warrants on the site of the alleged crime (the checkpoint) is, in my opinion, a violation of reasonable search and seizure: state law be damned.
Can you cite any particular reason the location of the magistrate issuing the warrant is relevant to the analysis of reasonableness? Is there some sort of spatial proximity field that renders a decision reasonable when in one location but unreasonable in another? Is there a temporal factor in which decisions are reasonable when rendered at 10AM but not 10PM?
These question are snarky, but my point is that I cannot identify what particular element offends your notion of reasonableness. From what I can tell, the procedure used here seems to be materially equivalent to the procedure used nationwide to obtain warrants -- the police gather evidence, they submit it upon oath to a judge, the judge decides if probable cause exists.
Now, maybe judges in FL are just rubber-stamping warrant applications in general, in which case that's an abdication of their duty to critically review each such application for probable cause. But if that's the case then it's not a complaint that's particular to this story.
The same reason you should refuse to provide the police with any information. False positives.
This is more analogous to refusing to let the police in your house after they've obtained a warrant to search it.
You aren't providing them with information, you are withholding pertinent physical evidence that one has probable cause to believe contains evidence of a crime.
And secondly, I don't agree. (I am not stating law, I am stating opinion of what I believe the law ought to be) I feel that my hair and my blood are just as much a part of "me" as my words and taking them to use to prosecute me should be considered forcing me to testify against myself.
Well, you are entitled to that position but it's not the law as it exists today, for better or worse. Broadly speaking, if it doesn't come out of your mouth in the form of words it's not testimony at all.
IMO, that position makes no sense. Suppose, for instance, that the police have collected DNA evidence from a rape or rape/murder victim and they want to compare it against a suspect for whom they already have objective reason* to suspect. It really does not seem unreasonable to compel that suspect to produce DNA solely for the purpose of comparison against the sample collected from the crime (I will even grant that the suspect has the right to insist that the test be performed and then his sample destroyed) either on its own merits or because production of that DNA is 'testimony'.
* You can invent whatever objective reason you want here. They have video evidence of him entering and leaving the crime scene at the time of the crime, they have witnesses that testify he was threatening the victim, whatever. Just imagine that you are convinced not that he's necessarily guilty but that it is objectively reasonable to suspect him given the evidence already on record.
What you have proposed stretches the definition of the word 'testify' beyond its breaking point. DNA doesn't testify, blood doesn't testify.
Providing the police with evidence that can be used against you at trial is covered by the 5th amendment. That is why we have the right to remain silent under miranda rights. We can remain silent with our words OR with our actions so that we do not incriminate themselves.
The fifth amendment's coverage for physical (not testimonial) evidence is not as strong as you seem to think it is (maybe it should be strong, but I'm merely stating my best read of what the last is, not how it ought to be). Quoting the Court in Schmerber v. California, 384 US 757 regarding compulsory blood tests
It could not be denied that in requiring petitioner to submit to the withdrawal and chemical analysis of his blood the State compelled him to submit to an attempt to discover evidence that might be used to prosecute him for a criminal offense. [...] The distinction which has emerged, often expressed in different ways, is that the privilege [in the 5A/14A/Miranda] is a bar against compelling "communications" or "testimony," but that compulsion which makes a suspect or accused the source of "real or physical evidence" does not violate it. [...] Petitioner's testimonial capacities were in no way implicated; indeed, his participation, except as a donor, was irrelevant to the results of the test, which depend on chemical analysis and on that alone. Since the blood test evidence, although an incriminating product of compulsion, was neither petitioner's testimony nor evidence relating to some communicative act or writing by the petitioner, it was not inadmissible on privilege grounds.
This comports with our (or at least my) common sense belief that there is a difference between the compulsion to turn over physical evidence (usually by warrant) and the compulsion to testify (impermissible period). The former might be incriminating but it's certainly not self-incriminating in the sense that testimony might be. If the police get a warrant and seize a dead hooker's body from my basement, surely that incriminates me but it can't be said that they have compelled me to testify against myself.
If there is evidence of DUI present, (glazed eyes, red eyes, slurred speech, open containers, ect) then I do believe that breathalyzers may be used. But I still do not think that refusing one is enough to allow a needle to be placed in a driver's arm.
If there is evidence of DUI and a judge determines upon that evidence that there is probable cause to believe that the defendant committed DUI and that evidence will be found in his blood, then that is enough to compel the defendant to give blood. In the same vein (harrr) if there is evidence of a murder and a judge determines that there is probable cause that it happened and that evidence will be found in my basement, I am compelled to allow the police to look around my basement.
I don't see how a judge can issue a warrant without evidence simply because someone exercises their rights.
It's not. On the other hand, upon sworn testimony from an officer that he observed multiple objective indicia of drunkenness (e.g. slurred speech, bloodshot eyes, smell of booze), a judge might conclude that it is more probable than not that you committed the crime of DUI.
That's been the standard for judging warrants since time immemorial -- the police gather evidence, they submit an sworn affidavit summarizing their evidence, the judge determines whether the materials in the affidavit suffice to establish probable cause that a crime was committed and that evidence of that crime is likely to be found in the place to be seized.
Of course, the entire procedure is amenable to criticism on a number of levels but none of those criticisms are specific to the particular manner its used here. This procedure is identical in nearly all respects to the manner used to kick down doors in murder investigations or seize computers in fraud investigations. Maybe they are all suspect, but I really don't see the distinction between them.
Thank you for reminding me again why I have a red dot next to your name. I hope to hell you never have to experience what he's going through simply for doing what you think is right, regardless of whatever your misshapen idea of right is.
The problem isn't doing what you think is right, it's about the honesty of saying one thing and then doing another. PFC Manning swore to the world that he would obey the orders of the President of the United States, the officers appointed above him, and the UCMJ. He then turned around and broke that oath.
To contrast, I have no beef with Assange and Wikileaks for doing what they think is right. Assange is an Australian and owes no loyalty to the US at any rate. Their moral convictions, right or wrong, are plain for all to see and judge. They have committed no perfidy, nor gained any trust by virtue of a lie. Whether or not anyone likes them, no one can accuse them of disloyalty because they never claimed to be loyal to anyone but themselves.
I'm still not sure how this doesn't violate the Fourth Amendment.
There has never, since the founding of this Republic, been any understanding that searches at a border are unreasonable under the 4A. Those searches might be wrong, or privacy-violating or even fascist, but they are certainly not contrary to the 4A. See, e.g. United States v. Arnold (9th Cir. 2007), 2007 WL 1407234 ("Computer devices are conceptually no different for Fourth Amendment purposes than other closed storage containers that are subject to suspicionless searches at the border.")
Use the commercial services for exceptions. Pay extra for when you need that. It will be cheaper than your monthly car payment.
I drive a $1900 truck I bought of Craigslist that I maintain myself. I pay another $410 yearly for insurance and about $100/month in gas. For this pittance of a sum, I have the power to drive wherever I want, whenever I want, hauling all manner my gear. Oh, and since it's an inline 4, I get 25-28mpg on the highway (depending on whether I'm going 85MPH or 55MPH). Heck, how the hell do you even expect me to get my groceries home when I live 5 blocks from the nearest bus stop? It snows here too, ya know.
Meanwhile, just to get all-access rail pass to the MBTA here costs $150/month = $1800 a year, or almost as much per year as it cost me to buy my truck once (been going for over 3 years now, for an annual capital cost under $600). Oh, and the MBTA makes a bit over a quarter of its operating revenue from fares, meaning that as a taxpayer I'm paying another $5000 a year towards it that pass. And the MBTA sucks -- it's frequently late and stops at 1AM, so if I want to take transit out to the city to socialize, I have to pay $50 for a cab ride back.
So your statement about it being less that the cost of my private transportation is just blatantly false. The public transit option is massively subsidized (70% of its budget) and still costs nearly twice as much. And that's before I pay extra for commercial services for shit I can already do whenever I want. No sale.
Fortunately, you can implement your own allocation strategy in C++ by overriding the new and delete operators. Admittedly, it's a bit more work but can in many cases easily result in a tenfold speedup.
Please don't do this. I work in a performance intensive field (physical/biological simulation) and every smart*** that's tried this has failed miserably or succeeded for a very modest (TCMalloc (google) or HOARD (uMass). Both are quite good, especially when dealing with heavily threaded applications.
If those don't solve your problem, then maybe you can think of starting to tinker with your own memory allocation. Maybe.
Each representative should vote in a manner that is consistent with the best interests of their constituents.
And I voted for the representative whose conception of what is best for my district matches mine. For instance, one of the candidates believes that gun control was best in the best interests of our district, the other one believes that gun rights are best for the district. Neither of them came out and said "I have a policy on guns that is wrong for you!".
Meanwhile, there is another problem: both candidates believe that economic growth is best for our district (gasps from the audience) but one believes the best way to achieve that is by cutting taxes and the other believes the best way to achieve it is to fund a second round of stimulus spending. Neither of them came out and said "I have a policy on the economy that is going to wreck it!".
If we all agreed on (a) what constitutes the best interests of the district and (b) which policies are most likely to achieve those interests, then there wouldn't be much point in a political process at all. As it happens, we are pretty divided as a nation both on (b) but more fundamentally on (a) -- on what outcomes are normatively preferable. That can't be resolved by a "best interests of the constituents" test because the constituents themselves don't agree.
The key word here is efficient. Specifically I am talking about operations per watt. If some combination of heat dissipation and cost to run the system are limiting factors, then this kind of efficiency is important.
But in the HPC world, the real limiting factor is the interconnect and the software interface. The interconnect latency determines how large of a job can finish in reasonable time, and is a fixed (high) cost per CPU. Meanwhile the software interface determines what off-the-shelf software will work with minimal investment. It's not worth spending programmer ($100k/year) or even graduate student ($40k/yr) time chasing a few watts when you have funding agencies expecting actual scientific results in the next quarterly review.
BTW, those custom supercomputers are neat and get to set records running highly specialized (and massively expensive in terms of programmer-hours), but look at https://www.teragrid.org/web/user-support/compute_resources and tell me how many TFLOPs are provided by x64 machines and how many are PowerPC and other weirdos.
TLDR: When you are pushing FLOPs, x64 (Intel and AMD alike) is king because it provides the most power per CPU. We care about watts, but don't want to screw either our scaling or networking-gear-budget.
Slashdot Mirror
I want to know why Obama hasn't closed the damn place yet. One of the major reasons I voted democratic in the last presidential election was to put an end to this sort of thing.
Because Obama is not the dictator of the United States but must faithfully execute[1] the laws passed by the Congress when they are within the power of Congress to regulate. As it happens, Congress has the explicit power to determine what happens to captures[2] during a time of war. So blaming Obama here is somewhat ridiculous as he is simply not in an office charged with
So far, Congress has forbidden the Executive from moving detainees from Guantanamo[3,4] by huge supermajority votes (90-6 in the Senate, for instance). The actual statutory language[5] is quite clear (quoted below). So if you want Obama to close Gitmo then you are essentially asking him to ask in open defiance of the law.
SEC. 1032. PROHIBITION ON THE USE OF FUNDS FOR THE TRANSFER OR RELEASE OF INDIVIDUALS DETAINED AT UNITED STATES NAVAL STATION, GUANTANAMO BAY, CUBA.
None of the funds authorized to be appropriated by this Act for fiscal year 2011 may be used to transfer, release, or assist in the transfer or release to or within the United States, its territories, or possessions of Khalid Sheikh Mohammed or any other detainee who--
(1) is not a United States citizen or a member of the Armed Forces of the United States; and
(2) is or was held on or after January 20, 2009, at United States Naval Station, Guantanamo Bay, Cuba, by the Department of Defense.
[1] Article II, Section 1.
[2] Article I, Section 8.
[3] http://www.nytimes.com/2009/05/21/us/politics/21detain.html?_r=1&pagewanted=all
[4] http://www.nytimes.com/2009/12/23/us/politics/23gitmo.html
[5] http://thomas.loc.gov/cgi-bin/query/D?c111:5:./temp/~c111aSU9NC::
TFS:
Worse, thanks to a recent California Supreme Court ruling, police officers may be able to search your smartphone for hours without a warrant if you're arrested for any reason.
First, not all Americans live in California. Other States can (and have) interpreted their 4A equivalents to provide more or less protection than the Federal one.
More to the point, it's probably not true that they can search your cellphone if you are arrested for any reason. Rather, the US Supreme Court explained recently in Gant[1], the idea is that the police can search for things "reasonably believed to contain evidence of the offense of arrest". So searching the cell phone of the CA drug dealer might come out differently than searching the cell phone of (say) a parole violator or a drunk driver.
To be fair, Gant was an automobile search and the court might distinguish a cellphone from a car in some important sense. Nevertheless, the blanket statement in the summary is not likely to hold up if the police do not have some nexus between the arresting crime and the cellphone.
And of course, Gant might be wrong as a matter of policy, although Orin Kerr has a very good writeup[2] of the extensive history of search incident to arrest in Anglo-Saxon law that's worth reading for some historical context.
[1] http://www.law.cornell.edu/supct/html/07-542.ZO.html
[2] http://volokh.com/2010/12/14/the-origins-of-the-search-incident-to-arrest-exception/
Your proposals are all technologically sound but, with all due respect, socially retarded. This is what I tried (and failed, apparently) to emphasize in my original reply: security protocols must respect the limits of user compliance and attempt to balance technological requirement with user tolerance for extra procedures. You appear to be evaluating 'best practices' from the point of view of assured user compliance instead of trying to find the set of practices that will actually result in the greatest security given the probable behavior of users when confronted with such a system.
Certainly requiring users to input a folder for what is essentially a disposable updater executable, then manually hunt that executable down, then run it with the appropriate privileges is going to reduce user compliance by a huge factor. Requiring users to copy newly downloaded applications from $USER/downloads to some other arbitrary directory before executing them is far more likely to yield Google searches on how to disable this draconian restriction than to enhance security in any meaningful way (for animated bunny screensaver, copy this to $USER/safe and double click!).
Programs can be executed from the default allowed directories. %programfiles% , %systemroot%\system32, etc, and some designated paths for placing executables in manually, in order to install them.
When Chrome closes it copies over the %ProgramFiles% version if the user have sufficient privileges to do so. That's the best place for it, but given that NTFS does not allow unlinking an exectuable when it is running, having it in %AppData% for the time being is the next best option.
User profile directories including appdata are specifically excluded, because this is best common practice. Programs/executables don't belong in any user's profile or appdata folder (Especially not in any folder used as a default download directory for saving files or temporary directory used by a mail application for opening attachments in a viewer). Contents of appdata is a data folder, and all of a user's profile are data folders, not program folders.
Wait, so if I instruct chrome to download an application, it shouldn't go in $USER/Downloads because executables aren't suppose to be in data folders? To where should setup.exe be downloaded then? In fact, how the heck is any updater supposed to work in this case? Even Firefox downloads an executable to %appdata%\Temp\ and then launches the process.
What you've described isn't best common practice, it's slavish attention to distinctions that are made for the sake of convenience -- allowing the a particular form to entirely straightjacket the function of software that keeps itself updated.
What's more, given that placing roadblocks to updating causes a huge decrease in user compliance, it's not even clear that such draconian measures even improve security. Having those 16 browser vulnerabilities patched as promptly as possible is far more important than adhering to whatever practices seem best in the abstract.
TL;DR: I'm very happy that Google does not adhere to 'best practices' that would result in more people using software with known vulnerabilities for longer.
As for your Firefox issue, go to Tools > Options > Advanced > Update and untick automatically update for Add-ons (and probably search engines). There, job done. Yes it isn't the best user interaction decision to update at startup and block the main UI from loading, but it doesn't mean you have to live with it when it clearly ticks you off so much.
So now I have to manually check for updates? And this is your idea of fixing things?
Is that updates take place silently and promptly without any user intervention even on systems with UAC activated (a copy is installed to %appdata%). Why can't other applications just keep themselves up to date automatically in that way? It's obviously not technologically impossible, we've seen it happen. Even Windows Update is vaguely alright in this respect once you disable the restart-nagging. Debian systems do fine after a simple 'apt-get update && apt-get upgrade -y' in the root crontab although the GUI will occasionally pester you.
Firefox has to be the worst offender in this respect, both in terms of actual software upgrades that block the UI and then add-ons that also block the main UI and then spawn a silly splash to inform you of the amazing upgrade rfom 2.1.6 to 2.1.6(b). Unless it requires a change in the terms of the license or more permissions (Android does this nicely), I don't care and I definitely don't need to be interrupted to see it.
Another free tip for the Mozilla team -- when I open an application is not the time to install any updates. In fact, that is the only time you can be nearly guaranteed that I want to use the application right this second. Schedule updates for when I close the app because it's pretty damn likely I don't need to use it for a few minutes.
Apple could learn the same thing about their infernal updates too, plus an extra special place in hell for pimping their other software at the same time. I still get calls from my parents "Do I need Safari?", hmm, no just upgrade iTunes when it asks you to. "What about quicktime?". Gah.
There's legitimate reasons why employees at a startup would need to put in tons of hours until things get up to speed. The flipside is that the potential for a large payday is significantly greater for the startup employer than for an established firm.
It seems therefore logical that the proper arrangement is to offer the employees a chunk of the profit in exchange for getting the push to release done on-time and with all the features. If your employer doesn't want to pay you like a startup, he has no right to ask for startup-esque sacrifices. Conversely, if the employees are not willing to push hard for release in exchange for such a bonus, they should find employment at a more well-established firm.
We're talking monitoring your supposedly private information behind the scenes
Well, here's the thing about US law (for better or worse, I'm just explaining it as I understand how it actually operates) is that there is no constitutional reasonable expectation of privacy in Facebook stuff, since my assumption you have already shared it with others (if only Facebook Inc). This is called "the third party doctrine", since it covers only information that an individual has voluntarily disclosed some third (non-government) entity. See, e.g. United States v. Miller (1976):
The Fourth Amendment does not prohibit the obtaining of information
revealed to a third party and conveyed by him to Government authorities,
even if the information is revealed on the assumption that it will be used
only for a limited purpose and the confidence placed in the third party will
not be betrayed.
The long and short of this is that the act of transmitting to Facebook establishes that you have no REP in whatever you transmit. A lot of ink has been spilled in debating the doctrine, both legally and normatively but that's past the scope of this post so I'll just point you to an article criticizing the doctrine and one defending it. Both contain excellent overviews of the law and the surrounding doctrinal argument.
More interestingly, however, Congress stepped in to provide even more protection than the Court when it passed the Stored Communications Act that provides an intermediate level of scrutiny past the normal scrutiny that attaches to any criminal subpoena[1]. In the SCA, Congress requires the government to prove "specific and articulable facts" that the information is relevant and material to a criminal investigation. That would be the standard applicable to a subpoena to Facebook.
Of course, if Facebook wanted to disclose information voluntarily, that would be well covered by the Third Party Doctrine (as it exists) except to the extent prohibited by the Facebook TOS.
[1] That would be, approximately, 'reasonable possibility that the materials sought will produce information relevant to the investigation'. See, e.g. United States v. R. Enterprises (1991) and FRCP 17.
[2] 18 U.S.C. 2703(d).
Maybe I'm missing something, but why does PHP have its own version of strtod()? It's a standard C99 function, so you'll find it in libc or equivalent in any C99-compliant platform (including Windows) and more effort has probably gone into optimising that version than the PHP version
Sometimes the overhead required to take values between languages actually makes doing it in the lower performance language slower, even if it's more highly optimized. This obviously manifests itself much more readily in quicker functions like strtod() where the overhead is likely to be large relative to the actual execution time. .
Just for a taste, in Python (I'm no php expert) in order to call a C function you must context switch (flush registers), unbox the value into the native type, perform the operation, rebox the value into the Python type, then context switch back to the interpreter. Here's the strdod() example in python (I think) with the error handling
int
Py_strtod(PyObject const *string, PyObject *target)
{
target = 0;
if ( !PyString_check(string) )
return -1;
const char* const cstr = PyString_AsString(string);
const double cdoub = strtod(cstr,NULL);
if ( 0 != errno )
target = PyFloat_FromDouble(cdoub);
if ( PyError_Occured() )
return 0; // Success
}
So in the process of using the faster C function, we've changed context a twice and had to check for a number of errors that are impossible to throw otherwise. Is it worth it for strtod()? Maybe, but not surely.
While it's one thing to charge people more to discourage excessive data use and maintain your network performance and the like, it's quite another thing to make it part of your business plan to charge unsuspecting users hundreds of dollars when they exceed that cap without realizing it. That's just exploiting people.
This is why AT&T send you a text message at 80%, 90% and 95% of your quota and gives you the option of cutting yourself off instead of getting overages. So they are proactively trying to make sure that you realize what's going on. And at $10/GB for those on the $25 plan, it would be seriously difficult to rack up hundreds of dollars of charges unless you went 500% over your quota.
In other words, what you are describing is exploitative but bears little relation to actual practice. Yes, they did this bullshit in 2008 but they don't any more AFAICT.
The article implies that the judge will just issue the warrant. Not that he will consider it, or hear arguments, or look at evidence: "issues a warrant."
Did it occur to you that the article might be simplifying things and that, in practice, the police would in fact make an affidavit and all that jazz? I mean, judges are not even allowed to issue a warrant without the police applying for one.
I concede without reservation that if the judge automatically issues the warrant without considering the facts sworn by the affiant then you are absolutely correct. Of course, he could that in his office or on the beach just as well as the roadblock. A corrupt judge is corrupt wherever he goes.
I think my point is that an honest (or at least well-meaning within the rules and constrains of the higher courts) judge is honest either in his chambers or on the road just as much.
The U.S. Constitution defines three distinct roles of the government, providing checks and balances. When two branches collude like this, it breaks down those checks. The judiciary (the magistrate) and the executive (the police) should not, in my opinion, be working together like this.
How is spatial proximity the same as collusion? Collusion requires an agreement towards nefarious end, not merely two people that are in the same place at the same time. There is no evidence in the record indicating that the magistrates have agreed to do anything nefarious by sitting at the roadblock and judging warrants by the exact same standard they always do.
Now, maybe you can argue that the judges at the roadblock are not applying the same standard they normally do in the office. But that's not a criticism of the program, it's a criticism of the performance of the judge that (allegedly, as no evidence of this claim is put forth) loosened his standards.
Of course, the argument remains that the standards they always apply are too weak or too deferential or whatever other complaint. But that complaint applies just as well to warrants issued while sitting in chambers as it does while sitting at a roadblock.
In other words, it's not clear exactly what substantive element of the process you think is broken, other than a vague insistence that some incidental elements are themselves evidence of malfeasance.
But in this case, there is not even an illusion of separation of powers.
I don't get it. The police gather evidence, they swear it to a judge, the judge decides. How is that not the same separation of powers that goes on all the time merely because of the spatial and temporal proximity?
Except that you'd be hard pressed to get a warrant to search every house in town as a police officer, but searching all traffic passing through a given intersection is entirely different.
Apples to oranges. Not every driver is being breathalyzed, only the ones for whom there are other indicia of DUI. The way you phrased this, you'd think the judges were on hand to issue a blood-draw warrant for every single car on the road, by analogy to 'every house in town'. Instead, each warrant application covers only a single individual, upon a sworn affidavit establishing probably cause to believe that evidence of a crime will be found in that individual's blood.
By your house analogy, it's like walking down the street looking at all the houses and then applying for a warrant for the ones that smell like s**t from the road on the grounds that they are likely to be violating the septic code. Some may, on closer inspection, be found to be violating the code, others might not. But no one will accuse the city of "searching every house in town".
That and there's a significant number of issues with the breathalyzer machines themselves that can lead to false positives. Not to mention that ni some states it is explicitly forbidden to mention the faults with a breathalyzer as part of your defense, and in at least one state receiving a 0.08 on a breathalyzer machine regardless of it's condition, calibration, any complicating medical issues, or the results of any following blood test makes you guilty of a DUI in and of itself, as the DUI law makes getting such a result on a breathalyzer a crime in and of itself.
The warrants here are for blood tests, not breath tests, which are a fair bit more reliable. In most States (incl. FL) the police cannot compel a blood test except in the case of an accident with serious injury or a search warrant, which is a major reason they are not usually deployed. Courts and juries tend to look more favorably on blood than breath, perhaps for the reasons you cited regarding false positives, so it makes sense that if you are going to get a warrant you should seize the most reliable possible evidence.
Having the judiciary working hand in hand with the police and rubberstamping warrants on the site of the alleged crime (the checkpoint) is, in my opinion, a violation of reasonable search and seizure: state law be damned.
Can you cite any particular reason the location of the magistrate issuing the warrant is relevant to the analysis of reasonableness? Is there some sort of spatial proximity field that renders a decision reasonable when in one location but unreasonable in another? Is there a temporal factor in which decisions are reasonable when rendered at 10AM but not 10PM?
These question are snarky, but my point is that I cannot identify what particular element offends your notion of reasonableness. From what I can tell, the procedure used here seems to be materially equivalent to the procedure used nationwide to obtain warrants -- the police gather evidence, they submit it upon oath to a judge, the judge decides if probable cause exists.
Now, maybe judges in FL are just rubber-stamping warrant applications in general, in which case that's an abdication of their duty to critically review each such application for probable cause. But if that's the case then it's not a complaint that's particular to this story.
The same reason you should refuse to provide the police with any information. False positives.
This is more analogous to refusing to let the police in your house after they've obtained a warrant to search it.
You aren't providing them with information, you are withholding pertinent physical evidence that one has probable cause to believe contains evidence of a crime.
And secondly, I don't agree. (I am not stating law, I am stating opinion of what I believe the law ought to be) I feel that my hair and my blood are just as much a part of "me" as my words and taking them to use to prosecute me should be considered forcing me to testify against myself.
Well, you are entitled to that position but it's not the law as it exists today, for better or worse. Broadly speaking, if it doesn't come out of your mouth in the form of words it's not testimony at all.
IMO, that position makes no sense. Suppose, for instance, that the police have collected DNA evidence from a rape or rape/murder victim and they want to compare it against a suspect for whom they already have objective reason* to suspect. It really does not seem unreasonable to compel that suspect to produce DNA solely for the purpose of comparison against the sample collected from the crime (I will even grant that the suspect has the right to insist that the test be performed and then his sample destroyed) either on its own merits or because production of that DNA is 'testimony'.
* You can invent whatever objective reason you want here. They have video evidence of him entering and leaving the crime scene at the time of the crime, they have witnesses that testify he was threatening the victim, whatever. Just imagine that you are convinced not that he's necessarily guilty but that it is objectively reasonable to suspect him given the evidence already on record.
What you have proposed stretches the definition of the word 'testify' beyond its breaking point. DNA doesn't testify, blood doesn't testify.
Providing the police with evidence that can be used against you at trial is covered by the 5th amendment. That is why we have the right to remain silent under miranda rights. We can remain silent with our words OR with our actions so that we do not incriminate themselves.
The fifth amendment's coverage for physical (not testimonial) evidence is not as strong as you seem to think it is (maybe it should be strong, but I'm merely stating my best read of what the last is, not how it ought to be). Quoting the Court in Schmerber v. California, 384 US 757 regarding compulsory blood tests
It could not be denied that in requiring petitioner to submit to the withdrawal and chemical analysis of his blood the State compelled him to submit to an attempt to discover evidence that might be used to prosecute him for a criminal offense.
[...]
The distinction which has emerged, often expressed in different ways, is that the privilege [in the 5A/14A/Miranda] is a bar against compelling "communications" or "testimony," but that compulsion which makes a suspect or accused the source of "real or physical evidence" does not violate it.
[...]
Petitioner's testimonial capacities were in no way implicated; indeed, his participation, except as a donor, was irrelevant to the results of the test, which depend on chemical analysis and on that alone. Since the blood test evidence, although an incriminating product of compulsion, was neither petitioner's testimony nor evidence relating to some communicative act or writing by the petitioner, it was not inadmissible on privilege grounds.
This comports with our (or at least my) common sense belief that there is a difference between the compulsion to turn over physical evidence (usually by warrant) and the compulsion to testify (impermissible period). The former might be incriminating but it's certainly not self-incriminating in the sense that testimony might be. If the police get a warrant and seize a dead hooker's body from my basement, surely that incriminates me but it can't be said that they have compelled me to testify against myself.
If there is evidence of DUI present, (glazed eyes, red eyes, slurred speech, open containers, ect) then I do believe that breathalyzers may be used. But I still do not think that refusing one is enough to allow a needle to be placed in a driver's arm.
If there is evidence of DUI and a judge determines upon that evidence that there is probable cause to believe that the defendant committed DUI and that evidence will be found in his blood, then that is enough to compel the defendant to give blood. In the same vein (harrr) if there is evidence of a murder and a judge determines that there is probable cause that it happened and that evidence will be found in my basement, I am compelled to allow the police to look around my basement.
Same procedure, same standard.
I don't see how a judge can issue a warrant without evidence simply because someone exercises their rights.
It's not. On the other hand, upon sworn testimony from an officer that he observed multiple objective indicia of drunkenness (e.g. slurred speech, bloodshot eyes, smell of booze), a judge might conclude that it is more probable than not that you committed the crime of DUI.
That's been the standard for judging warrants since time immemorial -- the police gather evidence, they submit an sworn affidavit summarizing their evidence, the judge determines whether the materials in the affidavit suffice to establish probable cause that a crime was committed and that evidence of that crime is likely to be found in the place to be seized.
Of course, the entire procedure is amenable to criticism on a number of levels but none of those criticisms are specific to the particular manner its used here. This procedure is identical in nearly all respects to the manner used to kick down doors in murder investigations or seize computers in fraud investigations. Maybe they are all suspect, but I really don't see the distinction between them.
Thank you for reminding me again why I have a red dot next to your name. I hope to hell you never have to experience what he's going through simply for doing what you think is right, regardless of whatever your misshapen idea of right is.
The problem isn't doing what you think is right, it's about the honesty of saying one thing and then doing another. PFC Manning swore to the world that he would obey the orders of the President of the United States, the officers appointed above him, and the UCMJ. He then turned around and broke that oath.
To contrast, I have no beef with Assange and Wikileaks for doing what they think is right. Assange is an Australian and owes no loyalty to the US at any rate. Their moral convictions, right or wrong, are plain for all to see and judge. They have committed no perfidy, nor gained any trust by virtue of a lie. Whether or not anyone likes them, no one can accuse them of disloyalty because they never claimed to be loyal to anyone but themselves.
Freedoms to speak (wikileaks)
Assange is an Australian, he doesn't have any rights under the US Constitution while not on American soil.
I'm still not sure how this doesn't violate the Fourth Amendment.
There has never, since the founding of this Republic, been any understanding that searches at a border are unreasonable under the 4A. Those searches might be wrong, or privacy-violating or even fascist, but they are certainly not contrary to the 4A. See, e.g. United States v. Arnold (9th Cir. 2007), 2007 WL 1407234 ("Computer devices are conceptually no different for Fourth Amendment purposes than other closed storage containers that are subject to suspicionless searches at the border.")
Unlike the UK, however, you cannot be jailed for failure to turn over an encryption key. http://www.theregister.co.uk/2010/10/06/jail_password_ripa/
Use the commercial services for exceptions. Pay extra for when you need that. It will be cheaper than your monthly car payment.
I drive a $1900 truck I bought of Craigslist that I maintain myself. I pay another $410 yearly for insurance and about $100/month in gas. For this pittance of a sum, I have the power to drive wherever I want, whenever I want, hauling all manner my gear. Oh, and since it's an inline 4, I get 25-28mpg on the highway (depending on whether I'm going 85MPH or 55MPH). Heck, how the hell do you even expect me to get my groceries home when I live 5 blocks from the nearest bus stop? It snows here too, ya know.
Meanwhile, just to get all-access rail pass to the MBTA here costs $150/month = $1800 a year, or almost as much per year as it cost me to buy my truck once (been going for over 3 years now, for an annual capital cost under $600). Oh, and the MBTA makes a bit over a quarter of its operating revenue from fares, meaning that as a taxpayer I'm paying another $5000 a year towards it that pass. And the MBTA sucks -- it's frequently late and stops at 1AM, so if I want to take transit out to the city to socialize, I have to pay $50 for a cab ride back.
So your statement about it being less that the cost of my private transportation is just blatantly false. The public transit option is massively subsidized (70% of its budget) and still costs nearly twice as much. And that's before I pay extra for commercial services for shit I can already do whenever I want. No sale.
Citations:
http://www.mbta.com/fares_and_passes/rail/ I live in Zone #2. They folks living out in the farther out suburbs pay a whopping $250/month for their rail passes.
http://en.wikipedia.org/wiki/Massachusetts_Bay_Transportation_Authority#Budget $400M in fares, $1400 budget. It's gotten much worse since 2008 too.
Fortunately, you can implement your own allocation strategy in C++ by overriding the new and delete operators. Admittedly, it's a bit more work but can in many cases easily result in a tenfold speedup.
Please don't do this. I work in a performance intensive field (physical/biological simulation) and every smart*** that's tried this has failed miserably or succeeded for a very modest (TCMalloc (google) or HOARD (uMass). Both are quite good, especially when dealing with heavily threaded applications.
If those don't solve your problem, then maybe you can think of starting to tinker with your own memory allocation. Maybe.
Each representative should vote in a manner that is consistent with the best interests of their constituents.
And I voted for the representative whose conception of what is best for my district matches mine. For instance, one of the candidates believes that gun control was best in the best interests of our district, the other one believes that gun rights are best for the district. Neither of them came out and said "I have a policy on guns that is wrong for you!".
Meanwhile, there is another problem: both candidates believe that economic growth is best for our district (gasps from the audience) but one believes the best way to achieve that is by cutting taxes and the other believes the best way to achieve it is to fund a second round of stimulus spending. Neither of them came out and said "I have a policy on the economy that is going to wreck it!".
If we all agreed on (a) what constitutes the best interests of the district and (b) which policies are most likely to achieve those interests, then there wouldn't be much point in a political process at all. As it happens, we are pretty divided as a nation both on (b) but more fundamentally on (a) -- on what outcomes are normatively preferable. That can't be resolved by a "best interests of the constituents" test because the constituents themselves don't agree.
The key word here is efficient. Specifically I am talking about operations per watt. If some combination of heat dissipation and cost to run the system are limiting factors, then this kind of efficiency is important.
But in the HPC world, the real limiting factor is the interconnect and the software interface. The interconnect latency determines how large of a job can finish in reasonable time, and is a fixed (high) cost per CPU. Meanwhile the software interface determines what off-the-shelf software will work with minimal investment. It's not worth spending programmer ($100k/year) or even graduate student ($40k/yr) time chasing a few watts when you have funding agencies expecting actual scientific results in the next quarterly review.
BTW, those custom supercomputers are neat and get to set records running highly specialized (and massively expensive in terms of programmer-hours), but look at https://www.teragrid.org/web/user-support/compute_resources and tell me how many TFLOPs are provided by x64 machines and how many are PowerPC and other weirdos.
TLDR: When you are pushing FLOPs, x64 (Intel and AMD alike) is king because it provides the most power per CPU. We care about watts, but don't want to screw either our scaling or networking-gear-budget.