Slashdot Mirror


User: BronsCon

BronsCon's activity in the archive.

Stories
0
Comments
8,054
First seen
Last seen
Profile
(view on slashdot.org)

Comments · 8,054

  1. And you just used the word, yourself... does that make you one, too?

  2. And the Intel CPU in your Mac wasn't designed on a Mac, it was designed on a Windows PC. You use a Windows PC by proxy every time you use your Mac.

    That, or your argument is just ridiculous.

  3. Re: Sounds like a CYA distraction statement on Tesla Issues Strongest Statement Yet Blaming Driver For Deadly Autopilot Crash (abc7news.com) · · Score: 1

    The best drivers are better than twice as good as average, a mathematical necessity to offset the fact that the worst drivers are less than half as good. It's not Tesla's fault that people read "at least twice as good" and think "as good as the best" when that's so far from what it means that you can't even reliably measure the distance.

  4. Re:I hope they fine Tesla. on FTC Warns Manufacturers That 'Warranty Void If Removed' Stickers Break the Law (vice.com) · · Score: 2

    Contracts don't trump laws.

    No, only Donald does that.

  5. It could also be that they were quoting their source. That's what quotation marks are for, after all.

  6. Re: Different outcome if you screw up on Ask Slashdot: Are Companies Under-Investing in IT? · · Score: 1

    COBOL most often ran on someone else's computer, not your local terminal. The cloud hasn't increased penetration, we're just back to running stuff on someone else's computer rather than our local terminals.

  7. I see your concern but wouldn't this require some method of UID/GID mapping to the windows user for it to work properly?

    They "set" everything to 0777 without mapping users or groups, why would they need to map them to display your effective permissions? Everything in the Windows mapped drives is owned by root anyway, so no, no user or group mapping is necessary. Just display what the current ACLs will allow my user to do.

    And there would have to be further distillation as the NTFS ACLs are more granular than Linux permissions (though doesn't SELinux allow ACLs?)

    Do the current ACLs allow my Windows user to execute this file? Set the execute bit accordingly. Do they allow my Windows user to read this file? Set the read bit accordingly. Do they allow my Windows user to write this file? Set the write bit accordingly. That's all the granularity that is needed if those permissions are read-only, and that would solve the problems faced but the vast majority of users complaining about this. I agree, some distillation would be necessary if we were asking that these permissions be made read/write, but the vast majority of us are just asking that they simply reflect the actual permissions we have for a given file.

    I'm guessing what they did to avoid some half-baked solution was to set everything to 777.

    So they didn't bake it at all?

    The correct way to do this would require UID/GID mapping

    Why? You can't sign in to WSL without first signing in to Windows. You don't actually get 0777 permissions on everything (now that would be super-bad), Windows ACLs still apply. How do they do that? Simply, you're signed in as your Windows user and, to Windows, your Linux user appears to be your Windows user and will inherit the same permissions.

    The "Why?" above was rhetorical. There is no why, because it wouldn't be necessary. Everything in Windows mapped drives is owned by root and nobody is asking for that to change, so there's nothing at all to map there. I could have a dozen Linux users and every single one of them would share the permissions of the currently-signed-in Windows user when accessing Windows mapped drives, so there's nothing at all to map there. This is a solved issue for removable media in Linux, and that's effectively what Windows mapped drives are; the solution is very much what the WSL userbase has been clamoring for: display the effective permissions.

    I don't think this can be done well natively today.

    Doesn't need to be.

    on OS X, a Samba share is mounted similarly under /Volumes/My_Mounted_Share; all files & directories are marked with mode 700 with UID of me & GID of the group 'staff'.....

    The difference is that one is a local resource being accessed via a local driver you can reasonably be sure the version and capabilities of and the other is a network resource being accessed via a remote driver or server where you cannot.

    Let's tackle the version and capabilities part of that difference, first. Samba doesn't provide permissions info along with a directory listing (I believe NFS does, but I'm not super familiar with it so I may be wrong), so you would have to request permissions on each file individually; not every Samba server even implements that functionality, so displaying "dummy" permissions makes sense in that situation. WSL only maps local drives, connected to (and mounted on) the same host in which it is running, and the version of WSL and the DrvFs driver are tied to the Windows release; neither can be installed or upgraded separately by the user. There is no reason DrvFs can't supply permissions info along with filename, size, ctime, atime, and mtime. The DrvFs access is done as the current Windows user, which means the DrvFs driver necessarily knows those permissions in advance.

    Now, let's tackle local vs network

  8. windows mapped drives are using the ACLs granted to your windows user access token when launching the shell

    Indeed they are, as you're still technically signed in as your Windows user, so you can only access what your Windows user can access, regardless of how you're accessing it. WSL still shows permissions of 0777 for all Windows mapped drives and all files and directories contained therein, regardless of your actual permissions. I'm not guessing at this; as I said, this is an issue I deal with on a daily basis. I can chmod those files all day long and the operation will appear to succeed, but when I list them they're all still 0777, regardless of the permissions I set or the permissions granted by Windows ACLs.

    That is the problem, and it's a big one considering that Git stores whether a given file is user executable or not. Semi-fortunately, Git has a configuration option that tells it to ignore that and either make any new files executable or not executable (based on the configured value), but that does make it a bit difficult to add a new file with an executable status other than the configured default. Yes, I know about "git update-index", but I work on a multitude of different systems, only one of which is WSL, and it's honestly a massive pain in my ass to have to remember to do that if I happen to be adding a new executable to a repo while I'm on that system.

    If you look at the number of people complaining about this on Github, I'm far from the only person who thinks they should, at the very least, default the Linux permissions that are displayed to something that reflects the Windows ACLs that are applied. Windows ACLs include "Read & execute", so doing this would allow me to solve my problem natively.

    To clarify, a file for which my Windows user has no permissions, e.g. I can not open, edit, write, execute, or delete that file, appears in WSL to have full permissions -- 0777. That, again, is the problem. Attempting to access the file, of course, will fail in that case, but you don't know it will fail until you try it.

    And things that shouldn't be executable are executable in Windows mapped drives in under WSL, because they all appear to have a permissions mask of 0777; if your Windows user can read it, your WSL user will happily execute it. Really not good if the file in question happens to be a potentially malicious binary or BASH script you copied off one of your servers or other systems for analysis (in which case, best practices dictate you "chmod -x" that sunuvabitch so you don't accidentally execute it). Whoops, can't do that in WSL if the file happens to reside on a Windows mapped drive. Again, that's the problem.

  9. No more than Microsoft's response makes them come off as a bunch of assholes, but okay Yours makes you seem like a swell guy, though.

  10. The issue has been discussed at length on Github and MS has voiced that they have no interest in doing anything about it. Where do you think I read Microsoft's "excuse" for not fixing it?

  11. Permissions? on Microsoft Open Source Tool Lets You 'Bring Your Own Linux' To Windows (microsoft.com) · · Score: 3, Interesting

    Have they fixed permissions mapping when accessing files on the Windows partition yet? If so, I'm excited for the update, because it's an issue I deal with daily.

    Their excuse is that they don't want to change Windows ACLs, and that's fine, I get that, but it's a poor excuse; WSL applies 0777 to all Windows files currently and, to add to it, doesn't seem to use Windows ACLs for the files within the lxss directory, which strongly implies that they're already storing Linux file permissions as metadata elsewhere, which is what they should be doing for Linux file permissions for Windows files -- defaulting to values mapped from Windows ACLs, of course.

  12. Re:Always suspected this. on Hot-Air Dryers Suck In Nasty Bathroom Bacteria, Shoot Them At Your Hands (arstechnica.com) · · Score: 2

    Resistance isn't the issue, the issue is that you're spreading the bacteria you don't kill around along with its food source. Since it now doesn't have to compete with the bacteria you did kill, that stronger bacteria will now multiply faster and there will be more if it spreading around every time you touch something than there would have been had you washed your hands or done nothing at all.

  13. Re: total fucking transparency on The FCC Is Refusing To Release Emails About Ajit Pai's 'Harlem Shake' Video (vice.com) · · Score: 1

    There's a market for it. You and I both know that. I say we put it up as bait and imprison anyone who shows interest.

  14. I assume you mean this to be an argument that because a bolt action rifle is more readily available in Australia that it would be possible to have a shooting like the Pulse shooting.

    Indeed that's what I mean, and it's 100% factually correct. People are allowed to get facts wrong, you know, and the better of us admit when we do and issue a correction. I did that in my prior post.

    So, an incident like Pulse is possible in Australia, yet it doesn't happen. Why? What's different about Australia, other than gun laws which wouldn't prevent Pulse from happening, that prevents an Australian rendition of Pulse?

    You'll point out that the bolt action rifle in question is heavily restricted and only available to a subset of the population and I'll point out that a 6-shooter (specifically named in OP's post) could maintain that fire rate as well, so let's just skip it, okay?

    No mass shooting in recent US history has relied on a high rate of fire. I'm limiting to the US because I'm admittedly ignorant of mass shootings elsewhere in the world, give the number of them we deal with over here making it somewhat hard to keep up if you have any interest in collecting all the facts -- coupled with the fact that I have other things going on in my life because I'm not exactly a gun nut. Every single mass shooting in recent US history could have been carried out with the guns available to anyone with a "genuine reason", which includes hunting.

    And my whole point in bringing up Pulse was to point out that you can mow down 50 people without an AK or AR; a bolt action could do it at the same rate as the Pulse shooter. Okay, so that was only 49, but the 53 wounded have to count for something, as well.

    So, what else changed in Australia that is preventing these mass shootings? I'ts certainly not the fact that guns are registered to their owners, as mass shooters typically die during the commission of their crime and, thus, don't give two shits if their gun is registered to them or not.

  15. Re: Or take away her gun on YouTube Will Increase Security At All Offices Worldwide Following Shooting (theverge.com) · · Score: 1

    I meant to say "hands and feet", I guess I deleted too much when I was rewording that statement.

  16. Re: Or take away her gun on YouTube Will Increase Security At All Offices Worldwide Following Shooting (theverge.com) · · Score: 1

    Yes. And I consider the threat of violence to be violence. But that's beside the point.

    So you consider the fact that, if you punch me in the face, I can and will punch you right back to be violence? Simply having a gun on my person is not a threat of violence. If you are not a threat to me, I am not a threat to you; if you are a threat to me, you damned well better believe I am a threat to you as well, with or without a gun. I don't carry, by the way.

    others have explicitly talked about said teachers being able to stop a shooter

    Yes, being able to is the deterrent. Mass shooters always choose soft targets, which aren't able to stop them. Change that and they either choose another target or the violence is avoided altogether. In either case, the goal of those people (with whom I don't necessarily agree, mind you) is to make our schools safer and they'll have done so without additional violence if they get their way. That, of course, assumes the teachers are able to maintain control of their weapons and don't, themselves, flip out -- which is why I don't necessarily agree with the idea.

    But there is nothing violent in possessing and carrying your own means of self defense. If there was, we would all be in a constant state of violence, as we all have hands and permanently attached to our bodies. Well, not "all" of us, quadriplegics might well be the only people in the world you consider nonviolent.

  17. [X] is a total that includes totals from [a,b,c and d]. You've summed [a], [b] and [c] and consider the lack of [d] or being able to show that [a,b,c and d] are the only elements is a problem with the source and that you can use the numbers the way you have? I repeat my claim. Your math is bad.

    Okay, fine, subtract [d] for yourself and see how little the result changes. That link was in the post you're complaining about, the numbers are right there for you.

    Hell, don't do it yourself (unless you really care to check my work, in which case please do), the new total is 1,152,455. That doesn't affect the murder rate percentage at all, as that whole number is still part of the new total, and you can see that the total has reduced by less than 100,000. The percentage of, well call them "qualified", violent crimes involving guns, then, is just a hair over 28%, well within the margin of error for most studies and, given that the statistics available only pertain to crimes reported, which will be a subset of crimes committed the margin of error here should be expected to be a bit higher; the statistical difference between 1,248,185 and 1,152,455 (or 26% and 28%) in this scenario is meaningless. The percentage of gun deaths (murders) relative to "qualified" violent crimes, since I know that's the number you probably really wanted, jumps way up, from 0.9% to 1%. I hope you could feel the sarcasm dripping from my words there. The ratio of homicides to total gun-related violent crime (3%) does not change because all of those are still in the new total.

    What a big difference that made, right? If you still think my math is wrong, please correct it for the sake of not spreading misinformation. This is an important subject and it deserves to be discussed openly and honestly. I provided my sources for a reason.

    No. You could look at places where guns have been removed or restricted and looking at the stats for violent crime.

    You mean like Oakland? I keep mentioning Oakland and nobody wants to respond.

    In Australia, for eg, there was a brief spike immediately following the buy-back/ban and then a steady decline.

    I did look at Australia. They had many fewer guns than we do and an entirely different social structure. The spike they had there should worry you for the US; ours would be much larger and, given our heritage with guns, would likely not decline. Even if it would eventually decline, it wouldn't really get a chance to as the regulations would be pulled so quickly your head would spin, we'd all get our guns back, and the subject of regulation would never be revisited; that initial spike would be all that was needed to shut up the "get rid of guns for the children" crowd and bolster the argument of the "only way to stop a bad guy with a gun is a good guy with a gun" crowd.

    In short, what worked in Australia worked only due to the relatively small number of guns they had in the first place and the relative obscurity of gun culture in the country to begin with. There wasn't a whole lot of debate before the regulations were enacted and the regulations weren't restricting or removing a constitutional right.

    Your inability to provide proof of your assertion doesn't make your assertion any more true. If you know what I'm saying.

    Again, there is only one way to prove it and it's not the kind of thing I actually want to be right about if that proof comes.

    You claimed an increase in gun crime will occur when/if guns are criminalised. It may seem simple to you; it may seem obvious. It's not. It's an assertion you haven't backed up with anything like evidence. For eg, show somewhere where guns were made illegal and where the use of guns by criminals increased.

    Note that I never said the

  18. Re: Or take away her gun on YouTube Will Increase Security At All Offices Worldwide Following Shooting (theverge.com) · · Score: 1

    Ever stop to think it might be a deterrent? After all, these mass shootings always seem to happen in "gun free zones". Nothing disingenuous about my remark at all.

  19. Re: Or take away her gun on YouTube Will Increase Security At All Offices Worldwide Following Shooting (theverge.com) · · Score: 1

    Call to arm teachers. It got political. You may have heard about it. I think that Trump guy mentioned it.

    Your joke aside, owning a gun is not a violent act. Neither is carrying one.

    Some people are characterising the call for arming teachers or increasing police on campus/schools or hiring more security guards as adding more guns. It's cheap rhetoric, but it only works because on some level it's true.

    Firing back at your humor with some of my own: You mean like YouTube piling on the armed guards in the wake of their crackdown on gun content on their platform?

  20. *bolt action

    Note to self: proofread BEFORE posting.

  21. Well it seems like he and I both got our facts wrong, then. The only mass shooting in recent history in the US with nearly 50 dead was Pulse (49 dead and 53 wounded) and it was not an AR-15 that was used. In fact, the fire rate averaged roughly one shot every 2 minutes, it just went on for a very long time. That could just as well have been accomplished with the bolt action I linked to. Hell, the same number of shots could have been made in a handful of minutes with that rifle, while the Pulse shooter took nearly 4 hours.

    Could the SigSauer MCX that was used have fired off the ~110 rounds that were fired that night much more quickly? Sure. Does that mean a bold action rifle couldn't have fired off the same ~110 rounds in 4 hours? No. It sure as hell could have.

  22. Straw man. No-one is saying guns make people kill.

    I neither built up, nor tore down, any argument there, ergo not a straw man. I was making a point; one which you, in your previous paragraph, claimed to agree with, but apparently completely missed.

    You omit rape and any other crimes included in the original that are not covered by those categories

    I tallied up the numbers in the statistics provided. If I didn't list a number for the category, there was no number provided by the FBI's statistics for that category. Blame the source, not my math.

    You go on to assume that crime is static and without easy access to a firearm (this may be true, but you simply assert it) violent crime would use another weapon.

    There's really only one way to study it and, well, if I'm right I sure don't want that study conducted. In short, I'd love to be proven wrong, but the chance of being proven right makes me really not want to see that study done. If you follow what I'm saying.

    With respect to the argument 'when it's a crime to own a gun, only criminals will own one' - this is scaremongering.

    Hardly. Criminals already don't care that their guns are illegal. Why would they suddenly start caring when our guns are also illegal? The simple answer is: they wouldn't. And how do I know the criminals' guns aren't legally purchased? Virtually nobody uses a legally purchased firearm in the commission of a violent crime. From the linked PDF:

    Most perpetrators (79%) were carrying a gun that did not belong to them. More than 30% of the guns recovered were reported stolen by owners when the FTU contacted them. For 44% of the guns, whether the gun was stolen was either unknown or not able to be determined.

    It would be extraordinarily difficult to remove guns from the US

    Indeed it would, a point that I made myself when I said "one might argue that getting rid of guns (a pipe dream, mind you, to say nothing of whether I would support it if it were possible) would save lives".

    When you don't _need_ a gun to rob a store why make it easy for the police to identify you as a criminal when you can get away with a knife or bat? As you say, a host of violent criminals do just fine with these already.

    When you've already got the gun... I'd suggest you think about that for a moment, but maybe it's better if you put yourself in the mindset of the average criminal and don't think. We're not talking about masterminds, here; we're talking about junkies and low-level scum who wouldn't already be in possession of an illegal firearm if they were thinking that far ahead. Yes, those people would use their guns unnecessarily; they already do, that's partly why we have this problem with guns in the first place. What more proof do you need than the current situation? But, again, there's only one way to know for sure and that's a study you really don't want to see conducted.

    Removing the pool of available weapons reduces the pool of weapons available illegally.

    You have no idea how many weapons enter this country illegally and I can't credibly speak to that number without naming my source, which may have repercussions for him. That said, Customs stops only about 10% (at their own estimate) of the guns coming into the country. That's on average a trunk full each day and a shipping container or two each month. Removing my legally purchased and owned, well looked after and unlikely to be stolen gun from my possession won't make a dent in that.

    No, the criminals who already have one probably won't give it up, but over time those that are held by criminals will be lost, destroyed and seized.

    That also won't make

  23. Re:botanist vs engineer on Apple's Redesigned Mac Pro is Coming in 2019 (theverge.com) · · Score: 1

    I haven't worked close enough to the metal for any of it to really matter since I was 12. That might color my perspective and understanding a bit differently than yours. You, on the other hand, seem to have worked so close to the metal for so long that it has literally sucked you in. I do ask the questions you pose at the end of your post, but I ask them of a great many things other than the lowest-level architecture of a CPU I will never bare-metal program, so I'll keep my perspective just as you'll keep yours. I feel a little more sane that way.

  24. In Australia its pretty well impossible to get something like an AK or AR , even illegally. They just are not around to obtain anymore.

    When you think AK and AR, do you think full-auto? Because the ones we can get here are just rifles, semi-auto, same as any other repeating rifle. We haven't been able to buy new-stock automatics since 1986. While we can still technically buy automatic weapons, they are heavily regulated, must be registered, must have been made and privately owned prior to the 1986 ban and, due to their relative rarity, sell for $10k or more. Nobody uses their $10k gun that's registered with the federal government to mow down 50 people.

    See how it works in the real USA? Not the retarded fantasy USA the rest of the world seems to think exists?

    Now, as for how it works in Australia...

    You can buy this which, with a 16 round capacity (15 round magazine + one in the chamber) sure ain't no six-shooter. I can buy a version of the same gun but, in California, I can on'y get it with a 10 round magazine. Who can kill more people, and quicker?

    Have you even looked at the guns available in Australia?

    That is not a single-shot shotgun, either. In fact, here is a 10 round semi-automatic 308 winmag rifle comparable to what I can buy here in the states.

    Funny, I thought you said you couldn't get shit like that Down Under.

    Perhaps you should learn your own country's gun laws before judging mine?

  25. Re: Or take away her gun on YouTube Will Increase Security At All Offices Worldwide Following Shooting (theverge.com) · · Score: 1

    Funny, "the only thing that can stop a bad guy with a gun is a good guy with a gun" doesn't mean we need more guns. It means we need to take care in restricting gun rights so that we're restricting the rights of those who are unfit to handle a weapon without abridging the rights of those who can do so safely. It's not the fault of responsible gun owners who support common sense regulations that actually work that you choose to misinterpret it to mean "we need more guns".

    Just because Arnie Grape might not be fit to handle a firearm doesn't mean I should be denied my constitutional right to do so if I've demonstrated that I am. If you want to argue that it shouldn't be a right, you're barking up the wrong tree seeking regulation and should, rather, seek to repeal the 2nd. Good luck with that, by the way.