Adwords is far more of a security risk than Windows 10's telemetry is, unless you have a keylogger installed. And for normal users, Chrome (with it's auto-save and auto-backup feature set) is just as bad.
Windows 3.11 has been discontinued for decades. By "all versions" a company/journalist only refers to actively supported products. (TYL = Today You Learned)
Well, the bigger question is what's going on with XP (and what affects XP probably affects Win2K). I can't really tell if this is dependent on the NT model or if it's truly a higher level Windows design flaw, which probably would then go back to Win 95.
"Printer configuration for Windows 95 and System 7.5.3" are things I've probably intentionally blocked out from my memory.
In my state it is required to tell the officer if you are concealed carrying when stopped. The officer just asks where it is in my experience. And if you are carrying you hand the officer both licenses. When your DL is run it will show you have a carry license though. I have heard it makes them a little nervous if you don't tell them you had one. So I always hand over both even when I am not carrying.
Yeah, the "hand both documents to the officer at once" advice is what I've generally heard is best. Simply remain still while they make the connection and perform whatever follow-up they deem necessary.
No one who intends on having a shootout with a couple of cops is going to tell them he's armed. And it's not like it would be the first time cops have ordered someone to produce ID, then shoot them for reaching for their ID.
It's standard advice given to ALL firearms owners to be extremely careful telling an officer that they have a firearm in a traffic stop situation. Accidents happen regardless of race, and if you say something like "I have a gun" an officer's mind is IMMEDIATELY going to fixate on the word "gun" that you just said.
DO NOT EVER DO THAT.
DO NOT EVER DO THAT WHILE REACHING FOR ANYTHING AT ALL.
The first is that people in the car are much, much more likely to be armed than they were a generation ago.
Citation needed on this one. First off, the degree of firearms ownership and armed drivers will vary VASTLY by State, jurisdiction, and neighborhood. In some areas, being armed in a vehicle is basically impossible unless you're Very Special. In other areas, a cop might only encounter a few people who *aren't* armed.
Crime, as a whole, has been dropping since the early 90's although we're not entirely sure why (that, or just don't want to talk about why), so it might be okay to hypothesize that "illegally having a gun in the car" might be dropping too, but that would need evidence. Is it outweighed by people carrying more often legally? Dunno.
Second, there is the sorry state of the relationship between police and minorities. This itself is nothing new, but the focus of minority anger on police is new. I think this has its genesis in new policing tactics like broken windows stop-and-frisk, which if you think about them are an experiment in intrusive government behavior control, albeit with good intentions.
With all due respect... a/s/l? If you're old enough to remember the early 90's and the LA Riots, you'll realize that "minority anger on police" is very much not new. This is how we got popular rap songs called "Cop Killer", and one of my favorite Snoop Dogg songs refers to killing undercover cops. This was long before broken windows and stop-and-frisk in NYC; those are red herrings. Herrings that are broadly supported by the overall community, however, and seem to correlate with dropping crime rates and residents' feelings of safety.
In an era of social media, the poisoned atmosphere created by clumsily intrusive police tactics spreads far beyond the places that employ them.
Actually, I think this is the root of the problem -- added and abetted by sensationalist media and our generally increasing cultural propensity for cocooning with like-minded individuals. Is there a specific reason this interaction deserved to be highlighted above others (in a neutral sense)? Actually, probably not. There are other police shootings, there's other loss of life, there are other issues coming up. Media coverage (George Zimmerman), Presidential commentary (basically everything... #ThanksObama), and live video of someone dying, all tend to remove the context that a relatively objective journalist might try to bring in to the situation.
There were two majorly publicized events last week causing BLM protests to resurge. Meanwhile, the crime rate in cities like St. Louis, Baltimore, and Chicago is skyrocketing out of control. ~50 people killed in Chicago last week, 14 just on the weekend.
I think technology may help with the first problem: approaching a car in which the occupants may very possibly be armed and either hostile or intoxicated. An app, sure, maybe even a robot.
No. Replacing human police officers with RoboCop will not help the situation, and will not help mend American culture going forward.
Sometimes I want to take all the app-focused millennials in Silicon Valley, sit them down, and force them to watch a bunch of back-to-back Dystopian science fiction films from the 70s and 80s.
If you have a non-lethal option you are obliged to take it. If you can run away or otherwise avoid a deadly fight, you are obliged to. At least that's how it is in most developed countries.
Actually, that depends on the jurisdiction in the US, and the specific circumstances of the situation. Although it would generally be preferable (from a humanistic standpoint) to use an available non-lethal option, the sticking point is who has the burden of proof that an option was available. This comes into effect in the so-called Castle Doctrine (in common law) as well as Stand Your Ground laws, which extend this to anywhere you have a lawful right to be in.
Lethality is distinct from self-defense, but in most jurisdictions if you have a reasonable fear of serious bodily injury then defense including deadly force is permitted. The "reasonable" clause there (as inevitably interpreted by a jury) pertains as to whether the escalation was justifiable or not. If a 10 year old kid is coming at you with a yellow wiffle ball bat and the jury feels that your fear of serious bodily injury was unreasonable, then the use of deadly force would become a manslaughter charge instead of justified homicide (at least in my state).
It's the study of the universe, which is interesting, and has some nifty things to say about the ontology of particles, but so does philosophy, to be honest.
If it can't be disproven, it's not science. It might still be a cool area of study, but not all fields of study are science.
Science is a tool and a methodology for acquisition and extrapolation of quantitative states.
What's interesting is that we've seen the same emphasis on quantitative states in the tech industry over the last decade or so. I wonder if the pedistalization of numeric "Data" over any other type of analysis is related to the fact that there are people who in some fit on insanity could possibly think that Rationalia is a good idea.
Big Data without domain knowledge is useless; and logic without philosophy is flat out dangerous.
I remember that shit being so bad that at one point I had to hunt-down a third party program just to remove it since it was clearly designed as a virus itself...
Yeah, Norton had to write a tool just to remove their own shit.
The "Norton Removal Tool" is still available from the Norton site, which should tell people all they need to know about Norton.
Why is that a bad thing? I'd prefer a separate tool to fully remove, rather than the normal Windows Uninstall being programatically accessible. Hell, if I install my AV, I'd love for a specific YubiKey being needed by some authoritative process to remove it.
We shouldn't be trying to get the computer to do things for us, because that makes things more vulnerable to malicious cyberspace actors. Pumping that back into meatspace (hey, how about we bring DIP switches back and require them to be flipped to write to BIOS again) forces humans back into the loop. Physical security and intrusion detection is a hell of a lot more of a solved problem than IoT security.
For people that don't open attachments, and are more resistant to Trojans, malvertising is probably the top infection vector there is.
Moral of the story: I can go without AV and have a clean system. AV doesn't do anything against malvertising, and with the advent of sites using Flash + EME to protect their content, AV only adds complexity, expands the attack surface, and does nothing.
BS. "Malvertising" doesn't exist fundamentally at a technical level any more than "malshareware" exists. The problems are, respectively, vulnerabilities in flash/imagemagick/browser software/etc and intentionally subversive code that doesn't do what it claims to do. "Restricting advertising" as an AV response is catching things in the dragnet, but that's much more just rationalizing the fact that you just don't want to see ads on websites.
We've all seen parents' and friends' computers that didn't have AV software installed and the sh*tshow they usually are, and it's not because they saw banner ads but because they got infected with viruses. Is AV foolproof or guaranteed to catch everything? Of course not. Does it run as a privileged process and thus require extra scrutiny in the privileged code sections? Of course. So does sudo. But most people are much better off with AV software than without, notwithstanding the fact that people at heightened risk should have even more layers of protection.
I was around, alive, and in office environments 40 years ago and there were only women in the typing pool and as AAs - secretaries, actually, back when ash trays were an office feature.
If you mean women got to do the punched cards, yes, but that isn't programming. It is data entry.
What you just foisted on us is a canard.
To punched cards era I'd add: database programming. Working in system administration at both startups and large corps, I've been struck by two facts:
1) The Oracle DBA group at three companies were almost entirely female 2) Almost everyone else in the technical/operations side of the house was male
I've known a number of women (older than I) that kicked ass (as far as I know) at Delphi, Filemaker Pro (back when small businesses were running on it), and Access.
With regards to "punched cards", though, I think that's part of the distinction between operator and administrator. When computers were primarily "business machines" it would be perfectly normal (even then) for it to be seen as a business administrative task. Is that different than "Linux Systems Engineering" as it was understood 6 years ago? I don't know. Is SysEng an awkward middle period between batch processing job operations + Oracle design and high-level cloud/container/dynamic app management? I don't know that either.
One of my semi-recent Android OS updates also became infested with flat icons. Looks like crap compared to the release prior to it. Used to have some very sharp 3D-look icons. All gone. Looks like a cartoon now.
To be fair, they were probably just copying Apple, though. So yeah.
You know what I'd like? A set of classic Mac OS interface icons ("Platinum" era, so Copland / Mac OS 8) to use. Despite insanely-high dpi on modern displays, the 72 dpi icons (scaled up, of course) were still great and distinctive visually, and gave a useful feel to what you're interacting with. Not everyone needs to subscribe to this bizarre roller coaster of trendy design and re-design that we've been stuck with over the past 15 years.
I'm lucky enough to live in an on-net building with them in San Diego (I've also worked at several locally San Diego ISPs in the past that did similar things for the point-to-point wireless market for businesses). Aside from a few hiccups, I've been very happy with them.
Unfortunately, I don't think I trust Alphabet to be running my direct uplink and to NOT mine every little bit of traffic to deduce even more about my life than they've already figured out.
I'll probably revert to using Cox Cable for my normal uplink and only do huge downloads (or PS/XBox and streaming stuff) through Google Fiber. A shame, because I've really liked Webpass and, as you mentioned, the price is amazing.
No one asked for systemd either, but look what happened.
That is patently false. SysV init is a fragile, horribly broken piece of shit that should have died 20 years ago. Having used systemd-based Kubuntu for some time now, I'm finding most things work far better than they ever did under SysV init:
1) Boot times on my virtual machines are much, much faster than they ever were under SysV init. The clever hacks piled on top of it to make Linux boot faster were so fragile that they broke at the drop of a pin.
I guarantee to you that that was 85-90% Ubuntu (or Debian) crap and maybe 10% "using rc scripts to do setup and then launch independent scripts in/rc#.d/ directories". RedHat-land init scripts were fine. They still are, if you're using RHEL/CentOS 6, which a metric crap ton of people still do... It's not like we can't boot using them, and if post-kernel boot times were that important for your VM (they're not) there are other solutions for streamlining things. Hell, using DashAsBinSh was something RH should have done a long time back.
2) Hardware interaction is far more reliable than it ever was under SysV init. This is much like boot speeds. There were hacks upon hacks to main dynamic hardware appear usable, but they were very fragile and painful to use. Desktop Linux under SysV init was horribly painful on the hardware interaction front between 1991 and the introduction of systemd.
These are the two main issues that had needed to be resolved since Linux's inception, and that nobody had ever been able to solve before. It's fashionable to dump on systemd (and Pulse Audio, for that matter), but they both solved a crucial failing of Linux desktops that nobody had done successfully -- ever.
Great. Make a laptop-focused OS that trades determinism and admin-controllability with dynamic response and trusting yourself into the arbitrary hands of fate and the guy who programmed udev.
It wasn't *that* painful -- it's not like people didn't use Linux on laptops in the 2000's -- and virtually all of it could be solved without replacing/sbin/init with a complex interlocking mesh of interdependent tools.
As a desktop user and server administrator, I love systemd. My only complaint about it is that it took way too long for it to kill SysV init.
There's very little besides automatic/free cgroup management that systemd's paradigm gives you that didn't already exist. Admins can and do use xinetd, service managers, tcp servers, and monitor programs to do their jobs. systemd wraps it all up into a nice, tiny, black-is-the-only-color bow for you... and if you think that's a good idea, you should consider why Embrace, Extend and Extinguish was such a successful strategy.
I want to write a lot of interesting and thought-provoking things here, but I'm just stuck on that one thing in the title.
Secret. Government. Proceedings.
Really, guys? Tell me again how your country is a free and democratic nation.
Well, because it's not a proceeding. Anyone who tells you otherwise doesn't understand Congressional proceedings.
The House is in recess. It's not even in the Committee of the Whole... So as far as proceedings go, they could just as easily be having a slumber party in their offices.
The Chairman didn't really have a choice... the Members were out of order. He could have: a) had the Sergeant-at-Arms "enforce order", meaning kicking them out of the room for not being in their chairs properly, b) call a recess
No one asked for systemd either, but look what happened.
Speaking more broadly, I honestly can't tell if this is because most of the major problems have been solved, people are too ignorant of the thought processes that went into the original tech, people don't know their history, people have too much faith in overly-complex technology that couldn't possible fail, people honestly, think they're just that much smarter than the installed base of users and want to increase "Quality of Life" (as one notable Borg put it), people want to make their own mark, or people are disingenuously trying to achieve lock-in on their newfangled contraption. No doubt, it's a mixture of all of the above.
Speaking as someone who's only been around in the industry for 15 years or so, I've already seen this pattern repeat way too frequently. I can only imagine what people who've been writing COBOL for the past 40 years think of it all...
Please, for the love of God, stop breaking sh*t that works fine.
The constitution doesn't grant you a right to use your cell phone
You're right. You know why? Because governments cannot grant rights. You are born with the right to absolutely anything you can imagine. The only thing any law can do is TAKE RIGHTS AWAY. The constitution doesn't have you grant you any right.
You're correct, but you have neither a civil right nor a Lockean or Hobbesian right to trespass on private property. If you're invited in, your invite can be revoked for any and all reasons not otherwise contrary to civil law relating to discrimination against Protected Classes in public establishments. "Cell phone user" is not one of those, so the owner of the hall is free to kick your ass out.
You can do all of this, already. Kinda makes you wonder, what else are you wrong about?
You can ask Google to do this. Once it's off your device, it's off your device.
Hopefully, we'll start demanding en masse for more and more data to remain on your device. F Google, Apple, and Facebook's cloud learning systems. For the actual processing and interpretation of data, we all have more than enough processing speed and more than enough space in the devices sitting in our pockets to parse through it offline.
Download Offline Maps, download a daily set of ads and rules, and let the CPU in my phone decide what to show me based on where I am now.
You don't want to switch. Nobody is saying use Snappy Core exclusively, it's not designed for that.
The "adios" in the headline would seem to imply that as a rhetorical, at least.
Its main purpose is for servers that require zero downtime, where you want to upgrade application X and/or its dependencies without breaking even the smallest functionality of application Y.
... -_- No server "requires" zero downtime. Anyone who put that into a req. document should be shot. You get more 9's by duplicating and providing physical, logical, and temporal redundancy. And we've gotten high redundancy for servers just fine using RPMs, thanks. (And.deb's, I assume.) Acting like there are no solutions out there is ludicrous. But then, a lot of the last six years of Linux-land have been like that.
It might broadcast itself to a nearby Internet-connected device by careful modulation of some sufficiently-long trace on the circuit board. People have managed to broadcast AM radio from graphing calculators (first example that came to mind), so it does not even require superintelligence. Now, maybe if you put it in a Faraday cage + audio shielding, it might require more intelligence to get out, but even then I would not expect a superintelligence to fail to find a way out.
What's crucial to point out is that "things AI can control" is exactly equivalent to "hackable thing".
InfoSec doubles as your AI Response team, because if there's a way to do it with software (or using software to trick a human into doing it), it can be done by a malevolent AI.
I swear, people haven't read enough science fiction in their lives.
Yea, this boggles my mind somewhat. I realize M$ is always a worthy target, but "privacy" means something very different now that there are orders of magnitudes more devices present in our lives with microphones and sensors than it did in 1995 when you were lucky if your lone work PC recognized your SoundBlaster card at all.
Railing against W10 is meaningless. You want to protest unnecessary telemetry and data collection, storage, and evaluation, it's Google (read: Alphabet) and Facebook you should be worried about, with Apple as a shifty-eyed honorable mention... not because of closed source, but because of unclear engineering.
Side note: We're all linux and OSS advocates here, and that's fine. But "open source" is meaningless at the consumer level -- consumers don't have the ability to do meaningful code review, and the scale of HOBE hacks means that the advanced gcc-is-programmed-to-replace-login-compile-code hacks are easy enough to create when the payoff is 100M vulnerabilities.
Slashdot Mirror
yup my windows 10 would never do that.
Adwords is far more of a security risk than Windows 10's telemetry is, unless you have a keylogger installed. And for normal users, Chrome (with it's auto-save and auto-backup feature set) is just as bad.
Windows 3.11 has been discontinued for decades. By "all versions" a company/journalist only refers to actively supported products. (TYL = Today You Learned)
Well, the bigger question is what's going on with XP (and what affects XP probably affects Win2K). I can't really tell if this is dependent on the NT model or if it's truly a higher level Windows design flaw, which probably would then go back to Win 95.
"Printer configuration for Windows 95 and System 7.5.3" are things I've probably intentionally blocked out from my memory.
In my state it is required to tell the officer if you are concealed carrying when stopped. The officer just asks where it is in my experience. And if you are carrying you hand the officer both licenses. When your DL is run it will show you have a carry license though. I have heard it makes them a little nervous if you don't tell them you had one. So I always hand over both even when I am not carrying.
Yeah, the "hand both documents to the officer at once" advice is what I've generally heard is best. Simply remain still while they make the connection and perform whatever follow-up they deem necessary.
No one who intends on having a shootout with a couple of cops is going to tell them he's armed. And it's not like it would be the first time cops have ordered someone to produce ID, then shoot them for reaching for their ID.
It's standard advice given to ALL firearms owners to be extremely careful telling an officer that they have a firearm in a traffic stop situation. Accidents happen regardless of race, and if you say something like "I have a gun" an officer's mind is IMMEDIATELY going to fixate on the word "gun" that you just said.
DO NOT EVER DO THAT.
DO NOT EVER DO THAT WHILE REACHING FOR ANYTHING AT ALL.
The first is that people in the car are much, much more likely to be armed than they were a generation ago.
Citation needed on this one. First off, the degree of firearms ownership and armed drivers will vary VASTLY by State, jurisdiction, and neighborhood. In some areas, being armed in a vehicle is basically impossible unless you're Very Special. In other areas, a cop might only encounter a few people who *aren't* armed.
Crime, as a whole, has been dropping since the early 90's although we're not entirely sure why (that, or just don't want to talk about why), so it might be okay to hypothesize that "illegally having a gun in the car" might be dropping too, but that would need evidence. Is it outweighed by people carrying more often legally? Dunno.
Second, there is the sorry state of the relationship between police and minorities. This itself is nothing new, but the focus of minority anger on police is new. I think this has its genesis in new policing tactics like broken windows stop-and-frisk, which if you think about them are an experiment in intrusive government behavior control, albeit with good intentions.
With all due respect... a/s/l? If you're old enough to remember the early 90's and the LA Riots, you'll realize that "minority anger on police" is very much not new. This is how we got popular rap songs called "Cop Killer", and one of my favorite Snoop Dogg songs refers to killing undercover cops. This was long before broken windows and stop-and-frisk in NYC; those are red herrings. Herrings that are broadly supported by the overall community, however, and seem to correlate with dropping crime rates and residents' feelings of safety.
In an era of social media, the poisoned atmosphere created by clumsily intrusive police tactics spreads far beyond the places that employ them.
Actually, I think this is the root of the problem -- added and abetted by sensationalist media and our generally increasing cultural propensity for cocooning with like-minded individuals. Is there a specific reason this interaction deserved to be highlighted above others (in a neutral sense)? Actually, probably not. There are other police shootings, there's other loss of life, there are other issues coming up. Media coverage (George Zimmerman), Presidential commentary (basically everything... #ThanksObama), and live video of someone dying, all tend to remove the context that a relatively objective journalist might try to bring in to the situation.
There were two majorly publicized events last week causing BLM protests to resurge. Meanwhile, the crime rate in cities like St. Louis, Baltimore, and Chicago is skyrocketing out of control. ~50 people killed in Chicago last week, 14 just on the weekend.
I think technology may help with the first problem: approaching a car in which the occupants may very possibly be armed and either hostile or intoxicated. An app, sure, maybe even a robot.
No. Replacing human police officers with RoboCop will not help the situation, and will not help mend American culture going forward.
Sometimes I want to take all the app-focused millennials in Silicon Valley, sit them down, and force them to watch a bunch of back-to-back Dystopian science fiction films from the 70s and 80s.
San Diego has a population of ~1.3 million but has a rate a third that of San Antonio (~1.4 million), why?
Why does Philadelphia have a rate twice that of Las Vegas? And that when Vegas has a rate 2.5x that of San Diego?
I think the answer to that is obvious... ;)
If you have a non-lethal option you are obliged to take it. If you can run away or otherwise avoid a deadly fight, you are obliged to. At least that's how it is in most developed countries.
Actually, that depends on the jurisdiction in the US, and the specific circumstances of the situation. Although it would generally be preferable (from a humanistic standpoint) to use an available non-lethal option, the sticking point is who has the burden of proof that an option was available. This comes into effect in the so-called Castle Doctrine (in common law) as well as Stand Your Ground laws, which extend this to anywhere you have a lawful right to be in.
Lethality is distinct from self-defense, but in most jurisdictions if you have a reasonable fear of serious bodily injury then defense including deadly force is permitted. The "reasonable" clause there (as inevitably interpreted by a jury) pertains as to whether the escalation was justifiable or not. If a 10 year old kid is coming at you with a yellow wiffle ball bat and the jury feels that your fear of serious bodily injury was unreasonable, then the use of deadly force would become a manslaughter charge instead of justified homicide (at least in my state).
Also... IANAL. So yeah.
It's the study of the universe, which is interesting, and has some nifty things to say about the ontology of particles, but so does philosophy, to be honest.
If it can't be disproven, it's not science. It might still be a cool area of study, but not all fields of study are science.
Tyson is nonsensical.
Science is a tool and a methodology for acquisition and extrapolation of quantitative states.
What's interesting is that we've seen the same emphasis on quantitative states in the tech industry over the last decade or so. I wonder if the pedistalization of numeric "Data" over any other type of analysis is related to the fact that there are people who in some fit on insanity could possibly think that Rationalia is a good idea.
Big Data without domain knowledge is useless; and logic without philosophy is flat out dangerous.
I remember that shit being so bad that at one point I had to hunt-down a third party program just to remove it since it was clearly designed as a virus itself...
Yeah, Norton had to write a tool just to remove their own shit.
The "Norton Removal Tool" is still available from the Norton site, which should tell people all they need to know about Norton.
Why is that a bad thing? I'd prefer a separate tool to fully remove, rather than the normal Windows Uninstall being programatically accessible. Hell, if I install my AV, I'd love for a specific YubiKey being needed by some authoritative process to remove it.
We shouldn't be trying to get the computer to do things for us, because that makes things more vulnerable to malicious cyberspace actors. Pumping that back into meatspace (hey, how about we bring DIP switches back and require them to be flipped to write to BIOS again) forces humans back into the loop. Physical security and intrusion detection is a hell of a lot more of a solved problem than IoT security.
For people that don't open attachments, and are more resistant to Trojans, malvertising is probably the top infection vector there is.
Moral of the story: I can go without AV and have a clean system. AV doesn't do anything against malvertising, and with the advent of sites using Flash + EME to protect their content, AV only adds complexity, expands the attack surface, and does nothing.
BS. "Malvertising" doesn't exist fundamentally at a technical level any more than "malshareware" exists. The problems are, respectively, vulnerabilities in flash/imagemagick/browser software/etc and intentionally subversive code that doesn't do what it claims to do. "Restricting advertising" as an AV response is catching things in the dragnet, but that's much more just rationalizing the fact that you just don't want to see ads on websites.
We've all seen parents' and friends' computers that didn't have AV software installed and the sh*tshow they usually are, and it's not because they saw banner ads but because they got infected with viruses. Is AV foolproof or guaranteed to catch everything? Of course not. Does it run as a privileged process and thus require extra scrutiny in the privileged code sections? Of course. So does sudo. But most people are much better off with AV software than without, notwithstanding the fact that people at heightened risk should have even more layers of protection.
I was around, alive, and in office environments 40 years ago and there were only women in the typing pool and as AAs - secretaries, actually, back when ash trays were an office feature.
If you mean women got to do the punched cards, yes, but that isn't programming. It is data entry.
What you just foisted on us is a canard.
To punched cards era I'd add: database programming. Working in system administration at both startups and large corps, I've been struck by two facts:
1) The Oracle DBA group at three companies were almost entirely female
2) Almost everyone else in the technical/operations side of the house was male
I've known a number of women (older than I) that kicked ass (as far as I know) at Delphi, Filemaker Pro (back when small businesses were running on it), and Access.
With regards to "punched cards", though, I think that's part of the distinction between operator and administrator. When computers were primarily "business machines" it would be perfectly normal (even then) for it to be seen as a business administrative task. Is that different than "Linux Systems Engineering" as it was understood 6 years ago? I don't know. Is SysEng an awkward middle period between batch processing job operations + Oracle design and high-level cloud/container/dynamic app management? I don't know that either.
30 comments and no one's brought up Blipverts yet? What is this world coming to... >.
One of my semi-recent Android OS updates also became infested with flat icons. Looks like crap compared to the release prior to it. Used to have some very sharp 3D-look icons. All gone. Looks like a cartoon now.
To be fair, they were probably just copying Apple, though. So yeah.
You know what I'd like? A set of classic Mac OS interface icons ("Platinum" era, so Copland / Mac OS 8) to use. Despite insanely-high dpi on modern displays, the 72 dpi icons (scaled up, of course) were still great and distinctive visually, and gave a useful feel to what you're interacting with. Not everyone needs to subscribe to this bizarre roller coaster of trendy design and re-design that we've been stuck with over the past 15 years.
I'm lucky enough to live in an on-net building with them in San Diego (I've also worked at several locally San Diego ISPs in the past that did similar things for the point-to-point wireless market for businesses). Aside from a few hiccups, I've been very happy with them.
Unfortunately, I don't think I trust Alphabet to be running my direct uplink and to NOT mine every little bit of traffic to deduce even more about my life than they've already figured out.
I'll probably revert to using Cox Cable for my normal uplink and only do huge downloads (or PS/XBox and streaming stuff) through Google Fiber. A shame, because I've really liked Webpass and, as you mentioned, the price is amazing.
No one asked for systemd either, but look what happened.
That is patently false. SysV init is a fragile, horribly broken piece of shit that should have died 20 years ago. Having used systemd-based Kubuntu for some time now, I'm finding most things work far better than they ever did under SysV init:
1) Boot times on my virtual machines are much, much faster than they ever were under SysV init. The clever hacks piled on top of it to make Linux boot faster were so fragile that they broke at the drop of a pin.
I guarantee to you that that was 85-90% Ubuntu (or Debian) crap and maybe 10% "using rc scripts to do setup and then launch independent scripts in /rc#.d/ directories". RedHat-land init scripts were fine. They still are, if you're using RHEL/CentOS 6, which a metric crap ton of people still do... It's not like we can't boot using them, and if post-kernel boot times were that important for your VM (they're not) there are other solutions for streamlining things. Hell, using DashAsBinSh was something RH should have done a long time back.
2) Hardware interaction is far more reliable than it ever was under SysV init. This is much like boot speeds. There were hacks upon hacks to main dynamic hardware appear usable, but they were very fragile and painful to use. Desktop Linux under SysV init was horribly painful on the hardware interaction front between 1991 and the introduction of systemd.
These are the two main issues that had needed to be resolved since Linux's inception, and that nobody had ever been able to solve before. It's fashionable to dump on systemd (and Pulse Audio, for that matter), but they both solved a crucial failing of Linux desktops that nobody had done successfully -- ever.
Great. Make a laptop-focused OS that trades determinism and admin-controllability with dynamic response and trusting yourself into the arbitrary hands of fate and the guy who programmed udev.
It wasn't *that* painful -- it's not like people didn't use Linux on laptops in the 2000's -- and virtually all of it could be solved without replacing /sbin/init with a complex interlocking mesh of interdependent tools.
As a desktop user and server administrator, I love systemd. My only complaint about it is that it took way too long for it to kill SysV init.
There's very little besides automatic/free cgroup management that systemd's paradigm gives you that didn't already exist. Admins can and do use xinetd, service managers, tcp servers, and monitor programs to do their jobs. systemd wraps it all up into a nice, tiny, black-is-the-only-color bow for you... and if you think that's a good idea, you should consider why Embrace, Extend and Extinguish was such a successful strategy.
I want to write a lot of interesting and thought-provoking things here, but I'm just stuck on that one thing in the title.
Secret. Government. Proceedings.
Really, guys? Tell me again how your country is a free and democratic nation.
Well, because it's not a proceeding. Anyone who tells you otherwise doesn't understand Congressional proceedings.
The House is in recess. It's not even in the Committee of the Whole... So as far as proceedings go, they could just as easily be having a slumber party in their offices.
The Chairman didn't really have a choice... the Members were out of order. He could have:
a) had the Sergeant-at-Arms "enforce order", meaning kicking them out of the room for not being in their chairs properly,
b) call a recess
On the whole, B seems like a simpler option.
No one asked for systemd either, but look what happened.
Speaking more broadly, I honestly can't tell if this is because most of the major problems have been solved, people are too ignorant of the thought processes that went into the original tech, people don't know their history, people have too much faith in overly-complex technology that couldn't possible fail, people honestly, think they're just that much smarter than the installed base of users and want to increase "Quality of Life" (as one notable Borg put it), people want to make their own mark, or people are disingenuously trying to achieve lock-in on their newfangled contraption. No doubt, it's a mixture of all of the above.
Speaking as someone who's only been around in the industry for 15 years or so, I've already seen this pattern repeat way too frequently. I can only imagine what people who've been writing COBOL for the past 40 years think of it all...
Please, for the love of God, stop breaking sh*t that works fine.
The constitution doesn't grant you a right to use your cell phone
You're right. You know why? Because governments cannot grant rights. You are born with the right to absolutely anything you can imagine. The only thing any law can do is TAKE RIGHTS AWAY. The constitution doesn't have you grant you any right.
You're correct, but you have neither a civil right nor a Lockean or Hobbesian right to trespass on private property. If you're invited in, your invite can be revoked for any and all reasons not otherwise contrary to civil law relating to discrimination against Protected Classes in public establishments. "Cell phone user" is not one of those, so the owner of the hall is free to kick your ass out.
You have no liberty right on private property.
You can do all of this, already. Kinda makes you wonder, what else are you wrong about?
You can ask Google to do this. Once it's off your device, it's off your device.
Hopefully, we'll start demanding en masse for more and more data to remain on your device. F Google, Apple, and Facebook's cloud learning systems. For the actual processing and interpretation of data, we all have more than enough processing speed and more than enough space in the devices sitting in our pockets to parse through it offline.
Download Offline Maps, download a daily set of ads and rules, and let the CPU in my phone decide what to show me based on where I am now.
Cake or Death!
(Sorry, couldn't resist.)
You don't want to switch. Nobody is saying use Snappy Core exclusively, it's not designed for that.
The "adios" in the headline would seem to imply that as a rhetorical, at least.
Its main purpose is for servers that require zero downtime, where you want to upgrade application X and/or its dependencies without breaking even the smallest functionality of application Y.
... -_- No server "requires" zero downtime. Anyone who put that into a req. document should be shot. You get more 9's by duplicating and providing physical, logical, and temporal redundancy. And we've gotten high redundancy for servers just fine using RPMs, thanks. (And .deb's, I assume.) Acting like there are no solutions out there is ludicrous. But then, a lot of the last six years of Linux-land have been like that.
It might broadcast itself to a nearby Internet-connected device by careful modulation of some sufficiently-long trace on the circuit board. People have managed to broadcast AM radio from graphing calculators (first example that came to mind), so it does not even require superintelligence. Now, maybe if you put it in a Faraday cage + audio shielding, it might require more intelligence to get out, but even then I would not expect a superintelligence to fail to find a way out.
What's crucial to point out is that "things AI can control" is exactly equivalent to "hackable thing".
InfoSec doubles as your AI Response team, because if there's a way to do it with software (or using software to trick a human into doing it), it can be done by a malevolent AI.
I swear, people haven't read enough science fiction in their lives.
That's not for the car itself, it's for the occupants. I don't live in my toaster.
Yea, this boggles my mind somewhat. I realize M$ is always a worthy target, but "privacy" means something very different now that there are orders of magnitudes more devices present in our lives with microphones and sensors than it did in 1995 when you were lucky if your lone work PC recognized your SoundBlaster card at all.
Railing against W10 is meaningless. You want to protest unnecessary telemetry and data collection, storage, and evaluation, it's Google (read: Alphabet) and Facebook you should be worried about, with Apple as a shifty-eyed honorable mention... not because of closed source, but because of unclear engineering.
Side note: We're all linux and OSS advocates here, and that's fine. But "open source" is meaningless at the consumer level -- consumers don't have the ability to do meaningful code review, and the scale of HOBE hacks means that the advanced gcc-is-programmed-to-replace-login-compile-code hacks are easy enough to create when the payoff is 100M vulnerabilities.