The typical way companies do this is to issue two call simultaneously, and disconnect whichever call connects to the switch first (under a second). The telco will roue the remaining call to voicemail, and this usually happens before the call notification worms its way through the cellular network to the handset.
The student loan scammers are already doing this. One company was calling-not-calling my phone five times an hour? Not a distraction? It certainly is! My phone still goes vrrt-vrrt, and I still look up from what I'm doing.
It's an utter waste of the recipient's time and fails the "What if some large number of people did this?" test.
It probably exists, but it really isn't necessary. The sorts of things that are loaded via the Registry on Windows are loaded on OS X by virtue of their location in the file system. Uninstalling things is a simple matter of deleting packages from the Finder, and most software is self-contained: delete the bundle from/Applications, and you're done! For particularly rude programs, arguing with launchctl can usually set things right.
I suppose I was imprecise in my post. The machines are exactly as fast as they were when I bought them (faster now with SSDs and more RAM), but the newer versions of Office, Photoshop, Quickbooks, Chrome, etc. are heavier than the versions from 2010. A fair amount of my work happens in virtual machines running Linux and Windows, too. While Windows 7 and later are much better-suited to running virtualized than previous versions, they still assume a more capable machine than XP did (and so does modern Windows software).
We used to jokingly refer to it as Gates' Law (every 18 months, the speed of software halves), but even FOSS software isn't immune. It's sad because, as we run out of silicon headroom to remove to make our computers faster and more power-efficient, software quickly tends to bear the lion's share of responsibility for how much power is used globally by IT equipment. On my laptop that means a difference of an hour of battery life, but what's that mean globally?
I went Mac-exclusive in 2001 and stopped buying Apple products entirely in 2014. No Apple laptop made since my 17" 2010 MacBook Pro is as durable or expandable. No Apple desktop holds as much storage as my 2010 Mac Pro, and iPhones are no fun to use if I have to run iTunes on Windows. I've made peace with the notion that Apple makes more money selling gateways to their 30%-commissioned walled garden than they do by selling tools to people who write code and run lots of virtual machines. Rather than selling me a $3000 machine every other year, they passively collect constant income from easily-distracted end-users. Even if the numbers didn't make sense, the reduced level of effort certainly does. See also: Valve and why Half Life 3 is vaporware.
In the time since it became really clear that Apple didn't want to chase the business of people like me, I've switched away from software that's OS X-specific. I built a CentOS desktop and a Windows 10 desktop to see which one I'd run next. Either is fine. I'd prefer FreeBSD, but graphics and power management are a little behind the curve.
You see, Apple's disdain for pro customers isn't new, and it comes in long stretches. When they had the educational market in the US sewn up, they didn't need professional users. When that dried up, they successfully sold GUI Unix to hackers. If they need us, they know how to get in touch, but until they need us, they won't.
That said, I do love my last two Macs. They mostly Just Work. They're not fast anymore (8 years of software bloat will do that), but they're acceptable. I lament that they won't be replaced by other Macs, but life goes on. In the interim, I have work to do that I can't do efficiently on a single-disk/single-screen machine or a tiny notebook with soldered-in storage.
5. Sites that black out content until you provide an email address or dismiss the not-a-dialog-box by clicking the nonstandard close button.
Who would return to a store if their first experience upon walking through the door was being blindfolded by a sneaky salesdroid demanding a telephone number? 10 years ago, the prevailing wisdom was that only old folks like me used email anymore, anyway. Why does this antisocial anti-pattern persist?
I can tell you that pop-ins-are-awful@yourdomain signs up for a lot of newsletters, though. no-thanks@example.org used to sign up for them before whatever prevailing validation script started catching RFC 2606 and 6717 domains.
The problem with "extremist" is that it's a subjective term.
When war is in the minds of the Western populace, people supporting the other side (or neutral to a sufficient level of vociferousness) are "extreme." When war is not on their minds, people supporting counter-culture are "extreme."
Would a frank discussion of life as a homosexual or a trans-person have been "extreme" in the 1980s? Would an academic discussion of religion have been "extreme" in the 1970s? Would an objective exposé of the Vietnam War have been "extreme" in the 1960s?
The slope doesn't need to be slippery for this to be a dangerous idea. A simple view of what life is like on the receiving end of Western "democratization" might be objectionable enough to be extreme today. However, as the companies involved are private, what we can do is complain to them, support their competitors, and shout hellaciously at any government functionary who might imply that participation in this nonsense becomes mandatory./p>
The sole reason I charge my mobile and pay my data plan is so that companies can use me to improve their business plans and profitability while providing me with zero compensation. Let me know if I can do something else to help!
Raymond Chen has a recurring theme on his Old New Thing blog of "What if applications other than yours did this?" What would battery life or capped data plans look like if every application felt a need to send location telemetry home all the time?
The we know what a functioning and healthy society looks like and, thus, know what next step we're missing to get there.
That there is anything like a consensus on what the "most" important problems are or what their approaches should be.
That these are problems with direct solutions in "technology," without cutting-edge domain-specific knowledge.
I'm not sure that any of these is strictly true, and I'm nearly positive that we'll only know most of those answers in hindsight.
How about race relations? There's no app for that. War? You can't solder-up a PCB that convinces governments to stop murdering each other's citizens over differences of opinion.
Speaking of governments, what would a "techie" solution to government oppression look like? We have Tor, cryptocurrencies, steganographic filesystems, and mobile devices that would destroy the data on them before giving it up to an intrusive search, and look at how governments react.
That said, how about some of the areas where technology absolutely has worked on big problems?
Do you think climate change is a big problem? Do you think that the amount of power consumed by information technology globally is a terrifying figure in the face of anthropogenic climate change? This is a problem we know how to fix in "tech," and we're working on it.
Deaths due to traffic accidents? Computer vision and distributed coordination algorithms are at the core of self-driving automobiles.
How about 3D-printed prosthetics, or the medical industry in general? Data processing revolutionized drug research and genome work. Sure, there are more people doing silly apps than designing new systems for doing drug interaction simulation because one requires connections to established research labs, years of work, very expensive studies of efficacy, a decade of postsecondary education to have the domain-specific knowledge, and a hardware budget that runs into the millions; the other requires a crappy $300 laptop and some free software.
If there's a big problem out there that you want solved, either put up, pay up, or shut up.
My primary work product is a C/C++ manufacturing process-control application with bits of Lua and Perl embedded in it. Surrounding it are some web services, which I generally write in Python and Perl, with the front-ends obviously in Javascript.. There are some backend data-crunching services that process XML, and I've written those in Java.
A totally separate media-oriented project is a spitball of shell scripts, Perl, Javascript, and C, which I'm slowly replacing with Python, C++, Javascript, and Go. It's less a case of tossing the old code to replace it with something nifty and more the case of refactoring the code into a better-fitting implementation, now that we have a decade or so of use-cases to reflect upon.
I used to be a "C or GTFO" sort of guy until I realized how much time I was wasting by reimplementing hash tables and B-trees wherever I needed them. Python and Perl are decidedly Not Fast, but if they don't have to be fast, I can save a lot of stress and reduce the technical debt of code maintenance by writing the program in something where the problem domain fits better idiomatically. Sure, Java is annoying, but if you need to fan-out a couple hundred threads of data crunching, and your source data and results are in XML, the only thing that might fit better is C#.
I once read the suggestion to learn a new programming language every year. Do this. Get past "Hello, World," and at least solve toy problems. If you came from C, your whole world will change when you "get" map. If you came from Javascript, you'll have a whole new appreciation for the machine when you grok pointers. Try writing a piece of code in a functional style in a nonfunctional language (especially C) and discover how the language works against you and your resulting code is woefully inefficient.
Then, you'll be in a great position to embrace whatever tools you find and select the best one for the task at hand.
The significance of the advisory isn't that the initial firmware can be replaced. As indicated, that's a standard feature not only with Cisco gear but just about any computing device.
This is what should change. Firmware being read-write without some significant intervention is a huge factor in the current generation of vulnerabilities. Why is ROMMON write-enabled without moving a jumper or flipping a physical switch on the chassis?
Why can we update firmware on our PCs without needing to reboot into some special mode first? That stuff should be read-only (preferably with a hardware latch on the write-enable pin that's only cleared by a processor reset) as early as possible in the boot sequence.
The general case is that we do not update firmware while running the device. Even if you did that thirty times in the lifetime of the computer, they'd still be relatively exceptional cases. Why is the default behavior to trust that the OS will be bug-free enough to protect something so critical?
Or maybe I'm just getting old. Break out the UV EPROM-eraser and get off my lawn!
In this particular case, yes. There will always be non-exploitable bugs.
The problem is that when you begin to dismiss bugs as non-exploitable (whether you've fixed them or not) and their reports as "overblown," you put yourself in the unfortunate position of only needing to be wrong once. Specifically, dismissing bug reports with the notion that the bug would never be exploitable—not because the bug is "beyond the airtight hatchway," but because no one would be dumb enough to write an application in a particularly boneheaded way discounts decades of examples of people writing software in amazingly boneheaded ways.
Whether it's true or not (and, in this case, it seems true), this is not a way to inspire confidence, and an SSL implementation needs every bit as much community confidence as it does technical correctness.
The person penalized did, or allowed to be done, something illegal but not especially malicious or very damaging. They face a penalty which will certainly be unwelcome and which will probably encourage them to act within the law. No huge court case, no lives wrecked, no lawyers riding the gravy train. *This is how a legal system is supposed to be.*
Granted, that's a far sight better than how things are here in the US, but to say that's how things are "supposed to be" is aiming pretty low. That's still a legal system that spends taxpayer money to defend the "property" of copyright holders from nebulous threats, and punishes people for activities that have no provable harm to anyone. Wouldn't it be far more preferable to have a system that spends its time restituting actual victims instead of collecting arbitrary fines from people who aren't hurting anyone, perhaps a system that considered impact instead of looking at who's coloring outside the lines drawn by politicians?
I will furthermore submit that "The Rule of Law" will always be "The Rule of Lawyers" so long as the lawyers are the ones constructing laws prohibiting whatever behavior the well-connected consider inappropriate.
"Back in the day" (mid-90s) when that was more common, the term for it was "stiction." I don't know if it's less common these days because disk mechanisms are more reliable, the lubricants are better, or machines have much shorter average service lifetimes.
SGI field-service engineers actually had a rubber mallet specifically dedicated to coaxing stictioned drives to run for long enough to get the data off them. The Micropolis disks they shipped in their workstations back then were notorious for that (among many other problems). The company I worked for at the time had such a service call, and the technician told me that the hard part wasn't getting the disk running again, but convincing the disk that whanging the disk with a hammer was a sane thing to do!
Have you heard of the Software Engineering Radio podcast? I've been listening to it for a few years, and I really enjoy it—even if I don't share Markus' enthusiasm for model-driven software. The web site is at http://www.se-radio.net/, and even the back issues are worth listening to (processes don't get dated nearly as rapidly as tools).
Users cannot sit on Firefox 4.x They will be updated to the latest version when they open the About dialog (or sooner) because all* but the current Firefox release are unsupported versions in the new rapid release cycle. Those not current versions do not not get critical security updates except via the current version. Firefox users will not be spread across Firefox 4, 5, 6, etc. They will be on the latest version or they will be about to be on the latest version.
Effective expiration, lack of bugfixes, and rapidly replaced by newer versions with bugfixes? By any practical definition, there is no stable version. They're all betas from here onwards. The whole notion of a release isn't that it's bug-free, but that it's supported for a reasonably-long period of time.
Seriously. Aren't we big enough, as a species, to realize that there are people out there that hate us (no matter whom the "us" are), and that, fundamentally, it's their right to do so? If you don't like me because I'm a male, an American, without a degree, overweight, a Christian, from Texas, or whatever, I just flat don't care. I have more important stuff to worry about, and criticism to notice from people whose opinions actually matter to me. A life so empty of strife an conflict that it can be shaken just by someone forming a group called "Fuck <some group that I happen to identify with>" is a life to be envied, I suppose.
Nobody is going to please everyone he meets in life, and if you don't make any enemies along the way, you're probably not doing anything meaningful. If someone is going to waste time and energy hating me, I don't feel threatened. I don't feel endangered. If there existed a group called "Fuck all fat egocentric Texan assholes," I'd get a good chuckle out of it. Because, really, we can be pretty overbearing at times; we all can, though you'd probably never see fat egocentric Texan assholes shooting up the place and lighting fires because someone circulated cartoons of Sam Houston or Stephen Austin.
I mean, really, what's the harm? Short of the US military, there isn't a single group of people organized and equipped to exterminate or even cause widespread inconvenience to all fat egocentric Texan assholes--or all adherents of Islam, for that matter. And, really, if I were Islamic, I'd be a lot more worried about the US military than a club on a social networking website.
We need to all grow up (grow a pair, as the saying goes). Every person on this planet is a pathetic loser in one way or another. Thankfully we're all pathetic losers in different ways. Grow from the worthwhile criticism, and laugh at the rest. Whining for censorship is picking a fight in the parking lot because you lost on the mat. Call me an asshole, and be happy you can; I'll gladly return the favor.
The day we can't is the day we really have something to worry about.
About dump. So, that's a freebsd command? I've always suspected it existed, doing the very thing the man page described, because of the dump field in/etc/fstab. But I have never actually seen a machine which had the dump command... It's possibly not very safe BTW. If it works like DOS's Archive bit, than it can't be trusted: it can be set manually. Some DOS apps even used them as copy protection mechanism...
Please don't take this the wrong way, but how in the world could you do any sort of proper research for a technical article on backing up Unix systems without having run across the dump command (and its OS-specific variants: ufsdump, xfsdump, efsdump, and AIX backup)? It's not a FreeBSD-specific command. It or a similarly-named variant exists just about everywhere except on Linux. Linux used to have a proper ext2dump, but Linus decided that dump was deprecated because it was too difficult to make it work in the grand new VM/disk-cache subsystems of recent Linux kernels.
It works nothing like MS-DOS backup programs that used the FAT archive bit. It uses date comparisons and dumps low-level filesystem structures to a storage medium. That means:
Backups are not portable between operating systems, and sometimes not between filesystems of different types on the same OS (though this is rarely a concern on a commercial Unix).
dump really wants to be run on a quiesced or unmounted filesystem or on a filesystem snapshot.
Backups do not contain files from more than one filesystem.
Extended attributes (ACLs, etc.) tend to be preserved best with dump because dump has to understand the innards of the filesystem to work, anyway.
To operate dump, you have dump "levels". Level 0 is a full filesystem dump. Level 1 contains files that changed since the last level 0 dump. Level 2 contains files that changed since the last level 1 dump, and so on. A file/etc/dumpdates contains a log of backup activity and is used for date comparisons when doing dumps at levels other than 0. In a classic tape rotation, you'd do a level 0 dump once a week, a level 1 the rest of the week (to separate tapes), a level 0 dump to a different tape the next week, and rotate through the level 1 backup tapes again.
Dump and restore are particularly useful for doing system images on systems like Solaris, where the native tar command doesn't always know about extended filesystem attributes.
Must have vehicle in reasonable working condition... makes sense.
Must register vehicle... you said that was required before, so check.
Must be of appropriate age, fitness, etc... just as you said.
Must know and follow the rules of safe driving... seems pretty logical to me.
Must certify all of the above... NO WAY!!!! That's just crazy! You need to just take my word that I'm a great driver.
And a driver's license proves this exactly how? The octogenerian driving 30 miles/hour below the speed limit in her Lincoln LTD probably has a license, as does the yuppie next to you weaving her SUV back and forth as she screams into her cell phone. What about the idiot in front of you spewing exhaust so badly that it chokes you even with your air-conditioning on? He might be a few months behind on his safety inspection, but he probably has a license. Never mind the utter morons we have around here in Austin. They're probably all licensed, but that doesn't mean they know what turn signals, lanes, or traffic lights mean. They just pay the policeman the $30 for a license renewal and go on their ways.
Like most sorts of government license, a driver's license does not effectively imply a level of competence nearly as much as it is used as a method of governmental control to gate who many and may-not participate in a particular activity or sort of business venture. Don't play by the rules? You'll lose your license! Those rules may or may-not have anything to do with safely operating an automobile, nor do they legally need to; the state can suspend your license for whatever reasons they wish because they've declared that you have no right to that license.
If driver's licenses had anything to do with operating a vehicle safely, you'd see cops pulling people over for infractions other than speeding, and you'd see a lot more people on mass transit because so few of them actually display an ability to operate a motor vehicle in a manner that doesn't endanger everyone nearby.
IBM Tivoli Storage Manager Just Works (after a rather complicated setup process), does its job in the background on whatever schedule you choose, does it without complaint, maintains excruciatingly detailed logs, maintains multiple back-revisions of files, works over a network, SAN, or shared-media, and talks to tape drives and optical drives and pools of cheap disk.
If you want, backups can be mirrored across multiple TSM serves, and you can always fire up the (simple, ugly, but effective) GUI to check up on things, initiate immediate backups, or start restores.
I don't think the "System Backup and Recovery" option is avaiable for PC Unix-alikes, but on AIX it lets you boot the system (across the network) to a known-good state and restore your backup deltas from there.
It's difficult to impress me, and TSM impresses me. For most OSes, IBM even includes tape-device drivers that are specifically tuned to the tape device you're using (provided it's on the supported device list) and is much faster than the OS's built-in device driver (even on AIX) because it's optimized for the read/write/caching patterns that make sense for TSM.
That was one of my favorite features of Eudora, when I used it. You could set up filters for inbound email. By default, none of my incoming mail made any noise unless it was sent to the email address associated with a project on which I was working, or it was from my boss or girlfriend.
My cow-orkers got quite a giggle out Denis Leary singing "I'm an Asshole" whenever my boss sent a message to me. If I'd have been in a cube environment, I wouldn't have indulged like that, but I don't think it was too rude of me to pull that sort of stunt in an office.
That's why we have belt clips. My telephone stays on my belt, and if it starts buzzing when I'm somewhere that wouldn't be prudent or polite for me to answer a call, I push the "go to voicemail" button. In a meeting, this takes something like half of a ring, due to the well-thought-out position of that button (at the hinge, right by the antenna). The fallout is that only the person sitting next to me hears a BZRT, but no one else is bothered.
There's no need for cellular telephones to be irritating, and I'm even "one of those damned Nextel cusotmers". You'll never hear "beep-beep (incomprehensible distorted shouting)" from my phone if I'm in a meeting or restaurant or shop or something like that because my speaker will be turned off. It's all about learning how to use your telephone politely.
That said, I think a lot of this problem can be solved by not taking telephones places where it'd be rude to use them or turning them off upon entering. Is politeness really that difficult a concept?
MSFT is not, as TFA summary indicates, "licencing the internet," in any meaningful way.
...
This is clearly, yet again, a story that is more about MSFT bashing than about anything real.
No, it isn't. Businesses fear the BSA and the Microsoft Piracy bunch. Ask the City of Austin how they feel having to buy over half their Windows licenses twice a couple of years ago because they couldn't (due partially to a decentralized IT infrastructure) present all their license certificates on command of the Microsoft thugs and didn't want to face MS in court.
What does this have to do with this license? I mean, anyone with two brain cells that occasionally knock together can figure out that most these technologies not only weren't invented at Microsoft but predate Microsoft's involvement
with internetworking.
The answer is: "suits".
To a lot of idiots and naive managers, Microsoft is the end-all and be-all of computing because that's all they use, that's all they've ever used, and that's all their suit friends and the "glossies" tell them to use. "Does that 'you nicks' LDAP server have a license from Microsoft to do LDAP? We wouldn't want have to pay a penalty fee."
It's most likely the first step of a scummy sales tactic from a company that's steadily losing server-side business to companies like IBM and Sun who actually listen to what their customers want and produce (or, in the case of Linux, distribute) software that can be connected, sans firewall, to the Internet for more than a couple of days without being compromised.
Slashdot Mirror
The typical way companies do this is to issue two call simultaneously, and disconnect whichever call connects to the switch first (under a second). The telco will roue the remaining call to voicemail, and this usually happens before the call notification worms its way through the cellular network to the handset.
The student loan scammers are already doing this. One company was calling-not-calling my phone five times an hour? Not a distraction? It certainly is! My phone still goes vrrt-vrrt, and I still look up from what I'm doing.
It's an utter waste of the recipient's time and fails the "What if some large number of people did this?" test.
It probably exists, but it really isn't necessary. The sorts of things that are loaded via the Registry on Windows are loaded on OS X by virtue of their location in the file system. Uninstalling things is a simple matter of deleting packages from the Finder, and most software is self-contained: delete the bundle from /Applications, and you're done! For particularly rude programs, arguing with launchctl can usually set things right.
I suppose I was imprecise in my post. The machines are exactly as fast as they were when I bought them (faster now with SSDs and more RAM), but the newer versions of Office, Photoshop, Quickbooks, Chrome, etc. are heavier than the versions from 2010. A fair amount of my work happens in virtual machines running Linux and Windows, too. While Windows 7 and later are much better-suited to running virtualized than previous versions, they still assume a more capable machine than XP did (and so does modern Windows software).
We used to jokingly refer to it as Gates' Law (every 18 months, the speed of software halves), but even FOSS software isn't immune. It's sad because, as we run out of silicon headroom to remove to make our computers faster and more power-efficient, software quickly tends to bear the lion's share of responsibility for how much power is used globally by IT equipment. On my laptop that means a difference of an hour of battery life, but what's that mean globally?
I went Mac-exclusive in 2001 and stopped buying Apple products entirely in 2014. No Apple laptop made since my 17" 2010 MacBook Pro is as durable or expandable. No Apple desktop holds as much storage as my 2010 Mac Pro, and iPhones are no fun to use if I have to run iTunes on Windows. I've made peace with the notion that Apple makes more money selling gateways to their 30%-commissioned walled garden than they do by selling tools to people who write code and run lots of virtual machines. Rather than selling me a $3000 machine every other year, they passively collect constant income from easily-distracted end-users. Even if the numbers didn't make sense, the reduced level of effort certainly does. See also: Valve and why Half Life 3 is vaporware.
In the time since it became really clear that Apple didn't want to chase the business of people like me, I've switched away from software that's OS X-specific. I built a CentOS desktop and a Windows 10 desktop to see which one I'd run next. Either is fine. I'd prefer FreeBSD, but graphics and power management are a little behind the curve.
You see, Apple's disdain for pro customers isn't new, and it comes in long stretches. When they had the educational market in the US sewn up, they didn't need professional users. When that dried up, they successfully sold GUI Unix to hackers. If they need us, they know how to get in touch, but until they need us, they won't.
That said, I do love my last two Macs. They mostly Just Work. They're not fast anymore (8 years of software bloat will do that), but they're acceptable. I lament that they won't be replaced by other Macs, but life goes on. In the interim, I have work to do that I can't do efficiently on a single-disk/single-screen machine or a tiny notebook with soldered-in storage.
5. Sites that black out content until you provide an email address or dismiss the not-a-dialog-box by clicking the nonstandard close button.
Who would return to a store if their first experience upon walking through the door was being blindfolded by a sneaky salesdroid demanding a telephone number? 10 years ago, the prevailing wisdom was that only old folks like me used email anymore, anyway. Why does this antisocial anti-pattern persist?
I can tell you that pop-ins-are-awful@yourdomain signs up for a lot of newsletters, though. no-thanks@example.org used to sign up for them before whatever prevailing validation script started catching RFC 2606 and 6717 domains.
The problem with "extremist" is that it's a subjective term.
When war is in the minds of the Western populace, people supporting the other side (or neutral to a sufficient level of vociferousness) are "extreme." When war is not on their minds, people supporting counter-culture are "extreme."
Would a frank discussion of life as a homosexual or a trans-person have been "extreme" in the 1980s? Would an academic discussion of religion have been "extreme" in the 1970s? Would an objective exposé of the Vietnam War have been "extreme" in the 1960s?
The slope doesn't need to be slippery for this to be a dangerous idea. A simple view of what life is like on the receiving end of Western "democratization" might be objectionable enough to be extreme today. However, as the companies involved are private, what we can do is complain to them, support their competitors, and shout hellaciously at any government functionary who might imply that participation in this nonsense becomes mandatory./p>
The sole reason I charge my mobile and pay my data plan is so that companies can use me to improve their business plans and profitability while providing me with zero compensation. Let me know if I can do something else to help!
Raymond Chen has a recurring theme on his Old New Thing blog of "What if applications other than yours did this?" What would battery life or capped data plans look like if every application felt a need to send location telemetry home all the time?
Your question implies the following:
I'm not sure that any of these is strictly true, and I'm nearly positive that we'll only know most of those answers in hindsight.
How about race relations? There's no app for that. War? You can't solder-up a PCB that convinces governments to stop murdering each other's citizens over differences of opinion.
Speaking of governments, what would a "techie" solution to government oppression look like? We have Tor, cryptocurrencies, steganographic filesystems, and mobile devices that would destroy the data on them before giving it up to an intrusive search, and look at how governments react.
That said, how about some of the areas where technology absolutely has worked on big problems?
Do you think climate change is a big problem? Do you think that the amount of power consumed by information technology globally is a terrifying figure in the face of anthropogenic climate change? This is a problem we know how to fix in "tech," and we're working on it.
Deaths due to traffic accidents? Computer vision and distributed coordination algorithms are at the core of self-driving automobiles.
How about 3D-printed prosthetics, or the medical industry in general? Data processing revolutionized drug research and genome work. Sure, there are more people doing silly apps than designing new systems for doing drug interaction simulation because one requires connections to established research labs, years of work, very expensive studies of efficacy, a decade of postsecondary education to have the domain-specific knowledge, and a hardware budget that runs into the millions; the other requires a crappy $300 laptop and some free software.
If there's a big problem out there that you want solved, either put up, pay up, or shut up.
My primary work product is a C/C++ manufacturing process-control application with bits of Lua and Perl embedded in it. Surrounding it are some web services, which I generally write in Python and Perl, with the front-ends obviously in Javascript.. There are some backend data-crunching services that process XML, and I've written those in Java.
A totally separate media-oriented project is a spitball of shell scripts, Perl, Javascript, and C, which I'm slowly replacing with Python, C++, Javascript, and Go. It's less a case of tossing the old code to replace it with something nifty and more the case of refactoring the code into a better-fitting implementation, now that we have a decade or so of use-cases to reflect upon.
I used to be a "C or GTFO" sort of guy until I realized how much time I was wasting by reimplementing hash tables and B-trees wherever I needed them. Python and Perl are decidedly Not Fast, but if they don't have to be fast, I can save a lot of stress and reduce the technical debt of code maintenance by writing the program in something where the problem domain fits better idiomatically. Sure, Java is annoying, but if you need to fan-out a couple hundred threads of data crunching, and your source data and results are in XML, the only thing that might fit better is C#.
I once read the suggestion to learn a new programming language every year. Do this. Get past "Hello, World," and at least solve toy problems. If you came from C, your whole world will change when you "get" map. If you came from Javascript, you'll have a whole new appreciation for the machine when you grok pointers. Try writing a piece of code in a functional style in a nonfunctional language (especially C) and discover how the language works against you and your resulting code is woefully inefficient.
Then, you'll be in a great position to embrace whatever tools you find and select the best one for the task at hand.
The significance of the advisory isn't that the initial firmware can be replaced. As indicated, that's a standard feature not only with Cisco gear but just about any computing device.
This is what should change. Firmware being read-write without some significant intervention is a huge factor in the current generation of vulnerabilities. Why is ROMMON write-enabled without moving a jumper or flipping a physical switch on the chassis?
Why can we update firmware on our PCs without needing to reboot into some special mode first? That stuff should be read-only (preferably with a hardware latch on the write-enable pin that's only cleared by a processor reset) as early as possible in the boot sequence.
The general case is that we do not update firmware while running the device. Even if you did that thirty times in the lifetime of the computer, they'd still be relatively exceptional cases. Why is the default behavior to trust that the OS will be bug-free enough to protect something so critical?
Or maybe I'm just getting old. Break out the UV EPROM-eraser and get off my lawn!
In this particular case, yes. There will always be non-exploitable bugs.
The problem is that when you begin to dismiss bugs as non-exploitable (whether you've fixed them or not) and their reports as "overblown," you put yourself in the unfortunate position of only needing to be wrong once. Specifically, dismissing bug reports with the notion that the bug would never be exploitable—not because the bug is "beyond the airtight hatchway," but because no one would be dumb enough to write an application in a particularly boneheaded way discounts decades of examples of people writing software in amazingly boneheaded ways.
Whether it's true or not (and, in this case, it seems true), this is not a way to inspire confidence, and an SSL implementation needs every bit as much community confidence as it does technical correctness.
Q: What do we call "contrived test programs" in the "real" word?
A: Exploits.
So, just like shopping any under-development technology around (or applying for research grants) in real life, then?
Granted, that's a far sight better than how things are here in the US, but to say that's how things are "supposed to be" is aiming pretty low. That's still a legal system that spends taxpayer money to defend the "property" of copyright holders from nebulous threats, and punishes people for activities that have no provable harm to anyone. Wouldn't it be far more preferable to have a system that spends its time restituting actual victims instead of collecting arbitrary fines from people who aren't hurting anyone, perhaps a system that considered impact instead of looking at who's coloring outside the lines drawn by politicians?
I will furthermore submit that "The Rule of Law" will always be "The Rule of Lawyers" so long as the lawyers are the ones constructing laws prohibiting whatever behavior the well-connected consider inappropriate.
"Back in the day" (mid-90s) when that was more common, the term for it was "stiction." I don't know if it's less common these days because disk mechanisms are more reliable, the lubricants are better, or machines have much shorter average service lifetimes.
SGI field-service engineers actually had a rubber mallet specifically dedicated to coaxing stictioned drives to run for long enough to get the data off them. The Micropolis disks they shipped in their workstations back then were notorious for that (among many other problems). The company I worked for at the time had such a service call, and the technician told me that the hard part wasn't getting the disk running again, but convincing the disk that whanging the disk with a hammer was a sane thing to do!
Have you heard of the Software Engineering Radio podcast? I've been listening to it for a few years, and I really enjoy it—even if I don't share Markus' enthusiasm for model-driven software. The web site is at http://www.se-radio.net/, and even the back issues are worth listening to (processes don't get dated nearly as rapidly as tools).
From the big Bugzilla thread about version numbers earlier this week:
Effective expiration, lack of bugfixes, and rapidly replaced by newer versions with bugfixes? By any practical definition, there is no stable version. They're all betas from here onwards. The whole notion of a release isn't that it's bug-free, but that it's supported for a reasonably-long period of time.
Okay, so petition signatures are public record? How about henceforth Congress is only permitted to pass legislation by roll call?
Government of the who by the huh for the what-now?
No, it's why you get a home inspection and title insurance, both of which are usually required by the mortgage company, anyhow.
Seriously. Aren't we big enough, as a species, to realize that there are people out there that hate us (no matter whom the "us" are), and that, fundamentally, it's their right to do so? If you don't like me because I'm a male, an American, without a degree, overweight, a Christian, from Texas, or whatever, I just flat don't care. I have more important stuff to worry about, and criticism to notice from people whose opinions actually matter to me. A life so empty of strife an conflict that it can be shaken just by someone forming a group called "Fuck <some group that I happen to identify with>" is a life to be envied, I suppose.
Nobody is going to please everyone he meets in life, and if you don't make any enemies along the way, you're probably not doing anything meaningful. If someone is going to waste time and energy hating me, I don't feel threatened. I don't feel endangered. If there existed a group called "Fuck all fat egocentric Texan assholes," I'd get a good chuckle out of it. Because, really, we can be pretty overbearing at times; we all can, though you'd probably never see fat egocentric Texan assholes shooting up the place and lighting fires because someone circulated cartoons of Sam Houston or Stephen Austin.
I mean, really, what's the harm? Short of the US military, there isn't a single group of people organized and equipped to exterminate or even cause widespread inconvenience to all fat egocentric Texan assholes--or all adherents of Islam, for that matter. And, really, if I were Islamic, I'd be a lot more worried about the US military than a club on a social networking website.
We need to all grow up (grow a pair, as the saying goes). Every person on this planet is a pathetic loser in one way or another. Thankfully we're all pathetic losers in different ways. Grow from the worthwhile criticism, and laugh at the rest. Whining for censorship is picking a fight in the parking lot because you lost on the mat. Call me an asshole, and be happy you can; I'll gladly return the favor.
The day we can't is the day we really have something to worry about.
Please don't take this the wrong way, but how in the world could you do any sort of proper research for a technical article on backing up Unix systems without having run across the dump command (and its OS-specific variants: ufsdump, xfsdump, efsdump, and AIX backup)? It's not a FreeBSD-specific command. It or a similarly-named variant exists just about everywhere except on Linux. Linux used to have a proper ext2dump, but Linus decided that dump was deprecated because it was too difficult to make it work in the grand new VM/disk-cache subsystems of recent Linux kernels.
It works nothing like MS-DOS backup programs that used the FAT archive bit. It uses date comparisons and dumps low-level filesystem structures to a storage medium. That means:
To operate dump, you have dump "levels". Level 0 is a full filesystem dump. Level 1 contains files that changed since the last level 0 dump. Level 2 contains files that changed since the last level 1 dump, and so on. A file /etc/dumpdates contains a log of backup activity and is used for date comparisons when doing dumps at levels other than 0. In a classic tape rotation, you'd do a level 0 dump once a week, a level 1 the rest of the week (to separate tapes), a level 0 dump to a different tape the next week, and rotate through the level 1 backup tapes again.
Dump and restore are particularly useful for doing system images on systems like Solaris, where the native tar command doesn't always know about extended filesystem attributes.
And a driver's license proves this exactly how? The octogenerian driving 30 miles/hour below the speed limit in her Lincoln LTD probably has a license, as does the yuppie next to you weaving her SUV back and forth as she screams into her cell phone. What about the idiot in front of you spewing exhaust so badly that it chokes you even with your air-conditioning on? He might be a few months behind on his safety inspection, but he probably has a license. Never mind the utter morons we have around here in Austin. They're probably all licensed, but that doesn't mean they know what turn signals, lanes, or traffic lights mean. They just pay the policeman the $30 for a license renewal and go on their ways.
Like most sorts of government license, a driver's license does not effectively imply a level of competence nearly as much as it is used as a method of governmental control to gate who many and may-not participate in a particular activity or sort of business venture. Don't play by the rules? You'll lose your license! Those rules may or may-not have anything to do with safely operating an automobile, nor do they legally need to; the state can suspend your license for whatever reasons they wish because they've declared that you have no right to that license.
If driver's licenses had anything to do with operating a vehicle safely, you'd see cops pulling people over for infractions other than speeding, and you'd see a lot more people on mass transit because so few of them actually display an ability to operate a motor vehicle in a manner that doesn't endanger everyone nearby.
Yes, there is, but it's expensive
IBM Tivoli Storage Manager Just Works (after a rather complicated setup process), does its job in the background on whatever schedule you choose, does it without complaint, maintains excruciatingly detailed logs, maintains multiple back-revisions of files, works over a network, SAN, or shared-media, and talks to tape drives and optical drives and pools of cheap disk. If you want, backups can be mirrored across multiple TSM serves, and you can always fire up the (simple, ugly, but effective) GUI to check up on things, initiate immediate backups, or start restores.
I don't think the "System Backup and Recovery" option is avaiable for PC Unix-alikes, but on AIX it lets you boot the system (across the network) to a known-good state and restore your backup deltas from there.
It's difficult to impress me, and TSM impresses me. For most OSes, IBM even includes tape-device drivers that are specifically tuned to the tape device you're using (provided it's on the supported device list) and is much faster than the OS's built-in device driver (even on AIX) because it's optimized for the read/write/caching patterns that make sense for TSM.
That was one of my favorite features of Eudora, when I used it. You could set up filters for inbound email. By default, none of my incoming mail made any noise unless it was sent to the email address associated with a project on which I was working, or it was from my boss or girlfriend.
My cow-orkers got quite a giggle out Denis Leary singing "I'm an Asshole" whenever my boss sent a message to me. If I'd have been in a cube environment, I wouldn't have indulged like that, but I don't think it was too rude of me to pull that sort of stunt in an office.
That's why we have belt clips. My telephone stays on my belt, and if it starts buzzing when I'm somewhere that wouldn't be prudent or polite for me to answer a call, I push the "go to voicemail" button. In a meeting, this takes something like half of a ring, due to the well-thought-out position of that button (at the hinge, right by the antenna). The fallout is that only the person sitting next to me hears a BZRT, but no one else is bothered.
There's no need for cellular telephones to be irritating, and I'm even "one of those damned Nextel cusotmers". You'll never hear "beep-beep (incomprehensible distorted shouting)" from my phone if I'm in a meeting or restaurant or shop or something like that because my speaker will be turned off. It's all about learning how to use your telephone politely.
That said, I think a lot of this problem can be solved by not taking telephones places where it'd be rude to use them or turning them off upon entering. Is politeness really that difficult a concept?
No, it isn't. Businesses fear the BSA and the Microsoft Piracy bunch. Ask the City of Austin how they feel having to buy over half their Windows licenses twice a couple of years ago because they couldn't (due partially to a decentralized IT infrastructure) present all their license certificates on command of the Microsoft thugs and didn't want to face MS in court.
What does this have to do with this license? I mean, anyone with two brain cells that occasionally knock together can figure out that most these technologies not only weren't invented at Microsoft but predate Microsoft's involvement with internetworking.
The answer is: "suits".
To a lot of idiots and naive managers, Microsoft is the end-all and be-all of computing because that's all they use, that's all they've ever used, and that's all their suit friends and the "glossies" tell them to use. "Does that 'you nicks' LDAP server have a license from Microsoft to do LDAP? We wouldn't want have to pay a penalty fee."
It's most likely the first step of a scummy sales tactic from a company that's steadily losing server-side business to companies like IBM and Sun who actually listen to what their customers want and produce (or, in the case of Linux, distribute) software that can be connected, sans firewall, to the Internet for more than a couple of days without being compromised.