same song (same verse even)... read the full thread over on NANOG; especially interesting is Randy Bush's followup, featuring some slides (specifically, slide 20) he presented recently that basically say "yeah, there's a problem; no, the sky is not falling; none of the forced-cutover plans thus far presented have fully taken into account operational and business issues. Careful thought and deliberate action (rather than panic and haste) are needed here to avoid creating problems we'll be living with for the next 30 years." (Apologies if Randy thinks I'm paraphrasing him incorrectly; I doubt he spends much time reading/. though.:))
"By using small meshes of many machines that share a pipeline to the net instead of relying on lots of parallel connections, experts say they can create a system that is more intelligent and less prone to attack."
so by reducing redundancy and parallelism (and creating more single points of failure that affect larger swaths of the Net), we are... doing what again? Intelligence at the edge, speed at the core - that's what makes the Internet scale. Academics and bureacrats, even when well-meaning, frequently confuse what's good in theory with that works in practice, due to a lack of operational experience.
"The Internet is failing! Film at 11!"
(please stop using it so we can build one that gives us more control... it's good for you, citizen!)
Windows (assorted versions), FreeBSD (>= 6.x July 2006 or later, prior to that there was a scsi enumeration bug between FreeBSD and ESX that made running 6-prerelease impossible), OpenBSD, RHEL 4 (a few Debian as well). Approaching 4 figures on windows VMs, and we have a single person that manages patching, anti-virus and whatnot.
It's not just that VMware removes (or rather, greatly reduces) the hardware administration requirements, it's that it makes managing infrastructures _much_ more scalable. Changes can be applied en masse; resources can be handled as pools rather than in terms of whatever throughput any individual server can manage. Virtualization (with decent mgmt software anyway; my posts have been from the perspective of VI3) makes infrastructure management more scalable. You do then need your admins to also be familiar with VMware, but we at least have found that this was not a roadblock at all (and because familiarity with VI3 presupposes familiarity with storage, networking and capacity planning, our admins were all in favor of learning VI3 because they realized they'd be increasing their knowledge of these other basic tools as well, making them more valuable sysadmins).
Anything that gives me more free time is a major plus in my book.:)
You are certainly right that the transition can be a hassle; one of the reasons ours was so successful is that we were building a new datacenter from scratch to replace an existing one, and we spent 18 months designing it and moving in (and mistakes were still made). If we'd been trying to migrate an existing DC in-place, it would have been a much bigger headache.
---- However, the inevitable cost of this is management. While you reduce physical footprint, there are more server instances to manage, thus you need a larger staff to manage your server infrastructure... not to mention the specialized staff managing the virtual environment itself. This is not in itself a bad thing, and generally might lead to better management tools, too... but this is something that needs to be considered in any virtualization strategy. ----
This is completely wrong - the increased scalability and ease of management means you can manage a much larger virtual infrastructure (in terms of hosts) than you could have if it were physical, without increasing staff. In fact, in our case the ops team dropped by about 40% (attrition and other issues) during the development and deployment of the new virtual infrastructure, even while the overall size of the environment increased. We manage more with fewer people and nobody has to work as hard, because allocating hardware (unless you work in storage engineering - SAN allocation still occasionally requires ordering new shelves - or datacenter ops - somebody has to rack and cable the blade chassis) is now a point and click operation. And because ESX has a decent CLI (and because we do all our installs with pubkey auth for root pre-configured), I can run global commands on the infrastructure with a simple shell loop and ssh.
is not consolidation (although that's popular with the CFO). In fact, I'd say personally it's not even the ability to build vastly more scalable and redundant infrastructures (although that's a close second). My favorite feature of virtualization is how much easier it makes life for sysadmins (a selfish perspective, but entirely valid, as sysadmins are the ones who will be doing the management work, whether it's on physical gear or virtual). Need a new server? Clone one from a template and you've got add'l compute capacity (CPU/RAM/net/disk) up and available and into the load balancer in 30 minutes. Compare that with the time required to bring a physical machine online (assuming you have what you need in stock, built and ready to be racked), cable it, configure it, install an OS (even with ghost, pxeboot, {kick,jump}start, etc.), and get it into production. Even if you're following all the great ideas from e.g. infrastructures.org regarding managing your physical infrastructure, it's still going to be an order of magnitude less scalable than managing VMs, blades (or non-blade servers), storage and network stuff from a single location (VirtualCenter, in the case of VMware VI3).
Virtualization lets admins finally treat their compute resources as a bucket from which they can allocate discrete amounts to projects or business units in an on-demand fashion (in fact, with a little work you can even do internal billing so that IT is no longer a cost center - the rest of the company can finally see how much, in dollars, they are consuming of compute resources that used to be provided "for free" by IT).
(the above is based on experience; the infrastructure my team built and designed last year was the world's largest production VI3 implementation when it went online fall 2006.)
been running Ubuntu dualhead as my primary desktop workstation for work for over a year. Encountered zero of the problems you listed above, and I don't tweak the system at all (last tweaking required was getting dualhead going with the nvidia driver, a classic problem well documented on the ubuntu wiki). I haven't used the system much in the past 6 months (macbook FTW!), but when I need it, it's there, it Just Works, and requires no hand-holding.
It's almost like it's BSD or something...
when I read your list, the comment that comes to mind is, "either this person is inventing problems ("the" music player is named movie player? c'mon...) - i.e. trolling - or just hasn't bothered to learn how to use the facilities at hand (virtual desktop setup, changing software update options, enabling/disabling icons like network connect, and incorrectly mixing *nix tasks - unpacking software archives, which should be done from a command line, if at all - there's a reason we have binary packages - with GUI tasks (using a file manager))."
If you are really having that much difficulty with the Ubuntu default interface, go buy a Mac. I have no problems running BSD or Linux (or Solaris) as my primary desktop, but I got a Mac anyway and it Just Works (and runs Windows in a VM, besides). If you want to run Linux on the desktop, you can expect a few oddities here and there. Don't like it? Run something else or submit a patch. If you're going ot whine, at least come up with valid problems to whine about.
(yes, I did conflate copyright and patents. Part of that was intentional, but I should probably have split some of that off into a separate comment/thread.)
I think copyright infringement is trivial in comparison to the other problems facing our nation at the moment (in case the initial sarcasm was somehow overlooked).
I think that even in the unlikely event that the Justice Dept. is acting for purely altruistic purposes, this action cannot help but cast them as little more than the enforcement arm of the entertainment industry.
I think that if industry (software, entertainment and otherwise) spent more time and money innovating and less time protecting and growing the deterrent power of their patent portfolio, consumers would be better off.
I think that artists (digital and otherwise, regardless of medium) have a right to reasonable compensation for their work, for a limited time, to encourage innovation before works fall into the public domain. I think that the intent of copyright and patent law has been almost entirely subverted by corporate interests, and that individual artists (in the case of copyrights) or engineers (in the case of software patents) receive little to none of the compensation those systems were created to ensure.
Mostly, I think that our federal law enforcement has a HOST of better things to be doing with its time (and our money) than pursuing an ultimately futile attempt to shore up a failing business model for an industry that is so busy looking backward that they can't be bothered to produce something people are willing to pay for.
(agreed that IP is an important and complex issue, but I think that in the US at least, we have spent decades with the government erring in a heavy-handed manner on the side of protection for rights holders, at the expense of encouraging innovation.)
that the scale of problems facing our nation is so trivial that federal law enforcement can afford to waste their time^W^W^Wgive this matter the attention it deserves...
or like backups, for that matter: the surest way to need them is to find out you don't have them. Or, as Rabbs put it, "Better to see a network that's working and a sysadmin that's not than the other way around."
If you're fortunate enough to be an admin working for management that understands the function of sysadmins (as with doctors, our goal is to work ourselves out of a job), be grateful. Most management can't seem to grasp the concept that a sysadmin who's sitting around playing Doom is a sign that things are going _well_... if there were problems, she'd be too busy fixing them to be reading slashdot or playing games or idling on IRC. If your job includes an architecture component, the analogy starts to break down, but still... always better to be idle because things are working smoothly than busy because they're not.
avronius is right - "IT" is a term so broad that it really doesn't accurately describe what _anybody_ does for a living. If what you're doing feels like more work than play, my advice is, look at what you do for fun when you're not working. Do you like to game? Like to build stuff? Like to run services out of your house? It may not be that you're burned out on technology in general, but rather on the particular aspects you've been stuck in for a while.
For instance: it would only take about a week of Windows desktop support to burn me out, but I'm pretty certain I'll be doing network/application architecture and hacking on UN*X and OSS apps until I'm permanently retired (and probably for fun thereafter). After all, this is what I was doing for fun before I figured out I could get paid for it...
You might also look at getting out of the "world's largest" anything... diff'rent strokes for diff'rent folks notwithstanding, nothing makes me burn out faster than having to deal with the mind-numbing, soul-crushing bureacracy of most large corporations.
In summary: find something you like to do (might even be in tech), and find a company to do it for that's small enough to be flexible, fun and still concerned about the individual. Maybe easier said than done, but there are certainly a lot of places hiring sysadmins and programmers lately...
Half-Life (plus mission packs, sequels, etc.) equates to an awful lot of playing time. I'm partial to FPS, but also to a good story... Thief/Thief II and Deus Ex (the original, not the lame sequel) would also be good medium-length options. For really killing time, the whole Final Fantasy series (and the consoles required to run them all) would be the ultimate, I'd think.:)
eliminates very near 100% of spam from zombie hosts, because they will never attempt to resubmit mail if the recipient mailserver is busy. All RFC-compliant mailservers will try back later if told to do so; zombies will not.
agreed - Linux is not "bad" or "wrong" or to be avoided because of the holes in the docs, it's just one of Linux's flaws. Every OS has them - pick which ones you are most able to live with for the project at hand. Generally the documentation flaw is a relatively minor one for enterprise apps, which is why we run so much Linux @work - but commercial support doesn't really mean much to me personally (consistent man pages are more valuable), which is why I run OpenBSD for my colo stuff and @home.
I think you may be the first person to have understood the sig (at least the first to say so:)).
Linux has had documentation of dubious quality as long as I've been using it, since before TLDP. Even back in Redhat 5 days (or earlier, on old Slackware) it was a crap shoot whether you'd get a man page returned for any arbitrary command or system call. More likely than not you'd get nothing returned for third-party software, and this has not improved with the advent of package management systems. I'm not sure why Linux has had such a hard time maintaining consistent, accurate and up-to-date manual pages, but I suspect the development model is at least partly to blame. So is the lack of coherent focus on what format documentation should take (e.g. the total waste of time that are "info" pages - if it's a better format, fine - just PICK A FORMAT, ANY FORMAT and ship complete and up-to-date docs IN THAT FORMAT. Users should not have to go troll the Intarwebs to find out how to use system tools and the like.)
In contrast, take a look at OpenBSD's man pages sometime - for users who grew up on Linux and haven't used a BSD, OpenBSD in particular will blow you away with the quality, accuracy and completeness of its man pages. _Every_ system command, system call and most programming artifacts have complete and well-written manual pages that ship with the system. Software from the ports tree, with few exceptions, also includes quality man pages. For those who are used to having to spend lots of time finding accurate and updated documentation, knowing that the man pages are always reliable and current is a godsend. (Not to mention the irritation of needing documentation on e.g. one's firewall software, and having to go to the Internet to find it, when your Internet connection is down due to firewall software misconfiguration...)
My employer converted all the sysadmins to hourly last summer (right before a big datacenter migration). It was the best thing that ever happened to my paycheck. You don't realize how much you're actually working as a salaried geek until you start making overtime for all of it...
I should have been more clear - "I suppose if you define 'owner' to include both those that purchased the item themselves and those that received it as a gift"...
-- "That is a sort of errant pedantry up with which I will not put." -- Winston Churchill
I'm scratching my head over the two-thirds of console owners that _aren't_ adults - what kid can afford $150-$500 for a console? I suppose if you define "owner" to include "recipient of a gift paid for by someone else", I could see those numbers being fairly accurate. I know when I was a kid, it would have taken almost 6 months' worth of allowance to acquire even a previous-gen console (PS2, Gamecube, Xbox), not to mention games and accessories. Even when I had a "real job" in high school, by the time I got done paying for gas and car insurance, it would have taken a lot of saving to come up with a console.
surprised nobody has yet mentioned either of the shorts from/The Animatrix/ that deal specifically with the rise of AI and the concomitant "sentient equals or mechanical slaves" issues...
I'm a sysadmin, and "where's the support contract?" is a common mantra among management. However... when was the last time _anybody_ called Microsoft for support with MS Office? Can anybody even name a single instance of this? I know I can't (granted, I haven't been in desktop support for ages, but I don't think most companies even bother to purchase a "support contract" for MS Office - they just buy the software and move on).
Anybody out there know of an instance of someone actually utilizing an MS Office (or any office software, for that matter) support contract? This argument strikes me as one that just doesn't hold water...
the fact that your list didn't contain largely ORA titles was kind of surprising (my bookshelf is largely ORA with some other notable exceptions). By far the most-used (and most useful) book in my technical library is UNIX Power Tools. I could probably get by with that one volume alone and really not miss any of the rest of it. 800 pages of tips, tricks and hacks for the all-encompassing environment that is "UNIX" today, and nearly all of it is cross-platform and uses tools that would have worked equally well 15-20 years ago. Highly recommended.
Slashdot Mirror
same song (same verse even) ... read the full thread over on NANOG; especially interesting is Randy Bush's followup, featuring some slides (specifically, slide 20) he presented recently that basically say "yeah, there's a problem; no, the sky is not falling; none of the forced-cutover plans thus far presented have fully taken into account operational and business issues. Careful thought and deliberate action (rather than panic and haste) are needed here to avoid creating problems we'll be living with for the next 30 years." (Apologies if Randy thinks I'm paraphrasing him incorrectly; I doubt he spends much time reading /. though. :))
"By using small meshes of many machines that share a pipeline to the net instead of relying on lots of parallel connections, experts say they can create a system that is more intelligent and less prone to attack."
... doing what again? Intelligence at the edge, speed at the core - that's what makes the Internet scale. Academics and bureacrats, even when well-meaning, frequently confuse what's good in theory with that works in practice, due to a lack of operational experience.
... it's good for you, citizen!)
so by reducing redundancy and parallelism (and creating more single points of failure that affect larger swaths of the Net), we are
"The Internet is failing! Film at 11!"
(please stop using it so we can build one that gives us more control
anybody who claims to be an admin and yet does not have an inherent frustration towards users is either lying, stoned or inexperienced.
Windows (assorted versions), FreeBSD (>= 6.x July 2006 or later, prior to that there was a scsi enumeration bug between FreeBSD and ESX that made running 6-prerelease impossible), OpenBSD, RHEL 4 (a few Debian as well). Approaching 4 figures on windows VMs, and we have a single person that manages patching, anti-virus and whatnot.
:)
It's not just that VMware removes (or rather, greatly reduces) the hardware administration requirements, it's that it makes managing infrastructures _much_ more scalable. Changes can be applied en masse; resources can be handled as pools rather than in terms of whatever throughput any individual server can manage. Virtualization (with decent mgmt software anyway; my posts have been from the perspective of VI3) makes infrastructure management more scalable. You do then need your admins to also be familiar with VMware, but we at least have found that this was not a roadblock at all (and because familiarity with VI3 presupposes familiarity with storage, networking and capacity planning, our admins were all in favor of learning VI3 because they realized they'd be increasing their knowledge of these other basic tools as well, making them more valuable sysadmins).
Anything that gives me more free time is a major plus in my book.
You are certainly right that the transition can be a hassle; one of the reasons ours was so successful is that we were building a new datacenter from scratch to replace an existing one, and we spent 18 months designing it and moving in (and mistakes were still made). If we'd been trying to migrate an existing DC in-place, it would have been a much bigger headache.
----
However, the inevitable cost of this is management. While you reduce physical footprint, there are more server instances to manage, thus you need a larger staff to manage your server infrastructure... not to mention the specialized staff managing the virtual environment itself. This is not in itself a bad thing, and generally might lead to better management tools, too... but this is something that needs to be considered in any virtualization strategy.
----
This is completely wrong - the increased scalability and ease of management means you can manage a much larger virtual infrastructure (in terms of hosts) than you could have if it were physical, without increasing staff. In fact, in our case the ops team dropped by about 40% (attrition and other issues) during the development and deployment of the new virtual infrastructure, even while the overall size of the environment increased. We manage more with fewer people and nobody has to work as hard, because allocating hardware (unless you work in storage engineering - SAN allocation still occasionally requires ordering new shelves - or datacenter ops - somebody has to rack and cable the blade chassis) is now a point and click operation. And because ESX has a decent CLI (and because we do all our installs with pubkey auth for root pre-configured), I can run global commands on the infrastructure with a simple shell loop and ssh.
Good comments, otherwise.
is not consolidation (although that's popular with the CFO). In fact, I'd say personally it's not even the ability to build vastly more scalable and redundant infrastructures (although that's a close second). My favorite feature of virtualization is how much easier it makes life for sysadmins (a selfish perspective, but entirely valid, as sysadmins are the ones who will be doing the management work, whether it's on physical gear or virtual). Need a new server? Clone one from a template and you've got add'l compute capacity (CPU/RAM/net/disk) up and available and into the load balancer in 30 minutes. Compare that with the time required to bring a physical machine online (assuming you have what you need in stock, built and ready to be racked), cable it, configure it, install an OS (even with ghost, pxeboot, {kick,jump}start, etc.), and get it into production. Even if you're following all the great ideas from e.g. infrastructures.org regarding managing your physical infrastructure, it's still going to be an order of magnitude less scalable than managing VMs, blades (or non-blade servers), storage and network stuff from a single location (VirtualCenter, in the case of VMware VI3).
Virtualization lets admins finally treat their compute resources as a bucket from which they can allocate discrete amounts to projects or business units in an on-demand fashion (in fact, with a little work you can even do internal billing so that IT is no longer a cost center - the rest of the company can finally see how much, in dollars, they are consuming of compute resources that used to be provided "for free" by IT).
(the above is based on experience; the infrastructure my team built and designed last year was the world's largest production VI3 implementation when it went online fall 2006.)
been running Ubuntu dualhead as my primary desktop workstation for work for over a year. Encountered zero of the problems you listed above, and I don't tweak the system at all (last tweaking required was getting dualhead going with the nvidia driver, a classic problem well documented on the ubuntu wiki). I haven't used the system much in the past 6 months (macbook FTW!), but when I need it, it's there, it Just Works, and requires no hand-holding.
...
...) - i.e. trolling - or just hasn't bothered to learn how to use the facilities at hand (virtual desktop setup, changing software update options, enabling/disabling icons like network connect, and incorrectly mixing *nix tasks - unpacking software archives, which should be done from a command line, if at all - there's a reason we have binary packages - with GUI tasks (using a file manager))."
It's almost like it's BSD or something
when I read your list, the comment that comes to mind is, "either this person is inventing problems ("the" music player is named movie player? c'mon
If you are really having that much difficulty with the Ubuntu default interface, go buy a Mac. I have no problems running BSD or Linux (or Solaris) as my primary desktop, but I got a Mac anyway and it Just Works (and runs Windows in a VM, besides). If you want to run Linux on the desktop, you can expect a few oddities here and there. Don't like it? Run something else or submit a patch. If you're going ot whine, at least come up with valid problems to whine about.
340 trillion
billion
million
gigabytes.
yeah, I think if you go out on a limb and say "never", you can rest assured that you will at least not be proven wrong in your lifetime.
(yes, I did conflate copyright and patents. Part of that was intentional, but I should probably have split some of that off into a separate comment/thread.)
I think copyright infringement is trivial in comparison to the other problems facing our nation at the moment (in case the initial sarcasm was somehow overlooked).
I think that even in the unlikely event that the Justice Dept. is acting for purely altruistic purposes, this action cannot help but cast them as little more than the enforcement arm of the entertainment industry.
I think that if industry (software, entertainment and otherwise) spent more time and money innovating and less time protecting and growing the deterrent power of their patent portfolio, consumers would be better off.
I think that artists (digital and otherwise, regardless of medium) have a right to reasonable compensation for their work, for a limited time, to encourage innovation before works fall into the public domain. I think that the intent of copyright and patent law has been almost entirely subverted by corporate interests, and that individual artists (in the case of copyrights) or engineers (in the case of software patents) receive little to none of the compensation those systems were created to ensure.
Mostly, I think that our federal law enforcement has a HOST of better things to be doing with its time (and our money) than pursuing an ultimately futile attempt to shore up a failing business model for an industry that is so busy looking backward that they can't be bothered to produce something people are willing to pay for.
(agreed that IP is an important and complex issue, but I think that in the US at least, we have spent decades with the government erring in a heavy-handed manner on the side of protection for rights holders, at the expense of encouraging innovation.)
that the scale of problems facing our nation is so trivial that federal law enforcement can afford to waste their time^W^W^Wgive this matter the attention it deserves ...
or like backups, for that matter: the surest way to need them is to find out you don't have them. Or, as Rabbs put it, "Better to see a network that's working and a sysadmin that's not than the other way around."
... if there were problems, she'd be too busy fixing them to be reading slashdot or playing games or idling on IRC. If your job includes an architecture component, the analogy starts to break down, but still ... always better to be idle because things are working smoothly than busy because they're not.
If you're fortunate enough to be an admin working for management that understands the function of sysadmins (as with doctors, our goal is to work ourselves out of a job), be grateful. Most management can't seem to grasp the concept that a sysadmin who's sitting around playing Doom is a sign that things are going _well_
avronius is right - "IT" is a term so broad that it really doesn't accurately describe what _anybody_ does for a living. If what you're doing feels like more work than play, my advice is, look at what you do for fun when you're not working. Do you like to game? Like to build stuff? Like to run services out of your house? It may not be that you're burned out on technology in general, but rather on the particular aspects you've been stuck in for a while.
...
... diff'rent strokes for diff'rent folks notwithstanding, nothing makes me burn out faster than having to deal with the mind-numbing, soul-crushing bureacracy of most large corporations.
...
For instance: it would only take about a week of Windows desktop support to burn me out, but I'm pretty certain I'll be doing network/application architecture and hacking on UN*X and OSS apps until I'm permanently retired (and probably for fun thereafter). After all, this is what I was doing for fun before I figured out I could get paid for it
You might also look at getting out of the "world's largest" anything
In summary: find something you like to do (might even be in tech), and find a company to do it for that's small enough to be flexible, fun and still concerned about the individual. Maybe easier said than done, but there are certainly a lot of places hiring sysadmins and programmers lately
ditto ... Thievery UT would be a good one to have along if that inter-island WAN was up and running. :)
Half-Life (plus mission packs, sequels, etc.) equates to an awful lot of playing time. I'm partial to FPS, but also to a good story ... Thief/Thief II and Deus Ex (the original, not the lame sequel) would also be good medium-length options. For really killing time, the whole Final Fantasy series (and the consoles required to run them all) would be the ultimate, I'd think. :)
hee ... that's the most I've laughed at a /. post in recent memory. Excellent choice of words, sir. :)
eliminates very near 100% of spam from zombie hosts, because they will never attempt to resubmit mail if the recipient mailserver is busy. All RFC-compliant mailservers will try back later if told to do so; zombies will not.
d
http://www.openbsd.org/cgi-bin/man.cgi?query=spam
agreed - Linux is not "bad" or "wrong" or to be avoided because of the holes in the docs, it's just one of Linux's flaws. Every OS has them - pick which ones you are most able to live with for the project at hand. Generally the documentation flaw is a relatively minor one for enterprise apps, which is why we run so much Linux @work - but commercial support doesn't really mean much to me personally (consistent man pages are more valuable), which is why I run OpenBSD for my colo stuff and @home.
:)).
I think you may be the first person to have understood the sig (at least the first to say so
cheers!
Linux has had documentation of dubious quality as long as I've been using it, since before TLDP. Even back in Redhat 5 days (or earlier, on old Slackware) it was a crap shoot whether you'd get a man page returned for any arbitrary command or system call. More likely than not you'd get nothing returned for third-party software, and this has not improved with the advent of package management systems. I'm not sure why Linux has had such a hard time maintaining consistent, accurate and up-to-date manual pages, but I suspect the development model is at least partly to blame. So is the lack of coherent focus on what format documentation should take (e.g. the total waste of time that are "info" pages - if it's a better format, fine - just PICK A FORMAT, ANY FORMAT and ship complete and up-to-date docs IN THAT FORMAT. Users should not have to go troll the Intarwebs to find out how to use system tools and the like.)
...)
In contrast, take a look at OpenBSD's man pages sometime - for users who grew up on Linux and haven't used a BSD, OpenBSD in particular will blow you away with the quality, accuracy and completeness of its man pages. _Every_ system command, system call and most programming artifacts have complete and well-written manual pages that ship with the system. Software from the ports tree, with few exceptions, also includes quality man pages. For those who are used to having to spend lots of time finding accurate and updated documentation, knowing that the man pages are always reliable and current is a godsend. (Not to mention the irritation of needing documentation on e.g. one's firewall software, and having to go to the Internet to find it, when your Internet connection is down due to firewall software misconfiguration
My employer converted all the sysadmins to hourly last summer (right before a big datacenter migration). It was the best thing that ever happened to my paycheck. You don't realize how much you're actually working as a salaried geek until you start making overtime for all of it ...
I should have been more clear - "I suppose if you define 'owner' to include both those that purchased the item themselves and those that received it as a gift" ...
--
"That is a sort of errant pedantry up with which I will not put." -- Winston Churchill
I'm scratching my head over the two-thirds of console owners that _aren't_ adults - what kid can afford $150-$500 for a console? I suppose if you define "owner" to include "recipient of a gift paid for by someone else", I could see those numbers being fairly accurate. I know when I was a kid, it would have taken almost 6 months' worth of allowance to acquire even a previous-gen console (PS2, Gamecube, Xbox), not to mention games and accessories. Even when I had a "real job" in high school, by the time I got done paying for gas and car insurance, it would have taken a lot of saving to come up with a console.
surprised nobody has yet mentioned either of the shorts from /The Animatrix/ that deal specifically with the rise of AI and the concomitant "sentient equals or mechanical slaves" issues ...
I'm a sysadmin, and "where's the support contract?" is a common mantra among management. However ... when was the last time _anybody_ called Microsoft for support with MS Office? Can anybody even name a single instance of this? I know I can't (granted, I haven't been in desktop support for ages, but I don't think most companies even bother to purchase a "support contract" for MS Office - they just buy the software and move on).
...
Anybody out there know of an instance of someone actually utilizing an MS Office (or any office software, for that matter) support contract? This argument strikes me as one that just doesn't hold water
the fact that your list didn't contain largely ORA titles was kind of surprising (my bookshelf is largely ORA with some other notable exceptions). By far the most-used (and most useful) book in my technical library is UNIX Power Tools. I could probably get by with that one volume alone and really not miss any of the rest of it. 800 pages of tips, tricks and hacks for the all-encompassing environment that is "UNIX" today, and nearly all of it is cross-platform and uses tools that would have worked equally well 15-20 years ago. Highly recommended.