Your environment has to be a complete mess if you need 1 desktop tech per 100 windows machines. I could see it for a Linux setup if it's your first go of it, I imagine your IT department would need a few years to streamline things to get that number down, but really, windows is easy.
My experience is the exact opposite. While Windows is easy for a non-techie user to use, it's always been a PITA to administer (at least IME).
We've got about 60 Linux desktops and about 120 Windows desktops where I work. The Windows machines constantly need work -- drivers, A/V, user hand-holding, pushing out updates, updates broke something, etc. We've got two full-time desktop-only support guys to handle the Windows machines. OTOH, I handle the Linux desktops (and the server work and not quite half of the network admin work) and I get *maybe* two or three calls per month for the Linux desktops (which are almost always printer issues). All of the Linux desktops are about 500 miles -- and a satellite Internet connection -- away, so the ability to SSH in to the machines and fix things via CLI rather than a slow-to-update GUI is invaluable.
Buy a KBox. We bought one about a year and a half, maybe two years, ago and it has made our lives much simpler. The learning curve is a little steep, and be prepared for your users to whine a lot during the first six months while you learn what settings are good and what settings will cause more problems than they solve. However, once you've learned how to make it sing and dance, it will make managing Windows, Mac or RHEL systems really, really easy.
Oh, btw...two sys/net admins, two desktop support, one billing system support person in our IT staff for ~180 desktops, maybe 20 servers and about 60 remote locations across a couple million square miles of some of the most remote country in the U.S. (you can't drive to most of our sites).
IT Administrators are there to do what they are told.
Isn't that the role of *any* employee in a business? Unless you are the CEO -- and if your company is publicly traded, not even then because you still have to answer to the share holders -- you report to someone, and whoever signs your paycheck has the right to tell you what to do. This does not, however, mean that you, as the prima dona developer, have the right to tell me, as the SA, what to do. You can report problems to me, and request my assistance in fixing them. If I have no higher priorities -- and from your post, I can guarantee you will get an automatic downgrade in priority for any problem you request my assistance in fixing -- I'll see what I can do for you. However, I don't work for you, and I WON'T ask "how high" every time you say, "jump". And I am fortunate enough to work for a manager who will back me when I say, "That's just stupid. No, we aren't doing that."
I'll agree that there are also certified airplanes that are just as limited as the LSA's; I was just pointing out that there are few, if any, choices for anything other than an LSA that isn't nearly as old as I am:(
I did forget about the Dimond and Cirrus airplanes, however...although they are rather expensive, too (and you are correct that I confused pilot certification and airplane certification. I should have known better <sheepish>).
What was the reason the FAA created the LSA category? Even they realized that aircraft certification had become too cumbersome and expensive for anyone to build and certify a new design.
Unfortunately, while the LSA category is a step in the right direction, there are still a number of restrictions that make it less than ideal for those who want to fly for recreation (i.e., not as a provider of air travel services): you can't fly at night; you can't fly above 10,000 feet MSL; an LSA can have no more than two seats; etc. If you want to do any of these things, you need to buy a *certified* airplane, which means either you are buying a 20+ year old design (Cessna 172/182/206, most of the Piper singles and twins) or you spending well over a million dollars for a VLJ or turboprop.
I fly an airplane that qualifies under the experimental LSA regs, and one of the biggest drawbacks (and the reason I am thinking of selling my airplane) is that I can't carry both my wife and my daughter in it, which certainly limits the utility of the airplane. If I want to take it out for solo practice, it's great, but the truth is, I can't really do anything useful with it.
The problem is that those numbers are devilishly difficult to calculate. It's easy enough to calculate the *direct* costs of litigation, but that is only telling part of the story. How many companies have pulled out of the aviation/avionics market because they don't want to take on the liability of building airplane parts? I can tell you story after story of parts manufacturers and service shops who refuse to produce or work on airplane parts because they are afraid of the potential liability. Econ 101 says that competition and prices are indirectly related -- less competition means higher prices. Do your 3-5% estimates reflect that lack of competition?
The idea of a amateur homebrew helicopter is truly frightening...
There are several flying examples in the U.S. of amateur-built helicopters. Granted, most of them are built from professionally manufactured kits (i.e., the manufacturer builds a prefab kit, and the "builder" assembles the parts), some of which are relatively sophisticated. Google Rotorway and Mini-500 (I think the Mini-500 is defunct now, but there were several built back in the '90s).
...but *anybody* can dream, especially if he doesn't realize how incompetent he really is.
I would argue that often times, those who don't realize how incompetent they really are are, in fact, the ones who make the biggest advancements. They don't know it's impossible before they try, therefore they are the ones who go ahead and do it anyway. "The Ark was built by amateurs; the Titanic was built by professionals.";)
The next step up is the human powered guy. He needs a long carbon fiber boom as the main longitudinal structural component, and probably another one to carry the lift generated along the wings to the main boom. That's pretty expensive. Once he has that, then the drivetrain is bicycle technology.
Human powered flight is about as difficult as aviation gets, at least for right now. The problem is that it takes more energy than most people realize to lift a human body; we simply don't have the strength or endurance to generate that much power for very long. I believe the current record is 199km -- not too shabby, and much better than I thought before I looked it up just now. FWIW, building the carbon fiber boom isn't difficult (although the carbon fiber is relatively expensive -- but less so than most power plants for experimental aviation). I did the calculations for a carbon fiber wing spar about a decade ago, and estimated a ~1500 pound airplane stressed for a max G-loading of +6/-4 (plus reserve) would cost around $1000 - $2000. It isn't cheap, but it's certainly within the realm of possibility for someone of reasonable means. There was also a guy in Arizona about ten years ago who built a foot-launched sailplane called the Carbon Dragon, that would probably be worth studying if you were interested in trying to build a DIY human powered airplane. He sold plans for the airplane, but I believe he died a few years ago, so they might be a little hard to find now.
The rest of the wings and pilot's nacelle are fairly sophisticated, but within the capability of a weekend tinkerer to construct. The key is the sophisticated materials you can buy. You make a basketwork out of lightweight wood and Kevlar tape, then heat shrink polyester sheeting on it to create a skin.
That's not a bad way to do it. I remember reading magazines dating from the late '60s (I read them in the '90s, though) about people who had build entire airplanes with this method. My experimental airplane uses a slightly different method, that also might work. The wing spar is an aluminum tube with foam ribs glued to it. The leading edge is 2024T3 aluminum wrapped over the ribs to make a stiff skin over the first 1/4 of the wing, then the entire wing is covered a material made by 3M for building greenhouses called Tedlar (as the story goes, 3M freaked and pulled Tedlar from the market after hearing people were building airplanes out of it; I haven't verified the story). It looks like a full-scale model airplane wing covered in clear Monokote;) For human powered flight, I would probably replace the aluminum spar with carbon fiber, like you mentioned earlier, but the principle is the same.
The guy who is working on a certifiable electric aircraft needs to have the most money. He needs a real mach
But then, that's what bounty hunters are for -- these people can take the risks required to grab him and get him to the border because they're not agents of the government paying them for the collar. You think those "$100,000 reward for capture" posters don't look appealing? A plane ticket, a little bit of research, and a criminal's ego is all it takes to bring home the bacon.
Nevertheless, none of those bounty hunters you mentioned have yet located this particular guy, despite the fact that he's posting on Facebook...
Not saying you're wrong, mind you, but if they can't find him while he is (presumably) still in the U.K., what makes you think they will be able to find him in one of these other countries?
I used to work at an ISP with a rather...ummm...rabid...abuse administrator. The dude literally had a zero tolerance policy towards spam from our network. I saw him shut down a number of Internet customers who probably had no intention of violating our AUP's, and (IMHO, at least) had no idea why what they were doing might be frowned upon.
Then we got a several-thousand dollar a month customer who claimed that he wanted to build a VoIP network, but either 1) did not understand anything at all about network security or 2) was lying about the primary source of income for his servers. His servers were hacked (so he says) about once a month, and every time I tried to shut down his network, I was told to reenable his account because he had "fixed the problem". Yeah, right.
The question itself is sexist. "Not enough females..." or "Too many males..." implies that there is some ratio of men:women that you would like to meet in the IT industry. If you are trying to reach some proportion of employees based on gender, race, or any other artificial division we have used to differentiate ourselves, then you have already created an environment where getting the job done is less important than meeting some trivial detail that has absolutely no benefit to the company, other than perhaps PR.
IMHO, unless you have some kind of evidence that gender, race, etc. is being used to discriminate against otherwise qualified employees, then stop worrying about it and let people naturally gravitate to the jobs they want.
I would say yes, it probably is. I am running a very old version of Slack (9.something?) on a machine at home with a Celeron 700MHz with maybe 512M of RAM, and it works reasonably well. There are probably a lot of security problems with that machine since the OS is so old, but it's off more than it's on anymore, so I'm not really worried about it. Anyway, I run blackbox for a window manager, and it's fairly snappy, even on such old hardware. I don't know how well a newer version of Slack would run (been running Gentoo lately), but I imagine you could trim it down to do what you need, especially if all you need is the CLI.
The only problem with that is that I've never encountered someone jumping to Linux without having prior usage of it.
Hi! Allow me to introduce myself.
In ~2001, I needed a database server because (I shudder to admit it) I was learning Visual Basic and wanted to write VB apps that queried a database backend. I couldn't afford a Windows server, so I bought (yes, bought) Slackware 7.1 disks from Amazon and built my first Linux machine. I had never used Linux or Unix before then, but quickly found myself loving Linux. Over the next two years, I found myself using Windows less and less until finally, I removed the Windows partition from my home computer, and have never looked back since;)
To each their own, but I always liked Gentoo's emerge. I'm still new to Debian based distros (Debian Sid courtesy of a Knoppix HD install -- long story -- and Ubuntu on two work laptops), so maybe I just haven't learned the technique. What bugs me about apt-get or dpkg is that I haven't found a way to search for a package, if I don't know what it's called in the repository except by going to Google. On Gentoo, for example, I run emerge --search CGI and I get a listing of all the packages that contain "CGI" in the package name (mod_cgi, fastcgi, etc.) I haven't figured out a similar feature for apt-get or dpkg, despite reading the man pages. As for RPM...I haven't spent much time with RH-based distros, so I don't feel qualified to answer.
I would add just one thing to your list above: Does the distro support your *wireless* network hardware without complicated (and reoccurring) setup procedures?.
This is the #1 reason why I have switched to Ubuntu from Gentoo on my laptop -- IME, Ubuntu detects wireless hardware, detects nearby access points, tells you what kind of encryption (if any) they are using, and prompts you to enter a password whenever necessary. Wireless on Debian (Sid) was a PITA, but ultimately worked (although I have to figure all of this out on my own, then update a sym-link in/etc/network/ to change between networks). Wireless on Gentoo would usually work, but every time I ran an emerge --update world, I would have to reinstall the Intel drivers and tools and spend a week or two tweaking all the settings before I could get wireless working again. On some networks, I *never* got it working. I've actually had better luck connecting Ubuntu to wireless networks than my wife or my daughter's Windows laptops.
In a corporate environment, I would assume that the sys admins are (or at least should be) knowledgeable enough to modify the sudoers file to provide more appropriate permissions. Problem solved.
The home user environment is more complicated, however. If you are building a PC for home use, where the end-user will also be the sys admin (and probably won't have a clue), I only see two choices. You can just prevent all home users from doing anything (like installing new software) that requires root access, but that won't work -- the machine will have to be updated eventually, and either the distro will be horribly bloated or will end up lacking some programs that at least some people will want. You have already raised concerns about giving regular users sudo permissions, so apparently that's not a good choice either. Unfortunately, I see no third option.
I myself have been caught by this trap, where I moved to a different distribution and tried to do things the way I knew how, but then somebody told me on some forum later that I wasn't doing it the right way.
lol, you and me both. First time I tried to use Gentoo, I tried to add a new init script like I had always done in Slack, until my coworker explained to me the Gentoo start-up system. I thought it was horribly over-complicated at first, but as I started using it with the tools Gentoo provides, I began to appreciate how elegant and well-designed the Gentoo init scripts really were;)
Debian and CentOS/RHEL are among the best Linuces for servers, while Gentoo is completely inappropriate (although this almost never comes up because Gentoo fans are also completely inappropriate as sysadmins)
Care to back that up with any citations, stats or even anecdotes? Although we are currently migrating to Ubuntu server to make our new corporate overlords happy -- you can at least get paid support from Canonical if I and the other admin get hit by a bus one day -- we have been a Gentoo shop for many, many years, and it seems to have been a good choice for us. I won't pretend that Gentoo is *the* server distro to use (if something breaks during an update/upgrade, it usually breaks Big Time), but I have found it to be a very reliable, very stable platform as long as you use good judgment to maintain it. For example, update on a test machine, or at least an expendable machine before running emerge --update world on the entire server farm, and always, always, always make sure you run emerge --pretend --update world first.
As someone who uses both Gentoo and Ubuntu, I have to say that your reply makes for a great 30-second sound bite, but has little to do with the real world. While to some degree, Linux is Linux is Linux, there are differences between distributions that aren't always immediately obvious -- even to experienced users. I'm still learning some of the ways of getting things done in Ubuntu after 3+ years of Gentoo (and ~8 years of Slackware). For example, if I'm trying to figure out which Ubuntu package contains some program I want to install, I *don't* want to hear from an LFS or Gentoo user; I want to hear from someone who actually knows the distribution I am asking about. As the saying goes, it's not what you don't know that will bite you in the butt; it's what you *think* you know that just ain't so...and from the tone of your post, I'm guessing that's probably quite a bit.
Slashdot Mirror
Your environment has to be a complete mess if you need 1 desktop tech per 100 windows machines. I could see it for a Linux setup if it's your first go of it, I imagine your IT department would need a few years to streamline things to get that number down, but really, windows is easy.
My experience is the exact opposite. While Windows is easy for a non-techie user to use, it's always been a PITA to administer (at least IME).
We've got about 60 Linux desktops and about 120 Windows desktops where I work. The Windows machines constantly need work -- drivers, A/V, user hand-holding, pushing out updates, updates broke something, etc. We've got two full-time desktop-only support guys to handle the Windows machines. OTOH, I handle the Linux desktops (and the server work and not quite half of the network admin work) and I get *maybe* two or three calls per month for the Linux desktops (which are almost always printer issues). All of the Linux desktops are about 500 miles -- and a satellite Internet connection -- away, so the ability to SSH in to the machines and fix things via CLI rather than a slow-to-update GUI is invaluable.
Buy a KBox. We bought one about a year and a half, maybe two years, ago and it has made our lives much simpler. The learning curve is a little steep, and be prepared for your users to whine a lot during the first six months while you learn what settings are good and what settings will cause more problems than they solve. However, once you've learned how to make it sing and dance, it will make managing Windows, Mac or RHEL systems really, really easy.
Oh, btw...two sys/net admins, two desktop support, one billing system support person in our IT staff for ~180 desktops, maybe 20 servers and about 60 remote locations across a couple million square miles of some of the most remote country in the U.S. (you can't drive to most of our sites).
IT Administrators are there to do what they are told.
Isn't that the role of *any* employee in a business? Unless you are the CEO -- and if your company is publicly traded, not even then because you still have to answer to the share holders -- you report to someone, and whoever signs your paycheck has the right to tell you what to do. This does not, however, mean that you, as the prima dona developer, have the right to tell me, as the SA, what to do. You can report problems to me, and request my assistance in fixing them. If I have no higher priorities -- and from your post, I can guarantee you will get an automatic downgrade in priority for any problem you request my assistance in fixing -- I'll see what I can do for you. However, I don't work for you, and I WON'T ask "how high" every time you say, "jump". And I am fortunate enough to work for a manager who will back me when I say, "That's just stupid. No, we aren't doing that."
It has easily been close to a decade since Windows was utter crap.
Vista?
I'll agree that there are also certified airplanes that are just as limited as the LSA's; I was just pointing out that there are few, if any, choices for anything other than an LSA that isn't nearly as old as I am :(
I did forget about the Dimond and Cirrus airplanes, however...although they are rather expensive, too (and you are correct that I confused pilot certification and airplane certification. I should have known better <sheepish>).
What was the reason the FAA created the LSA category? Even they realized that aircraft certification had become too cumbersome and expensive for anyone to build and certify a new design.
Unfortunately, while the LSA category is a step in the right direction, there are still a number of restrictions that make it less than ideal for those who want to fly for recreation (i.e., not as a provider of air travel services): you can't fly at night; you can't fly above 10,000 feet MSL; an LSA can have no more than two seats; etc. If you want to do any of these things, you need to buy a *certified* airplane, which means either you are buying a 20+ year old design (Cessna 172/182/206, most of the Piper singles and twins) or you spending well over a million dollars for a VLJ or turboprop.
I fly an airplane that qualifies under the experimental LSA regs, and one of the biggest drawbacks (and the reason I am thinking of selling my airplane) is that I can't carry both my wife and my daughter in it, which certainly limits the utility of the airplane. If I want to take it out for solo practice, it's great, but the truth is, I can't really do anything useful with it.
The problem is that those numbers are devilishly difficult to calculate. It's easy enough to calculate the *direct* costs of litigation, but that is only telling part of the story. How many companies have pulled out of the aviation/avionics market because they don't want to take on the liability of building airplane parts? I can tell you story after story of parts manufacturers and service shops who refuse to produce or work on airplane parts because they are afraid of the potential liability. Econ 101 says that competition and prices are indirectly related -- less competition means higher prices. Do your 3-5% estimates reflect that lack of competition?
The idea of a amateur homebrew helicopter is truly frightening...
There are several flying examples in the U.S. of amateur-built helicopters. Granted, most of them are built from professionally manufactured kits (i.e., the manufacturer builds a prefab kit, and the "builder" assembles the parts), some of which are relatively sophisticated. Google Rotorway and Mini-500 (I think the Mini-500 is defunct now, but there were several built back in the '90s).
...but *anybody* can dream, especially if he doesn't realize how incompetent he really is.
I would argue that often times, those who don't realize how incompetent they really are are, in fact, the ones who make the biggest advancements. They don't know it's impossible before they try, therefore they are the ones who go ahead and do it anyway. "The Ark was built by amateurs; the Titanic was built by professionals." ;)
The next step up is the human powered guy. He needs a long carbon fiber boom as the main longitudinal structural component, and probably another one to carry the lift generated along the wings to the main boom. That's pretty expensive. Once he has that, then the drivetrain is bicycle technology.
Human powered flight is about as difficult as aviation gets, at least for right now. The problem is that it takes more energy than most people realize to lift a human body; we simply don't have the strength or endurance to generate that much power for very long. I believe the current record is 199km -- not too shabby, and much better than I thought before I looked it up just now. FWIW, building the carbon fiber boom isn't difficult (although the carbon fiber is relatively expensive -- but less so than most power plants for experimental aviation). I did the calculations for a carbon fiber wing spar about a decade ago, and estimated a ~1500 pound airplane stressed for a max G-loading of +6/-4 (plus reserve) would cost around $1000 - $2000. It isn't cheap, but it's certainly within the realm of possibility for someone of reasonable means. There was also a guy in Arizona about ten years ago who built a foot-launched sailplane called the Carbon Dragon, that would probably be worth studying if you were interested in trying to build a DIY human powered airplane. He sold plans for the airplane, but I believe he died a few years ago, so they might be a little hard to find now.
The rest of the wings and pilot's nacelle are fairly sophisticated, but within the capability of a weekend tinkerer to construct. The key is the sophisticated materials you can buy. You make a basketwork out of lightweight wood and Kevlar tape, then heat shrink polyester sheeting on it to create a skin.
That's not a bad way to do it. I remember reading magazines dating from the late '60s (I read them in the '90s, though) about people who had build entire airplanes with this method. My experimental airplane uses a slightly different method, that also might work. The wing spar is an aluminum tube with foam ribs glued to it. The leading edge is 2024T3 aluminum wrapped over the ribs to make a stiff skin over the first 1/4 of the wing, then the entire wing is covered a material made by 3M for building greenhouses called Tedlar (as the story goes, 3M freaked and pulled Tedlar from the market after hearing people were building airplanes out of it; I haven't verified the story). It looks like a full-scale model airplane wing covered in clear Monokote ;) For human powered flight, I would probably replace the aluminum spar with carbon fiber, like you mentioned earlier, but the principle is the same.
The guy who is working on a certifiable electric aircraft needs to have the most money. He needs a real mach
But then, that's what bounty hunters are for -- these people can take the risks required to grab him and get him to the border because they're not agents of the government paying them for the collar. You think those "$100,000 reward for capture" posters don't look appealing? A plane ticket, a little bit of research, and a criminal's ego is all it takes to bring home the bacon.
Nevertheless, none of those bounty hunters you mentioned have yet located this particular guy, despite the fact that he's posting on Facebook...
Not saying you're wrong, mind you, but if they can't find him while he is (presumably) still in the U.K., what makes you think they will be able to find him in one of these other countries?
Seconded.
I used to work at an ISP with a rather...ummm...rabid...abuse administrator. The dude literally had a zero tolerance policy towards spam from our network. I saw him shut down a number of Internet customers who probably had no intention of violating our AUP's, and (IMHO, at least) had no idea why what they were doing might be frowned upon.
Then we got a several-thousand dollar a month customer who claimed that he wanted to build a VoIP network, but either 1) did not understand anything at all about network security or 2) was lying about the primary source of income for his servers. His servers were hacked (so he says) about once a month, and every time I tried to shut down his network, I was told to reenable his account because he had "fixed the problem". Yeah, right.
Money talks, unfortunately.
I hate you... :D
Oh, you use Gentoo! ;D
"Irregardless" is a lot less annoying (to me, anyway) than, say, "ginormous."
I don't know why, but I loathe that word.
What should I do, server?
Move to Postfix ;)
On a completely off-topic, well...topic, I love the "#include" in your sig
The question itself is sexist. "Not enough females..." or "Too many males..." implies that there is some ratio of men:women that you would like to meet in the IT industry. If you are trying to reach some proportion of employees based on gender, race, or any other artificial division we have used to differentiate ourselves, then you have already created an environment where getting the job done is less important than meeting some trivial detail that has absolutely no benefit to the company, other than perhaps PR.
IMHO, unless you have some kind of evidence that gender, race, etc. is being used to discriminate against otherwise qualified employees, then stop worrying about it and let people naturally gravitate to the jobs they want.
Not snarky at all, and thanks!
I would say yes, it probably is. I am running a very old version of Slack (9.something?) on a machine at home with a Celeron 700MHz with maybe 512M of RAM, and it works reasonably well. There are probably a lot of security problems with that machine since the OS is so old, but it's off more than it's on anymore, so I'm not really worried about it. Anyway, I run blackbox for a window manager, and it's fairly snappy, even on such old hardware. I don't know how well a newer version of Slack would run (been running Gentoo lately), but I imagine you could trim it down to do what you need, especially if all you need is the CLI.
The only problem with that is that I've never encountered someone jumping to Linux without having prior usage of it.
Hi! Allow me to introduce myself.
;)
In ~2001, I needed a database server because (I shudder to admit it) I was learning Visual Basic and wanted to write VB apps that queried a database backend. I couldn't afford a Windows server, so I bought (yes, bought) Slackware 7.1 disks from Amazon and built my first Linux machine. I had never used Linux or Unix before then, but quickly found myself loving Linux. Over the next two years, I found myself using Windows less and less until finally, I removed the Windows partition from my home computer, and have never looked back since
To each their own, but I always liked Gentoo's emerge. I'm still new to Debian based distros (Debian Sid courtesy of a Knoppix HD install -- long story -- and Ubuntu on two work laptops), so maybe I just haven't learned the technique. What bugs me about apt-get or dpkg is that I haven't found a way to search for a package, if I don't know what it's called in the repository except by going to Google. On Gentoo, for example, I run emerge --search CGI and I get a listing of all the packages that contain "CGI" in the package name (mod_cgi, fastcgi, etc.) I haven't figured out a similar feature for apt-get or dpkg, despite reading the man pages. As for RPM...I haven't spent much time with RH-based distros, so I don't feel qualified to answer.
I would add just one thing to your list above: Does the distro support your *wireless* network hardware without complicated (and reoccurring) setup procedures?.
/etc/network/ to change between networks). Wireless on Gentoo would usually work, but every time I ran an emerge --update world, I would have to reinstall the Intel drivers and tools and spend a week or two tweaking all the settings before I could get wireless working again. On some networks, I *never* got it working. I've actually had better luck connecting Ubuntu to wireless networks than my wife or my daughter's Windows laptops.
This is the #1 reason why I have switched to Ubuntu from Gentoo on my laptop -- IME, Ubuntu detects wireless hardware, detects nearby access points, tells you what kind of encryption (if any) they are using, and prompts you to enter a password whenever necessary. Wireless on Debian (Sid) was a PITA, but ultimately worked (although I have to figure all of this out on my own, then update a sym-link in
What do you recommend, then?
In a corporate environment, I would assume that the sys admins are (or at least should be) knowledgeable enough to modify the sudoers file to provide more appropriate permissions. Problem solved.
The home user environment is more complicated, however. If you are building a PC for home use, where the end-user will also be the sys admin (and probably won't have a clue), I only see two choices. You can just prevent all home users from doing anything (like installing new software) that requires root access, but that won't work -- the machine will have to be updated eventually, and either the distro will be horribly bloated or will end up lacking some programs that at least some people will want. You have already raised concerns about giving regular users sudo permissions, so apparently that's not a good choice either. Unfortunately, I see no third option.
I myself have been caught by this trap, where I moved to a different distribution and tried to do things the way I knew how, but then somebody told me on some forum later that I wasn't doing it the right way.
lol, you and me both. First time I tried to use Gentoo, I tried to add a new init script like I had always done in Slack, until my coworker explained to me the Gentoo start-up system. I thought it was horribly over-complicated at first, but as I started using it with the tools Gentoo provides, I began to appreciate how elegant and well-designed the Gentoo init scripts really were ;)
Debian and CentOS/RHEL are among the best Linuces for servers, while Gentoo is completely inappropriate (although this almost never comes up because Gentoo fans are also completely inappropriate as sysadmins)
Care to back that up with any citations, stats or even anecdotes? Although we are currently migrating to Ubuntu server to make our new corporate overlords happy -- you can at least get paid support from Canonical if I and the other admin get hit by a bus one day -- we have been a Gentoo shop for many, many years, and it seems to have been a good choice for us. I won't pretend that Gentoo is *the* server distro to use (if something breaks during an update/upgrade, it usually breaks Big Time), but I have found it to be a very reliable, very stable platform as long as you use good judgment to maintain it. For example, update on a test machine, or at least an expendable machine before running emerge --update world on the entire server farm, and always, always, always make sure you run emerge --pretend --update world first.
I think you've just proven Harry Squatter's point for him. Congrats!
As someone who uses both Gentoo and Ubuntu, I have to say that your reply makes for a great 30-second sound bite, but has little to do with the real world. While to some degree, Linux is Linux is Linux, there are differences between distributions that aren't always immediately obvious -- even to experienced users. I'm still learning some of the ways of getting things done in Ubuntu after 3+ years of Gentoo (and ~8 years of Slackware). For example, if I'm trying to figure out which Ubuntu package contains some program I want to install, I *don't* want to hear from an LFS or Gentoo user; I want to hear from someone who actually knows the distribution I am asking about. As the saying goes, it's not what you don't know that will bite you in the butt; it's what you *think* you know that just ain't so...and from the tone of your post, I'm guessing that's probably quite a bit.
Furthermore, before you go bashing on Fedora and Ubuntu users too much, you might want to check which distro Torvalds says he prefers.