"A Linux System Administrator who doesn't build his own box is like a Jedi who doesn't build his own lightsaber."
Besides, when building for a non-Windows PC, one has to exercise extra care to ensure that one gets hardware that will be supported by the OS one intends to install. Unfortunately, Linux just doesn't have nearly as much vendor buy-in for hardware support as MicroSoft Windows. Thus, while many pre-assembled PC's may have hardware suitable for use with a non-MicroSoft OS it's still easier IMHO to just "bare-bones" an appropriate system together. YMMV.
If Greenpeace had been around when T. Edison was doing all his most brilliant work, we'd never have gotten electricity for public use (too dangerous - although Edison's DC system was less dangerous than Westinghouse's AC system, which is in use to this day).
If Greenpeace had been around when Herr Bayer did his thing with acetylsalicylic acid, we'd all still be chewing on tree-bark every time we got a headache. I mean, have you looked at the side effects of that stuff?
Praise the Maker, Greenpeace wasn't around when H. Ford started churning out cheap automobiles. All that wood, all that metal - hell, Ford couldn't even figure out what to do with the char-coal left over from firing his plants; his stepbrother Mr. Kingsford found a market for it. Greenpeace would've gone ape at the thought of encouraging Americans to cook over an artificial fuel source, I'm sure.
No wonder we can't build stuff like the Grand Coulee Dam anymore.
Only requires a moment to stop by the "Software Selection" portion of the initial installation and remove X11 et. al. from the list. I guess that puts me in the top 1.00%, eh?
If it's a server, your default init state should be runlevel 3, right? Doesn't matter if you've installed all the XGL stuff in the world if it never gets run (for the most part; carrying unnecessary executables/packages/services around on your system does potentially open vulnerabilities in your system).
You should never take a server to runlevel 5 unless it's been taken out of service for maintenance - and not even then! Just because a GUI may make you able to more quickly or more simply maintain your server doesn't mean that it's okay to run X on a server. GUI's tend to "dumb down" user tasks (that is their function, after all). GUI's have progressed over the last decade, but they still carry their penalties in system load, "dumb-down" factor and increased vulnerability to exploitation.
As for using RHEL as a desktop, I agree wholeheartedly. Everyone knows that Gnome under OpenSuSE 10.2 is the ultimate XGL desktop experience!
Or are you just making this up? Even clear back to the 1.0.25 kernel, I can recall seeing uptimes in excess of 200 days. If some specific Linux distro ever had such a bug, it was almost certainly short-lived.
Wait a minute . . . that was a Windows bug - Win95, Win98 and (IIRC) NT4.0SP2. Boot yer box and let it do nothing, some kernel pointer associated with timekeeping would overflow at 49.7 days. Hellfire, MicroSoft squashed that one years ago!
So the only question that's left is: are you a Luddite or a Fudite?
No, I didn't know these things - although it only makes sense: "The SCO group" has nothing to do with the Santa Cruz Operation, except that they had a business dealing once.
I guess I'm still saddened to hear how the Santa Cruz Operation has ended up; but at least they didn't come to this (even if their name did).
I remember when I first saw Xenix, as provided by the Santa Cruz Operation. I was utterly and completely enthralled. This thing ran on a '286 and it smoked. It had a respectable filesystem, decent performance, excellent stability - it was aces, and an excellent learning environment for a lot of us old-timers. SCO's licensing made sense, their support was top-flight and their OS was undeniably the direct descendant of AT&T UNIX.
Now? About fifteen years out of date (funny, how SCO's ability to innovate stopped around the same time they started litigating).
But (as you say) the I-17 corridor isn't a hotly contested business opportunity. At the end of the day, companies would rather turn a 5% profit on a $10,000 investment than a 15% profit on a $1,000 investment. Since they can't get that 5% of $10,000 thing going, it isn't worth their attention to try for 15% of $1,000; they have plenty of other 5% on $10,000 opportunities to investigate around populated areas.
Consider - in Namibia and Nigeria, I doubt there is much of a paying customer base to pay for installation and upkeep of such facilities; and I'll wager that in India, the towers under discussions are meant to serve areas of the country with little of interest to commercial providers. They need a system of towers which they can install, configure and walk away from. Doubtless the governments of the respective countries are directly involved in the planning, implementation and financing of the towers in question; hence their drive to have an efficient, self-sufficient implementation.
Here in the US, if somebody wants something, they'd better damned well be ready to pay for it. Here, "you gets what you pays for!". Now, if Cellulite-1 (or whoever) decides to put a tower up, you can bet it'll be the classic sort which requires power and periodic maintainance to remain operational; expenses which will not only be passed along to the consumer, but which will be passed along at a profit! If the cell towers generate their own electricity, there's that much less expense to make a profit on.
Yes, it seems counterintuitive, but why develop a technology which costs less to implement if that technology is going to cut into your bottom line? Current, technologies with a lower TCO provide less profit than conventional technologies.
you can't play a game, distract yourself or even "think" yourself out of it - rather like expecting a drunk to become sober by performing some task or "think" themselves sober.
Of course, if the game takes long enough, it could be somewhat distracting while the effects of depression work themselves out. The equivalent there would be our drunkard concentrating on some task until they were sober.
Aren't most *NIX users (and especially Perl-geeks) fond of saying that there's more than one way to get the job done? I'd say the only difference here is choosing the right tool for the right job.
It'll be a rocket, right up until they get the payload attached, then it'll be a missile - probably with a name like "qassam" or "al-qud" or something essentially similar.
Shouldn't try to stop 'em - it's their sovereign right to do whatever research they see fit. If they haven't built their own nukes yet or their own ICBM's to carry 'em, they will. We can't punish them for what we fear they will do with these technologies, only hold them fully accountable for what they actually do once they have them.
I wish it were otherwise - I'd love to see what our military could do to yet another smaller, less technically equipped and supported force. I really would. Unfortunately, our much publicized goal of "spreading freedom" runs right into a brick wall when we deal with entire societies which don't see it our way - so . . . do we obey our moral and ethical obligation to permit the sovereign nation of Iran to do whatever research they like and publish whatever news they like (regardless of the facts; the facts have nothing to do with this), OR do we start with the political/economic/military pressure which seems to be in our best interests but conflicts with our basic assertions regarding national sovereignity?
Sorta like evolution - you know it's correct, I know it's correct, but Darwin's Theory of Evolution is a theory - as is the existence of extrasolar life.
Personally, I even believe there is intelligent life out there. Unfortunately, I'm often forced to wonder - what if our current understanding of physics is fundamentally correct and meaningful interstellar contact is and will always remain physically infeasible?
extrasolar planets. Conditions favorable for life, maybe.
Barring a signal that we can recognize as both extrasolar and artificial in origin, I'm afraid we're still going to have to go and look for ourselves if we want a definitive answer to the question of life "out there".
I have over a hundred machines to inventory - the hardware clocks are all set to UTC (as it should be), but the DST information still has to be correct so that the OS will report/record time properly. Also, I have (depending on the host) either one or two JVM's to ensure will be able to handle it (why does Java not simply trust the underlying OS to report time accurately?).
I don't mind checking the OS, but when I have to start looking at apps, that kinda gets my goat a little.
Slashdot Mirror
Besides, when building for a non-Windows PC, one has to exercise extra care to ensure that one gets hardware that will be supported by the OS one intends to install. Unfortunately, Linux just doesn't have nearly as much vendor buy-in for hardware support as MicroSoft Windows. Thus, while many pre-assembled PC's may have hardware suitable for use with a non-MicroSoft OS it's still easier IMHO to just "bare-bones" an appropriate system together. YMMV.
Easily detectable (by their electromagnetic signature), but fast and quiet - no moving parts.
If Greenpeace had been around when T. Edison was doing all his most brilliant work, we'd never have gotten electricity for public use (too dangerous - although Edison's DC system was less dangerous than Westinghouse's AC system, which is in use to this day).
If Greenpeace had been around when Herr Bayer did his thing with acetylsalicylic acid, we'd all still be chewing on tree-bark every time we got a headache. I mean, have you looked at the side effects of that stuff?
Praise the Maker, Greenpeace wasn't around when H. Ford started churning out cheap automobiles. All that wood, all that metal - hell, Ford couldn't even figure out what to do with the char-coal left over from firing his plants; his stepbrother Mr. Kingsford found a market for it. Greenpeace would've gone ape at the thought of encouraging Americans to cook over an artificial fuel source, I'm sure.
No wonder we can't build stuff like the Grand Coulee Dam anymore.
Here.
Getting modded up for feeding the troll was just bonus.
Only requires a moment to stop by the "Software Selection" portion of the initial installation and remove X11 et. al. from the list. I guess that puts me in the top 1.00%, eh?
Or use the System Administration Tool (YaST?) to explicitly set a default runlevel of 3.
Or edit /etc/inittab and change the default runlevel to 3.
Other solutions (and I can think of two right off the top of my head) are left as an exercise for the reader.
You should never take a server to runlevel 5 unless it's been taken out of service for maintenance - and not even then! Just because a GUI may make you able to more quickly or more simply maintain your server doesn't mean that it's okay to run X on a server. GUI's tend to "dumb down" user tasks (that is their function, after all). GUI's have progressed over the last decade, but they still carry their penalties in system load, "dumb-down" factor and increased vulnerability to exploitation.
As for using RHEL as a desktop, I agree wholeheartedly. Everyone knows that Gnome under OpenSuSE 10.2 is the ultimate XGL desktop experience!
Wait a minute . . . that was a Windows bug - Win95, Win98 and (IIRC) NT4.0SP2. Boot yer box and let it do nothing, some kernel pointer associated with timekeeping would overflow at 49.7 days. Hellfire, MicroSoft squashed that one years ago!
So the only question that's left is: are you a Luddite or a Fudite?
You must be thinking of Fedora.
I guess I'm still saddened to hear how the Santa Cruz Operation has ended up; but at least they didn't come to this (even if their name did).
Now? About fifteen years out of date (funny, how SCO's ability to innovate stopped around the same time they started litigating).
But (as you say) the I-17 corridor isn't a hotly contested business opportunity. At the end of the day, companies would rather turn a 5% profit on a $10,000 investment than a 15% profit on a $1,000 investment. Since they can't get that 5% of $10,000 thing going, it isn't worth their attention to try for 15% of $1,000; they have plenty of other 5% on $10,000 opportunities to investigate around populated areas.
Here in the US, if somebody wants something, they'd better damned well be ready to pay for it. Here, "you gets what you pays for!". Now, if Cellulite-1 (or whoever) decides to put a tower up, you can bet it'll be the classic sort which requires power and periodic maintainance to remain operational; expenses which will not only be passed along to the consumer, but which will be passed along at a profit! If the cell towers generate their own electricity, there's that much less expense to make a profit on.
Yes, it seems counterintuitive, but why develop a technology which costs less to implement if that technology is going to cut into your bottom line? Current, technologies with a lower TCO provide less profit than conventional technologies.
<insert your own joke here>
I've still got quite a few of them, sizes ranging from 20MB - 2GB. Still operational (I presume). I wonder if those'd count towards the average?
Of course, if the game takes long enough, it could be somewhat distracting while the effects of depression work themselves out. The equivalent there would be our drunkard concentrating on some task until they were sober.
Aren't most *NIX users (and especially Perl-geeks) fond of saying that there's more than one way to get the job done? I'd say the only difference here is choosing the right tool for the right job.
Shouldn't try to stop 'em - it's their sovereign right to do whatever research they see fit. If they haven't built their own nukes yet or their own ICBM's to carry 'em, they will. We can't punish them for what we fear they will do with these technologies, only hold them fully accountable for what they actually do once they have them.
I wish it were otherwise - I'd love to see what our military could do to yet another smaller, less technically equipped and supported force. I really would. Unfortunately, our much publicized goal of "spreading freedom" runs right into a brick wall when we deal with entire societies which don't see it our way - so . . . do we obey our moral and ethical obligation to permit the sovereign nation of Iran to do whatever research they like and publish whatever news they like (regardless of the facts; the facts have nothing to do with this), OR do we start with the political/economic/military pressure which seems to be in our best interests but conflicts with our basic assertions regarding national sovereignity?
My God, hasn't anybody here seen Planet of the Apes?
Must remember that next time I'm profiled on AMW . . .
Personally, I even believe there is intelligent life out there. Unfortunately, I'm often forced to wonder - what if our current understanding of physics is fundamentally correct and meaningful interstellar contact is and will always remain physically infeasible?
Barring a signal that we can recognize as both extrasolar and artificial in origin, I'm afraid we're still going to have to go and look for ourselves if we want a definitive answer to the question of life "out there".
I don't mind checking the OS, but when I have to start looking at apps, that kinda gets my goat a little.
Also, Lather, Rinse, Repeat.