An entire organization running Linux would seem open to being completely shut out from a Microsoft shop.
Irony in the phrasing aside ("open to being... shut"), this is bad how?
Bear in mind, the converse is more likely to be true. Windows will only run "trusted" (by whom?) apps that are signed with a certificate bought from Microsoft, Linux will happily run anything that the user trusts enough to install. Microsoft users will find themselves shut into a jail of their (or rather, Microsoft's) own choosing.
the multi-volume X Windows Reference series (which is a quite COMPLETE reference for X if you run it bareboned with good old TWM)
Yep, amongst the first tech stuff O'Reilly ever published, as I recall. (Glances over at bookshelf...) The "Rainbow Series" (each volume a different color, no animals) takes up just shy of two feet of shelf space. The stuff on Motif and XView may be a little dated, but the rest is still invaluable if you're doing X Windows coding.
No, nothing like Cowboy Bepop. Well, a tiny bit, maybe. And there are bounty hunters, but the Serenity crew/passengers are the hunted. Somewhat reminiscent (if that's the right term) of, say, Han Solo (and/or Lando Calrissian) before Skywalker, without the Wookiee (or any other aliens), and in the universe where Solo shot first.
No, not really. It does have comedic moments, but it's no Red Dwarf. The humor tends to the ironic.
(Example, from an unaired episode of the series: scene opens with Mal (the captain) sitting alone on a rock in the middle of nowhere, buck naked. He sighs and says "Yep, that went well.")
That's a valid point. I'd consider updates to those (especially libc) as equivalent to kernel updates. You don't have to reboot, but you're probably in a world of hurt if you don't.
We can install basically any linux distribution (which we've typically included the latest updates in already) on basically any hardware in under 15 minutes.
I'm tempted to call BS. I've got this 8-way hyperthreaded Xeon with a bunch of array controllers, NICs, etc that I can barely get to boot in 15 minutes. (Okay, it's a couple of minutes from power up to the boot prompt, and five minutes from the OS loading until a login: prompt).
More seriously, I just did a RH AS 3 kickstart install over gigabit ethernet (a lot faster than the CD) in, yeah, about 15 minutes. And an RH AS 4 install with most of the packages from CD in about 2 hours (unattended except to change discs). So either number works. As for two weeks to install an application -- crikey, even Oracle at its worst (years ago, when you had to rebuild the OS (SunOS, back then) with compiled in shared memory, etc, parameters), on slow (by today's standards) hardware, didn't take more than a day, maybe a day and a half.
Been there, etc. Three new servers, identical in hardware and software. Fine for a month or so while the systems were being set up and configured and the apps installed (not a rush job). Then during pre-production testing one began randomly crashing every few days. Probably an intermittent hardware problem, load related, but running a full night of diagnostics didn't reveal it. Finally somebody happened to be looking at the console when it crashed in the middle of the diagnostics, while it was testing the RAID controller. We replaced that as well as all the memory (just to be sure), and the box has been fine ever since.
Took us a while to figure it out though, and if we hadn't had other identical setups that worked (tending to eliminate a software problem) I hate to think how much longer it might have taken.
If you're rebooting because of patches that often, you're running the wrong OS.
The only reason to reboot a 'nix system is if you're patching the kernel. We've got systems (bastion systems that do email filtering, very stripped down) that have uptimes coming up on two years. Yes, the applications and services have been upgraded and/or patched a few times, but the (linux) kernel hasn't needed it, and it's still going strong. (Heck, we even discovered one crazy process that leaked about a meg of memory a day -- a cron job to restart it every month took care of that until we got an update.)
The solution to that -- and it can certainly happen -- is to have monitoring software running to detect things like running out of memory or swap being more than half (or whatever) full, and alert the sysadmin as appropriate. Email his pager, or something. (On up to elaborate software suites like HP's OpenView that not only monitors the hardware, the OS and the applications (depending on congfig) but will also generate trouble tickets and track them.)
This is why you need sysadmins that not only know the relevant technical details of the platform, but who also understand that keeping a set of critical servers running is a little different from running a home lan and a couple of servers in the basement.
What he said. GFS is similar (in concept, anyway, I haven't looked at GFS's innards) to OCFS (Oracle Cluster Filesystem). It lets multiple servers shared the same "drive". (Typically a high end SAN, but OCFS -- and I'm presuming GFS -- will work with multiple computers plugged into the same single FireWire drive.)
Well, HP already has too -- HPUX. But a lot of HP customers (or potential customers) want an x86-based solution, or at least Linux based. (There are a few people out there running Linux on Itanium.)
Did you try entering "linux drivers" in the search field on the www.hp.com page? The fourth link down in the results is printer drivers, and there are links for various other HP (and ex-Compaq) hardware drivers.
If it's sunny, you don't even need a hand-cranked device. Improvised solar stills -- all you need is a couple of square meters of clear plastic sheet -- can easily yield a liter or more of water a day, even in the desert. With water around, as in New Orleans, more than that is easy.
See here for an improvised desert still, and here for a commerical inflatable one (it floats! handy in NOLA). (The latter site also offers a hand-cranked water maker that will make over a gallon an hour -- but it costs nearly $2K.)
Exactly. We need to cut down more trees and use the wood in ways (such as books) that will keep their carbon out of the atmosphere. Growing replacement trees will help suck up more CO2.
Do your bit to reduce greenhouse gases, cut down a tree! (And plant a new one.)
Most new paper pulp comes from tree farms, and has for decades.
Plague is endemic in the prairie dogs and such in the western US. There are a couple of cases of it in humans in Colorado every year or so. Personally I worry more about hantavirus.
Plague isn't the big deal it was a few centuries ago. Most of us are descended from people who were exposed to plague (considering how widespread the epidemics were) and survived. We're just not as tasty to Yersinia pestis as those who didn't. It has about a 6% mortality rate worldwide these days vs 60% to 100%, depending on the infection mode, during the widespread epidemics of old.
The effects of "any single genetic change" can range from none at all to far more significant than 1%. One example is the change in the haemoglobin molecule that causes (if expressed by two copies of the gene) sickle cell anemia. A single copy confers enough advantage in terms of malaria resistance that the gene survives in the population even though a double copy can be fatal.
One reason for periods of rapid change is if a whole lot of ecological niches suddenly open up, either because their previous occupants died off (eg dinosaurs) or because of geological changes, or the isolation of a few 'seed' species in an isolated set of niches (eg Galapagos Islands).
Of course "rapid" is relative. It took several million years from the extinction of the dinosaurs to produce significant diversity in mammalian species. (Ie, recognizable prototypes of the major mammalian lines.) In fact mammals had been around through the entire era of the dinosaurs, which outcompeted them for most niches.
But now or soon, when almost everyone born reaches maturity (so to speak...) and reproduces,
False premise. True, infant mortality is way down, and -- barring teenage drunk driving, which probably kills more pre-reproduction individuals than anything else (no, I don't have source, it's a wild ass guess) -- most people born do reach maturity, but by no means do they all reproduce. Hell, look at the slashdot crowd;-) Seriously, lots of people don't have kids, for one reason or another. Sometimes it's a concious choice, sometimes it's (possibly environmentally induced) partial sterility. To the extent those reasons are influenced even slightly by genetics (susceptiblity to environmental chemicals, perhaps), there's evolution in action.
for human brains to grow any more, human females would be saddled with hips which would prevent efficient walking.
True (well, C-sections aside), but brains don't necessarily have to get bigger to get better. A number of studies (eg positron emission brain scans of people doing various mental activities) have shown that "more intelligent" individuals tend to have more efficient brains, i.e. they use less energy to solve the given problem. It turns out that indeed for some people, thinking really is hard work.
Slashdot Mirror
Trusted Computing (this has to be one the more ironic names ever,
No, people are just spelling it wrong. It's really Trussed Computing -- you know, like how you lace up a turkey before putting it into the oven.
An entire organization running Linux would seem open to being completely shut out from a Microsoft shop.
... shut"), this is bad how?
Irony in the phrasing aside ("open to being
Bear in mind, the converse is more likely to be true. Windows will only run "trusted" (by whom?) apps that are signed with a certificate bought from Microsoft, Linux will happily run anything that the user trusts enough to install. Microsoft users will find themselves shut into a jail of their (or rather, Microsoft's) own choosing.
the multi-volume X Windows Reference series (which is a quite COMPLETE reference for X if you run it bareboned with good old TWM)
Yep, amongst the first tech stuff O'Reilly ever published, as I recall. (Glances over at bookshelf...) The "Rainbow Series" (each volume a different color, no animals) takes up just shy of two feet of shelf space. The stuff on Motif and XView may be a little dated, but the rest is still invaluable if you're doing X Windows coding.
Ah. Yes, that makes sense. Thanks.
I've personally watched a temperature reading cycle between -40F to -40C.
You realize, of course, that that's not much of a variation. In fact, "none at all" would describe it perfectly. (-40F is -40C).
"There's no indication in the series that she had been turned into the ultimate killing machine".
Huh ? What about that time when she killed several guards inside Niska's station with her eyes closed ?
Or when she tells Jayne "I can kill you with my mind" -- and neither he nor we know if she's kidding.
No, nothing like Cowboy Bepop. Well, a tiny bit, maybe. And there are bounty hunters, but the Serenity crew/passengers are the hunted. Somewhat reminiscent (if that's the right term) of, say, Han Solo (and/or Lando Calrissian) before Skywalker, without the Wookiee (or any other aliens), and in the universe where Solo shot first.
No, it's an intelligent space comedy.
No, not really. It does have comedic moments, but it's no Red Dwarf. The humor tends to the ironic.
(Example, from an unaired episode of the series: scene opens with Mal (the captain) sitting alone on a rock in the middle of nowhere, buck naked. He sighs and says "Yep, that went well.")
From the blurb:
That's a valid point. I'd consider updates to those (especially libc) as equivalent to kernel updates. You don't have to reboot, but you're probably in a world of hurt if you don't.
We can install basically any linux distribution (which we've typically included the latest updates in already) on basically any hardware in under 15 minutes.
I'm tempted to call BS. I've got this 8-way hyperthreaded Xeon with a bunch of array controllers, NICs, etc that I can barely get to boot in 15 minutes. (Okay, it's a couple of minutes from power up to the boot prompt, and five minutes from the OS loading until a login: prompt).
More seriously, I just did a RH AS 3 kickstart install over gigabit ethernet (a lot faster than the CD) in, yeah, about 15 minutes. And an RH AS 4 install with most of the packages from CD in about 2 hours (unattended except to change discs). So either number works. As for two weeks to install an application -- crikey, even Oracle at its worst (years ago, when you had to rebuild the OS (SunOS, back then) with compiled in shared memory, etc, parameters), on slow (by today's standards) hardware, didn't take more than a day, maybe a day and a half.
Been there, etc. Three new servers, identical in hardware and software. Fine for a month or so while the systems were being set up and configured and the apps installed (not a rush job). Then during pre-production testing one began randomly crashing every few days. Probably an intermittent hardware problem, load related, but running a full night of diagnostics didn't reveal it. Finally somebody happened to be looking at the console when it crashed in the middle of the diagnostics, while it was testing the RAID controller. We replaced that as well as all the memory (just to be sure), and the box has been fine ever since.
Took us a while to figure it out though, and if we hadn't had other identical setups that worked (tending to eliminate a software problem) I hate to think how much longer it might have taken.
If you're rebooting because of patches that often, you're running the wrong OS.
The only reason to reboot a 'nix system is if you're patching the kernel. We've got systems (bastion systems that do email filtering, very stripped down) that have uptimes coming up on two years. Yes, the applications and services have been upgraded and/or patched a few times, but the (linux) kernel hasn't needed it, and it's still going strong. (Heck, we even discovered one crazy process that leaked about a meg of memory a day -- a cron job to restart it every month took care of that until we got an update.)
The solution to that -- and it can certainly happen -- is to have monitoring software running to detect things like running out of memory or swap being more than half (or whatever) full, and alert the sysadmin as appropriate. Email his pager, or something. (On up to elaborate software suites like HP's OpenView that not only monitors the hardware, the OS and the applications (depending on congfig) but will also generate trouble tickets and track them.)
This is why you need sysadmins that not only know the relevant technical details of the platform, but who also understand that keeping a set of critical servers running is a little different from running a home lan and a couple of servers in the basement.
What he said. GFS is similar (in concept, anyway, I haven't looked at GFS's innards) to OCFS (Oracle Cluster Filesystem). It lets multiple servers shared the same "drive". (Typically a high end SAN, but OCFS -- and I'm presuming GFS -- will work with multiple computers plugged into the same single FireWire drive.)
Well, HP already has too -- HPUX. But a lot of HP customers (or potential customers) want an x86-based solution, or at least Linux based. (There are a few people out there running Linux on Itanium.)
Did you try entering "linux drivers" in the search field on the www.hp.com page? The fourth link down in the results is printer drivers, and there are links for various other HP (and ex-Compaq) hardware drivers.
If it's sunny, you don't even need a hand-cranked device. Improvised solar stills -- all you need is a couple of square meters of clear plastic sheet -- can easily yield a liter or more of water a day, even in the desert. With water around, as in New Orleans, more than that is easy.
See here for an improvised desert still, and here for a commerical inflatable one (it floats! handy in NOLA). (The latter site also offers a hand-cranked water maker that will make over a gallon an hour -- but it costs nearly $2K.)
Exactly. We need to cut down more trees and use the wood in ways (such as books) that will keep their carbon out of the atmosphere. Growing replacement trees will help suck up more CO2.
Do your bit to reduce greenhouse gases, cut down a tree! (And plant a new one.)
Most new paper pulp comes from tree farms, and has for decades.
Plague is endemic in the prairie dogs and such in the western US. There are a couple of cases of it in humans in Colorado every year or so. Personally I worry more about hantavirus.
Plague isn't the big deal it was a few centuries ago. Most of us are descended from people who were exposed to plague (considering how widespread the epidemics were) and survived. We're just not as tasty to Yersinia pestis as those who didn't. It has about a 6% mortality rate worldwide these days vs 60% to 100%, depending on the infection mode, during the widespread epidemics of old.
Hey, sounds like a great plot for a movie.
(Okay, so it wasn't New Orleans. But the town was rendered uninhabitable.)
No, that's a use for a dead cat. Which brings the count up to 102.
The effects of "any single genetic change" can range from none at all to far more significant than 1%. One example is the change in the haemoglobin molecule that causes (if expressed by two copies of the gene) sickle cell anemia. A single copy confers enough advantage in terms of malaria resistance that the gene survives in the population even though a double copy can be fatal.
One reason for periods of rapid change is if a whole lot of ecological niches suddenly open up, either because their previous occupants died off (eg dinosaurs) or because of geological changes, or the isolation of a few 'seed' species in an isolated set of niches (eg Galapagos Islands).
Of course "rapid" is relative. It took several million years from the extinction of the dinosaurs to produce significant diversity in mammalian species. (Ie, recognizable prototypes of the major mammalian lines.) In fact mammals had been around through the entire era of the dinosaurs, which outcompeted them for most niches.
But now or soon, when almost everyone born reaches maturity (so to speak ...) and reproduces,
;-) Seriously, lots of people don't have kids, for one reason or another. Sometimes it's a concious choice, sometimes it's (possibly environmentally induced) partial sterility. To the extent those reasons are influenced even slightly by genetics (susceptiblity to environmental chemicals, perhaps), there's evolution in action.
False premise. True, infant mortality is way down, and -- barring teenage drunk driving, which probably kills more pre-reproduction individuals than anything else (no, I don't have source, it's a wild ass guess) -- most people born do reach maturity, but by no means do they all reproduce. Hell, look at the slashdot crowd
for human brains to grow any more, human females would be saddled with hips which would prevent efficient walking.
True (well, C-sections aside), but brains don't necessarily have to get bigger to get better. A number of studies (eg positron emission brain scans of people doing various mental activities) have shown that "more intelligent" individuals tend to have more efficient brains, i.e. they use less energy to solve the given problem. It turns out that indeed for some people, thinking really is hard work.