RHS 3 is a pretty solid server IMHO, after using it for a few months on a web server and finding it far superior and simpler to manage than the Solaris box the company has its other website on.
The funny part is that if you're running on a small machine and you can live without phone support, a Solaris right-to-use license is actually cheaper than a RHEL "support agreement," also known as "Redhat's extortion scheme since if the updates and download were freely available nobody would buy the fucking thing".
A lot of applications that people are buying for these servers are certified torun on RHEL and sometimes Suse's enterprise linux. Things like oracle may not run on debian.
More importantly, Oracle simply won't support you if you run on anything other than an "enterprise" Linux.
Exactly, Red Hat is marketing Red Hat Enterprise Linux to the enterprise
SuSE is a fine choice for the small and medium business customers.
While both those statements are true, it's missing the point. SuSE is good for a lot more than "small and medium" businesses. Keep in mind, SuSE maintains a close relationship with IBM and has been first with ports for IBM mainframes, minis and the Power architecture, which are not exactly mom-and-pop systems.
The price is too high, that is why some of us have been using White Box Linux for some time.
Or CentOS. Or Taolinux. Or Scientific Linux/Fermi Linux. Or probably some others I'm not remembering...
My favorite is Scientific Linux. That's primarily because it seems the least likely to be suddenly unsupported one day, as it is maintained by a national lab. (not that this is a big deal, obviously, with all RHEL clones being so closely compatible) They also have a nice "contrib" section and fast servers.
I know you're trolling, and sorry to disappoint you, but I work in the central IT organization here, and I haven't exaggerated anything. This are tremendously, ridiculously more Mac OS X installations than Linux or any other *nix here, and this is the same at other US-based large academic research institutions I regularly deal with (and I'm NOT just talking about public student computing labs). And the vast, vast majority of these people don't use them for graphic design.
Maybe your organization is like every other organization in the world, and Linux users end up not bothering to tell the pinheaded IT guys that they are going to remove the NT/OS X/whatever that came with their machine before they install Linux.
Or maybe there really are "tremendously, ridiculously, wankingly adjectiveoverusingly" more OS X installations than *nix.
Well, the question is: Is the PIII really that much different from the PII? Don't they have basically identical everything?
Sure, they probably run hotter, but you could always underclock them.
What you couldn't (necessarily) do would be to update the microcode (typically in the BIOS) of whatever devices you're looking to replace CPUs in. There were lots of PII motherboards that ended up supporting PIII CPUs, since they used voltage regulators that could adjust accordingly. But they invariably required BIOS updates (if they were bought early in their life cycle) so the system could load the microcode appropriate to the CPU. Many embedded devices won't have the facility for that kind of upgrade, I would think.
You could sort of limp along without the appropriate microcode in some cases but hopefully you see the difficulty. You couldn't do it with a lot of confidence, and any devices that are still using a PII are either throwaways or devices that someone wants to have a lot of confidence in.
I agree that the optimization thing is kind of silly, but your critique is wrong on a number of points.
Considering much of what a mail client does is either disk or display, and not very repetitive, processor-specific optimizations will do little to no good.
Insofar as a lot of "display" stuff (HTML rendering) is CPU-intensive, and a lot of people are using HTML in their email, these optimizations are likely to be equally useful here as in the web browser. (not especially useful IMO, but still)
Even search functions are largely disk constrained if the mailbox is big enough that search time becomes an issue on any modern system.
Wrong, sort of. Even if a user's mailbox is big enough to blow the disk cache, which is fairly unlikely in these days of machines with 512MB of RAM or more, searching is quite CPU intensive. I imagine there's some goofing around you could do with grep to convince yourself of the truth of this. Whether these optimizations will help with that or not, I'm pretty doubtful.
In fact I'd sort of like to understand why an "optimized" web browser should be an faster at all when rendering pages. It's not like special features of the processor like SSE3 or whatever are important to the rendering function, so... why? But I believe the tests when they indicate speed differences. Something kooky about the compiler?
It seems to me that the problem is far more how the rest of the world behaves to the US
Ha. I'm not sure that's a meaningful distinction, but okay.
But I agree with the parent - if you want to discuss NASA missions, great. If you want to whine incessantly about George Bush, go to MoveOn or Democratic Underground or some other such site.
Who cares? It seems to me the problem isn't what people are talking about, which is their own business, it's what the moderators are willing to moderate up to 5. Right now dissing Bush is as easy a score boost as dissing Bill Gates.
Others will go into great detail about how these are neither "standard" Unix cores but some variant of some variant of some variant of some microkernel but nobody cares anymore, that is really so early 90's
Tru64 is every bit as much a Microkernel OS as GNU/Hurd or OS X, they all use Mach. Hope pointing that out isn't too "early 90s".:)
And most commercial applications have been somehow hoodwinked into the proprietary hooks.
Why do you say "somehow"? How many alternatives would the early adopters of Tru64's clustering features have had?
Now instead of discussing the technical aspects OF THE NASA MISSIONS, we discuss politics, like I cant go discuss politics somewhere else..I need a new nerd haven, plz slashdot, kill the politics or it will kill you.
I imagine this is just a taste of what it's like in any community that spans national boundaries nowadays. Not talking politics is like ignoring an elephant in the room.
It sucks, but with the way the US behaves toward the rest of the world nowadays, it ought to be expected.
You'd think that by now, more people would've figured out a basic trait of Steve Jobs: he's got his own will, and he'll hit eject before he compromises it.
You can do *much* better error checking with a turing-complete programming language than you can with SQL. That's what we do with our applications even though we use Oracle. The biggest problem is that you must have good programmers.
Which you obviously don't...
You're right that pre-processing is a good way to do constraints.
Sure, this sends the spammers the message that we don't like what they're doing... but one has to assume that they already know that. I guess I don't see what practical purpose this is serving.
I'm not sure precisely what is meant by "spammer's websites." If it refers to those sites the really gullible one-in-ten-thousand clicks on to get their penis enlarger or ponzi scheme or whatever, then you're arguably helping that gullible person.
Or keeping them from getting a fun new toy. Not that I am saying that it would be fun. That's not my bag, baby.
That a bunch of general purpose commodity hardware is going to replace their highly engineered, specialized hardware.
Yes, and SGI probably never thought that PC hardware would drive them out of business either.
PC hardware didn't damage their business - NT and Linux did. It's a lousy analogy anyhow...
Cisco is a hardware company. If XORP is great and their customers want it, Cisco can provide it, and make just as much money as they are making now. (more, maybe) There's probably no reason why Cisco can't put XORP on their next generation of Linksys stuff for starters, if there is any interest at all in XORP and it's ready.
If Cisco had a business selling shrink-wrapped bundles of IOS this might be a worry for them.
My understanding is that a school which calls itself, for instance, a Catholic College, can't receive state or federal aid. My alma mater, however, is a "College in the Dominican Tradition", which fits through the loophole.
I guess it depends on what you mean by "aid," AFAIK any Catholic university would be eligable to accept grants from the NSF, for example, or to get the tax breaks other schools get. That's pretty different than the state paying to create a college, like the land-grant schools. So I guess there's some ambiguity there.
I don't know anything about the distinction you're drawing above with school naming, although I know we have plenty of hospitals with "Catholic" and "Methodist" in their names, so I don't know what legal principle might be involved. Maybe it's a state thing.
I've observed that, on the elementary-school level, people are often surprised when they find out how much money their parochial schools get from the state and county.
Considering you have to give Valve all the rights to any map you create for HLDM, you're basically being paid to create a HL DM map for them.
Incidentally, mappers give up the rights to their maps when they use Unrealed to make maps for ut2k4 (or whatever), unless the Epic EULA has changed recently. I don't think this is true of Radiant or whatever the editor for d3 is called, but I don't really know for sure.
Among competitive gamers there is a shorthand for these people, they're often called "cheaters." In the estimation of a large part of the user community, banning these people is just fine, and having a looming threat of these people losing their $50 or whatever is also just fine.
I agree with most of the remainder of what you said. I don't think fears of Steam suddenly changing to a subscription model and pulling existing HL2 users into this model are very well-founded, though. Even if they were completely amoral and contemptuous of their user community, (and their strong support for modders indicates this is not the case) it wouldn't be a very smart business move for Valve.
Except that a large number of universities receive state and federal money.
There is, in fact, only one college in the US that receives no federal money. I can't remember its name but it is the subject of a certain amount of wankery on right-wing talk radio sometimes.
I don't know exactly why colleges all get federal money. If they do any research they'll probably want money from the NSF. Pure teaching schools are typically community colleges or state colleges, so there's that.
In a different subject, I think that the problem with embryonic stem cell research is its potential to undermine human dignity. What would the world look like if we knowingly bred people just to harvest their organs/cells/meat(?) out?
Not any worse than a world in which people hypocritically lust for war and executions while affecting moral outrage over the killing of fetuses. You're worried about smacking "human dignity" upside the head, meanwhile we are slicing it into bloody little bits.
And no, I don't believe in breeding "people" to harvest organs, I'm just saying...
Slashdot Mirror
The funny part is that if you're running on a small machine and you can live without phone support, a Solaris right-to-use license is actually cheaper than a RHEL "support agreement," also known as "Redhat's extortion scheme since if the updates and download were freely available nobody would buy the fucking thing".
More importantly, Oracle simply won't support you if you run on anything other than an "enterprise" Linux.
Which would be ironic, since IBM also pushes SuSE harder than Redhat. Though both companies will sell you either OS, apparently.
Why? If there's one company capable of managing stupid-customer expectations it's Dell.
Because you're pulling those figures out of your ass.
While both those statements are true, it's missing the point. SuSE is good for a lot more than "small and medium" businesses. Keep in mind, SuSE maintains a close relationship with IBM and has been first with ports for IBM mainframes, minis and the Power architecture, which are not exactly mom-and-pop systems.
Or CentOS. Or Taolinux. Or Scientific Linux/Fermi Linux. Or probably some others I'm not remembering...
My favorite is Scientific Linux. That's primarily because it seems the least likely to be suddenly unsupported one day, as it is maintained by a national lab. (not that this is a big deal, obviously, with all RHEL clones being so closely compatible) They also have a nice "contrib" section and fast servers.
https://www.scientificlinux.org/
Maybe your organization is like every other organization in the world, and Linux users end up not bothering to tell the pinheaded IT guys that they are going to remove the NT/OS X/whatever that came with their machine before they install Linux.
Or maybe there really are "tremendously, ridiculously, wankingly adjectiveoverusingly" more OS X installations than *nix.
Unnecessary Capitalization makes you look like a Big Dork.
What you couldn't (necessarily) do would be to update the microcode (typically in the BIOS) of whatever devices you're looking to replace CPUs in. There were lots of PII motherboards that ended up supporting PIII CPUs, since they used voltage regulators that could adjust accordingly. But they invariably required BIOS updates (if they were bought early in their life cycle) so the system could load the microcode appropriate to the CPU. Many embedded devices won't have the facility for that kind of upgrade, I would think.
You could sort of limp along without the appropriate microcode in some cases but hopefully you see the difficulty. You couldn't do it with a lot of confidence, and any devices that are still using a PII are either throwaways or devices that someone wants to have a lot of confidence in.
Insofar as a lot of "display" stuff (HTML rendering) is CPU-intensive, and a lot of people are using HTML in their email, these optimizations are likely to be equally useful here as in the web browser. (not especially useful IMO, but still)
Wrong, sort of. Even if a user's mailbox is big enough to blow the disk cache, which is fairly unlikely in these days of machines with 512MB of RAM or more, searching is quite CPU intensive. I imagine there's some goofing around you could do with grep to convince yourself of the truth of this. Whether these optimizations will help with that or not, I'm pretty doubtful.
In fact I'd sort of like to understand why an "optimized" web browser should be an faster at all when rendering pages. It's not like special features of the processor like SSE3 or whatever are important to the rendering function, so... why? But I believe the tests when they indicate speed differences. Something kooky about the compiler?
Ha. I'm not sure that's a meaningful distinction, but okay.
Who cares? It seems to me the problem isn't what people are talking about, which is their own business, it's what the moderators are willing to moderate up to 5. Right now dissing Bush is as easy a score boost as dissing Bill Gates.
Tru64 is every bit as much a Microkernel OS as GNU/Hurd or OS X, they all use Mach. Hope pointing that out isn't too "early 90s".
Why do you say "somehow"? How many alternatives would the early adopters of Tru64's clustering features have had?
I imagine this is just a taste of what it's like in any community that spans national boundaries nowadays. Not talking politics is like ignoring an elephant in the room.
It sucks, but with the way the US behaves toward the rest of the world nowadays, it ought to be expected.
Baloney. They had more than their fair share of production problems back when their CPUs were coming from Motorolla.
Don't you mean, he'll drag it into the trash can?
Which you obviously don't...
If your database has one user.
I'm not sure precisely what is meant by "spammer's websites." If it refers to those sites the really gullible one-in-ten-thousand clicks on to get their penis enlarger or ponzi scheme or whatever, then you're arguably helping that gullible person.
Or keeping them from getting a fun new toy. Not that I am saying that it would be fun. That's not my bag, baby.
PC hardware didn't damage their business - NT and Linux did. It's a lousy analogy anyhow...
Cisco is a hardware company. If XORP is great and their customers want it, Cisco can provide it, and make just as much money as they are making now. (more, maybe) There's probably no reason why Cisco can't put XORP on their next generation of Linksys stuff for starters, if there is any interest at all in XORP and it's ready.
If Cisco had a business selling shrink-wrapped bundles of IOS this might be a worry for them.
I guess it depends on what you mean by "aid," AFAIK any Catholic university would be eligable to accept grants from the NSF, for example, or to get the tax breaks other schools get. That's pretty different than the state paying to create a college, like the land-grant schools. So I guess there's some ambiguity there.
I don't know anything about the distinction you're drawing above with school naming, although I know we have plenty of hospitals with "Catholic" and "Methodist" in their names, so I don't know what legal principle might be involved. Maybe it's a state thing.
I've observed that, on the elementary-school level, people are often surprised when they find out how much money their parochial schools get from the state and county.
Incidentally, mappers give up the rights to their maps when they use Unrealed to make maps for ut2k4 (or whatever), unless the Epic EULA has changed recently. I don't think this is true of Radiant or whatever the editor for d3 is called, but I don't really know for sure.
Only if he's changed his name from "zoid" to "zonk."
Among competitive gamers there is a shorthand for these people, they're often called "cheaters." In the estimation of a large part of the user community, banning these people is just fine, and having a looming threat of these people losing their $50 or whatever is also just fine.
I agree with most of the remainder of what you said. I don't think fears of Steam suddenly changing to a subscription model and pulling existing HL2 users into this model are very well-founded, though. Even if they were completely amoral and contemptuous of their user community, (and their strong support for modders indicates this is not the case) it wouldn't be a very smart business move for Valve.
There is, in fact, only one college in the US that receives no federal money. I can't remember its name but it is the subject of a certain amount of wankery on right-wing talk radio sometimes.
I don't know exactly why colleges all get federal money. If they do any research they'll probably want money from the NSF. Pure teaching schools are typically community colleges or state colleges, so there's that.
And as their assfucked support of Kylix proves, your statement is particularly true if that one platform happens to be Linux.
Not any worse than a world in which people hypocritically lust for war and executions while affecting moral outrage over the killing of fetuses. You're worried about smacking "human dignity" upside the head, meanwhile we are slicing it into bloody little bits.
And no, I don't believe in breeding "people" to harvest organs, I'm just saying...