I've heated fish tanks larger than that, with a small heater (you know the little fish tank kind).
So we all have.
However,
This site suggests a 75 watt heater to keep a 20 gallon tank 18 degrees F above room temperature, or a 150 watt heater for 36 degrees F.
The computer probably puts out less than 150 watts total. Even assuming an 80 degree F room, that would put the computer at 116 degrees F, which wouldn't upset the computer at all. Granted, the heater you put in a fish tank has a thermostat, and so it's not on all the time, but your computer will not have any problems at 116 degrees F inside, and could go a good deal higher safely.
But I do agree with the other guy to respond to your post -- I don't see the fans even turning, let alone turning enough to move the oil around. Perhaps if they were cut down some...
Of course, I have no idea how well heat flows through oil, or how well it's transferred from oil to the air. But I imagine that the heat generated is low enough for it to not be a problem.
Of course, looking at Chungles (awful name, by the way)... it doesn't look like anything more than a GUI wrapped around a ftp client and server.
(Well, I assume it doesn't use the ftp protocol, but it serves the same purpose -- it copies files from host A to host B. rsync, scp, rcp and sftp all do the same thing.)
If so, then it's not even really giving the same functionality as SMB, NFS and others, and they're really doing Samba a disservice by comparing themselves to it.
Chungles appears to just be a GUI for copying files from host to host, but you can't use the files on a remote host as if they were local, like SMB and NFS and others let you do. The distinction is subtle, but important.
Sharing files amongst different platforms has most always resulted in using samba
It has? Since when?
SMB is only one of many file sharing protocols. Samba is generally only used when you want a Windows machine to use filesystems on a *nix fileserver. That's it.
But Windows is hardly the whole world. There's also NFS (pretty much the standard in the *nix world), Appleshare (what Macs use. To mount *nix filesystems on a Mac, you generally run
netatalk on the *nix box, though OSX supports NFS natively too.)
Actually, you can get NFS clients and servers for Windows from several places, and they generally work pretty well once one realizes how NFS differs from SMB.
To mount a Windows file system on a *nix box, you can use smbfs if it's a Linux box -- and note that smbfs is not the same as Samba. Another option is Sharity (previously Rumba) which takes SMB and converts it to NFS.
And in addition to SMB, Appleshare and NFS, there's also Netware, Coda, InterMezzo, DFS, AFS, and many others, and nowadays they're all pretty much multi-platform.
This page gives a nice summary of some of the more popular network file sharing methods.
Of course, network file sharing is not the only way. Sneakernet still works, for example. People have been taking floppy disks out of one computer type and putting them into another computer type for as long as computers have had floppy drives, though it probably didn't usually work until both Macs and PCs had 3.5" drives and the Macs learned to talk FAT. And today, USB and firewire drives, if formatted FAT, generally work in Mac, Windows and *nix systems and probably others, and are often used to share files...
Seriously though, there's a huge difference. Do you really think that working in a concentration camp is remotely close to slaving away in a cubical trying to find out why Microsoft Excel crashes when somebody clicks on a specific cell, but only on SMP P4 boxes?
I certainly do believe that most of the German soldiers in WWI and WWII were just average guys, trying to make it through the war in one piece. But war is a bit different than computer programming -- soldiers generally do things like point guns at people and kill them, which is generally considered to be `evil' unless there's a very good reason for it. I suspect that many (most?) Microsoft employees, on the other hand, are developers and support personel and such, and so their job is to actually make their products better (fewer bugs, more features, etc.), which would generally be considered `good'.
(Though I guess that one could consider improving software to be `bad' if the software was bad. Malware or spamware might be considered bad, as might software used to keep track of prisoners in your concentration camp.)
Overall, Microsoft may be the `enemy', but the individual employees certainly aren't.
So you consider Vinod Valloppillil and Josh Cohen just average working people like those working at any other software company?
Thank you for reminding me that I forgot the phrase most of in the sentence of mine that you quoted. Obviously Microsoft has a few `bad' (or at least somewhat unethical, dishonest or something) employees -- most every company does.
He certainly has a lot of the microsoft marketdroid-speak.
And that sort of marketdroid-speak is hardly specific to Microsoft.
Bullshit! - Microsoft on more than 1 occasion had to hire people through other companies because they couldnt find true Linux experts that would work for them.
Ok, fine -- perhaps Microsoft couldn't get Linus or Alan Cox to work for them, at any price. Maybe, maybe not.
But most of the `true Linux experts' out there still have mouths to feed and like their sports cars and such, and if Microsoft were to offer them enough money, at least some of them would work for Microsoft. I would if they offered enough money, though I'm somewhere between L++ and L+++ in my geek code block so I probably don't count.
And I can easily see how a `Linux expert' might even justify working for Microsoft -- he might see it as a way to infiltrate Microsoft from within, perhaps influence their corporate culture, help make them a better company. (Yes, I think this is unlikely, but he might justify it like that anyways.) And while I'm certainly a OSS advocate, I still think that any improvements in Microsoft software will benefit the computing community as a whole, and if a Linux expert can help improve their software, I think this would be a good thing for him to do.
The individuals who work at Microsoft are either ignorant, desperate or believe there should be a fascist monopoly in the operating system market.
There may be a few people there like that, but I certainly do believe that most people there are just normal, every day people who do their job so they will get paid. And if they can do something to improve the world, that's a nice bonus, but mostly they're just there to get paid, just like the rest of the world.
And I don't think that Microsoft is so much more evil than any other software company, it's just that they are in a position where their actions have a lot more effect on things, so people watch them more closely.
The good average guys are thrown out for mentioning Microsoft uses Apple Mac.
Are you referring
to this incident? If so, that's a long stretch from what you said. Yes, Microsoft's reaction was extreme, but it's not quite what you suggested. He certainly did more than mention that Microsoft uses Macs.
It's nice to know that at least somebody there has some understanding of open source/Linux/alternatives.
Microsoft is a large company. I'll bet there's hundreds of employees there that have a good understanding of open source alternatives. There's probably even some employees who regularly contribute to some open source projects, unless Microsoft policy actively prohibits it.
The marketing stuff that you see from them is written by a small subset of the company, and it's generally written with one goal in mind -- to benefit Microsoft. They aren't worried about giving the alternatives a fair treatment, unless they think that that will benefit them somehow.
Overall, Microsoft may be the `enemy', but the individual employees certainly aren't. They're just average working people like those working at any other software company.
Right, just as cheap IDE drives aren't designed to serve the Enterprise.
Cheap IDE drives serve the `Enterprise' (I assume we're talking about a business, not a naval ship, space shuttle or star ship, even though you've capitialized it like it's a proper name) far better than cheap ink jets do. When you work out the per-page cost, low-end ink jets are *expensive*, mostly due to the ink which costs thousands of dollars per ounce when you do the math.
Even the cheap modern IDE drives are pretty good. SCSI (and maybe fiber channel, though I have no experience with FC) is still king, but even the consumer grade IDE drives can generally handle a good database pounding.
I'd go all SCSI, but IDE is almost an order of magnitude cheaper now, so in many cases it makes good sense to just do RAID-1 with IDE, even in the `enterprise'.
Libel does not have to be `accusing somebody of doing something that's illegal' -- all it has to do is damage somebody's reputation. (It (the claim made) also has to be false.)
If I accuse you of having lots of one night stands with random women, that might damage your reputation, especially if you're a priest or married. But having that sex is not against the law, but my claim may still be libel -- as long as it's not true.
Truth is an absolute defense against libel and slander claims, though our legal system can make it very expensive to even defend yourself against a claim of libel, even if you have that proof.
Its more or less required that you can have an engine fail and still be able to fly and make it to land safely.
That's the goal, but it depends on where and how the engine fails, and the skill of the pilot.
For example, there was a case some time ago where the propeller came apart in a commercial plane (this was not obvious at the time) and the engine vibrated itself to bits in a few seconds (THAT was obvious, once the pilots actually looked at the engine. Before that, they'd thought they'd just lost the engine in a more `normal' way.) The plane had enough thrust to maintain altitude after losing an engine under normal conditions, but in this case the destroyed engine had much more drag than a normal non-functional engine, and the plane could not maintain altitude, and eventually crashed.
The crash was bad, but controlled. Nobody was killed by the crash, but the plane was basically destroyed. Unfortunately, destroyed planes tend to leak fuel, and this one did, which promptly ignited, and ultimately about half the people aboard died due to the fire.
This was all described in a show about aviation disasters on the Discovery Channel. Maybe somebody else can provide more details.
Or, if a two-engine plane were to lose an engine in the middle of the ocean, that would increase drag as the plane would slip due to mismatched thrust, and might not have enough fuel to make it to land. Hopefully they plan for this, and provide enough extra fuel for this kind of emergency.
But yes -- pretty much most 2+ engine airplanes can maintain altitude and even climb slightly if they lose one engine in a `normal' way. (Exploding/disintegrating engines don't really count. Fortunately they're very rare.)
However, when you talk about private planes, the fatal accident per flying hour ratio is signifigantly actually higher for two engine planes than single engine planes. This is because a two engine plane is quite difficult to fly with an engine out, and this tends to cause fatal accidents. In a single engine plane, when you lose your engine, all you can do is look for a good place to land (or crash, if you can't find a good place.) But even if you crash, you're likely to be in control of the plane and while the plane is likely to be destroyed, you're likely to survive.
But in a dual engine plane, what often happens is that the sudden yaw as the engine is lost causes the plane to turn into a lawn dart, killing all aboard. Yes, a good and alert pilot can prevent this from happening, but mistakes are often made.
... though I did forget to comment on that -- while morse code conversations are abbreviated as often as SMS ones are, he did not abbreviate.
To be fair, he probably used a good quality morse code key or paddle and not the tiny keyboard on a phone like the girl did. It makes a big difference -- you can send morse code with pretty much anything, even something like a flash light, but a good key makes a huge difference.
Morse code is full of shorthand slang too. It just doesn't quite look so illiterate to the untrained eye as SMS shorthand, and you're not likely to see it turned in on homework, unlike SMS slang.
If you get slashdotted, can't you just coralize your own site for a while?
Coral looks like an obvious solution to this sort of problem -- and to some degree it is. However, there are problems too --
1) If/. has linked to your site, that means your site still needs to serve up the main page. You could coralize your images and such and save some bandwidth that way, but if your web server can't even serve that first page under the load, you're screwed. And if you do find yourself/.ed, and go and coralize your site real quick, then it'll be a while before the traffic slacks off enough for the coral servers to even reach your site to get the images that you've coralized.
Many sites do replace their fancy dynamic pages with a `hi slashdotters!' page after getting/.ed... saves a lot of cpu on the box. But if what's special about your site is the dynamic aspect of it, well, that won't work.
2) Coral won't do files over 50 or 100 MB. So if you've got some large download, you'd better set up a Bit Torrent instead... and fast.
3) Currently, Coral uses some non-standard ports that some places may not be able to access due to restrictive firewalls. I understand that this is to change.
4) Coral uses some DNS tricks that don't work with the entire world. Specially, Windows DNS servers tend to have problems with it.
But still, mentioning coral as a way of reducing the/. effect is an excellent idea. It's not the perfect solution, but it's pretty good.
But how does slashdot itself cope with the high traffic?
Lots of bandwidth, lots of hardware. Since it gets `slashdotted' every single day, it'll be pretty easy to predict how much traffic you'll get tomorrow -- approximately the same as you got yesterday, perhaps a bit more.
But when you're running your own server, and it normally gets 50 hits/day, and then suddenly a Slashdot listing hits it with millions of hits in one day, well, that's harder to prepare for, because 1) you often don't know you're going to be on/. until it's already happened, and 2) is it even worth preparing for? It's just one or two days, and then things will go back to normal. More hardware and bandwidth may cost lots of money, money that you're not going to spend just so people can see pictures of whatever neat thing you did.
Really, the only sites that get/.ed are the smaller ones. The larger ones already have the hardware and bandwidth needed to handle it. Sure, a/.ing probably shows up on their mrtg reports, but it's probably just a 20% or so increase in traffic, not a 1000x fold increase.
are abandoned after the shopper goes to netcraft and discovers that he's being asked to put his credit card number into an NT/IIis system
Probably a very, very small percentage. Why?
1) few people look that up at all (and why use Netcraft? nmap works fine.)
2) A poorly configured *nix system or application is a lot more insecure than a similar well configured Windows system. Merely being *nix does not mean `secure', and merely being `Windows' does not mean `insecure'. There's far more important variables, variables that Netcraft does not track.
3) the risk of exposure is pretty small. If I see charges on my card I didn't make, I report them to my card company, and I don't pay. The biggest problem is that I have to wait a week or so for them to send me a new card. As long as my credit card is physically in my possession, I have zero liability for fraudulent use of it. (If I lose it and report it in a reasonable amount of time, the liability limit is only $50.)
they tend to insist "there's no risk; this is a secure system."
That's pretty common. People see that little lock at the bottom of the screen (the one that indicates SSL is in use) and think that means it's secure!. You know better, I know better, but most people do not.
I really hate it when a site tries to make me jump through all the hoops of buying a product before telling me the price of the product.
Agreed, though it's a minor inconvenience, so I don't really care too much.
What really annoys me is sites that make it impossible to determine the shipping costs. Many require that you go almost all of the way through the ordering process, often even going past the point where I enter my credit card, before I'm given any clue what the shipping will be.
Here's a free clue for you, online retailers -- when I run into a site that doesn't tell me what shipping will be without making an order, I usually go somewhere else. If you won't tell me the shipping cost up front, my reasoning is that it probably sucks anyways. In reality, it may not, but it's not worth my time, and I'm certainly not going to enter my credit card number before I know how much I'm going to spend.
As for complaining that people abandon shopping carts, well, we do that because we don't think of them as shopping carts. Because they're not. It's a list of items we might want to buy, stuff we're interested in. Not a list of things we are going to buy, at least not until we start checking out. And really, if a real brick and mortar store did some of the annoying tricks that online stores do to `trap' me into buying from them, I'd abandon my real shopping cart there too, though in that case somebody would have to put the stuff back so I might feel a tad guilty about it.
Probably the best thing that an online retailer can do to encourage people to not forget about what they had in their `shopping cart' before is to make sure it persists. If we come back tomorrow or two weeks or two months later, remind us that we'd left some stuff in our `shopping cart'. Since we're not impulse buyers, if we really want something, we'll probably come back later. Don't make us find the stuff again.
4) Learn that screwing somebody you just met in a bar just might have a negative effect on the rest of your life
As might walking across the street. Or starting your car. Or volunteering at the local hospital, getting a job, raking the grass. Life is dangerous.
Sex, however, is generally believed to be normal, even if our society tends to demonize it. For example, the odds are pretty good that your parents have had it at least once.
(forget the fact more than 1/4 have genital warts and that it is not prevented by condoms
Nothing is absolutely prevented by condoms, not HIV/AIDS or pregnancy. However, they are still believed to be at least somewhat effective,
even against genital warts. They're not perfect, but they're far better than nothng.
there is no cure, and it can cause a woman to be infertile).
So you've heard of STDs. Good. But what does that have to do with screwing a woman you just met in a bar? If I recall correctly, the first time I met my wife was in a bar. That was perhaps 12 years and two kids ago...
You could meet a woman in church (or pick some other place for finding wholesome, God fearing women), get to know her for a few months, fall in love, and finally have sex and then get genital warts from her -- she may not even realize that she has it. And then you learn that she's only 17, get arrested, go to prison, and when you're released you get labeled as a sex offender and have to wear a GPS tracker for the rest of your life. Which may not be very long, as some vigilante finds out that there's a sex offender living in his neighborhood on the Intraweb, and he breaks in and kills you in your sleep. (Hopefully they'll take the GPS tracker off before they bury you.)
And genital warts aren't the worst thing you can get, and not the only thing that cant' be cured. And you can also get them without even having sex (kissing could pass them from mouth to mouth.)
Nobody said life was fair. But in theory, our legal system ought to be, and treating `sex offenders' like we do, making them register, tracking their movements, especially when their crimes are stupid things like `public urination' (it varies from state to state, but some do treat that as full fledged `sex offenders'), when we don't do similar things for people convicted of murder, assault and battery, armed robbery, etc. is about as far from `fair' as you can get
But all the politicians have to do is play the `think about the children' card, and everybody involved seems to stop thinking and start jerking their knees instead...
"Apparantly. And the RIAA is using civil law to punish people (and profit!) for what has traditionally been the sort of thing that has been a criminal case."
Uhhh... no? Copyright infringement has only recently been punishable under criminal law, and only then in limited circumstances.
Sorry if I didn't make myself clear. Generally, the RIAA and MPAA and others would like us to look at `copyright infringement' as stealing. (Many agree with this view, and many do not. I won't get into that here.) But stealing has generally been a crime, and when you're caught stealing men in blue come and lock you up. That's what I meant by `this sort of thing'.
But copyright infringement has generally been a civil issue, not a criminal issue.
OJ was found "not guilty". He was not found to be innocent. The family has the right to sue for damages.
The Constition is still the law in this land, right? (I realize that GWB and others are chipping away at that, but...)
`Innocent until proven guilty'. He was not proven guilty, so he's still innocent in the eyes of our legal system. You can nitpick about `innocent' not being the same as `not guilty', but as far as our criminal system is concerned, as far as I know, they're the same thing.
You'd probably do the same thing.
Just because I'd probably do the same thing, that doesn't mean it's right. If somebody had killed my daughter, and I got the opportunity, I'd probably kill them. Especially if I wouldn't get held accountable for it (like if nobody knew I did it.) Even if a court had found them `not guilty' (but I knew they were.) I'd know it's wrong, but I might do it anyways.
Except that they aren't. The RIAA is suing for damages. They can't put you in jail. Moreover, you aren't fighting the government. It's not the same thing.
If I go into a store and steal a few CDs and get arrested, I'm not likely to go to jail. (As long as the dollar value is low enough.) Instead, I'll probably get a ticket, and a summons to appear in court. It'll work almost exactly like a traffic ticket, where I can just pay the fine (and therefore plead `guity' or `no contest') or go to court and fight it. (Yes, the officer could take me to jail, but this rarely happens here (Texas) for minor offenses, and shoplifting is one.)
If I put a CD up for download, and the RIAA sues me, I won't go to jail. I'll end up either settling for a few thousand dollars, or get sued for a few hundred thousand dollars.
To make matters worse, if the RIAA sues me, not only are the penalties much higher, but apparantly they don't even have to be reasonably sure that they have the right person, and I don't have the fifth amendment to protect me. As far as the defendant is concerned, it's like a criminal case -- but worse in every possible way, except that it's harder for him to end up in jail. (He could still ultimately end up in jail -- but for things like not paying a judgement, or contempt of court, etc.)
As always, our legal system is out of control, and the RIAA and others are taking advantage of this. And my whining about it on/. isn't going to fix it, and even talking to the ACLU, EFF and others isn't going to do it. But at least they're generally fighting the good fight, so I send them money every year so I can tell myself I'm doin something to help.
Better if you're the plaintiff, of course. Not better if you're the guy who happened to be using the library computer on the same day where the allegeded copyright violation happened, but didn't do it.
You'll find that US law is quite favorable to civil plaintiffs.
Apparantly. And the RIAA is using civil law to punish people (and profit!) for what has traditionally been the sort of thing that has been a criminal case. [Though I'm not sure that sharing some mp3s or movies is even a criminal matter at all, just a civil one (I'd assume that you know, however). And if it is, it's not something the police are likely to care about it, not on a small scale anyways.]
It seems to be the new legal `thing' -- when criminal charges fail, go with civil charges. OJ was found innocent of murder (I certainly agree that he seemed guilty, but `beyond a reasonable doubt' ? Probably not...) but he lost the civil suit, and made the family of his dead wife lots of money.
Maybe it's time to start applying criminal trial standards (i.e. you can't be forced to inciminate yourself, `beyond a reasonable doubt', court appointed attorney for those that can't afford their own, etc.) to certain sorts of civil cases, because they're being used like criminal cases. Of course, merely mentioning this on/. isn't going to make it happen. But I wonder what the ACLU, EFF and others think about it... [hmmmm]...
If this answer doesn't prove I'm a lawyer, probably nothing will.
It doesn't prove it. It makes me perhaps 70% sure you're a lawyer. (The other 30% could be a guy who googles a lot, or somebody who didn't make it through law school, or...:) )
As for `probably nothing will', I imagine there's some paperwork that goes along with passing the bar -- a certificate, a diploma or something similar. That would probably provide better `I'm a lawyer!' proof than merely not answering a question in a direct manner:)
That is, if it was marginally more likely that you did it, based on the IP log, than that someone else did it, it is proof that it was you.
That's a very interesting viewpoint. So if the the alleged copyright infringement was tracked down to a specific computer, and person A used the computer for 4 hours that day, and person B for 3 hours that day, and they can't narrow it down any more than that, person A is guilty?
I thought it was all about convincing the judge and/or jury, and if I honestly knew that the odds were 51% that the guy was guilty, I'm not so sure I'd say he was `guilty' (though this is a civil thing, so I'm not sure that `guilty' is the right term.)
Absolute proof is not required.
Absolute proof is not even required in a criminal trial. In that case, it's `beyond a reasonable doubt'. (But this is a civil matter, so things are different.)
I'll bet it doesn't. This sort of place seems to have been ditching the PC games lately, replacing them with more console games. Probably because there's more money to be made in console games.
I get most of my PC games at Frys now it seems. (And the only consoles I own are an Atari 2600, Intelivision and a Dreamcast. They don't seem to have many games for those:)
Minix was a teaching tool to teach OS design concepts - with source.
Interesting. You are correct - I hadn't considered that it was open source -- I guess I was confusing it with Xenix or Coherent or something.
However, it looks like redistribution of Minix is permitted as long as you keep the Minix license in the file. So if you're allowed to use the code elsewhere -- so Linus could legally use as much of the code as he wanted, and didn't need to rewrite any of it later if he didn't want to.
The only way I'd see that using this code wasn't OK would be if it wasn't OK for Andy Tanenbaum
and Kees J. Bot and any other authors of Minix to give away, because it wasn't their code. Which seems odd, but it does sound like about par for the SCO course:)
Linus Torvalds, for example, didn't actually try to write Linux from scratch. Instead, he started by reusing code and ideas from Minix, a tiny Unix-like operating system for PC clones. Eventually all the Minix code went away or was completely rewritten -- but while it was there, it provided scaffolding for the infant that would eventually become Linux.
I've heard this claim, or at least similar claims, made before (usually by SCO or SCO supporters.) Wasn't Minix closed source? If so, how could Linus have used any of it's code? How would Linus have gotten access to the code?
Sure, he used some ideas from Minix (and *nix in general), but I don't see how he could have used actual code. Or am I missing some way that he would have had access to the code?
Restoring from rsync would require you to create the partition, format the partition and the restore the files.
Sure, but that's not difficult. Systemimager for Linux keeps images of disks of remote systems via rsync, and has scripts that take care of partition tables and such.
Yes, it's written for Linux, but it wouldn't be difficult to update it to work with NetBSD or any other OS. The reason it's Linux specific is that it makes some efforts to customize the image to match the destination machines.
Also, if you need the MBR...
It's not like you can't just handle the MBR seperately. It's not difficult.
As the article says, this is drive imaging whereas rsync is file copying.
Whatever you want to call it. In any event, file `copying' is more flexible than merely keeping dd'd images of disks -- you can update systems on the fly (without even rebooting), you can use normal *nix commands on the contents of the images themselves, you can do incremental backups on the images themselves (and only get the changes) and the list goes on.
The big advantage to making images with dd or a similar tool and using that is that 1) it can deal with raw partition formats, where you can't just mount them -- I guess this would be useful for a Tivo, or maybe for an Oracle database (but in that case, you'd be better off using the Oracle backup utilities) or 2) If you had an application that required that files not move around on the disk (pretty much unheard of in *nix, but somewhat common as a copy protection on Windows) dd'ing images would be better.
Overall, I'd think that rsync would be a lot better, and while Systemimager isn't perfect, it's architecture is pretty sound and I'd start there.
And, THAT's easily mitigated by proper configuration of border routers. It's a simple rule, easily implemented, and seriously reduces the odds that you won't be part of a DDOS attack...
I was just pointing out that it (easily forged source addresses of packets) is a limitation of IPv4. To be fair, I'm not sure if it's an easy limitation to fix, and I doubt IPv6 fixes it.
Yes, egress filtering is a very good thing, and the Internet would be a better place if everybody did it properly. But even so, it's not a complete fix to the ability to forge source IP addresses of attacks. For example, if one box in a company is being used to launch an attack on a remote site, and that company is doing proper egress filtering, that box could just pick another IP address within the organization and spoof that address. The attack would then happily go through the egress filters, and still we wouldn't know the exact source of the attack (and may in fact start tearing down the wrong machine trying to figure out what's going on.) To really find the source, you'd need to look at the routers in the organization, see where the traffic is really coming from, probably tracking it down to the wire where it's coming from. Can be lots of work for a large organization.
Slashdot Mirror
However, This site suggests a 75 watt heater to keep a 20 gallon tank 18 degrees F above room temperature, or a 150 watt heater for 36 degrees F.
The computer probably puts out less than 150 watts total. Even assuming an 80 degree F room, that would put the computer at 116 degrees F, which wouldn't upset the computer at all. Granted, the heater you put in a fish tank has a thermostat, and so it's not on all the time, but your computer will not have any problems at 116 degrees F inside, and could go a good deal higher safely.
But I do agree with the other guy to respond to your post -- I don't see the fans even turning, let alone turning enough to move the oil around. Perhaps if they were cut down some ...
Of course, I have no idea how well heat flows through oil, or how well it's transferred from oil to the air. But I imagine that the heat generated is low enough for it to not be a problem.
(Well, I assume it doesn't use the ftp protocol, but it serves the same purpose -- it copies files from host A to host B. rsync, scp, rcp and sftp all do the same thing.)
If so, then it's not even really giving the same functionality as SMB, NFS and others, and they're really doing Samba a disservice by comparing themselves to it.
Chungles appears to just be a GUI for copying files from host to host, but you can't use the files on a remote host as if they were local, like SMB and NFS and others let you do. The distinction is subtle, but important.
SMB is only one of many file sharing protocols. Samba is generally only used when you want a Windows machine to use filesystems on a *nix fileserver. That's it.
But Windows is hardly the whole world. There's also NFS (pretty much the standard in the *nix world), Appleshare (what Macs use. To mount *nix filesystems on a Mac, you generally run netatalk on the *nix box, though OSX supports NFS natively too.)
Actually, you can get NFS clients and servers for Windows from several places, and they generally work pretty well once one realizes how NFS differs from SMB.
To mount a Windows file system on a *nix box, you can use smbfs if it's a Linux box -- and note that smbfs is not the same as Samba. Another option is Sharity (previously Rumba) which takes SMB and converts it to NFS.
And in addition to SMB, Appleshare and NFS, there's also Netware, Coda, InterMezzo, DFS, AFS, and many others, and nowadays they're all pretty much multi-platform. This page gives a nice summary of some of the more popular network file sharing methods.
Of course, network file sharing is not the only way. Sneakernet still works, for example. People have been taking floppy disks out of one computer type and putting them into another computer type for as long as computers have had floppy drives, though it probably didn't usually work until both Macs and PCs had 3.5" drives and the Macs learned to talk FAT. And today, USB and firewire drives, if formatted FAT, generally work in Mac, Windows and *nix systems and probably others, and are often used to share files ...
Seriously though, there's a huge difference. Do you really think that working in a concentration camp is remotely close to slaving away in a cubical trying to find out why Microsoft Excel crashes when somebody clicks on a specific cell, but only on SMP P4 boxes?
I certainly do believe that most of the German soldiers in WWI and WWII were just average guys, trying to make it through the war in one piece. But war is a bit different than computer programming -- soldiers generally do things like point guns at people and kill them, which is generally considered to be `evil' unless there's a very good reason for it. I suspect that many (most?) Microsoft employees, on the other hand, are developers and support personel and such, and so their job is to actually make their products better (fewer bugs, more features, etc.), which would generally be considered `good'.
(Though I guess that one could consider improving software to be `bad' if the software was bad. Malware or spamware might be considered bad, as might software used to keep track of prisoners in your concentration camp.)
But most of the `true Linux experts' out there still have mouths to feed and like their sports cars and such, and if Microsoft were to offer them enough money, at least some of them would work for Microsoft. I would if they offered enough money, though I'm somewhere between L++ and L+++ in my geek code block so I probably don't count.
And I can easily see how a `Linux expert' might even justify working for Microsoft -- he might see it as a way to infiltrate Microsoft from within, perhaps influence their corporate culture, help make them a better company. (Yes, I think this is unlikely, but he might justify it like that anyways.) And while I'm certainly a OSS advocate, I still think that any improvements in Microsoft software will benefit the computing community as a whole, and if a Linux expert can help improve their software, I think this would be a good thing for him to do.
There may be a few people there like that, but I certainly do believe that most people there are just normal, every day people who do their job so they will get paid. And if they can do something to improve the world, that's a nice bonus, but mostly they're just there to get paid, just like the rest of the world.And I don't think that Microsoft is so much more evil than any other software company, it's just that they are in a position where their actions have a lot more effect on things, so people watch them more closely.
Are you referring to this incident? If so, that's a long stretch from what you said. Yes, Microsoft's reaction was extreme, but it's not quite what you suggested. He certainly did more than mention that Microsoft uses Macs.The marketing stuff that you see from them is written by a small subset of the company, and it's generally written with one goal in mind -- to benefit Microsoft. They aren't worried about giving the alternatives a fair treatment, unless they think that that will benefit them somehow.
Overall, Microsoft may be the `enemy', but the individual employees certainly aren't. They're just average working people like those working at any other software company.
Even the cheap modern IDE drives are pretty good. SCSI (and maybe fiber channel, though I have no experience with FC) is still king, but even the consumer grade IDE drives can generally handle a good database pounding.
I'd go all SCSI, but IDE is almost an order of magnitude cheaper now, so in many cases it makes good sense to just do RAID-1 with IDE, even in the `enterprise'.
Libel does not have to be `accusing somebody of doing something that's illegal' -- all it has to do is damage somebody's reputation. (It (the claim made) also has to be false.)
If I accuse you of having lots of one night stands with random women, that might damage your reputation, especially if you're a priest or married. But having that sex is not against the law, but my claim may still be libel -- as long as it's not true.
Truth is an absolute defense against libel and slander claims, though our legal system can make it very expensive to even defend yourself against a claim of libel, even if you have that proof.
For example, there was a case some time ago where the propeller came apart in a commercial plane (this was not obvious at the time) and the engine vibrated itself to bits in a few seconds (THAT was obvious, once the pilots actually looked at the engine. Before that, they'd thought they'd just lost the engine in a more `normal' way.) The plane had enough thrust to maintain altitude after losing an engine under normal conditions, but in this case the destroyed engine had much more drag than a normal non-functional engine, and the plane could not maintain altitude, and eventually crashed.
The crash was bad, but controlled. Nobody was killed by the crash, but the plane was basically destroyed. Unfortunately, destroyed planes tend to leak fuel, and this one did, which promptly ignited, and ultimately about half the people aboard died due to the fire.
This was all described in a show about aviation disasters on the Discovery Channel. Maybe somebody else can provide more details.
Or, if a two-engine plane were to lose an engine in the middle of the ocean, that would increase drag as the plane would slip due to mismatched thrust, and might not have enough fuel to make it to land. Hopefully they plan for this, and provide enough extra fuel for this kind of emergency.
But yes -- pretty much most 2+ engine airplanes can maintain altitude and even climb slightly if they lose one engine in a `normal' way. (Exploding/disintegrating engines don't really count. Fortunately they're very rare.)
However, when you talk about private planes, the fatal accident per flying hour ratio is signifigantly actually higher for two engine planes than single engine planes. This is because a two engine plane is quite difficult to fly with an engine out, and this tends to cause fatal accidents. In a single engine plane, when you lose your engine, all you can do is look for a good place to land (or crash, if you can't find a good place.) But even if you crash, you're likely to be in control of the plane and while the plane is likely to be destroyed, you're likely to survive.
But in a dual engine plane, what often happens is that the sudden yaw as the engine is lost causes the plane to turn into a lawn dart, killing all aboard. Yes, a good and alert pilot can prevent this from happening, but mistakes are often made.
To be fair, he probably used a good quality morse code key or paddle and not the tiny keyboard on a phone like the girl did. It makes a big difference -- you can send morse code with pretty much anything, even something like a flash light, but a good key makes a huge difference.
Some examples? `de' for `from', `cq' for `anybody out there?', `73' for `warm regards', etc. Here's some more examples for you.
Just thought I'd be fair ...
AD5RH (yes, I passed the morse code 5 wpm test. But only barely :)
1) If /. has linked to your site, that means your site still needs to serve up the main page. You could coralize your images and such and save some bandwidth that way, but if your web server can't even serve that first page under the load, you're screwed. And if you do find yourself /.ed, and go and coralize your site real quick, then it'll be a while before the traffic slacks off enough for the coral servers to even reach your site to get the images that you've coralized.
Many sites do replace their fancy dynamic pages with a `hi slashdotters!' page after getting /.ed ... saves a lot of cpu on the box. But if what's special about your site is the dynamic aspect of it, well, that won't work.
2) Coral won't do files over 50 or 100 MB. So if you've got some large download, you'd better set up a Bit Torrent instead ... and fast.
3) Currently, Coral uses some non-standard ports that some places may not be able to access due to restrictive firewalls. I understand that this is to change.
4) Coral uses some DNS tricks that don't work with the entire world. Specially, Windows DNS servers tend to have problems with it.
But still, mentioning coral as a way of reducing the /. effect is an excellent idea. It's not the perfect solution, but it's pretty good.
But when you're running your own server, and it normally gets 50 hits/day, and then suddenly a Slashdot listing hits it with millions of hits in one day, well, that's harder to prepare for, because 1) you often don't know you're going to be on /. until it's already happened, and 2) is it even worth preparing for? It's just one or two days, and then things will go back to normal. More hardware and bandwidth may cost lots of money, money that you're not going to spend just so people can see pictures of whatever neat thing you did.
Really, the only sites that get /.ed are the smaller ones. The larger ones already have the hardware and bandwidth needed to handle it. Sure, a /.ing probably shows up on their mrtg reports, but it's probably just a 20% or so increase in traffic, not a 1000x fold increase.
1) few people look that up at all (and why use Netcraft? nmap works fine.)
2) A poorly configured *nix system or application is a lot more insecure than a similar well configured Windows system. Merely being *nix does not mean `secure', and merely being `Windows' does not mean `insecure'. There's far more important variables, variables that Netcraft does not track.
3) the risk of exposure is pretty small. If I see charges on my card I didn't make, I report them to my card company, and I don't pay. The biggest problem is that I have to wait a week or so for them to send me a new card. As long as my credit card is physically in my possession, I have zero liability for fraudulent use of it. (If I lose it and report it in a reasonable amount of time, the liability limit is only $50.)
That's pretty common. People see that little lock at the bottom of the screen (the one that indicates SSL is in use) and think that means it's secure!. You know better, I know better, but most people do not.What really annoys me is sites that make it impossible to determine the shipping costs. Many require that you go almost all of the way through the ordering process, often even going past the point where I enter my credit card, before I'm given any clue what the shipping will be.
Here's a free clue for you, online retailers -- when I run into a site that doesn't tell me what shipping will be without making an order, I usually go somewhere else. If you won't tell me the shipping cost up front, my reasoning is that it probably sucks anyways. In reality, it may not, but it's not worth my time, and I'm certainly not going to enter my credit card number before I know how much I'm going to spend.
As for complaining that people abandon shopping carts, well, we do that because we don't think of them as shopping carts. Because they're not. It's a list of items we might want to buy, stuff we're interested in. Not a list of things we are going to buy, at least not until we start checking out. And really, if a real brick and mortar store did some of the annoying tricks that online stores do to `trap' me into buying from them, I'd abandon my real shopping cart there too, though in that case somebody would have to put the stuff back so I might feel a tad guilty about it.
Probably the best thing that an online retailer can do to encourage people to not forget about what they had in their `shopping cart' before is to make sure it persists. If we come back tomorrow or two weeks or two months later, remind us that we'd left some stuff in our `shopping cart'. Since we're not impulse buyers, if we really want something, we'll probably come back later. Don't make us find the stuff again.
Sex, however, is generally believed to be normal, even if our society tends to demonize it. For example, the odds are pretty good that your parents have had it at least once.
Nothing is absolutely prevented by condoms, not HIV/AIDS or pregnancy. However, they are still believed to be at least somewhat effective, even against genital warts. They're not perfect, but they're far better than nothng. So you've heard of STDs. Good. But what does that have to do with screwing a woman you just met in a bar? If I recall correctly, the first time I met my wife was in a bar. That was perhaps 12 years and two kids agoYou could meet a woman in church (or pick some other place for finding wholesome, God fearing women), get to know her for a few months, fall in love, and finally have sex and then get genital warts from her -- she may not even realize that she has it. And then you learn that she's only 17, get arrested, go to prison, and when you're released you get labeled as a sex offender and have to wear a GPS tracker for the rest of your life. Which may not be very long, as some vigilante finds out that there's a sex offender living in his neighborhood on the Intraweb, and he breaks in and kills you in your sleep. (Hopefully they'll take the GPS tracker off before they bury you.)
And genital warts aren't the worst thing you can get, and not the only thing that cant' be cured. And you can also get them without even having sex (kissing could pass them from mouth to mouth.)
Nobody said life was fair. But in theory, our legal system ought to be, and treating `sex offenders' like we do, making them register, tracking their movements, especially when their crimes are stupid things like `public urination' (it varies from state to state, but some do treat that as full fledged `sex offenders'), when we don't do similar things for people convicted of murder, assault and battery, armed robbery, etc. is about as far from `fair' as you can get
But all the politicians have to do is play the `think about the children' card, and everybody involved seems to stop thinking and start jerking their knees instead ...
But copyright infringement has generally been a civil issue, not a criminal issue.
The Constition is still the law in this land, right? (I realize that GWB and others are chipping away at that, but`Innocent until proven guilty'. He was not proven guilty, so he's still innocent in the eyes of our legal system. You can nitpick about `innocent' not being the same as `not guilty', but as far as our criminal system is concerned, as far as I know, they're the same thing.
Just because I'd probably do the same thing, that doesn't mean it's right. If somebody had killed my daughter, and I got the opportunity, I'd probably kill them. Especially if I wouldn't get held accountable for it (like if nobody knew I did it.) Even if a court had found them `not guilty' (but I knew they were.) I'd know it's wrong, but I might do it anyways. If I go into a store and steal a few CDs and get arrested, I'm not likely to go to jail. (As long as the dollar value is low enough.) Instead, I'll probably get a ticket, and a summons to appear in court. It'll work almost exactly like a traffic ticket, where I can just pay the fine (and therefore plead `guity' or `no contest') or go to court and fight it. (Yes, the officer could take me to jail, but this rarely happens here (Texas) for minor offenses, and shoplifting is one.)If I put a CD up for download, and the RIAA sues me, I won't go to jail. I'll end up either settling for a few thousand dollars, or get sued for a few hundred thousand dollars.
To make matters worse, if the RIAA sues me, not only are the penalties much higher, but apparantly they don't even have to be reasonably sure that they have the right person, and I don't have the fifth amendment to protect me. As far as the defendant is concerned, it's like a criminal case -- but worse in every possible way, except that it's harder for him to end up in jail. (He could still ultimately end up in jail -- but for things like not paying a judgement, or contempt of court, etc.)
As always, our legal system is out of control, and the RIAA and others are taking advantage of this. And my whining about it on /. isn't going to fix it, and even talking to the ACLU, EFF and others isn't going to do it. But at least they're generally fighting the good fight, so I send them money every year so I can tell myself I'm doin something to help.
It seems to be the new legal `thing' -- when criminal charges fail, go with civil charges. OJ was found innocent of murder (I certainly agree that he seemed guilty, but `beyond a reasonable doubt' ? Probably not ...) but he lost the civil suit, and made the family of his dead wife lots of money.
Maybe it's time to start applying criminal trial standards (i.e. you can't be forced to inciminate yourself, `beyond a reasonable doubt', court appointed attorney for those that can't afford their own, etc.) to certain sorts of civil cases, because they're being used like criminal cases. Of course, merely mentioning this on /. isn't going to make it happen. But I wonder what the ACLU, EFF and others think about it ... [hmmmm] ...
It doesn't prove it. It makes me perhaps 70% sure you're a lawyer. (The other 30% could be a guy who googles a lot, or somebody who didn't make it through law school, orAs for `probably nothing will', I imagine there's some paperwork that goes along with passing the bar -- a certificate, a diploma or something similar. That would probably provide better `I'm a lawyer!' proof than merely not answering a question in a direct manner :)
(No, I'm not asking for proof.)
I thought it was all about convincing the judge and/or jury, and if I honestly knew that the odds were 51% that the guy was guilty, I'm not so sure I'd say he was `guilty' (though this is a civil thing, so I'm not sure that `guilty' is the right term.)
Absolute proof is not even required in a criminal trial. In that case, it's `beyond a reasonable doubt'. (But this is a civil matter, so things are different.)A 51% change of being guilty is far more than a `reasonable doubt'. Apparantly the definition of the phrase preponderance of the evidence is rather subjective as well.
Ok, you're a lawyer. (I'm certainly not.) What is your specialty?I get most of my PC games at Frys now it seems. (And the only consoles I own are an Atari 2600, Intelivision and a Dreamcast. They don't seem to have many games for those :)
However, it looks like redistribution of Minix is permitted as long as you keep the Minix license in the file. So if you're allowed to use the code elsewhere -- so Linus could legally use as much of the code as he wanted, and didn't need to rewrite any of it later if he didn't want to.
The only way I'd see that using this code wasn't OK would be if it wasn't OK for Andy Tanenbaum and Kees J. Bot and any other authors of Minix to give away, because it wasn't their code. Which seems odd, but it does sound like about par for the SCO course :)
Sure, he used some ideas from Minix (and *nix in general), but I don't see how he could have used actual code. Or am I missing some way that he would have had access to the code?
Yes, it's written for Linux, but it wouldn't be difficult to update it to work with NetBSD or any other OS. The reason it's Linux specific is that it makes some efforts to customize the image to match the destination machines.
It's not like you can't just handle the MBR seperately. It's not difficult. Whatever you want to call it. In any event, file `copying' is more flexible than merely keeping dd'd images of disks -- you can update systems on the fly (without even rebooting), you can use normal *nix commands on the contents of the images themselves, you can do incremental backups on the images themselves (and only get the changes) and the list goes on.The big advantage to making images with dd or a similar tool and using that is that 1) it can deal with raw partition formats, where you can't just mount them -- I guess this would be useful for a Tivo, or maybe for an Oracle database (but in that case, you'd be better off using the Oracle backup utilities) or 2) If you had an application that required that files not move around on the disk (pretty much unheard of in *nix, but somewhat common as a copy protection on Windows) dd'ing images would be better.
Overall, I'd think that rsync would be a lot better, and while Systemimager isn't perfect, it's architecture is pretty sound and I'd start there.
Yes, egress filtering is a very good thing, and the Internet would be a better place if everybody did it properly. But even so, it's not a complete fix to the ability to forge source IP addresses of attacks. For example, if one box in a company is being used to launch an attack on a remote site, and that company is doing proper egress filtering, that box could just pick another IP address within the organization and spoof that address. The attack would then happily go through the egress filters, and still we wouldn't know the exact source of the attack (and may in fact start tearing down the wrong machine trying to figure out what's going on.) To really find the source, you'd need to look at the routers in the organization, see where the traffic is really coming from, probably tracking it down to the wire where it's coming from. Can be lots of work for a large organization.