So you will be unable to purchase media that uses it (legally). Talk to the vendors. They are the ones who use it.
The presence of a feature provides an excuse for its usage. If nobody had Flash preinstalled, few designers would use it without really needing it. Witness how the wide presence of Flash, and the scarcity of support for other video playing plugins like quicktime resulted in flash being massively used to play video.
Plus the presence of it, regardless of whether it's being used or not involves things I'm not willing to tolerate.
So it either updates without your permission, or doesn't update
I'm not only talking about Windows, but various programs that popup message boxes when they want to get updated. This is a consequence of having no package manager.
3rd party. Talk to the AV makers.
But the overall behavior is a Windows culture thing. Try an Ubuntu box sometime. Notice how with a disk full of software you still don't end up with 50 icons in the systray, and applications popping up various notifications and trying to get your attention.
Wait, i thought we're supposed to complain about too many versions of Windows. You want more.
Why more? I want that instead of XP and Vista with all their versions.
Commercial license vs GPL or no license. Some of us have jobs and like getting paid for our work.
The GPL doesn't need to be a clickthrough, as it's not an EULA. So no reason to show it in an installer.
BTW, I make money by improving GPL licensed software. I still would even if it was BSD.
I'm sorry but I had to laugh out loud when I read that. I don't think anyone actually reads the legalese to install a program. Further, whenever I try to install someone non-trivial on Linux, I wish I got questions. Instead, I get standard error output! I usually spend an hour or so trying to resolve some dependency error, or debug on obtuse error when trying to use some very well-intentioned but buggy (in my experience) utility for automating it (e.g. apt-get).
The issue there is that it's not unattended. On Linux I can install all of KDE, in one command, pulling in dozens of packages, and have all that happen in a completely unattended manner. On Windows I'd have to do manual dependency resolution, just like what you get with programs that require you to have SP4, IE6 and DirectX 9 installed first.
I ask a very simple thing: That software be installed. Once I ask that I want it to be just installed. I absolutely hate babysitting the thing. My favourite are the ones that include some extra junk I don't want (google toolbar, itunes, etc).
And while dependency problems do exist with apt-get it's in my experience a very infrequent thing. I don't remember having any in the last 6 months or so. One dependency problem I remember having was due to installing the latest version of KDE from a third party source, but some risks have to be assumed if you want to be on the bleeding edge.
I use Linux as my primary OS at work and I have been using it for years, but I spend much more time at work tweaking my machine than I do at home. And further, I don't know any non-zealot who believes the whole "Linux is easier to maintain and use on the desktop" nonsense. Hell, even Linus doesn't. RMS might, but he hasn't used a non-GNU OS since System V;)
Who said anything about ease? I'm talking about convenience. I turn my computer on, start KDE, then start kdevelop. My work isn't interrupted by an antivirus slowing things down and popping up notifications about updates and some random box on the internet that decided to ping mine. If my hardware fails I can move the disk to new hardware with very minimal hassle.
There's some nice folks at the the pirate bay that can help you with that....
And they probably very kindly include a trojan in there as well. I can't know for sure, there's no source for any of that.
Actually it's a funny thing, but these days I use Linux not so much because of what it does, but because of what it doesn't.
For me, an OS is a base system that's just there to run my applications. It's supposed to do its thing, be unobtrusive, and then get the hell out of my way.
Linux: * My current install doesn't have performance degrading pointless effects. * It doesn't have activation, or require entering serial numbers * It doesn't have DRM * It doesn't popup message boxes when it wants to get updated. * It doesn't try to REBOOT without my consent. Seriously, WTF is up with that? * It doesn't require an antivirus which slows down performance, and constantly pops up message boxes announcing gleefully how it now can detect 3 viruses more. * Installing programs doesn't require clicking through legalese, and refusing offers to register. They install, no questions asked. * Software doesn't ship with spyware, and doesn't nag to be updated/registered * It doesn't require a full OS reinstall if I want to get a feature added in the latest version. On Windows, you can't get ClearType without upgrading to XP. On Linux all you need is to update the necessary components and everything else stays the same.
Trying to sell me Vista because it has features is a pointless endeavor. Here's what I want: Win2K with kernel improvements, DX10 and all that. No DRM, no Aero, no activation, no interface changes. Until MS makes that, I'm not buying.
Who is concerned about matters like the fire code? Engineers. And I'd say engineers fit into the "geek" category quite nicely.
Who is concerned about elections on a technical level? Discussing subjects like approval voting, condorcet, hardware and methods used, etc, is a very geeky thing.
Also, IMO "geek" is a very general category that isn't limited to computers, or even technology.
This isn't a physics exercise. That 1cm deep layer of water isn't sitting there in complete magical isolation. It's got a LOT of water under it, a LOT of air over it, and then some more water and air around it, and all that isn't just staying there, but constantly moving. During that hour you're heating that water it'll be mixing with the surrounding one. The water also has currents, which will carry some of that hot water away.
Then there's that there's no such thing as 100% efficient transfer of heat, so you'll need a good deal more to compensate for the loss.
Luckily most modern file systems are "journaling" which helps prevent against corruption. Either the data has been committed, or it hasn't.
My point exactly. Any kernel made in the last few years supports a journalling filesystem which recovers in seconds from a dirty shutdown. It probably even recovers faster on larger disks as the journal is usually of a fixed size, and bigger disks are usually faster than smaller ones.
The grandparent is talking about what would happen if you used ext2 for a drive with a size in the terabytes range. But no current distribution I know uses ext2 by default, so this isn't really a problem.
LILO copies a kernel image to its boot area. It doesn't matter if you change the kernel on the hard drive, because LILO's installed image won't change until you invoke the "lilo" command. I've actually seen LILO successfully boot a kernel and initrd (which panicked) after I had formatted a drive and removed all of the partitions, because I hadn't bothered to wipe the MBR.
No, it doesn't. You can read the LILO technical documentation if you don't believe me.
The fundamental thing about LILO is that unlike Grub, it's incapable of actually reading the filesystem the kernel is on. The way it works is that the boot sector contains the location of the map file, and the map file contains the list of sectors that make the kernel. There's no "boot area" as such. It's trivial to verify that the map file isn't the kernel, as it's tiny (78KB on one of my boxes)
The reason LILO booted for you is simply because a format didn't overwrite the data areas of the disk. Since LILO doesn't read the FS itself, it doesn't matter to it that all the metadata is gone. So long the boot sector is there, the map's data is there (even without metadata indicating the filename, etc), and the kernel's data is there, it'll boot.
LILO however will completely mess up if you are unlucky. Overwriting the kernel without calling "lilo" afterwards might work if it just happens to write over the same sectors and uses the same number of them. Or maybe the new kernel is written somewhere else entirely, in which case you'll boot the old kernel and it'll break later when something reuses the space taken by the old version.
The recovery console is useful, but as I'm often tinkering with things, I prefer to have a bootloader that's static, and won't change until I explicitly tell it to. There's also the "lilo -v -t" command to test when I make a change to/etc/lilo.conf.
The problem with LILO is that you can screw it up without touching LILO itself. For example, delete the active kernel. It'll probably work anyway, right until something reuses the space previously taken by the kernel. Then boom, doesn't work anymore. With grub it doesn't matter if you make a bad config file, or delete a needed kernel. So long there's a kernel on disk, grub can boot it.
Grub has a console available right from the boot menu.
That means that as long as grub works at all, you can load and boot any kernel from any disk with a FS grub can recognize. This is really great for the cases where for example there's a typo in the configuration and the kernel's name is wrong in the file, or it's been mistakenly deleted, but there's an old kernel on the disk not listed in the config file.
LILO, on the other hand relies on a map file that tells it where on the disk the kernel is. That means it's unable to load anything it's not been configured for, and breaks if the kernel on the disk changes but the map file isn't updated.
These days I always install grub if possible, as it means that if something goes wrong I don't have to look for a rescue CD.
For games, IMO, the question is not so much about the code, but about the polish. To be honest, your games don't look like something that'd make me run and download them. That doesn't mean there aren't promising open source games, it's just that writing a good game is HARD. What you should do is teaming up with a good artist. Good code + decent graphics + playability = you're 90% there.
Small games I liked:
Closed source: Starscape, Lugaru. Open source: Scorched 3D, Tux Racer (IMO the proof of that it's possible to make something small but still fun), Frozen Bubble, Chromium, Vegastrike (incomplete but very promising last time I tried it), UFO: Alien Invasion (same as Vegastrike)
For games it doesn't even need to be original. IMO, a requirement for success is filling a niche (I'd LOVE a Magical Drop style game if there isn't one already), an amount of polish (even if the code is great, paintbrush graphics will turn pretty much everybody off), and if the project is still not playable, enough progress that everybody can see it's promising. UFO: AI, Vegastrike are there. Anybody can see how those could become excellent games, as they're 90% there (or they were last time I checked)
I bet they could come up with a way of applying very considerable pressure. Especially as the rest of the world seems to be less and less happy with the US' position.
People said the EU couldn't fine Microsoft. Well, they did. Now they say the WTO can't fine the US. I'm pretty sure they'll find a way.
What I don't like about Linux is the amount of unnecessary services installed by default. For example, I have an old computer (P133/16MB RAM), and Debian 3.1 on it. Debian demands that I run Sendmail (or Exim, or Postfix). Why can't I live without a MTA?
Because the standard way to send mail on an unix box is to send it through a 'sendmail' program, which is provided by Sendmail (or Exim or Postfix)
I doubt that regular home users actually send their e-mail through their local MTA, they probably use their ISP's SMTP, which, I think, it's the proper way to do it.
That's because for example cron doesn't know how to talk SMTP. It calls sendmail which does know. With the configuration you mention the way to do it is to run a mail server locally which then relays the mail to your ISP's one.
The advantage of doing it that way is that you don't have to code a full SMTP implementation into every program that wants to send mail, and don't need to bother handling retransmission and every possible error condition, your mail server already knows how to do all that and can do it for you.
The disadvantage is an extra server, but you could run it from inetd, so that it's not really running unless something needs to be sent.
Privacy and Internet mix very well actually. A lot better than real life at least.
You can easily have multiple identities online (I do, this isn't my only slashdot account), and there are multiple methods available to confuse and hide your true identity that wouldn't be possible if you had to interact with people directly.
In fact, it's quite possible to do anonymous business these days. Through Second Life for instance -- people may know me for my avatar's name, but my real name and location aren't known to most people. By using proxies, tor, open access points, free webhosting and such, it's possible to make finding out your real identity very difficult for normal people.
Social networking doesn't eliminate privacy. I could be very social as "Dale Glass" (which isn't my real name) if I wanted, then suddenly vanish and reappear somewhere else with a different name, age, gender and interests. With enough effort I could probably even rejoin the same places under a different identity and maintain them separately.
I find it interesting how online you can consider somebody to be a friend without knowing their real name, location, age, gender or what they look like. And what they present themselves as can eventually become their real self in my mind -- I'd probably be less surprised now at seeing an antropomorphic fox than the actual person if I met in reality some of the people I know in SL.
Don't see what's wrong with looking it up on Wikipedia. If you look at the wikipedia page, you'll notice it has lots of links on the bottom which look like the "more reliable domain specific sources" you want.
Have you ever used Memtest386? At my job I've only ever encountered two or three instances of bad ram and this is the tool that's done it. Replace the ram, perfect functionality. I've had very good luck with it. For servers everyone says don't dick around, make sure you get ECC, so that seems like smart advice.
Yes, and it failed. See the "This took me quite a while to diagnose, because memtest86 passed perfectly fine." in my post.
Memory affects pretty much everything, so it's hard to isolate it from everything else. Bad RAM can result in disk corruption, making it hard to determine it's the memory and not the disk that's broken.
For example, take Nero, burn a CD, then verify it. If the RAM is bad it may well happen that a few bits you read from the CD got flipped, and now the verification fails. Obvious conclusion: The CD-R was bad. After a few of those, obvious conclusion: the drive is bad. That the computer crashes ocasionally can be attributed to spyware or viruses. A tech working for cheap isn't going to spend hours to test every possible case.
RAM is also one of the most annoying things to try to diagnose. Disks at least have SMART, so if it got to the point where it's really broken, SMART will tell you about that quickly. And once it breaks it tends to do so very obviously. Now memory can pass tests and still be bad, and be marginal enough to work most of the time.
I had several problems with RAM that firmly convinced me to always buy ECC.
First one was when my Linux firewall, which ran for months without a hitch suddenly had a kernel panic. I thought it was strange, but oh well, nothing is perfect. Rebooted it, expecting that the new kernel installed weeks ago probably has a fix for that. A couple days later it crashed again. Rebooted it again making a note to investigate later. A day later it crashed yet again, but didn't boot this time due to disk corruption. Turns out the RAM was loose in the slot, and somehow stopped making proper contact. The module itself was good and passed memtest86 just fine when I set up the box.
Second one was when I was buying a new shiny box, and started having strange crashes. This took me quite a while to diagnose, because memtest86 passed perfectly fine. Yet "memtester", an userspace tool did catch it finally, after running for 8 hours straight, and even then with about 50% accuracy. On repeated 8 hour runs sometimes it'd catch it, and sometimes not, while testing the whole memory several times during that period.
Something like that probably won't be diagnosed correctly by tech support. Even if they do test the memory they're almost certainly not going to bother running it for a day straight, just to make really sure it's not a marginal case.
Error correction on a physical channel that's transmitting digital data works by "guessing" what the most likely transmitted symbol is, and it's not outside the realm of possibility that the algorithm guesses wrong.
No it doesn't. Error correction works by including redundant data, verifying that it's consistent with the data transmitted, and detecting, if possible, where the error is.
For example a trivial error correction algorithm is to transmit data in blocks of say, 8x8, plus a parity bit per column, plus a parity bit per row. Then you check the parity for the rows and columns, and from that determine where the error is. Then you flip the bit at that location, and voila, the original data.
Sane implementations can detect when there are too many errors to correct the information. For example, it may detect and correct 1 bit in 16, and detect but not correct 2 bits in 16. In the later case it drops it, asks for a retransmit, etc.
But digital data generally includes more than the raw audio/video, so yes, if the cable is so incredibly horrible that the error correction can't deal with it, then that'll result in a very obvious malfunction, and not silent and almost unnoticeable data corruption.
The problem here isn't just that you're pissing off people by bricking their phones. The problem is you are locking down a device that would otherwise be a killer development platform.
If what you wanted is an open development platform, why did you go with Apple, especially when they made their intentions quite explicit?
It's not like there's lack of choice in the area. If you want complete openness, there's OpenMoko. If that's too hardcore, there's plenty other hardware that'll run whatever you like on it.
IMO, those represent very well the two different approaches to the problem. The rest are a needless complication. Besides, their meaning and implications are understood very well, so I don't see what Google is going to achieve by creating their own.
Nobody seems to agree on what exactly it means. It seems to include from people who just like to happen anthro art or cartoons, to people completely obsessed with the subject.
I would say "furry" is a very fuzzy (no pun intended) term, like "geek" for instance. It has certain connotations, hints at certain things, points at a range of likely interests, but is so unspecific that it doesn't really say all that much about the person it's applied to.
True, but now you've lost your previous boast about it being accessible from anywhere.
Why? Offline IMAP is just like normal IMAP, except messages are cached locally. All the mail is still on the server, except that instead of loading the message only when I click on it, they're downloaded and kept on the disk automatically. I can access it through IMAP over SSL, through SSH, and through a plain web interface.
And you've obviously never actually run beagle before. It's bad news.
Beagle, Meta Tracker, whatever. There's more than one app capable of it, and I even rolled my own program of the sort already, could use that. Generally I don't search large amounts of mail anyway, and got everything sorted out in folders. Grep would probably suffice.
Slashdot Mirror
The presence of a feature provides an excuse for its usage. If nobody had Flash preinstalled, few designers would use it without really needing it. Witness how the wide presence of Flash, and the scarcity of support for other video playing plugins like quicktime resulted in flash being massively used to play video.
Plus the presence of it, regardless of whether it's being used or not involves things I'm not willing to tolerate.
I'm not only talking about Windows, but various programs that popup message boxes when they want to get updated. This is a consequence of having no package manager.
But the overall behavior is a Windows culture thing. Try an Ubuntu box sometime. Notice how with a disk full of software you still don't end up with 50 icons in the systray, and applications popping up various notifications and trying to get your attention.
Why more? I want that instead of XP and Vista with all their versions.
The GPL doesn't need to be a clickthrough, as it's not an EULA. So no reason to show it in an installer.
BTW, I make money by improving GPL licensed software. I still would even if it was BSD.
The issue there is that it's not unattended. On Linux I can install all of KDE, in one command, pulling in dozens of packages, and have all that happen in a completely unattended manner. On Windows I'd have to do manual dependency resolution, just like what you get with programs that require you to have SP4, IE6 and DirectX 9 installed first.
I ask a very simple thing: That software be installed. Once I ask that I want it to be just installed. I absolutely hate babysitting the thing. My favourite are the ones that include some extra junk I don't want (google toolbar, itunes, etc).
And while dependency problems do exist with apt-get it's in my experience a very infrequent thing. I don't remember having any in the last 6 months or so. One dependency problem I remember having was due to installing the latest version of KDE from a third party source, but some risks have to be assumed if you want to be on the bleeding edge.
Who said anything about ease? I'm talking about convenience. I turn my computer on, start KDE, then start kdevelop. My work isn't interrupted by an antivirus slowing things down and popping up notifications about updates and some random box on the internet that decided to ping mine. If my hardware fails I can move the disk to new hardware with very minimal hassle.
And they probably very kindly include a trojan in there as well. I can't know for sure, there's no source for any of that.
Actually it's a funny thing, but these days I use Linux not so much because of what it does, but because of what it doesn't.
For me, an OS is a base system that's just there to run my applications. It's supposed to do its thing, be unobtrusive, and then get the hell out of my way.
Linux:
* My current install doesn't have performance degrading pointless effects.
* It doesn't have activation, or require entering serial numbers
* It doesn't have DRM
* It doesn't popup message boxes when it wants to get updated.
* It doesn't try to REBOOT without my consent. Seriously, WTF is up with that?
* It doesn't require an antivirus which slows down performance, and constantly pops up message boxes announcing gleefully how it now can detect 3 viruses more.
* Installing programs doesn't require clicking through legalese, and refusing offers to register. They install, no questions asked.
* Software doesn't ship with spyware, and doesn't nag to be updated/registered
* It doesn't require a full OS reinstall if I want to get a feature added in the latest version. On Windows, you can't get ClearType without upgrading to XP. On Linux all you need is to update the necessary components and everything else stays the same.
Trying to sell me Vista because it has features is a pointless endeavor. Here's what I want: Win2K with kernel improvements, DX10 and all that. No DRM, no Aero, no activation, no interface changes. Until MS makes that, I'm not buying.
Who is concerned about matters like the fire code? Engineers. And I'd say engineers fit into the "geek" category quite nicely.
Who is concerned about elections on a technical level? Discussing subjects like approval voting, condorcet, hardware and methods used, etc, is a very geeky thing.
Also, IMO "geek" is a very general category that isn't limited to computers, or even technology.
You're forgetting something there.
This isn't a physics exercise. That 1cm deep layer of water isn't sitting there in complete magical isolation. It's got a LOT of water under it, a LOT of air over it, and then some more water and air around it, and all that isn't just staying there, but constantly moving. During that hour you're heating that water it'll be mixing with the surrounding one. The water also has currents, which will carry some of that hot water away.
Then there's that there's no such thing as 100% efficient transfer of heat, so you'll need a good deal more to compensate for the loss.
My point exactly. Any kernel made in the last few years supports a journalling filesystem which recovers in seconds from a dirty shutdown. It probably even recovers faster on larger disks as the journal is usually of a fixed size, and bigger disks are usually faster than smaller ones.
The grandparent is talking about what would happen if you used ext2 for a drive with a size in the terabytes range. But no current distribution I know uses ext2 by default, so this isn't really a problem.
There are plenty filesystems that don't do that, which are included with any kernel released in the last few years.
No, it doesn't. You can read the LILO technical documentation if you don't believe me.
The fundamental thing about LILO is that unlike Grub, it's incapable of actually reading the filesystem the kernel is on. The way it works is that the boot sector contains the location of the map file, and the map file contains the list of sectors that make the kernel. There's no "boot area" as such. It's trivial to verify that the map file isn't the kernel, as it's tiny (78KB on one of my boxes)
The reason LILO booted for you is simply because a format didn't overwrite the data areas of the disk. Since LILO doesn't read the FS itself, it doesn't matter to it that all the metadata is gone. So long the boot sector is there, the map's data is there (even without metadata indicating the filename, etc), and the kernel's data is there, it'll boot.
LILO however will completely mess up if you are unlucky. Overwriting the kernel without calling "lilo" afterwards might work if it just happens to write over the same sectors and uses the same number of them. Or maybe the new kernel is written somewhere else entirely, in which case you'll boot the old kernel and it'll break later when something reuses the space taken by the old version.
The problem with LILO is that you can screw it up without touching LILO itself. For example, delete the active kernel. It'll probably work anyway, right until something reuses the space previously taken by the kernel. Then boom, doesn't work anymore. With grub it doesn't matter if you make a bad config file, or delete a needed kernel. So long there's a kernel on disk, grub can boot it.
Grub has a console available right from the boot menu.
That means that as long as grub works at all, you can load and boot any kernel from any disk with a FS grub can recognize. This is really great for the cases where for example there's a typo in the configuration and the kernel's name is wrong in the file, or it's been mistakenly deleted, but there's an old kernel on the disk not listed in the config file.
LILO, on the other hand relies on a map file that tells it where on the disk the kernel is. That means it's unable to load anything it's not been configured for, and breaks if the kernel on the disk changes but the map file isn't updated.
These days I always install grub if possible, as it means that if something goes wrong I don't have to look for a rescue CD.
For games, IMO, the question is not so much about the code, but about the polish. To be honest, your games don't look like something that'd make me run and download them. That doesn't mean there aren't promising open source games, it's just that writing a good game is HARD. What you should do is teaming up with a good artist. Good code + decent graphics + playability = you're 90% there.
Small games I liked:
Closed source: Starscape, Lugaru.
Open source: Scorched 3D, Tux Racer (IMO the proof of that it's possible to make something small but still fun), Frozen Bubble, Chromium, Vegastrike (incomplete but very promising last time I tried it), UFO: Alien Invasion (same as Vegastrike)
For games it doesn't even need to be original. IMO, a requirement for success is filling a niche (I'd LOVE a Magical Drop style game if there isn't one already), an amount of polish (even if the code is great, paintbrush graphics will turn pretty much everybody off), and if the project is still not playable, enough progress that everybody can see it's promising. UFO: AI, Vegastrike are there. Anybody can see how those could become excellent games, as they're 90% there (or they were last time I checked)
I don't get this kind of post. It seems to be saying "big deal, everybody does that, why did this even get posted?"
While I'm not surprised at all this sort of thing happens this IMO doesn't mean it shouldn't be reported on. Maybe then it'll happen a bit less often.
Lots of people said that on Slashdot for example.
Here's one post like that. I remember seeing a few of those in every discussion on the subject, some of which were quite highly modded up.
This is the WTO we're talking about.
I bet they could come up with a way of applying very considerable pressure. Especially as the rest of the world seems to be less and less happy with the US' position.
People said the EU couldn't fine Microsoft. Well, they did. Now they say the WTO can't fine the US. I'm pretty sure they'll find a way.
Because the standard way to send mail on an unix box is to send it through a 'sendmail' program, which is provided by Sendmail (or Exim or Postfix)
That's because for example cron doesn't know how to talk SMTP. It calls sendmail which does know. With the configuration you mention the way to do it is to run a mail server locally which then relays the mail to your ISP's one.
The advantage of doing it that way is that you don't have to code a full SMTP implementation into every program that wants to send mail, and don't need to bother handling retransmission and every possible error condition, your mail server already knows how to do all that and can do it for you.
The disadvantage is an extra server, but you could run it from inetd, so that it's not really running unless something needs to be sent.
Privacy and Internet mix very well actually. A lot better than real life at least.
You can easily have multiple identities online (I do, this isn't my only slashdot account), and there are multiple methods available to confuse and hide your true identity that wouldn't be possible if you had to interact with people directly.
In fact, it's quite possible to do anonymous business these days. Through Second Life for instance -- people may know me for my avatar's name, but my real name and location aren't known to most people. By using proxies, tor, open access points, free webhosting and such, it's possible to make finding out your real identity very difficult for normal people.
Social networking doesn't eliminate privacy. I could be very social as "Dale Glass" (which isn't my real name) if I wanted, then suddenly vanish and reappear somewhere else with a different name, age, gender and interests. With enough effort I could probably even rejoin the same places under a different identity and maintain them separately.
I find it interesting how online you can consider somebody to be a friend without knowing their real name, location, age, gender or what they look like. And what they present themselves as can eventually become their real self in my mind -- I'd probably be less surprised now at seeing an antropomorphic fox than the actual person if I met in reality some of the people I know in SL.
Well, go and buy a slave in the US, then. Oops, you can't.
There's no free market in the US either.
Don't see what's wrong with looking it up on Wikipedia. If you look at the wikipedia page, you'll notice it has lots of links on the bottom which look like the "more reliable domain specific sources" you want.
Yes, and it failed. See the "This took me quite a while to diagnose, because memtest86 passed perfectly fine." in my post.
Memory affects pretty much everything, so it's hard to isolate it from everything else. Bad RAM can result in disk corruption, making it hard to determine it's the memory and not the disk that's broken.
For example, take Nero, burn a CD, then verify it. If the RAM is bad it may well happen that a few bits you read from the CD got flipped, and now the verification fails. Obvious conclusion: The CD-R was bad. After a few of those, obvious conclusion: the drive is bad. That the computer crashes ocasionally can be attributed to spyware or viruses. A tech working for cheap isn't going to spend hours to test every possible case.
RAM is also one of the most annoying things to try to diagnose. Disks at least have SMART, so if it got to the point where it's really broken, SMART will tell you about that quickly. And once it breaks it tends to do so very obviously. Now memory can pass tests and still be bad, and be marginal enough to work most of the time.
I had several problems with RAM that firmly convinced me to always buy ECC.
First one was when my Linux firewall, which ran for months without a hitch suddenly had a kernel panic. I thought it was strange, but oh well, nothing is perfect. Rebooted it, expecting that the new kernel installed weeks ago probably has a fix for that. A couple days later it crashed again. Rebooted it again making a note to investigate later. A day later it crashed yet again, but didn't boot this time due to disk corruption. Turns out the RAM was loose in the slot, and somehow stopped making proper contact. The module itself was good and passed memtest86 just fine when I set up the box.
Second one was when I was buying a new shiny box, and started having strange crashes. This took me quite a while to diagnose, because memtest86 passed perfectly fine. Yet "memtester", an userspace tool did catch it finally, after running for 8 hours straight, and even then with about 50% accuracy. On repeated 8 hour runs sometimes it'd catch it, and sometimes not, while testing the whole memory several times during that period.
Something like that probably won't be diagnosed correctly by tech support. Even if they do test the memory they're almost certainly not going to bother running it for a day straight, just to make really sure it's not a marginal case.
No it doesn't. Error correction works by including redundant data, verifying that it's consistent with the data transmitted, and detecting, if possible, where the error is.
For example a trivial error correction algorithm is to transmit data in blocks of say, 8x8, plus a parity bit per column, plus a parity bit per row. Then you check the parity for the rows and columns, and from that determine where the error is. Then you flip the bit at that location, and voila, the original data.
Sane implementations can detect when there are too many errors to correct the information. For example, it may detect and correct 1 bit in 16, and detect but not correct 2 bits in 16. In the later case it drops it, asks for a retransmit, etc.
But digital data generally includes more than the raw audio/video, so yes, if the cable is so incredibly horrible that the error correction can't deal with it, then that'll result in a very obvious malfunction, and not silent and almost unnoticeable data corruption.
If what you wanted is an open development platform, why did you go with Apple, especially when they made their intentions quite explicit?
It's not like there's lack of choice in the area. If you want complete openness, there's OpenMoko. If that's too hardcore, there's plenty other hardware that'll run whatever you like on it.
I think that the GPLv2 and GPLv3 are the same underlying concept, except for GPLv3 closes some loopholes.
I'd choose based on compatibility issues. If possible, GPLv3, if that would cause a compatibility issue, then GPLv2.
GPL and BSD.
IMO, those represent very well the two different approaches to the problem. The rest are a needless complication. Besides, their meaning and implications are understood very well, so I don't see what Google is going to achieve by creating their own.
As a furry, I'm not really sure.
Nobody seems to agree on what exactly it means. It seems to include from people who just like to happen anthro art or cartoons, to people completely obsessed with the subject.
I would say "furry" is a very fuzzy (no pun intended) term, like "geek" for instance. It has certain connotations, hints at certain things, points at a range of likely interests, but is so unspecific that it doesn't really say all that much about the person it's applied to.
You could try checking the definition on WikiFur to see what kind of things may be involved.
Why? Offline IMAP is just like normal IMAP, except messages are cached locally. All the mail is still on the server, except that instead of loading the message only when I click on it, they're downloaded and kept on the disk automatically. I can access it through IMAP over SSL, through SSH, and through a plain web interface.
Beagle, Meta Tracker, whatever. There's more than one app capable of it, and I even rolled my own program of the sort already, could use that. Generally I don't search large amounts of mail anyway, and got everything sorted out in folders. Grep would probably suffice.