Nothing is free. It will take time and/or money to acquire and configure the apps that allow you to use cloud storage. Likewise, if you are an app writer, you have to budget for coding to use the cloud storage. In both cases, you have to plan for what to do if that cloud storage is unavailable (either temporarily or permanently).
Just don't forget redundancy. Don't put all your eggs in one basket, and all that.
This makes the use of any such storage even more costly. The end user now has to develop a plan to sync to multiple cloud storage systems (assuming they want the "available anywhere" functionality regardless of the state of a single storage provider), and find apps that support that plan. If apps that do this automatically for all storage providers don't exist on every platform the user needs, then they will have to plan for manual copy of at least some types of data.
If you want to (by way of completely random example) do a translation of Les Miserables and then make an english-language musical out of it, there's nobody to stop you because the original source is long out of copyright.
This is not strictly true, as if there is anything in copyright that was based on the same out-of-copyright source, then you might be in trouble.
This happens a lot with Disney, as they have used many public domain or out-of-copyright works as sources for their movies, and then sue if there is another work that might possibly in some way be confused with the Disney work.
So all of the server manufacturers were willing to give up on the 65% of the server market that is NOT Windows? If you're going to make crap up, at least make it plausible.
In the business world, the vast majority of operating systems being run either don't or soon won't care about the physical hardware because they are running on top of a hypervisor. As long as the hypervisor can secure boot, then that's all that matters.
Now, there are obviously physical servers being used to run various operating systems directly, but again the vast majority in the business world (which is where most hardware dollars are spent) are going to be running Solaris, BSD,, Red Hat, SuSE, Microsoft Windows, etc. In other words, nothing that a business needs to run on physical hardware won't have a signed bootloader available.
It only ever could be done by staff of the airport (or someone who manages to get into the staff-only areas some other way), at least at all airports that I know. The tag gets attached by the service person when you give up your luggage and then you'll not have access to your luggage again until you get it back at your destination airport.
At the Reno, NV, airport when flying Southwest, they attach the tag, but only sometimes will they take the bag. For any item that is moderately heavy (even if well within the weight limits), the passenger must carry the bag to a different location where a "professional" (i.e., strong but not really bright) will lift the bag onto the belt. It sounds like a joke, but that's the system.
So, if the check in desk is busy at all, the "professional" doesn't know if you came from the desk, straight in from outside, or if you had time to tamper with the tag that was placed on the bag by the agents.
Southwest in Reno is strange as far as baggage is concerned in other ways, too, as I have flown in there with bags that they won't let me fly out with unless I waive damage/loss claims because they believe the bag could "come apart" (which, admittedly, it does, but can't accidentally unless the bag were completely destroyed). This same bag has been flown on Southwest to many other destinations with no problem, yet the people in Reno don't like it.
I've never seen a TV, US or otherwise that couldn't play any HDMI source, even if not "proper" to the region.
There are a decent number of TVs sold in the US that can't sync to 50Hz input. Most are fairly cheap, but some are top of the line (like the Panasonic plasmas).
But, since every media player can do a decent frame rate conversion for output, it really doesn't matter. I watch 25 and 50 fps content all the time by letting the media player convert the output to 60fps.
GP used "gerrymander" because of the author's vague feeling that gerrymandering is a bad thing that bad people do.
Wikipedia says:
In the process of setting electoral districts, gerrymandering is a practice that attempts to establish a political advantage for a particular party or group by manipulating district boundaries to create partisan advantaged districts.
So let's try:
In the process of returning search results, gerrymandering is a practice that attempts to establish an advantage for a particular party or group by manipulating the results to create a more positive view of the party or group.
No, it's not the same thing, but the point behind "gerrymandering" is "manipulating borders to gain an advantage", and if you assume that "good" search results and "bad" results are on opposite sides of a border (yes, I know it's really a spectrum and not just a single line, but tell that to someone who has a stake in any emotional argument), then the term isn't really that bad.
And, oh, yeah, by definition, "gerrymandering" is a "bad thing that bad people do'. Re-apportioning districts isn't a bad thing, but doing so to gain a political advantage is.
If in doubt, the determining factor between Jr. and Sr. is "do you ask people questions, or do people ask questions of you?" If your peers are always coming to you for help/advice, you're a Sr. If you're always having to go ask for help from your peers, you're a Jr.
So, what you're saying is that a "Senior" knows (or think they know) everything, while a "Junior" knows they don't know everything and but does know who to ask. Or, it could just be a personality thing. Some people will ask for "help" even if they are pretty sure they know the answer. Maybe they lack self-confidence or maybe they are just very careful.
"Senior" and "Junior" have a lot more to do with how the management of your company sees you. If they expect other people to look to you for guidance, you're a "Senior", while if they expect you to look to someone for guidance, then "Junior" it is. Note that all bets are off if you've only been at a position for a short period of time, because at that point, even if you are the most skilled on the planet at whatever, you are still learning how your new employer does things.
Same as cloning a 10TB disk array (which I already do now).
Sure you can "just wait until it's done", but I sure wouldn't hold my breath. At the relatively insane write rate of 1GB/sec (although many of the disk arrays where I work can handle this on a single stream write), a petabyte will take about 11 days to write.
With the coaster rate on current optical media, I also wouldn't count on this new system completing a write successfully. But if it really works, I'd like to order a few cases of the discs, as I am just starting on a project where we need to back up 40PB of data.
Also, somebody teach this idiot how incredibly difficult it is to adjust data retrieval algorithms is*, especially on Google's scale.
Actually, the algorithms adjust themselves, in real time, all the time, based on trending searches.
In addition, the algorithm is personalized based on your previous searches (although you can turn this off).
So, an attorney general who accepted the auto-complete for "prescription drugs online without a prescription" would be more likely to see that as the first recommendation when subsequently typing "prescription drugs online" into the Google search box.
Otherwise, admins sit with the users and are aware immediately of the user's problems - and vice versa: users are aware of the administrative problems. I have seen that kind of structure implemented only once, in a mid-sized company, and it worked pretty well (from admin's own words).
It would often be good from the administrator's point of view, but likely quite bad for the company overall.
For one thing, there would always be one group that breaks everything they touch, and their sysadmin is working 60 hour weeks, while another group doesn't have many issues and their sysadmin is on Slashdot most of the time. Then, when vacation time comes for a group's sysadmin, nobody else in the company knows the group well enough to fill in, and wouldn't have the time anyway, as they have to tend to their own group.
It's far better to have sysadmins who cross all boundaries all the time, so that they can be familiar with all the projects and back each other up. There's still going to be the "Exchange specialist", or "the guy who knows how antique system works", but that's just the guy who does that best, not the only one who works with it.
In light of recent news about the NSA spying on everyone, these limitations do not fill me with confidence. Put another way, I don't trust the federal government to follow the law anymore.
Oh, I absolutely agree, and that does make my post more wishful thinking than reality.
But, if pre-9/11 law is followed, I really don't have an issue with drones or other high-tech tools in the hands of law enforcement. Get a judge to sign a warrant because you have probable cause of a specific crime, and that's as close to what the founding fathers envisioned as we can reasonably expect.
what happens when they can develop swarming nanobot flying insects with cameras and microphones on them that dont need to charge and are attracted to noise. always swarming above peoples heads and fully autonomous.
It's real simple regardless of the technology. If they have a warrant issued by a real judge for that one specific purpose (which means probable cause for a specific crime), even long-term surveillance that violates privacy is OK. You may not like this, but that's the way it has been for many decades.
On the other hand, no warrant, no privacy-violating surveillance. They can still watch/listen to you when you are in public, but they can't legally listen to your phone calls, listen to conversations inside your house that are not loud enough to be heard without augmentation, etc. What this means is that they can't fly a drone over your house in the middle of your 40 acres of land to watch or listen to you unless they have a warrant. I'd even argue that a drone looking into your fenced backyard that can't otherwise be seen from public property would require a warrant.
That's what I was thinking. So would-be pirates get together, buy a couple copies of the book, compare them for differences, and post a version of the book that combines both alterations in a random method thus ruining the tracking.
The way this system might work is to have 32 possible differences per book, where each difference can be thought of as a single bit (e.g., "is not" in one version and "isn't" in another). The combination of differences would allow 4 billion different versions, more than enough for even the most insane best seller.
So, if you only have 3 versions, you might see as few as two bits of information. You could get lucky and see a lot more, but even so, the publisher could likely localize the leak to a very small number of buyers.
Monty is angry over the fact that he can no longer take the MySQL man pages, modify them so as to apply to MariaDB, and then redistribute them at little or no cost to himself with his product.
Monty already has the last set of MySQL man pages that he would ever need, and can apply all changes to those to make them correct for the changes in MariaDB.
The only situation that the MySQL man pages might be needed is if there is a change made to MySQL by Oracle that would be beneficial to MariaDB. But, if that change isn't GPL, then it doesn't matter if the man page is GPL, as Monty would have to implement the feature without using the original code.
Everyone saw the writing on the wall and switched to MariaDB a few months ago. In for a repeat show?
What do you mean "repeat show"? Nobody is going to switch from MariaDB because of this.
TFA is a MariaDB blog, but it concerns the MySQL documentation/license as published by Oracle, not the MariaDB docs/license. So, this is just more "writing on the wall", but now it's in bold text.
The reason that the others don't work as well is that TiVo have patents on some of the best features. For instance, when you are fast forwarding and hit Play, it goes back a little bit to account for reaction times. No-one else does this because TiVo hold the patent on it.
The non-TiVo DirecTV DVRs do this, but only when you are at the 30x or higher speeds. This is even better as far as I am concerned, as at the slower speeds I can hit play exactly where I want it to start.
I have both a DirecTiVo and DirecTV DVR, and the only thing I really miss on a daily basis from the TiVo UI is the "TiVo-#" shortcuts. I have only had two recordings fail in 9 months, so the reliability of the DirecTV DVR is about the same as the TiVo. And, the DirecTV DVR already has support for drives greater than 2TB. I haven't bothered to upgrade, as the 1TB drive in the unit is only half full, with about 120 hours of HD recordings stored on it.
Lock it in a safe. Nuke the safe from orbit. That's the only way to be sure.
I would assume that a safe can withstand more damage than a refrigerator, and we know a refrigerator can protect its contents from a nuke, so I don't think your idea will work.
What do any of those devices have to do with an actual reliable internet connection?
Even though I have an extremely reliable Internet connection, I still wouldn't buy any gaming system that required a connection to play.
That's because even though I've had less than 12 hours total downtime in the 4 years I've had FiOS (ignoring power outages, which make it hard to play a console, anyway), I suspect that any authentication servers will have a lot more "downtime" (either really down, or too slow to respond so as to not allow me to authenticate) that it would swamp any downtime that was my ISP's fault.
Slavery means you can indiscriminately kill them? I didn't know that.
Generally, yes, it does, as a slave is property.
If you can dispose of a piece of equipment that you don't need or want any more, then you can do the same thing to a slave. As long as you dispose of it in a way that is otherwise legal (e.g., no dumping it on your neighbor's lawn), there is no problem, so you can toss a dead slave in the recycle bin if you can figure out which category they belong to.
For them to be a tracker, they would have to host torrent files, and they haven't in years.
This is completely wrong.
A bittorrent tracker is just a service that responds to requests that contain the hash that identifies the torrent (and optionally statistics about the state of the torrent on the client that is making the request), and returns IP addresses of other clients that have asked to be stored in the tracker. The tracker doesn't need to know anything about the torrent other than the hash. The only information the tracker needs to know about the clients is whether they want to be added to the list of available peers, then it needs their completion percentage.
A bittorrent client connects to a tracker and then connects to peers that the tracker told it about to transfer the actual data described in the torrent file. Once a bittorrent client has information about other peers, it never has to connect to the tracker again, as long as the peers it knows about satisfy the client's requirement for download (availability, speed, etc.).
TPB is just a database of user-generated and uploaded torrent files, a search engine for that database, a modified bittorrent client (that only queries trackers, but does not register as being available for connections), along with the ability to download the torrent file that is found from a search. TPB still stores all torrent files uploaded to it, but only sometimes allows downloading of the file, based on the number of peers that are currently registered with the tracker(s) for that torrent.
TL; DR version: knowing that some IP address is doing one of (hosting torrent files or running a tracker or transferring data using a bittorrent client) doesn't give you any information about whether that IP address is also doing any of the other things.
You can only "download" web pages and magnet links from piratebay now. Not even torrent files are available, other then linked from torcache.net.
TBP still serves torrent files for torrents that don't meet the minimum requirements for seeder and leecher count. Check a link and you will see that the file comes from a.piratebay.sx domain. It's also easy to show that it's not coming from torcache, as there are torrent files available on TPB that aren't on torcache.
I can't recall the exact numbers used as thresholds, but it's about a total of 10 peers.
Hmmm, that's true of course... I forgot RAID controllers.:)
Which is exactly the application that I had in mind, and use.
I've got WD Black 2TB drives in a 7-drive RAID-6 and get around 380MB/sec on a rebuild, while I'm transferring 200-400MB/sec to/from the array over the network cards. With 125MB/sec the theoretical max for 1Gbps, it doesn't take much (any single SSD, or almost any striping of more than 2 mechanical disks, for example) to break that limit.
Slashdot Mirror
If it's free, who cares?
Nothing is free. It will take time and/or money to acquire and configure the apps that allow you to use cloud storage. Likewise, if you are an app writer, you have to budget for coding to use the cloud storage. In both cases, you have to plan for what to do if that cloud storage is unavailable (either temporarily or permanently).
Just don't forget redundancy. Don't put all your eggs in one basket, and all that.
This makes the use of any such storage even more costly. The end user now has to develop a plan to sync to multiple cloud storage systems (assuming they want the "available anywhere" functionality regardless of the state of a single storage provider), and find apps that support that plan. If apps that do this automatically for all storage providers don't exist on every platform the user needs, then they will have to plan for manual copy of at least some types of data.
If you want to (by way of completely random example) do a translation of Les Miserables and then make an english-language musical out of it, there's nobody to stop you because the original source is long out of copyright.
This is not strictly true, as if there is anything in copyright that was based on the same out-of-copyright source, then you might be in trouble.
This happens a lot with Disney, as they have used many public domain or out-of-copyright works as sources for their movies, and then sue if there is another work that might possibly in some way be confused with the Disney work.
So all of the server manufacturers were willing to give up on the 65% of the server market that is NOT Windows? If you're going to make crap up, at least make it plausible.
In the business world, the vast majority of operating systems being run either don't or soon won't care about the physical hardware because they are running on top of a hypervisor. As long as the hypervisor can secure boot, then that's all that matters.
Now, there are obviously physical servers being used to run various operating systems directly, but again the vast majority in the business world (which is where most hardware dollars are spent) are going to be running Solaris, BSD,, Red Hat, SuSE, Microsoft Windows, etc. In other words, nothing that a business needs to run on physical hardware won't have a signed bootloader available.
It only ever could be done by staff of the airport (or someone who manages to get into the staff-only areas some other way), at least at all airports that I know. The tag gets attached by the service person when you give up your luggage and then you'll not have access to your luggage again until you get it back at your destination airport.
At the Reno, NV, airport when flying Southwest, they attach the tag, but only sometimes will they take the bag. For any item that is moderately heavy (even if well within the weight limits), the passenger must carry the bag to a different location where a "professional" (i.e., strong but not really bright) will lift the bag onto the belt. It sounds like a joke, but that's the system.
So, if the check in desk is busy at all, the "professional" doesn't know if you came from the desk, straight in from outside, or if you had time to tamper with the tag that was placed on the bag by the agents.
Southwest in Reno is strange as far as baggage is concerned in other ways, too, as I have flown in there with bags that they won't let me fly out with unless I waive damage/loss claims because they believe the bag could "come apart" (which, admittedly, it does, but can't accidentally unless the bag were completely destroyed). This same bag has been flown on Southwest to many other destinations with no problem, yet the people in Reno don't like it.
Actually, although the ribbon was frustrating at first, the ability to apply 3-4 styles at once WYSIWYG is a great benefit.
This isn't a function of the ribbon, but rather new interaction between the control and the document.
There is no reason the same functionality could not be used with conventional toolbars or floating style windows.
I've never seen a TV, US or otherwise that couldn't play any HDMI source, even if not "proper" to the region.
There are a decent number of TVs sold in the US that can't sync to 50Hz input. Most are fairly cheap, but some are top of the line (like the Panasonic plasmas).
But, since every media player can do a decent frame rate conversion for output, it really doesn't matter. I watch 25 and 50 fps content all the time by letting the media player convert the output to 60fps.
GP used "gerrymander" because of the author's vague feeling that gerrymandering is a bad thing that bad people do.
Wikipedia says:
So let's try:
No, it's not the same thing, but the point behind "gerrymandering" is "manipulating borders to gain an advantage", and if you assume that "good" search results and "bad" results are on opposite sides of a border (yes, I know it's really a spectrum and not just a single line, but tell that to someone who has a stake in any emotional argument), then the term isn't really that bad.
And, oh, yeah, by definition, "gerrymandering" is a "bad thing that bad people do'. Re-apportioning districts isn't a bad thing, but doing so to gain a political advantage is.
If in doubt, the determining factor between Jr. and Sr. is "do you ask people questions, or do people ask questions of you?" If your peers are always coming to you for help/advice, you're a Sr. If you're always having to go ask for help from your peers, you're a Jr.
So, what you're saying is that a "Senior" knows (or think they know) everything, while a "Junior" knows they don't know everything and but does know who to ask. Or, it could just be a personality thing. Some people will ask for "help" even if they are pretty sure they know the answer. Maybe they lack self-confidence or maybe they are just very careful.
"Senior" and "Junior" have a lot more to do with how the management of your company sees you. If they expect other people to look to you for guidance, you're a "Senior", while if they expect you to look to someone for guidance, then "Junior" it is. Note that all bets are off if you've only been at a position for a short period of time, because at that point, even if you are the most skilled on the planet at whatever, you are still learning how your new employer does things.
Just set it off an wait until it's done.
Same as cloning a 10TB disk array (which I already do now).
Sure you can "just wait until it's done", but I sure wouldn't hold my breath. At the relatively insane write rate of 1GB/sec (although many of the disk arrays where I work can handle this on a single stream write), a petabyte will take about 11 days to write.
With the coaster rate on current optical media, I also wouldn't count on this new system completing a write successfully. But if it really works, I'd like to order a few cases of the discs, as I am just starting on a project where we need to back up 40PB of data.
Also, somebody teach this idiot how incredibly difficult it is to adjust data retrieval algorithms is*, especially on Google's scale.
Actually, the algorithms adjust themselves, in real time, all the time, based on trending searches.
In addition, the algorithm is personalized based on your previous searches (although you can turn this off).
So, an attorney general who accepted the auto-complete for "prescription drugs online without a prescription" would be more likely to see that as the first recommendation when subsequently typing "prescription drugs online" into the Google search box.
Otherwise, admins sit with the users and are aware immediately of the user's problems - and vice versa: users are aware of the administrative problems. I have seen that kind of structure implemented only once, in a mid-sized company, and it worked pretty well (from admin's own words).
It would often be good from the administrator's point of view, but likely quite bad for the company overall.
For one thing, there would always be one group that breaks everything they touch, and their sysadmin is working 60 hour weeks, while another group doesn't have many issues and their sysadmin is on Slashdot most of the time. Then, when vacation time comes for a group's sysadmin, nobody else in the company knows the group well enough to fill in, and wouldn't have the time anyway, as they have to tend to their own group.
It's far better to have sysadmins who cross all boundaries all the time, so that they can be familiar with all the projects and back each other up. There's still going to be the "Exchange specialist", or "the guy who knows how antique system works", but that's just the guy who does that best, not the only one who works with it.
In light of recent news about the NSA spying on everyone, these limitations do not fill me with confidence. Put another way, I don't trust the federal government to follow the law anymore.
Oh, I absolutely agree, and that does make my post more wishful thinking than reality.
But, if pre-9/11 law is followed, I really don't have an issue with drones or other high-tech tools in the hands of law enforcement. Get a judge to sign a warrant because you have probable cause of a specific crime, and that's as close to what the founding fathers envisioned as we can reasonably expect.
what happens when they can develop swarming nanobot flying insects with cameras and microphones on them that dont need to charge and are attracted to noise. always swarming above peoples heads and fully autonomous.
It's real simple regardless of the technology. If they have a warrant issued by a real judge for that one specific purpose (which means probable cause for a specific crime), even long-term surveillance that violates privacy is OK. You may not like this, but that's the way it has been for many decades.
On the other hand, no warrant, no privacy-violating surveillance. They can still watch/listen to you when you are in public, but they can't legally listen to your phone calls, listen to conversations inside your house that are not loud enough to be heard without augmentation, etc. What this means is that they can't fly a drone over your house in the middle of your 40 acres of land to watch or listen to you unless they have a warrant. I'd even argue that a drone looking into your fenced backyard that can't otherwise be seen from public property would require a warrant.
That's what I was thinking. So would-be pirates get together, buy a couple copies of the book, compare them for differences, and post a version of the book that combines both alterations in a random method thus ruining the tracking.
The way this system might work is to have 32 possible differences per book, where each difference can be thought of as a single bit (e.g., "is not" in one version and "isn't" in another). The combination of differences would allow 4 billion different versions, more than enough for even the most insane best seller.
So, if you only have 3 versions, you might see as few as two bits of information. You could get lucky and see a lot more, but even so, the publisher could likely localize the leak to a very small number of buyers.
Monty is angry over the fact that he can no longer take the MySQL man pages, modify them so as to apply to MariaDB, and then redistribute them at little or no cost to himself with his product.
Monty already has the last set of MySQL man pages that he would ever need, and can apply all changes to those to make them correct for the changes in MariaDB.
The only situation that the MySQL man pages might be needed is if there is a change made to MySQL by Oracle that would be beneficial to MariaDB. But, if that change isn't GPL, then it doesn't matter if the man page is GPL, as Monty would have to implement the feature without using the original code.
Everyone saw the writing on the wall and switched to MariaDB a few months ago. In for a repeat show?
What do you mean "repeat show"? Nobody is going to switch from MariaDB because of this.
TFA is a MariaDB blog, but it concerns the MySQL documentation/license as published by Oracle, not the MariaDB docs/license. So, this is just more "writing on the wall", but now it's in bold text.
The reason that the others don't work as well is that TiVo have patents on some of the best features. For instance, when you are fast forwarding and hit Play, it goes back a little bit to account for reaction times. No-one else does this because TiVo hold the patent on it.
The non-TiVo DirecTV DVRs do this, but only when you are at the 30x or higher speeds. This is even better as far as I am concerned, as at the slower speeds I can hit play exactly where I want it to start.
I have both a DirecTiVo and DirecTV DVR, and the only thing I really miss on a daily basis from the TiVo UI is the "TiVo-#" shortcuts. I have only had two recordings fail in 9 months, so the reliability of the DirecTV DVR is about the same as the TiVo. And, the DirecTV DVR already has support for drives greater than 2TB. I haven't bothered to upgrade, as the 1TB drive in the unit is only half full, with about 120 hours of HD recordings stored on it.
We need a "Facebook is not the Internet" campaign.
Facebook is the new AOL.
Lock it in a safe. Nuke the safe from orbit. That's the only way to be sure.
I would assume that a safe can withstand more damage than a refrigerator, and we know a refrigerator can protect its contents from a nuke, so I don't think your idea will work.
What do any of those devices have to do with an actual reliable internet connection?
Even though I have an extremely reliable Internet connection, I still wouldn't buy any gaming system that required a connection to play.
That's because even though I've had less than 12 hours total downtime in the 4 years I've had FiOS (ignoring power outages, which make it hard to play a console, anyway), I suspect that any authentication servers will have a lot more "downtime" (either really down, or too slow to respond so as to not allow me to authenticate) that it would swamp any downtime that was my ISP's fault.
He probably read the comma sutra
That would be the Chicago Manual of Style, right?
Slavery means you can indiscriminately kill them? I didn't know that.
Generally, yes, it does, as a slave is property.
If you can dispose of a piece of equipment that you don't need or want any more, then you can do the same thing to a slave. As long as you dispose of it in a way that is otherwise legal (e.g., no dumping it on your neighbor's lawn), there is no problem, so you can toss a dead slave in the recycle bin if you can figure out which category they belong to.
For them to be a tracker, they would have to host torrent files, and they haven't in years.
This is completely wrong.
A bittorrent tracker is just a service that responds to requests that contain the hash that identifies the torrent (and optionally statistics about the state of the torrent on the client that is making the request), and returns IP addresses of other clients that have asked to be stored in the tracker. The tracker doesn't need to know anything about the torrent other than the hash. The only information the tracker needs to know about the clients is whether they want to be added to the list of available peers, then it needs their completion percentage.
A bittorrent client connects to a tracker and then connects to peers that the tracker told it about to transfer the actual data described in the torrent file. Once a bittorrent client has information about other peers, it never has to connect to the tracker again, as long as the peers it knows about satisfy the client's requirement for download (availability, speed, etc.).
TPB is just a database of user-generated and uploaded torrent files, a search engine for that database, a modified bittorrent client (that only queries trackers, but does not register as being available for connections), along with the ability to download the torrent file that is found from a search. TPB still stores all torrent files uploaded to it, but only sometimes allows downloading of the file, based on the number of peers that are currently registered with the tracker(s) for that torrent.
TL; DR version: knowing that some IP address is doing one of (hosting torrent files or running a tracker or transferring data using a bittorrent client) doesn't give you any information about whether that IP address is also doing any of the other things.
You can only "download" web pages and magnet links from piratebay now. Not even torrent files are available, other then linked from torcache.net.
TBP still serves torrent files for torrents that don't meet the minimum requirements for seeder and leecher count. Check a link and you will see that the file comes from a .piratebay.sx domain. It's also easy to show that it's not coming from torcache, as there are torrent files available on TPB that aren't on torcache.
I can't recall the exact numbers used as thresholds, but it's about a total of 10 peers.
Hmmm, that's true of course... I forgot RAID controllers. :)
Which is exactly the application that I had in mind, and use.
I've got WD Black 2TB drives in a 7-drive RAID-6 and get around 380MB/sec on a rebuild, while I'm transferring 200-400MB/sec to/from the array over the network cards. With 125MB/sec the theoretical max for 1Gbps, it doesn't take much (any single SSD, or almost any striping of more than 2 mechanical disks, for example) to break that limit.