Mod parent up! I do a lot of video installations all over the world, and no matter what, I can always use VGA.
What resolution are the monitors you're installing? 1024x768? I've set up several 1920*1080 and 1920*1200 monitors for myself, and I very much prefer being able to see the exact pixels on the monitor that the graphics card is trying to send through. For anything that resolution or above, I only use digital (be it DVI, HDMI, etc.).
The world would be a better place if these space cadets would read and understand Kant, or MLK, or Ghandi, rather than trying to find the most expensive way imaginable to kill themselves.
Those writers you mention... do they have some sort of feasible backup plan to asteroids, nuclear war, or other ecological catastrophes making the earth uninhabitable? There is no humanitarianism without humanity... and that is the fundamental reason why space colonization is important.
I am not into databases, so that just sounds like fault tolerance to me. What's wrong with that?
I can't comment on the illegal hostname bit. However, the general idea behind databases is that your data is extremely important. Enter the data in once, use it in any possible way you want, often in ways you had not anticipated at the time of design. The data should be as constrained as possible at the point of entry to ensure that it is consistent and correct - this is an infinitely better time to rectify errors than say, several years after the fact when you are performing operations on the data and the original records may be lost, or the employees are missing or have forgotten the information. From that perspective, if there is ambiguity or error in what you are trying to tell the database to do with the data, it MUST fail (with a helpful error message) rather than guess what you are trying to do in the name of "fault tolerance".
Sometimes I think the best thing that could possibly happen for the human race is to be invaded by aliens with far superior weaponry. Maybe then we could get along?
Isn't it obvious that this has already happened, though not as you suggest? Put yourself in their position. Do you think the aliens are going to send some generation ship with a thousand crew members aboard and take us out, 1 vs 10 million, conquistador style? No, they are going to use a far more sophisticated approach than that - why kick over the beehive when you can just lay down a little smoke and lift out all that honey. The fact that we expend orders of more magnitude more GDP on attacking each other than exploring what's out there... DIVIDE AND CONQUER my friend. All you have to do is trace the levels of NASA funding back to 1966 when it halved and never recovered to pinpoint the exact year the alien occupation began.
FWIW I agree with most of what you say. I think for SME, HDD is definitely the way to go, put them in silicone covers in a padded bag if you are worried about dropping them. There should be redundancy too, if dropping your only set of backup HDDs is going to wipe out your backups you are doing it wrong.
There is still silent data corruption to worry about - tapes take care of this. If you use HDDs, using ZFS will solve that problem. The checksums will let you know if there is an error and automatically fix it if you have configured enough redundancy into the system. Which you can do easily enough; RAIDZ, mirror or on the filesystem level itself set copies=2+. If you scrub after every transfer, you will know the moment you start seeing bad sectors to replace the HDD.
If your primary systems are running ZFS as well, you won't get silent data corruption being introduced to the backups either. Unless you run without ECC RAM - you might get errant data in RAM that is correctly recorded on the HDD, verifiably so with checksums.
Consider trying to do that with current hard drives in 2040, I doubt they'll even spin up.
If you are using something like ZFS with snapshots on the HDD, you would keep your ancient copies of data in old snapshots (rather than old tapes) but in current HDDs. Backups should use multiple HDDs or sets of HDDs for redundancy anyway. If you drop one of the HDDs in a rotation, the other(s) should be fine. When the individual HDDs are obsolete, you get rid of them. So there you've got equivalent redundancy as tape without the $4k AUD. You don't have to search for drivers, and you get random access restoration capability. It's worth considering if you are cash strapped. And as SSD get cheap enough to use, you can seamlessly transition to using those for your backups, as they are much more vibration tolerant, probably more robust than tape even.
many USB drives are slow
It would be nearly as quick as tape to use e-SATA rather than USB 2.0. Potentially quicker in RAIDZ or RAIDZ2. And cheaper (if you are buying internal HDDs to use).
For most users that can't afford it, it's cheaper and more reliable to run ZFS capable servers in multiple locations with some kind of backup and snapshot jobs running to keep data and history data available.
Depending on how much data needs to be backed up, you could get away with just transporting HDD offsite, perhaps mirrored or RAIDZed. No need to have a separate server, and the offline nature of it means that if you are hacked then your backup is safer than it otherwise would be. As you say, the snapshots will take care of data that is deleted and needs to be recovered down the road, and the checksums will ensure the any silent data corruption will be known about, and if you have redundancy, corrected.
It cost $30 for a caddy that takes the naked 5.25" drives.
Surely you mean 3.5"? The caddy system itself might be 5.25". The only place I can think of to get a 5.25" HDD from would be an attic or a museum.
Does a caddy have any advantages over an e-SATA HDD dock? I like that the latter can be used for multiple machines, but it's more clutter. And I guess that usually machines tend to only have one e-SATA port, so if you had two caddies you could double the throughput. Hmmm. Thanks for the tip, I've got several 5.25" bays doing absolutely nothing.
Actually, there are millions of people who really do need a larger vehicle.
For that reason, the vehicle registration pricing structure in most countries is usually nonsensical. You have to pay additional vehicle registration for each vehicle you own, registration which will cancel out any economic benefit from owning other vehicles with better gas mileage. e.g. if you own a small car, it could conceivably save a lot of money, fuel and CO2 pollution to also own a 100cc motorbike (e.g. Honda Supercub) for local trips in good weather. Depending on where you live this could be up to 90% of the total distance travelled, and only using 1/6 the fuel. This works out to be no additional burden on the infrastructure (usually less), and better for everyone. However, unless the registration policy changes, it is a very economically marginal thing to do.
The same concept applies to owning a big vehicle for work purposes or infrequent use. There is little incentive to buy a smaller car to run around in, because the registration cost is large. It would make more sense to pay one registration per person if you own a vehicle, and that includes any vehicles you own. Or even a subsidy for owning certain classes of vehicles, e.g. 100cc motorbikes. If a registration subsidy for owning a Supercub was almost the depreciation + maintenance cost for owning one, a lot more people would own one, and use one. Every km travelled in such a vehicle is saving 90% of the fuel that would otherwise be used. That one administrative change could potentially shave a double digit percentage reduction off petroleum usage.
This would also mean LESS waste in the production of new cars. A bike like the Supercub uses a fraction of the resources to create as a car. And if the car is only getting half the annual km on it, then it is going to last a lot longer.
Probably my favorite EQ 1 story of addiction (favorite as in the most telling, imo) was 2 wizards (husband and wife) in my guild, who were hooked on meth - started playing EverQuest 1 - and just one day completely forgot to score more meth because they couldn't leave EQ - and quit cold turkey, having completely forgotten about Meth.
O M F G. Something more addictive than a hard core drug, and it's not even a drug. Not even slashdot is that addictive.
That story is awesome. Thanks for sharing it. It only confirms my decision to never, ever play an MMO. An FPS clan is bad enough - being part of one nearly caused me to drop out of college. We competed at a similar level too. With such a situation, not only do you have the draw of the challenge of the online sport, able to be played at a moment's notice, with an intense amount of fun and adrenalin, you've got all your friends clamoring for you to get online with them. I'm guessing that with drugs like meth and heroin, you wouldn't have your friends calling you up as what happens with computer games.
And that's without the draw of an RPG - they endless "one more level" and "one more quest". At least every map on an FPS you have played before. It will still be there the next day. I remember talking to graduated engineers that had lost their jobs because of addiction to Evercrack. No thanks. It was telling that most of the members of another FPS clan we were friends with dropped out of college the next year. And these are people who would probably be at the top of their field and earning good money if they were born say, 10 years earlier.
A key clue to this being a joke is the use of the word 'fail' which these days is often associated with LOLcats. Those damn cats have raised the use of deliberately bad grammar to an artform in and of itself.
I did give the meaning of his post a good deal of thought before replying. I am familiar with LOLcats, LOLCODE and more than my share of despair.com inspired "fail" jpegs. Whether his reply was a joke or not could go either way. On one hand, structuring the response so that the "joke" is where a punchline would be is one clue. The "fail" is another minor indication, but not enough to sway. If he had made the degree of epicicity of the fail explicit (even in a binary fashion), that would be sufficient for me to conclude that it was a joke - but he didn't.
OTOH, there was nothing particularly egregious about the OP's grammar or post to warrant the overly pedantic and snarky reply. If the OP hadn't used the semicolon, there wouldn't have been enough of a pause for "unlike yours truly." to have much comedic effect. While not Strunk and White kosher, surely it's acceptable for slashdot. At this point, you have a pointless post and what may or may not be a joke that succeeds or fails based on whether the OP has ridiculously bad grammar. Which equates to pointless post + unfunny joke (fail), or pointless post + unintentional grammatical error in the context of a grammatical correction (epic fail). Which in turn further condenses down to pointless post + fail, and if you consider a pointless post to be a fail, then fail + fail = fail, QED.
This is why I long ago resolved to never, ever, ever correct someone else's grammar on slashdot. The risk in inadvertently failing to grammar is unacceptable.
Jupiter only has one season. It is called Crush-To-Death-And-Fry-With-Radiation.
I agree - of all the planets, Jupiter does suck the most. A couple planets further though, and it's arguably more comfortable and inviting... though not much light reaches there.
I'd rather read about other people's processes than anything else.
I do agree. My only problem with this thread is that so far no one has come up with a magical way to deal with this problem - the only real answer I've seen so far (and it's unstated) is "man up and deal with it". If you are the sort of person who needs to put an end statement right after every begin statement just so you won't forget to put it in after you've nutted out what should go in between, having tidy little application specific keyrings just won't cut it, no matter how much mangling they do to your pockets. If simply remembering "wallet, keys, phone" is almost too taxing for your organizational skills, guaranteed the only time you will remember to bring your post office box key with you when you leave the house is the moment when you pull up at the post office. And by the time you get home you will have forgotten what it was exactly that you forgot before.
The other solution is to leave your house or mother's basement so few times per year that most of the time your keys live on your desk. This way, your clothes will wear out faster than your pockets. If you intend to maintain the practice of living in mom's basement, another poster had an excellent suggestion to keep your keys ensconced in a snot-rag. This will be at least as effective a ward against future girlfriends/wives as garlic is to vampires. For extra efficacy, make sure the handkerchief emblazoned with either the D&D or Star Trek logo, your choice. You can't be too careful.
Does the "derivative work" function as a substitute for the original work? If it serves a different function, it should be considered fair use.
I would say no. The derivative work function is power. If all we have is the power function, we can't integrate back to find the original work function because we are now missing a constant. It will therefore not function as a substitute. QED
Your complaints about bad pixels - and your refusal to accept even one - is the best explanation among all these comments as to why screen resolution stopped scaling.
Yes. It's all about the yields. AFAIK, there is no way to bin screens with dead pixels (other than "good" or "trash"). You can't just allocate the bad pixels away with a low level format, or bin them - no one wants that screen with the dead/stuck pixels. There is also probably some relation between DPI and error rate that is greater than 1:1.
The 2560 x 1440 27" monitors aren't too bad, either.
Except that they cost 3 times as much as the 24 inch 1920x1200. And not sure about the others, but the Dell U2711 doesn't pivot, so the stand is going to cost extra. Not sure if any others exist yet. And configuration might be a pain - it's very easy with the 1920x1200 being supported for so long. I'll let the early adopters work out the bugs and bring the price down for the next few years.
You know what, you almost have a point. What's with some websites requiring more than 1200 pixels in width? A portrait setup works fine for what I do, but I hope the trend does not continue.
Thanks for looking at each of those links. If this is the best FUD that can be brought to bear on ZFS, it's actually very encouraging. I searched myself for horror stories a few weeks ago, and did not find any that I thought were conclusive.
I've been using FreeBSD for several months now, specifically for ZFS. The more I play with it, the more I like it. Of course, if it wasn't for the few years of running Ubuntu, I wouldn't have built up the skillset or the patience to tinker with things until they work, which is what I've had to do (not so much ZFS, but the FreeBSD side of it).
As part of the design process, Jeff Bonwick questioned pretty much every convention that went with filesystem design, threw out everything that no long applied, and instituted everything he thought was a good idea. It shows. Dealing with ZFS is different, but in a much better way. The most beautiful thing is the checksums/hashes for every block, so that you KNOW when something has corrupted, and more importantly, you should have a backup, and hopefully, some redundancy so fixing it is as simple as swapping in another drive. Why should we, in 2010, when prices per gigabyte are dirt cheap, be still dealing with silent data corruption? There is no reason for it. Everywhere I look with ZFS shows that it has been sensibly architected.
Slashdot Mirror
What resolution are the monitors you're installing? 1024x768? I've set up several 1920*1080 and 1920*1200 monitors for myself, and I very much prefer being able to see the exact pixels on the monitor that the graphics card is trying to send through. For anything that resolution or above, I only use digital (be it DVI, HDMI, etc.).
If you meet him some day, and you think this stuff is worth it, buy him a beer.
Those writers you mention... do they have some sort of feasible backup plan to asteroids, nuclear war, or other ecological catastrophes making the earth uninhabitable? There is no humanitarianism without humanity... and that is the fundamental reason why space colonization is important.
I can't comment on the illegal hostname bit. However, the general idea behind databases is that your data is extremely important. Enter the data in once, use it in any possible way you want, often in ways you had not anticipated at the time of design. The data should be as constrained as possible at the point of entry to ensure that it is consistent and correct - this is an infinitely better time to rectify errors than say, several years after the fact when you are performing operations on the data and the original records may be lost, or the employees are missing or have forgotten the information. From that perspective, if there is ambiguity or error in what you are trying to tell the database to do with the data, it MUST fail (with a helpful error message) rather than guess what you are trying to do in the name of "fault tolerance".
Isn't it obvious that this has already happened, though not as you suggest? Put yourself in their position. Do you think the aliens are going to send some generation ship with a thousand crew members aboard and take us out, 1 vs 10 million, conquistador style? No, they are going to use a far more sophisticated approach than that - why kick over the beehive when you can just lay down a little smoke and lift out all that honey. The fact that we expend orders of more magnitude more GDP on attacking each other than exploring what's out there... DIVIDE AND CONQUER my friend. All you have to do is trace the levels of NASA funding back to 1966 when it halved and never recovered to pinpoint the exact year the alien occupation began.
http://en.wikipedia.org/wiki/NASA_Budget#Annual_budget.2C_1958-2009
Oh shit... it appears that they have triangulated my positJDGPJ&*GP$J#Jm{V#WJ9{W
NO CARRIER
FWIW I agree with most of what you say. I think for SME, HDD is definitely the way to go, put them in silicone covers in a padded bag if you are worried about dropping them. There should be redundancy too, if dropping your only set of backup HDDs is going to wipe out your backups you are doing it wrong.
There is still silent data corruption to worry about - tapes take care of this. If you use HDDs, using ZFS will solve that problem. The checksums will let you know if there is an error and automatically fix it if you have configured enough redundancy into the system. Which you can do easily enough; RAIDZ, mirror or on the filesystem level itself set copies=2+. If you scrub after every transfer, you will know the moment you start seeing bad sectors to replace the HDD.
If your primary systems are running ZFS as well, you won't get silent data corruption being introduced to the backups either. Unless you run without ECC RAM - you might get errant data in RAM that is correctly recorded on the HDD, verifiably so with checksums.
If you are using something like ZFS with snapshots on the HDD, you would keep your ancient copies of data in old snapshots (rather than old tapes) but in current HDDs. Backups should use multiple HDDs or sets of HDDs for redundancy anyway. If you drop one of the HDDs in a rotation, the other(s) should be fine. When the individual HDDs are obsolete, you get rid of them. So there you've got equivalent redundancy as tape without the $4k AUD. You don't have to search for drivers, and you get random access restoration capability. It's worth considering if you are cash strapped. And as SSD get cheap enough to use, you can seamlessly transition to using those for your backups, as they are much more vibration tolerant, probably more robust than tape even.
It would be nearly as quick as tape to use e-SATA rather than USB 2.0. Potentially quicker in RAIDZ or RAIDZ2. And cheaper (if you are buying internal HDDs to use).
Depending on how much data needs to be backed up, you could get away with just transporting HDD offsite, perhaps mirrored or RAIDZed. No need to have a separate server, and the offline nature of it means that if you are hacked then your backup is safer than it otherwise would be. As you say, the snapshots will take care of data that is deleted and needs to be recovered down the road, and the checksums will ensure the any silent data corruption will be known about, and if you have redundancy, corrected.
http://hothardware.com/Articles/Definitive-2TB-Hard-Drive-Roundup/?page=7
You appear to be right. The best write average is about 100MB/s. It's the reads that are near 120.
Surely you mean 3.5"? The caddy system itself might be 5.25". The only place I can think of to get a 5.25" HDD from would be an attic or a museum.
Does a caddy have any advantages over an e-SATA HDD dock? I like that the latter can be used for multiple machines, but it's more clutter. And I guess that usually machines tend to only have one e-SATA port, so if you had two caddies you could double the throughput. Hmmm. Thanks for the tip, I've got several 5.25" bays doing absolutely nothing.
For that reason, the vehicle registration pricing structure in most countries is usually nonsensical. You have to pay additional vehicle registration for each vehicle you own, registration which will cancel out any economic benefit from owning other vehicles with better gas mileage. e.g. if you own a small car, it could conceivably save a lot of money, fuel and CO2 pollution to also own a 100cc motorbike (e.g. Honda Supercub) for local trips in good weather. Depending on where you live this could be up to 90% of the total distance travelled, and only using 1/6 the fuel. This works out to be no additional burden on the infrastructure (usually less), and better for everyone. However, unless the registration policy changes, it is a very economically marginal thing to do.
The same concept applies to owning a big vehicle for work purposes or infrequent use. There is little incentive to buy a smaller car to run around in, because the registration cost is large. It would make more sense to pay one registration per person if you own a vehicle, and that includes any vehicles you own. Or even a subsidy for owning certain classes of vehicles, e.g. 100cc motorbikes. If a registration subsidy for owning a Supercub was almost the depreciation + maintenance cost for owning one, a lot more people would own one, and use one. Every km travelled in such a vehicle is saving 90% of the fuel that would otherwise be used. That one administrative change could potentially shave a double digit percentage reduction off petroleum usage.
This would also mean LESS waste in the production of new cars. A bike like the Supercub uses a fraction of the resources to create as a car. And if the car is only getting half the annual km on it, then it is going to last a lot longer.
O M F G. Something more addictive than a hard core drug, and it's not even a drug. Not even slashdot is that addictive.
That story is awesome. Thanks for sharing it. It only confirms my decision to never, ever play an MMO. An FPS clan is bad enough - being part of one nearly caused me to drop out of college. We competed at a similar level too. With such a situation, not only do you have the draw of the challenge of the online sport, able to be played at a moment's notice, with an intense amount of fun and adrenalin, you've got all your friends clamoring for you to get online with them. I'm guessing that with drugs like meth and heroin, you wouldn't have your friends calling you up as what happens with computer games.
And that's without the draw of an RPG - they endless "one more level" and "one more quest". At least every map on an FPS you have played before. It will still be there the next day. I remember talking to graduated engineers that had lost their jobs because of addiction to Evercrack. No thanks. It was telling that most of the members of another FPS clan we were friends with dropped out of college the next year. And these are people who would probably be at the top of their field and earning good money if they were born say, 10 years earlier.
http://www.silentpcreview.com/article986-page6.html
http://www.jonnyguru.com/modules.php?name=NDReviews&op=Story5&reid=169
Well played sir, well played.
gj, lol
I did give the meaning of his post a good deal of thought before replying. I am familiar with LOLcats, LOLCODE and more than my share of despair.com inspired "fail" jpegs. Whether his reply was a joke or not could go either way. On one hand, structuring the response so that the "joke" is where a punchline would be is one clue. The "fail" is another minor indication, but not enough to sway. If he had made the degree of epicicity of the fail explicit (even in a binary fashion), that would be sufficient for me to conclude that it was a joke - but he didn't.
OTOH, there was nothing particularly egregious about the OP's grammar or post to warrant the overly pedantic and snarky reply. If the OP hadn't used the semicolon, there wouldn't have been enough of a pause for "unlike yours truly." to have much comedic effect. While not Strunk and White kosher, surely it's acceptable for slashdot. At this point, you have a pointless post and what may or may not be a joke that succeeds or fails based on whether the OP has ridiculously bad grammar. Which equates to pointless post + unfunny joke (fail), or pointless post + unintentional grammatical error in the context of a grammatical correction (epic fail). Which in turn further condenses down to pointless post + fail, and if you consider a pointless post to be a fail, then fail + fail = fail, QED.
This is why I long ago resolved to never, ever, ever correct someone else's grammar on slashdot. The risk in inadvertently failing to grammar is unacceptable.
I agree - of all the planets, Jupiter does suck the most. A couple planets further though, and it's arguably more comfortable and inviting... though not much light reaches there.
just get married instead?
I do agree. My only problem with this thread is that so far no one has come up with a magical way to deal with this problem - the only real answer I've seen so far (and it's unstated) is "man up and deal with it". If you are the sort of person who needs to put an end statement right after every begin statement just so you won't forget to put it in after you've nutted out what should go in between, having tidy little application specific keyrings just won't cut it, no matter how much mangling they do to your pockets. If simply remembering "wallet, keys, phone" is almost too taxing for your organizational skills, guaranteed the only time you will remember to bring your post office box key with you when you leave the house is the moment when you pull up at the post office. And by the time you get home you will have forgotten what it was exactly that you forgot before.
The other solution is to leave your house or mother's basement so few times per year that most of the time your keys live on your desk. This way, your clothes will wear out faster than your pockets. If you intend to maintain the practice of living in mom's basement, another poster had an excellent suggestion to keep your keys ensconced in a snot-rag. This will be at least as effective a ward against future girlfriends/wives as garlic is to vampires. For extra efficacy, make sure the handkerchief emblazoned with either the D&D or Star Trek logo, your choice. You can't be too careful.
I would say no. The derivative work function is power. If all we have is the power function, we can't integrate back to find the original work function because we are now missing a constant. It will therefore not function as a substitute. QED
Yes. It's all about the yields. AFAIK, there is no way to bin screens with dead pixels (other than "good" or "trash"). You can't just allocate the bad pixels away with a low level format, or bin them - no one wants that screen with the dead/stuck pixels. There is also probably some relation between DPI and error rate that is greater than 1:1.
Except that they cost 3 times as much as the 24 inch 1920x1200. And not sure about the others, but the Dell U2711 doesn't pivot, so the stand is going to cost extra. Not sure if any others exist yet. And configuration might be a pain - it's very easy with the 1920x1200 being supported for so long. I'll let the early adopters work out the bugs and bring the price down for the next few years.
You know what, you almost have a point. What's with some websites requiring more than 1200 pixels in width? A portrait setup works fine for what I do, but I hope the trend does not continue.
I've been using FreeBSD for several months now, specifically for ZFS. The more I play with it, the more I like it. Of course, if it wasn't for the few years of running Ubuntu, I wouldn't have built up the skillset or the patience to tinker with things until they work, which is what I've had to do (not so much ZFS, but the FreeBSD side of it).
As part of the design process, Jeff Bonwick questioned pretty much every convention that went with filesystem design, threw out everything that no long applied, and instituted everything he thought was a good idea. It shows. Dealing with ZFS is different, but in a much better way. The most beautiful thing is the checksums/hashes for every block, so that you KNOW when something has corrupted, and more importantly, you should have a backup, and hopefully, some redundancy so fixing it is as simple as swapping in another drive. Why should we, in 2010, when prices per gigabyte are dirt cheap, be still dealing with silent data corruption? There is no reason for it. Everywhere I look with ZFS shows that it has been sensibly architected.