Interesting, I can't say that I've ever touched Hitachi and it sounds like I was right to be wary of them. Maxtor has always been the trouble maker for me with about an 80% failure rate after 4 years. Western Digital is the primary brand I use and some Seagate. I've only had 1 WD fail in less than 5 years, though I'm not operating on nearly the same scale. Only a paltry 57 computers to backup in a non-intensive environment.
Agreed, there is definitely a data threshold for the cost/benefit calculations. I peg it around 30TB rather than 100 but I also air on the side of caution and an extra layer of redundancy that most would consider excessive.
For a spinning disk setup I'd beg and plead before going to a RAID array. They're wonderful things in the short term but long term there are just too many things that can (and do) go wrong with RAID. I can't argue that they do solve some storage issues that would just not be practical or financially feasible otherwise.
If I may ask, what's brand of HDD that you've seen fail most often at your plant?
Exactly that, high quality tape storage designed for archiving is not cheap. It's in fact quite expensive in comparison. The typical consumer grade tape is what we're talking about here and in an uncontrolled home environment. In that use-case drives typically last about 50% longer than tape. And I completely agree, I've seen more drives fail than tapes which is why I also recommended backing up to drives in triplicate in case of such a failure. This does several things:
1) Protects against mechanical failures 2) Protects against human error (drops/improper handling) 3) In case of mechanical failure across all 3 drives, if the data is important enough, they could be taken to a professional for repair; even if that meant salvaging parts from one drive to fix the other
In a professional archiving environment I 100% agree with you, tape is the best choice for now. In a home environment the quality and control is just not there to make tape as reliable as HDDs.
lol linking to company's claims does not make it true. The data presented at the IEEE storage conference found the actual life of tape is 5-10 years depending on the quality.
Not to mention, CDC budget for 2012: $11,255,301,000. Imagine the budget required for something like this? There's a lot of areas to spend on, this is not one of them.
It wasn't about a specific feature. Yes I did not like it, I found it distracting and it rarely brought up what I wanted or if it did it would be near the bottom or require me to type nearly the entire url anyway. That said, the reason I believe that was their break from what users wanted to what devs wanted is because it was the first major UI change that did not have customization options.
As instagib points out you can now set a flag to turn it off but that removes the urlbar autocomplete entirely. From the start is should have had configuration options as to what you wanted included when you used auto-complete. Some users may just want urls, some may want to include bookmarks, others yet may not want bookmarks but they do want titles to be included. There is just no control over the feature, it's devs spoon feeding what they want to the masses. That's fine in the background/technical end where standards need to be met, UI and behaviour changes should always maintain customization options.
From themes/personas, optional toolbars, addons, the ability to switch button sizes/positions, etc. Firefox has always been about personalization until that feature. Since then FF4 removed text menus affecting the visually impaired and less computer literate, removed small buttons as a UI option for UI minimizers, and removed much of the previous control users had over the UI.
"Heavenly Sword feels like a summer action flick. It's full of nonstop action, and it looks terrific. Unfortunately, it's over far too quickly." - Gamespot
I'm sure there are some like yourself that didn't enjoy the game, that's a given for any game. Overall it got fairly high ratings (8.1 average review on gamespot, 79 metacritic) but that quality didn't translate into sales the way it typically does because everywhere you looked people were saying it was too short at 6-7 hours so not worth the $$$ to pick it up. It ended up with 1-1.5million units sold which is pathetic for the amount of money that was invested in it.
Tape is great as a backup medium there's no denying that, backups are actively refreshed. The question was talking about archiving which is a completely different set of issues. The biggest issue is how long can the object sit on a shelf unused before it starts to corrupt the data. HDD has an active life of 5 years but a shelf life of 10 years, tape has a shelf life of 5 to 10 years. Tape also has the disadvantage of sequential access. It's the preferred method for professional archives because of this fact. It's very easy to transfer all the data sequentially from one tape to another, the data transfer rates are faster, and compression makes a big difference. For home archives though those aren't really considerations, cost, ease of access, and reliability long term are the more important issues.
These guys are so far off it's scary. The fact is there's data readily available which shows that AAA games are the ones that most people complete. L.A. Noire as their example, 16% of people have completed to 100% it despite it being a fairly new game. That doesn't just mean completing the story, that means completing all of the following: - Solve all 21 story cases - Find all 95 vehicles in the game - Read all 13 Newspaper collectibles - Finish all 40 street crimes - Discover all 30 landmarks - Find all 50 hidden film reels
By contrast, 47% of people have done half of the "40 street crimes", 28% have done them all.
These numbers are from a sample size of 34,675 people, though with a bias towards completion.
Looking at the most completed AAA game, Assassin's Creed 2, 78% of people have completed the story, only 35% have completed everything to do in the game (sample size 77,526). Both games are fairly similar in length to do everything.
The fact is people get bored, they get distracted by newer shinier things, and they often have small chunks of time to devote to something. It doesn't mean you should stop making quality games that have length and depth. By example Gears of War, 5% of 105,452 people completed Seriously - 10,000 kills online, requiring a minimum of 127 hours. It could be done in very small chunks though, a couple hours here and there. The campaign as well, 78% of those people completed it despite it being a lengthy campaign.
The design is the big difference. Gears is broken up into acts but then again into many small checkpoints that you can start the game from any one of them. It was also designed to be local and online co-op friendly. L.A. Noir cannot claim the same, it required larger chunks of time, being able to remember where you were in the story and unlike Assassin's Creed 2 is not designed to be able to be able to just pickup and play.
One need only look at Heavenly Sword to see what a major flop short AAA games are, and then again at Enslaved: Odyssey to the West which was equally as short, but with collectibles to attempt to extend gameplay.
Online storage makes no sense. Give data you obviously care about to a company who's going to implement the cheapest "backup" system they can get away with and won't give a crap if they lose your data because they're protected by their terms of service. That and uploading 60gigs of data would take forever let alone multiple sets of 60gigs.
Put it in the cloud! *waves arms like it's something mystical*
Seriously though, there is no great solution. Burned discs separate over time, there's not enough data on SSDs yet but it's not looking promising, platter drives are susceptible to radiation, tape to magnetic fields and degradation. HDD in triplicate, replace every 7-10 years is the "best" method right now. So despite being modded down, serkit is right. Hard drives.
Again, not the same. Microwaves as an example - it's a physical property and the original scientists probably could have thought of dozens of ways they could be harnessed. Each individual invention which harnessed them became patentable from radar to microwave ovens to gps. As an example WiMAX vs Wireless LAN vs Bluetooth vs traditional microwave tower transmission are all different ways of utilizing microwaves to transmit data. Each one is an invention on it's own but would not be possible if someone was granted a patent on something so general as "the transmission of data using microwaves". This is effectively what is happening with software.
Yeah, I put that down to not enough options for modding. I do think they should be voided but not because they're 0s and 1s. Copyright is a far more appropriate form of protection imo.
Being able to be expressed as an equation is not the same as being made up of equations. You can't take your equations and have a physical product at the end, at a certain point you must actually make that object. You can take your equations and make a piece of software though.
Copyright would still apply to software patents so you couldn't take the same code/visual design and create a copy using a different language. But you you could demonstrate that you have a different way of coming about the same function, perhaps a faster way or away that allows additional flexibility or features, then that is something new.
Not the same at all. In the case of software the algorithms at play are merely applications of well established math. Someone else figured it out years ago and they are merely applying it in their code.
In the case of physical objects you are creating unique parts or structures to serve a function. These structures cannot have existed before.
If you were to apply that standard to software then the unique way that you build the software to serve the function could be patented but it would not prevent someone else from creating another unique way to build the software to serve that same function. Instead the patents are patenting the function no matter how it was derived.
Slashdot Mirror
That was the launch price of the PS3... you're an idiot. http://www.urbandictionary.com/define.php?term=you're%20an%20idiot
Interesting, I can't say that I've ever touched Hitachi and it sounds like I was right to be wary of them. Maxtor has always been the trouble maker for me with about an 80% failure rate after 4 years. Western Digital is the primary brand I use and some Seagate. I've only had 1 WD fail in less than 5 years, though I'm not operating on nearly the same scale. Only a paltry 57 computers to backup in a non-intensive environment.
wow... can't believe I just typed that. "err on the side of caution".
Agreed, there is definitely a data threshold for the cost/benefit calculations. I peg it around 30TB rather than 100 but I also air on the side of caution and an extra layer of redundancy that most would consider excessive.
For a spinning disk setup I'd beg and plead before going to a RAID array. They're wonderful things in the short term but long term there are just too many things that can (and do) go wrong with RAID. I can't argue that they do solve some storage issues that would just not be practical or financially feasible otherwise.
If I may ask, what's brand of HDD that you've seen fail most often at your plant?
Exactly that, high quality tape storage designed for archiving is not cheap. It's in fact quite expensive in comparison. The typical consumer grade tape is what we're talking about here and in an uncontrolled home environment. In that use-case drives typically last about 50% longer than tape. And I completely agree, I've seen more drives fail than tapes which is why I also recommended backing up to drives in triplicate in case of such a failure. This does several things:
1) Protects against mechanical failures
2) Protects against human error (drops/improper handling)
3) In case of mechanical failure across all 3 drives, if the data is important enough, they could be taken to a professional for repair; even if that meant salvaging parts from one drive to fix the other
In a professional archiving environment I 100% agree with you, tape is the best choice for now. In a home environment the quality and control is just not there to make tape as reliable as HDDs.
lol linking to company's claims does not make it true. The data presented at the IEEE storage conference found the actual life of tape is 5-10 years depending on the quality.
Not to mention, CDC budget for 2012: $11,255,301,000. Imagine the budget required for something like this? There's a lot of areas to spend on, this is not one of them.
I also find a couple other things curious:
1) Epic starts by attacking the person not the argument
2) Epic goes on a random rant about there being no definition of "tracking"
ooo - can I have some of this magic money that appears out of thin air?
It wasn't about a specific feature. Yes I did not like it, I found it distracting and it rarely brought up what I wanted or if it did it would be near the bottom or require me to type nearly the entire url anyway. That said, the reason I believe that was their break from what users wanted to what devs wanted is because it was the first major UI change that did not have customization options.
As instagib points out you can now set a flag to turn it off but that removes the urlbar autocomplete entirely. From the start is should have had configuration options as to what you wanted included when you used auto-complete. Some users may just want urls, some may want to include bookmarks, others yet may not want bookmarks but they do want titles to be included. There is just no control over the feature, it's devs spoon feeding what they want to the masses. That's fine in the background/technical end where standards need to be met, UI and behaviour changes should always maintain customization options.
From themes/personas, optional toolbars, addons, the ability to switch button sizes/positions, etc. Firefox has always been about personalization until that feature. Since then FF4 removed text menus affecting the visually impaired and less computer literate, removed small buttons as a UI option for UI minimizers, and removed much of the previous control users had over the UI.
Overwhelmingly it was:
"Heavenly Sword feels like a summer action flick. It's full of nonstop action, and it looks terrific. Unfortunately, it's over far too quickly." - Gamespot
http://ps3.ign.com/articles/815/815721p3.html
http://www.google.ca/search?hl=en&q=heavenly+sword+"too+short"
I'm sure there are some like yourself that didn't enjoy the game, that's a given for any game. Overall it got fairly high ratings (8.1 average review on gamespot, 79 metacritic) but that quality didn't translate into sales the way it typically does because everywhere you looked people were saying it was too short at 6-7 hours so not worth the $$$ to pick it up. It ended up with 1-1.5million units sold which is pathetic for the amount of money that was invested in it.
Pretty much. Mozilla stopped being about the user with the crapsome bar.
Tape is great as a backup medium there's no denying that, backups are actively refreshed. The question was talking about archiving which is a completely different set of issues. The biggest issue is how long can the object sit on a shelf unused before it starts to corrupt the data. HDD has an active life of 5 years but a shelf life of 10 years, tape has a shelf life of 5 to 10 years. Tape also has the disadvantage of sequential access. It's the preferred method for professional archives because of this fact. It's very easy to transfer all the data sequentially from one tape to another, the data transfer rates are faster, and compression makes a big difference. For home archives though those aren't really considerations, cost, ease of access, and reliability long term are the more important issues.
These guys are so far off it's scary. The fact is there's data readily available which shows that AAA games are the ones that most people complete. L.A. Noire as their example, 16% of people have completed to 100% it despite it being a fairly new game. That doesn't just mean completing the story, that means completing all of the following:
- Solve all 21 story cases
- Find all 95 vehicles in the game
- Read all 13 Newspaper collectibles
- Finish all 40 street crimes
- Discover all 30 landmarks
- Find all 50 hidden film reels
By contrast, 47% of people have done half of the "40 street crimes", 28% have done them all.
These numbers are from a sample size of 34,675 people, though with a bias towards completion.
Looking at the most completed AAA game, Assassin's Creed 2, 78% of people have completed the story, only 35% have completed everything to do in the game (sample size 77,526). Both games are fairly similar in length to do everything.
The fact is people get bored, they get distracted by newer shinier things, and they often have small chunks of time to devote to something. It doesn't mean you should stop making quality games that have length and depth. By example Gears of War, 5% of 105,452 people completed Seriously - 10,000 kills online, requiring a minimum of 127 hours. It could be done in very small chunks though, a couple hours here and there. The campaign as well, 78% of those people completed it despite it being a lengthy campaign.
The design is the big difference. Gears is broken up into acts but then again into many small checkpoints that you can start the game from any one of them. It was also designed to be local and online co-op friendly. L.A. Noir cannot claim the same, it required larger chunks of time, being able to remember where you were in the story and unlike Assassin's Creed 2 is not designed to be able to be able to just pickup and play.
One need only look at Heavenly Sword to see what a major flop short AAA games are, and then again at Enslaved: Odyssey to the West which was equally as short, but with collectibles to attempt to extend gameplay.
Online storage makes no sense. Give data you obviously care about to a company who's going to implement the cheapest "backup" system they can get away with and won't give a crap if they lose your data because they're protected by their terms of service. That and uploading 60gigs of data would take forever let alone multiple sets of 60gigs.
Put it in the cloud! *waves arms like it's something mystical*
Seriously though, there is no great solution. Burned discs separate over time, there's not enough data on SSDs yet but it's not looking promising, platter drives are susceptible to radiation, tape to magnetic fields and degradation. HDD in triplicate, replace every 7-10 years is the "best" method right now. So despite being modded down, serkit is right. Hard drives.
100% of the internet population is over-opinionated and under-informed
Ditto. I refuse to upgrade until they fix the UI. Once the security updates stop I may have to look at Opera of all things.
Again, not the same. Microwaves as an example - it's a physical property and the original scientists probably could have thought of dozens of ways they could be harnessed. Each individual invention which harnessed them became patentable from radar to microwave ovens to gps. As an example WiMAX vs Wireless LAN vs Bluetooth vs traditional microwave tower transmission are all different ways of utilizing microwaves to transmit data. Each one is an invention on it's own but would not be possible if someone was granted a patent on something so general as "the transmission of data using microwaves". This is effectively what is happening with software.
Exactly!
Yeah, I put that down to not enough options for modding. I do think they should be voided but not because they're 0s and 1s. Copyright is a far more appropriate form of protection imo.
Being able to be expressed as an equation is not the same as being made up of equations. You can't take your equations and have a physical product at the end, at a certain point you must actually make that object. You can take your equations and make a piece of software though.
The 0s and 1s was supposed to be silly not taken seriously. :P
Copyright would still apply to software patents so you couldn't take the same code/visual design and create a copy using a different language. But you you could demonstrate that you have a different way of coming about the same function, perhaps a faster way or away that allows additional flexibility or features, then that is something new.
Not the same at all. In the case of software the algorithms at play are merely applications of well established math. Someone else figured it out years ago and they are merely applying it in their code.
In the case of physical objects you are creating unique parts or structures to serve a function. These structures cannot have existed before.
If you were to apply that standard to software then the unique way that you build the software to serve the function could be patented but it would not prevent someone else from creating another unique way to build the software to serve that same function. Instead the patents are patenting the function no matter how it was derived.