Intel offered an enterprise version of the 320 series drive in the 710. The only difference was in expected lifetime, which ripples out to impact write performance guarantees too. See my 320 vs. 710 lifespan article for details.
I've done a lot more of this type of testing than this article, and the only other drive on the market to consider is Seagate's 600 Pro. That hasn't been out long enough for me to have a strong opinion, it seems fine. This article is correct--if you look at drives where there is more field data available, Intel's 320/710 and DCS3500/DCS3700 are the only models that do power protection and have solid lifespan monitoring data too. Crucial's M500 seems to get power protection write, but there's no SMART data on wearout.
The expensive part isn't the capacitor, it's writing the firmware to detect power loss and do the right thing every time. Intel struggled with this on their early 320 series models, with the 8MB bug as one example where things became corrupt when power loss protection didn't work correctly. One of the reasons Intel drives do this well where no one else does is that they had broad consumer adoption of the 320 series to shake out these firmware issues.
The article isn't shilling for Intel, it's pointing out the facts. I publish recommendations on reliable SSDs for PostgreSQL, including a public disk plug pull test guide with similar properties to this one. The only drives that consistently pass my tests and have solid drive wear SMART data are the Intel 320, 710, DCS3500, and DCS3700. Every other inexpensive SSD on the market fails if your criteria for power loss handling and lifespan monitoring are high.
Adding a UPS to one of these systems doesn't really help, it just makes the problem slightly less likely to occur. In environments where data corruption is expensive, UPSes are not reliable enough that it makes the concern go away altogether. The ugly part about how SSDs lose data is that it will corrupt every copy of your data. Having multiple drives in a RAID array for example won't save you; every copy will be toast.
The power loss protection on the Crucial M500 worked fine in my initial tests. It can't be taken seriously as a reliable drive because it doesn't have any SMART data on longevity. There's no way to know when the drive is wearing out, so it's pretty much useless for serious work. The one I bought for testing is in my laptop, it's a fine drive there. See Tech Report for a review complaining about the missing SMART data, I'm not the only one who noticed.
Intel's data on wear is very good, see my look at the 320 vs. 710 lifetime for example. The replacement models, DCS3500 and DCS3700, are even better drives in every way.
Conservative Donkey Kong play does not get you a world record. The bonus for finishing a board goes down faster than you can sustain earning points once all the hammers are gone. That means that the faster you get through the low scoring parts of the board, the higher your score. And with only a limited number of boards to play before the kill screen, the core of top tier play is balancing how much risk you're willing to take on to finish the board with higher points. If you are too conservative, even if you make it to the kill screen you'll lose too many bonus points to approach record territory. That's why runs all the way to the kill screen used to be good if they hit 850,000 points, but nowadays people are pulling over a million through the same path. Players are increasing their risk to get more points.
At the first Kong Off I bought a cheap DVD from Steve Wiebe with a video of his entire world record game, that's available for other people to study. Mixed in with some boring waiting games are amazingly skilled sections of jumping and timing.
The vulnerable ones are the ReadyNAS x86 based models that currently are running firmware with version numbers like 4.2.X. Things like the ReadyNAS Duo are either ARM based with versions 5.3.X, or SPARC based with versions like 4.1.X. The buggy feature here looks like it's only on the more expensive models.
The profit==quality issue is real, and sometimes it's not even a bad way to make decisions. But everyone gets the accountability thing completely backwards.
This mindset of having a vendor to blame is nonsense when talking about one of these programs. Microsoft doesn't care whether any one company is happy with Office or not. If you run into a bug with the program, unless you are a gigantic company they will not to give a single fuck about fixing it for you.
As you noted yourself, it's the contractors in the middle who are getting the vendor abuse. And the reality here is that a vendor who is supporting an open-source solution can genuinely understand and fix things. Any free software consultants of good skill level can fix things when you make them do the vendor dance. The Microsoft resellers of the world get to collect up workarounds, but when the shit hits the fan hard they're powerless to change Microsoft's code.
There are some C-level executives who appreciate having maximum control over their own fate. The easiest way to sell that crowd on free software is to point out they really do have the personal power to make things better when source code is available. I work doing PostgreSQL support, and I fix bugs in the program regularly. When people run into bugs in Oracle/SQL Server/etc, it's far more difficult to get someone at those companies to actually care about them. If a problem is impacting your production system, and Microsoft support doesn't feel like fixing it, you are screwed; there is no plan B for fixing the code.
Large companies require resumes in Word formats because they run them through screening/grading programs before a human even gets to them. By the time an interviewer sees a print quality resume in big business now, it's usually gone a round of computer screening and a round of HR messing with things.
Making the name heavily American centric is a pretty bad idea, given both the lack of worldwide respect for the US's foreign policies and the decreasing percentage the country makes up in the global market. "FreedomOffice" might have worked as a less US-centric name. Maybe give each copy out with a coupon for some Freedom Fries!
Dude, it's 2013 now. A large chunk of the world has a Linux based smartphone in their pocket, Facebook and Google are everywhere with free services having commercial angles, and both ad support and pay to upgrade applications are everywhere. The idea that people will think free software is automatically of poor quality is pretty outdated FUD at this point. "FreeOffice" was the obvious renaming choice, and whatever low quality association "free" has is surely outweighed by the low recognition of "libre" as an uncommon word.
To be fair, even being a fully paying Microsoft lock-in customer doesn't always eliminate this problem. Try to collaborate when one person has Office 2003 and the other Office 2010. Mix in Office 2011 for Mac if you want to make things really interesting. You'll see the same class of formatting glitches as moving between Office and [Open|Libre]Office. You have to exactly match Office version, platform, and sometimes even installed fonts to make collaboration seamless, even when you only have Office to deal with. The sort of Powerpoint flaws you decsribe are certainly worse moving between the free and Microsoft offerings--the one I get all the time is OO not displaying slide titles when I edit them, so I have to fix typos blind--but those same PPT files do milder funny stuff when I move them between Office programs too.
Word hasn't given me as many gross formatting issues, but there are so many badly designed Word documents that even its subtle problems are numerous, from the perspective of Office version compatibility. You can easily destroy some people's careful "let's do all formatting with spaces!" editing stupidity just by switching printers in Word. I've seen plenty of them made in older versions as.doc files where the formatting was eaten trying to move to.docx in newer versions. I know how to carefully use style sheets to minimize these issues, but none of these programs are close to reliable for character-level formatting in non-expert hands. And sometimes, even if you know the program well. I wrote a book using Office 2003 a few years ago. Each chapter passed through a few reviewers using a mix of Office versions and then came back to me with feedback. Every one of those files came back to me messed up in some way. For some of them I had to extract the reviewer feedback but roll back to my original file to get a usefully formatted version again. (Amusingly, the publisher exported out of Word and did all the real typesetting level formatting in another program, so the end result wasn't even that sensitive to formatting changes)
The idea that someone "has to be compatible with the rest of the world" at the formatting level via an Office purchase is really an impossible goal. There are too many version, platform, font, and printer quirks in the world for WYSIWYG document formatting to ever be perfect. OfficeOffice has less such seams than OOOffice, but it doesn't have zero.
Matt is done for. He accepted the EULA when he signed up for Facebook and they know it, and that mistake paints him into an impossible to escape corner. I'm actually glad. I couldn't trust his one-man operation anyway, but his code good enough that most people were unwilling to work on an alternative as long as SocialFixer was available.
I think that keeping this running will take a clean room style approach. One group of people describes the latest ads; someone who hasn't signed up for Facebook publishes open-source code (outside of the US) to squash them based on that spec.
I use Facebook and Adblock and block almost every ad they carry. Right now, that nukes the entire set from the right hand bar. Facebook knows perfectly well how many people block those easy to filter right hand side ads. It's low enough that they don't care, because they have a few ways to give you internal ads instead.
What they are doing now is putting more and more ads in the main section instead. If for example I click to "Like" a post from a group, the minute I do that it rewrites the page to add an inline "If you like that you might also like..." set of ads. These aren't blocked by Adblock because they're all internal links toward other pages on Facebook. As they get more an more infrastructure for that sort of thing, they don't have to leave their regular content to serve you an ad. That makes eliminating ads an increasingly tricky game of detection and rewriting the middle of the main page. And that's exactly the thing that Social Fixer did that Adblock doesn't try. That's why Matt Kruse is being targeted while Adblock isn't. He's the only popular source for code that can block all their ads, even the internally directed ones, and that they won't tolerate.
The Duck brand has nothing to do with the early history of the tape. The whole story is covered in duct tape. The first duck tape was developed by Revolite. The name "duck" was a mix of being made from cotton duck, named based on the "Dutch word doek, which refers to a linen canvas once used for sailors’ white trousers and outerwear", and that the resulting product had a duck-like resistance to water.
"Ductape" was the originally trademarked name for a heat-resistant version of the tape sold for duct work, developed by a different company.
The Duck brand is the current owner of the original tape design. But they didn't acquire that name until almost 50 years after the "duck cotton" name started.
The missing episodes don't start until a few serials in. There are decent quality copies of all of the first three serials floating around. Almost all of the Second Doctor Patrick Troughton episodes are missing though. A few of the key ones are intact--"The Tomb of the Cybermen" and "The War Games" for example--but for the most part his entire run is gone.
If only we had a way to go back and keep this happening, by using some sort of "time machine"...
Your attempt to simplify the math of pre-existing conditions based on conditional probability ignores a behavior pattern your first citation points out: "A Higher Proportion of People with Employer-Sponsored Insurance Have Health Issues". People who have chronic health issues actively make decisions that lead to them being more likely to have health insurance.
As a simple example, imagine someone who is 25 years old and is laid off from work. If they are healthy, they'll likely just drop coverage and be uninsured until they find their next job. But someone with a pre-existing condition will always purchase a COBRA policy if they can afford it. This factor, what the report calls "job lock", is big enough that you can't just apply logic and assume people with pre-existing conditions who have been denied coverage are the only ones who you have to count on this balance.
The idea that no one is losing coverage due to the ACA changes is pretty optimistic too. Here's the first examples I remembered reading this week. I have two immediate family members going through a similar story. They're losing their employer coverage, and it's fair to blame the ACA because makes it easy for employers to abstain from providing coverage. They used to worry about people quitting if they did that, now that's less of a risk. But it's not clear if they will be able to get coverage through their state exchange in time to replace what they're losing.
And then there's North Carolina, where 61% of the counties have one option. You've described the best case scenario. The worst case one involves your employer cancelling your group plan, due solely to issues introduced by the ACA, and then finding out that your state exchange is a monopoly. The anecdotes that balance yours out are reports with a doubling of premiums from the bad combinations possible here.
Insurance providers are effectively cherry-picking only the markets where they can make good profits (ones near the upper limit of how much profit they can make under the ACA). And since some part of insurance company profit is based on extracting more money from the customers, that means the only policies that are going to be available are ones that are not that good of a deal for the buyer. Employers used to be forced into making that work anyway, because a decent health plan was a necessary component to hiring high quality workers. Now they feel it's optional and can drop it if the price is unreasonable, and insurers can drop offering plans if the price is reasonable. That's the reinforcing pair that's led to the North Carolina mess, and I expect to hear a lot more of those stories in the upcoming months.
It turns out that if you search for software help on Google, by the third hit it's already tired of dealing with these bad decisions and suggests you get a Mac instead.
Re:If Gnome 3 team design a car, then...
on
GNOME 3.10 Released
·
· Score: 1
If the GNOME 3 team designed a car, they'd build a moped and say simpler designs are preferred by walkers.
Ooh, can we do car analogies next? GNOME3 routes the exhaust pipe into the ventilation system based on the fact that it's a superior design for a bike.
Partial credit for guessing "rocK", it's actually Amarok. Which also became popular, did a major redesign, was forked by people who liked the old way better, and only recently is starting to be as useful as the original version again. The main difference from the GNOME history is that eventually Amarok realized they should reconsider their older version's features, instead of just continuing to ignore community feedback forever.
Slashdot Mirror
Intel offered an enterprise version of the 320 series drive in the 710. The only difference was in expected lifetime, which ripples out to impact write performance guarantees too. See my 320 vs. 710 lifespan article for details.
I've done a lot more of this type of testing than this article, and the only other drive on the market to consider is Seagate's 600 Pro. That hasn't been out long enough for me to have a strong opinion, it seems fine. This article is correct--if you look at drives where there is more field data available, Intel's 320/710 and DCS3500/DCS3700 are the only models that do power protection and have solid lifespan monitoring data too. Crucial's M500 seems to get power protection write, but there's no SMART data on wearout.
The expensive part isn't the capacitor, it's writing the firmware to detect power loss and do the right thing every time. Intel struggled with this on their early 320 series models, with the 8MB bug as one example where things became corrupt when power loss protection didn't work correctly. One of the reasons Intel drives do this well where no one else does is that they had broad consumer adoption of the 320 series to shake out these firmware issues.
The article isn't shilling for Intel, it's pointing out the facts. I publish recommendations on reliable SSDs for PostgreSQL, including a public disk plug pull test guide with similar properties to this one. The only drives that consistently pass my tests and have solid drive wear SMART data are the Intel 320, 710, DCS3500, and DCS3700. Every other inexpensive SSD on the market fails if your criteria for power loss handling and lifespan monitoring are high.
Adding a UPS to one of these systems doesn't really help, it just makes the problem slightly less likely to occur. In environments where data corruption is expensive, UPSes are not reliable enough that it makes the concern go away altogether. The ugly part about how SSDs lose data is that it will corrupt every copy of your data. Having multiple drives in a RAID array for example won't save you; every copy will be toast.
The power loss protection on the Crucial M500 worked fine in my initial tests. It can't be taken seriously as a reliable drive because it doesn't have any SMART data on longevity. There's no way to know when the drive is wearing out, so it's pretty much useless for serious work. The one I bought for testing is in my laptop, it's a fine drive there. See Tech Report for a review complaining about the missing SMART data, I'm not the only one who noticed.
Intel's data on wear is very good, see my look at the 320 vs. 710 lifetime for example. The replacement models, DCS3500 and DCS3700, are even better drives in every way.
Conservative Donkey Kong play does not get you a world record. The bonus for finishing a board goes down faster than you can sustain earning points once all the hammers are gone. That means that the faster you get through the low scoring parts of the board, the higher your score. And with only a limited number of boards to play before the kill screen, the core of top tier play is balancing how much risk you're willing to take on to finish the board with higher points. If you are too conservative, even if you make it to the kill screen you'll lose too many bonus points to approach record territory. That's why runs all the way to the kill screen used to be good if they hit 850,000 points, but nowadays people are pulling over a million through the same path. Players are increasing their risk to get more points.
At the first Kong Off I bought a cheap DVD from Steve Wiebe with a video of his entire world record game, that's available for other people to study. Mixed in with some boring waiting games are amazingly skilled sections of jumping and timing.
The vulnerable ones are the ReadyNAS x86 based models that currently are running firmware with version numbers like 4.2.X. Things like the ReadyNAS Duo are either ARM based with versions 5.3.X, or SPARC based with versions like 4.1.X. The buggy feature here looks like it's only on the more expensive models.
The profit==quality issue is real, and sometimes it's not even a bad way to make decisions. But everyone gets the accountability thing completely backwards.
This mindset of having a vendor to blame is nonsense when talking about one of these programs. Microsoft doesn't care whether any one company is happy with Office or not. If you run into a bug with the program, unless you are a gigantic company they will not to give a single fuck about fixing it for you.
As you noted yourself, it's the contractors in the middle who are getting the vendor abuse. And the reality here is that a vendor who is supporting an open-source solution can genuinely understand and fix things. Any free software consultants of good skill level can fix things when you make them do the vendor dance. The Microsoft resellers of the world get to collect up workarounds, but when the shit hits the fan hard they're powerless to change Microsoft's code.
There are some C-level executives who appreciate having maximum control over their own fate. The easiest way to sell that crowd on free software is to point out they really do have the personal power to make things better when source code is available. I work doing PostgreSQL support, and I fix bugs in the program regularly. When people run into bugs in Oracle/SQL Server/etc, it's far more difficult to get someone at those companies to actually care about them. If a problem is impacting your production system, and Microsoft support doesn't feel like fixing it, you are screwed; there is no plan B for fixing the code.
Large companies require resumes in Word formats because they run them through screening/grading programs before a human even gets to them. By the time an interviewer sees a print quality resume in big business now, it's usually gone a round of computer screening and a round of HR messing with things.
Making the name heavily American centric is a pretty bad idea, given both the lack of worldwide respect for the US's foreign policies and the decreasing percentage the country makes up in the global market. "FreedomOffice" might have worked as a less US-centric name. Maybe give each copy out with a coupon for some Freedom Fries!
Dude, it's 2013 now. A large chunk of the world has a Linux based smartphone in their pocket, Facebook and Google are everywhere with free services having commercial angles, and both ad support and pay to upgrade applications are everywhere. The idea that people will think free software is automatically of poor quality is pretty outdated FUD at this point. "FreeOffice" was the obvious renaming choice, and whatever low quality association "free" has is surely outweighed by the low recognition of "libre" as an uncommon word.
To be fair, even being a fully paying Microsoft lock-in customer doesn't always eliminate this problem. Try to collaborate when one person has Office 2003 and the other Office 2010. Mix in Office 2011 for Mac if you want to make things really interesting. You'll see the same class of formatting glitches as moving between Office and [Open|Libre]Office. You have to exactly match Office version, platform, and sometimes even installed fonts to make collaboration seamless, even when you only have Office to deal with. The sort of Powerpoint flaws you decsribe are certainly worse moving between the free and Microsoft offerings--the one I get all the time is OO not displaying slide titles when I edit them, so I have to fix typos blind--but those same PPT files do milder funny stuff when I move them between Office programs too.
Word hasn't given me as many gross formatting issues, but there are so many badly designed Word documents that even its subtle problems are numerous, from the perspective of Office version compatibility. You can easily destroy some people's careful "let's do all formatting with spaces!" editing stupidity just by switching printers in Word. I've seen plenty of them made in older versions as .doc files where the formatting was eaten trying to move to .docx in newer versions. I know how to carefully use style sheets to minimize these issues, but none of these programs are close to reliable for character-level formatting in non-expert hands. And sometimes, even if you know the program well. I wrote a book using Office 2003 a few years ago. Each chapter passed through a few reviewers using a mix of Office versions and then came back to me with feedback. Every one of those files came back to me messed up in some way. For some of them I had to extract the reviewer feedback but roll back to my original file to get a usefully formatted version again. (Amusingly, the publisher exported out of Word and did all the real typesetting level formatting in another program, so the end result wasn't even that sensitive to formatting changes)
The idea that someone "has to be compatible with the rest of the world" at the formatting level via an Office purchase is really an impossible goal. There are too many version, platform, font, and printer quirks in the world for WYSIWYG document formatting to ever be perfect. OfficeOffice has less such seams than OOOffice, but it doesn't have zero.
Matt is done for. He accepted the EULA when he signed up for Facebook and they know it, and that mistake paints him into an impossible to escape corner. I'm actually glad. I couldn't trust his one-man operation anyway, but his code good enough that most people were unwilling to work on an alternative as long as SocialFixer was available.
I think that keeping this running will take a clean room style approach. One group of people describes the latest ads; someone who hasn't signed up for Facebook publishes open-source code (outside of the US) to squash them based on that spec.
I use Facebook and Adblock and block almost every ad they carry. Right now, that nukes the entire set from the right hand bar. Facebook knows perfectly well how many people block those easy to filter right hand side ads. It's low enough that they don't care, because they have a few ways to give you internal ads instead.
What they are doing now is putting more and more ads in the main section instead. If for example I click to "Like" a post from a group, the minute I do that it rewrites the page to add an inline "If you like that you might also like..." set of ads. These aren't blocked by Adblock because they're all internal links toward other pages on Facebook. As they get more an more infrastructure for that sort of thing, they don't have to leave their regular content to serve you an ad. That makes eliminating ads an increasingly tricky game of detection and rewriting the middle of the main page. And that's exactly the thing that Social Fixer did that Adblock doesn't try. That's why Matt Kruse is being targeted while Adblock isn't. He's the only popular source for code that can block all their ads, even the internally directed ones, and that they won't tolerate.
I predict the way you're using two digits to count the errors is going to turn into a scalability limit.
The Duck brand has nothing to do with the early history of the tape. The whole story is covered in duct tape. The first duck tape was developed by Revolite. The name "duck" was a mix of being made from cotton duck, named based on the "Dutch word doek, which refers to a linen canvas once used for sailors’ white trousers and outerwear", and that the resulting product had a duck-like resistance to water.
"Ductape" was the originally trademarked name for a heat-resistant version of the tape sold for duct work, developed by a different company.
The Duck brand is the current owner of the original tape design. But they didn't acquire that name until almost 50 years after the "duck cotton" name started.
You can pay real money to get The War Machines on DVD.
The missing episodes don't start until a few serials in. There are decent quality copies of all of the first three serials floating around. Almost all of the Second Doctor Patrick Troughton episodes are missing though. A few of the key ones are intact--"The Tomb of the Cybermen" and "The War Games" for example--but for the most part his entire run is gone.
If only we had a way to go back and keep this happening, by using some sort of "time machine"...
Your attempt to simplify the math of pre-existing conditions based on conditional probability ignores a behavior pattern your first citation points out: "A Higher Proportion of People with Employer-Sponsored Insurance Have Health Issues". People who have chronic health issues actively make decisions that lead to them being more likely to have health insurance.
As a simple example, imagine someone who is 25 years old and is laid off from work. If they are healthy, they'll likely just drop coverage and be uninsured until they find their next job. But someone with a pre-existing condition will always purchase a COBRA policy if they can afford it. This factor, what the report calls "job lock", is big enough that you can't just apply logic and assume people with pre-existing conditions who have been denied coverage are the only ones who you have to count on this balance.
The idea that no one is losing coverage due to the ACA changes is pretty optimistic too. Here's the first examples I remembered reading this week. I have two immediate family members going through a similar story. They're losing their employer coverage, and it's fair to blame the ACA because makes it easy for employers to abstain from providing coverage. They used to worry about people quitting if they did that, now that's less of a risk. But it's not clear if they will be able to get coverage through their state exchange in time to replace what they're losing.
My state has over 40 options on the exchange.
And then there's North Carolina, where 61% of the counties have one option. You've described the best case scenario. The worst case one involves your employer cancelling your group plan, due solely to issues introduced by the ACA, and then finding out that your state exchange is a monopoly. The anecdotes that balance yours out are reports with a doubling of premiums from the bad combinations possible here.
Insurance providers are effectively cherry-picking only the markets where they can make good profits (ones near the upper limit of how much profit they can make under the ACA). And since some part of insurance company profit is based on extracting more money from the customers, that means the only policies that are going to be available are ones that are not that good of a deal for the buyer. Employers used to be forced into making that work anyway, because a decent health plan was a necessary component to hiring high quality workers. Now they feel it's optional and can drop it if the price is unreasonable, and insurers can drop offering plans if the price is reasonable. That's the reinforcing pair that's led to the North Carolina mess, and I expect to hear a lot more of those stories in the upcoming months.
It turns out that if you search for software help on Google, by the third hit it's already tired of dealing with these bad decisions and suggests you get a Mac instead.
If the GNOME 3 team designed a car, they'd build a moped and say simpler designs are preferred by walkers.
Ooh, can we do car analogies next? GNOME3 routes the exhaust pipe into the ventilation system based on the fact that it's a superior design for a bike.
Partial credit for guessing "rocK", it's actually Amarok. Which also became popular, did a major redesign, was forked by people who liked the old way better, and only recently is starting to be as useful as the original version again. The main difference from the GNOME history is that eventually Amarok realized they should reconsider their older version's features, instead of just continuing to ignore community feedback forever.
Next GNOME is going politically correct as Magical Little Person.
Pick them a very special apology SHA1 hash.