Re:Drive reliability/backups are major factors
on
IDE RAID Examined
·
· Score: 2
I know you're just joking, but 12x burners are a thing of the past. 24x burners can be had for little more than an antiquated 12x and can squeeze out a new CD-R in under 4 minutes. For a paltry $500, you can buy 8 drives and squeeze out a new CD every 30 seconds, which does a better job of keeping the MCSE busy and dramatically reduces the costs of your hypothetical backup!
You're preaching to the choir, my friend... As of last week, I've had a Tivo for two years. I have mine upgaded and pulling data via a serial port. Tivo is decidedly superior to any of the ATI AIW cards and totally blows away this home-grown solution, but if the choice is between a home grown similar to what this guy has and the AIW, I'll go with the AIW.
I completely agree. There's a lot of criticism from people that have never used a Tivo. I used to be one of those on-the-fence people that never thought it was an ok idea, but I would never need one. Pausing live TV just didn't seem like it was that good a feature. I saw the displays at the store, but didn't really think much of it. It wasn't until I actually sat down and used one that I understood its true power. I could sit here all day and explain to you how good a Tivo is but it's nearly impossible for you to gain the understanding you would if you spent one day with one, recording shows and playing with the UI. So before many readers go out and try to build their own Tivos with Mountain Dew and duct tape, I encourage them to find a friend who has one and use it to understand what they should be shooting for. It'd be great to permanently archive shows to CD/DVD and do all kinds of crazy stuff, but sometimes it's just nice to be able to buy a component that works. You can use that old PC with a radio card and spend a few nights writing code to turn it into an awesome clock/radio that will play mp3s or grab streaming radio for the morning news, but it might just be better to buy that $19 Sony clock radio.
Why wouldn't you just use an ATI All-In-Wonder card, then? It's not as nice as Tivo, but you're leaps and bounds ahead of this guy's work: pause live TV, record shows (at much better than 15 fps), instant replay, a graphical UI, and an RF remote. I believe Tivo is available in the UK, but for the rest of the world, this seems like a better intermediate solution then trying to roll-your-own.
Why is this such a difficult idea to understsand on Slashdot? It's not a TCO study that was conducted over 5 years. It's a study to measure the expected TCO using Windows and Linux for 5 years. It's an estimation of the future, not a report of what's happened. This whole story is full of comments wondering how this could have happened when it's painfully obvious that people aren't reading the article.
Looking at the history, it seems that they actually changed their server TODAY from Windows NT to Windows 2000. This morning, their site was so badly Slashdotted, I never saw it. Looks like they migrated it over to Windows 2000 pretty quickly, possibly as a result of the Slashdotting. As far as I can tell, their server is flying right now.
That's not true. Windows XP isn't a server OS. Windows 2000 is still the current server software until Windows.NET Server comes out. The article specifically uses Windows 2000 Server in server roles.
The article is written under the supposition that the situation presented is the situation that's being studied for the TCO. So while your argument is perfectly correct that Linux is priceless to you, it does nothing to deteriorate the article, as you haven't addressed the situation at hand (which is unfortunately nearly impossible because the situation isn't presented in the article!).
Cost is the deciding factor if there's a factor to decide. You didn't have a choice, so there's nothing to consider.
You're also vastly simplifying the cost of training upon users. You and I can figure out any browser that's thrown at us, but you'd be amazed at how stupid people are sometimes.
Your last line says it best. Linux users will gravitate toward certain means of help and Windows users others. IRC is almost inherently more geek oriented, then Usenet, then the web, and then AOL. You can find tons of help on the web and other messageboards. IRC is an old system, but it's still considered underground. Ask 10 people off the street and 9 won't have heard of IRC. 8 won't have heard of Usenet.
Why does Windows 2000 have to be out for five years? The study is for how long it's expected to cost in the NEXT five years, not how much it cost in the last five years! Don't you understand the concept of a TCO study?!
You're completely misunderstanding what a 5 year TCO study is. It's a study of what it's expected to cost in next 5 years. It's the same kind of study that's done all the time, every time you or anyone else does something. That $1,000 TV looks nice, but is it worth it? You do a quick 5 year study in your head of how well you're going to use it over the next 5 years, not how much it was worth to you in the past 5 years (which is presumably $0). The military conducts 25 year studies of how it will use a new piece of armament in the field starting from delivery, not how useful it was 25 years in the past.
A TCO study is designed to help make decisions today about tomorrow, not evaluate the results of yesterday's decisions today. You misread it as being a TCO study that took 5 years to compile.
Find ways that Linux helps them to that in a cost-effective and friendly way and I'm sure more and more business will bite.
Could not have succinctly said it better myself. The truth is that no OS is a panacea. Like screwdrivers and hammers, you have to find the right tool for the right job. There all kinds of factors which can rapidly skew any study or opinion: "Need RAD? Windows 2000 & IIS is the answer. Oh, wait, you just need to serve up static HTML pages? Linux/Apache is what you really need." No one OS/solution is best -- that's why there's more than one!
Well, in all fairness, you didn't have to call Microsoft. You could have done what the OSS developers would have done at that time: scour the Internet for help from other people. Other Usenet groups have been around for a lot longer than the microsoft.public.* groups. And the MSDN subscription is totally optional. I've developed MSDN subscriptions for years and only in the last year received one from my employer. All of the MSDN documentation is available on their website. If the MSDN subscription is paid for, then the cost of development tools is eliminated, so it's one or the other for TCO. Without details, it's impossible to tell how they calculated it, but there are valid counterpoints to the equally valid points that you raised.
Especially in a corporate environment, you're more likely to see old operating systems running old systems because it costs money to migrate and upgrade the system rather than keep the old stuff running. I've seen large companies keep VAX machines up and running from 20 years ago. Sounds silly, but it performs a function and does it well. The cost of development time to migrate it to a spankin' new Windows or Linux machine far outweighs the cost of maintaining the old system once every solar eclipse.
The study didn't take five years to compile. The study analyzes the costs of what it takes to run Windows 2000 for five years -- big difference. Either the quote is poor or the quoter is reading what he wants to read, not what's written.
Sorry, but it doesn't work that way. Moore's law is for processing power, not for work achieved with that power. Otherwise, that will be like saying that computer-based productivity, weather forecasting accuracy, and entertainment value from computer games would also double. Spammer's resources might double in power, but their output might not. Bandwidth is still required (is bandwidth doubling every 18 months? If so, I'm sure that Worldcom has some fiber to sell you), policies change, filters get better, and all kinds of other mitigating factors.
That sounds like it's an excellent project -- good to see a practical implementation of a sound theory. However, nothing precludes a commercial solution from working just as well or being just as successful. There's a low barrier to entry, so ease of use or availabilty may be a greater deciding factor. This one particular one seems like it'll do very well, but it's not a "bright future" for OSS as a whole -- they will all face the challenges that OSS faces today.
You're right -- average fuel economy is not increasing, but manufacturers are definitely introducing more fuel efficient cars. So while they're developing them and producing them, they're not selling nearly as well. CAFE is partially responsible for creating the higher efficiency cars that the public doesn't really want to offset the production of the shitty SUVs that people do want. At least it got the option there. I'm not sure if taxing the hell out of gasoline is the right move. I don't like the idea of sin taxes on cigarettes, either, at ~$8/pack in NYC.
You'll have to explain that one for me. What do you mean by "short-sighted"? In what way?
Perhaps I misspoke, but what I meant was that looking at the prices of gasoline the consumer pays doesn't really yield all the information about why those prices are set that way and how much. You're right in that taxes can be used for all kinds of purposes -- it's just that I don't think that should be the case.:)
I'm a huge fan of cars, so I'd live in Europe just for the cars alone! I don't like SUVs, so they can all go! Just gimme some of those sportscars that you just can't get here! If only that gasoline was a li'l bit cheaper!
My comment wasn't wrong -- but it was admittedly incomplete. SUVs and trucks aren't included in CAFE because when CAFE was first legislated, SUVs didn't exist (they at least weren't named as such). However, now, with top pick-up trucks more than doubling the sales of top consumer vehicles, it's difficult situation. Should vehicles used for commercial traffic be included in CAFE? Should 18 wheelers and farm tractors? Or should it be just those limited to passenger vehicles? How do we count just those sold to consumers? How do states report those vehicles that are commercially registered vs. those that are privately registered? It's a sticky issue with no clear answers. So, yes, I'll concede that the US hasn't covered EVERY single option to completion, but you can't do everything at once. There's always something to criticize about everyone if you hunt long enough (such as why aren't catalytic converters mandatory in Europe?). BTW, road surfacing isn't really considered a viable option for the US because of its high cost per mile. With many more miles of roadway than any other country in the world, it's a daunting task. It's too easy to oversimplify situations in these posts, but it's suffice to say that everyone could be doing more.
Europe uses its fuel tax for all sorts of things, not just related to transportation. As much as 80% of the price per liter that consumers pay goes to the government. That's very high and makes it almost intolerable. Europe could make their fuel much more affordable if they wanted to, but they've chosen to structure their tax system in a manner which makes it difficult. I have admittedly limited understanding of Europe, but all I ever hear is complaints about the cost of driving. I just don't feel that taxing to encourage certain things (such as sin taxes for cigarettes -- apparently it's about $8/pack in NYC) are the proper way to deal with such things.
Yes, SUVs are exempt from CAFE, which is a huge loophole that many SIGs are interested in closing. There's a big controversey about 6,000 lb trucks and the exemptions that they get.
Re:Wrong country
on
239 MPG Car
·
· Score: 4, Interesting
The US is actually very interested in fuel economy: Corporate Average Fuel Economy is very much responsible for pushing manufacturers to increase fuel economy. Note that this isn't just American manufacturers, but all auto manufacturers that sell in the US. In fact, this may be a primary reason that BMW launched the MINI brand: to boost the company's overall fuel economy in its largest single market. With talk of it being pushed over 40 mpg, it could be a real race for fuel economy very soon.
Comparing fuel costs in the US to those in Europe is just short-sighted. I agree that the US has interests in keeping oil petroleum prices down, just as every other country in the world, but Europe has chosen to tax their fuel very heavily, making non-gasoline options more attractive. It's not really an apples-to-apples comparison, as those taxes subsidize all kinds of other efforts and don't really reflect the true cost of driving on the consumer.
Re:You Americans are funny sometimes...
on
239 MPG Car
·
· Score: 2
European taxes on fuels is extraordinary and it makes "us Americans" happy about having low cost fuel. However, the diesel isn't an alternative fuel... most heavy machinery and trucks run on diesel fuel. The alternative fuel is biodiesel and ultra-low sulphur diesel, not regular diesel. Europe has many more diesel-fueled consumer cars and there are only a few offerings available here in the US with diesel, but that may change soon.
Are you sure about that? I've seen the same thing at other companies, but they often switch applications to something else for a different purpose. They will lookup prices in up to three different pricing systems, each accessed via a different method. Just minimize one window and open up another. Besides, there's always the question of is it worth maintaining two different kinds of systems vs keeping Windows up and running everywhere. I think HSBC they're really doing this, then it probably doesn't make sense, but hiring a group of guys to maintain a separate OS is a lot more money than it would cost to just use the same OS. 500 installations of a $100 OS is only $50,000 -- which is not enough money to pay for just one guy. With at least 1 guy in each of 4 shifts (rotating weekends) being paid $50k each (which is more like $75k cost to the company), you'll have to install at least 2,000 installations in order to exceed that amount. I understand your argument, but without knowing the details, it's very difficult to gauge whether or not it's more cost effective for them to do that.
Slashdot Mirror
I know you're just joking, but 12x burners are a thing of the past. 24x burners can be had for little more than an antiquated 12x and can squeeze out a new CD-R in under 4 minutes. For a paltry $500, you can buy 8 drives and squeeze out a new CD every 30 seconds, which does a better job of keeping the MCSE busy and dramatically reduces the costs of your hypothetical backup!
You're preaching to the choir, my friend ... As of last week, I've had a Tivo for two years. I have mine upgaded and pulling data via a serial port. Tivo is decidedly superior to any of the ATI AIW cards and totally blows away this home-grown solution, but if the choice is between a home grown similar to what this guy has and the AIW, I'll go with the AIW.
Aw, damn ... I knew that Preview button was there for a reason!
I completely agree. There's a lot of criticism from people that have never used a Tivo. I used to be one of those on-the-fence people that never thought it was an ok idea, but I would never need one. Pausing live TV just didn't seem like it was that good a feature. I saw the displays at the store, but didn't really think much of it. It wasn't until I actually sat down and used one that I understood its true power. I could sit here all day and explain to you how good a Tivo is but it's nearly impossible for you to gain the understanding you would if you spent one day with one, recording shows and playing with the UI. So before many readers go out and try to build their own Tivos with Mountain Dew and duct tape, I encourage them to find a friend who has one and use it to understand what they should be shooting for. It'd be great to permanently archive shows to CD/DVD and do all kinds of crazy stuff, but sometimes it's just nice to be able to buy a component that works. You can use that old PC with a radio card and spend a few nights writing code to turn it into an awesome clock/radio that will play mp3s or grab streaming radio for the morning news, but it might just be better to buy that $19 Sony clock radio.
Why wouldn't you just use an ATI All-In-Wonder card, then? It's not as nice as Tivo, but you're leaps and bounds ahead of this guy's work: pause live TV, record shows (at much better than 15 fps), instant replay, a graphical UI, and an RF remote. I believe Tivo is available in the UK, but for the rest of the world, this seems like a better intermediate solution then trying to roll-your-own.
Why is this such a difficult idea to understsand on Slashdot? It's not a TCO study that was conducted over 5 years. It's a study to measure the expected TCO using Windows and Linux for 5 years. It's an estimation of the future, not a report of what's happened. This whole story is full of comments wondering how this could have happened when it's painfully obvious that people aren't reading the article.
Looking at the history, it seems that they actually changed their server TODAY from Windows NT to Windows 2000. This morning, their site was so badly Slashdotted, I never saw it. Looks like they migrated it over to Windows 2000 pretty quickly, possibly as a result of the Slashdotting. As far as I can tell, their server is flying right now.
That's not true. Windows XP isn't a server OS. Windows 2000 is still the current server software until Windows.NET Server comes out. The article specifically uses Windows 2000 Server in server roles.
The article is written under the supposition that the situation presented is the situation that's being studied for the TCO. So while your argument is perfectly correct that Linux is priceless to you, it does nothing to deteriorate the article, as you haven't addressed the situation at hand (which is unfortunately nearly impossible because the situation isn't presented in the article!).
Cost is the deciding factor if there's a factor to decide. You didn't have a choice, so there's nothing to consider.
You're also vastly simplifying the cost of training upon users. You and I can figure out any browser that's thrown at us, but you'd be amazed at how stupid people are sometimes.
Your last line says it best. Linux users will gravitate toward certain means of help and Windows users others. IRC is almost inherently more geek oriented, then Usenet, then the web, and then AOL. You can find tons of help on the web and other messageboards. IRC is an old system, but it's still considered underground. Ask 10 people off the street and 9 won't have heard of IRC. 8 won't have heard of Usenet.
Why does Windows 2000 have to be out for five years? The study is for how long it's expected to cost in the NEXT five years, not how much it cost in the last five years! Don't you understand the concept of a TCO study?!
You're completely misunderstanding what a 5 year TCO study is. It's a study of what it's expected to cost in next 5 years. It's the same kind of study that's done all the time, every time you or anyone else does something. That $1,000 TV looks nice, but is it worth it? You do a quick 5 year study in your head of how well you're going to use it over the next 5 years, not how much it was worth to you in the past 5 years (which is presumably $0). The military conducts 25 year studies of how it will use a new piece of armament in the field starting from delivery, not how useful it was 25 years in the past.
A TCO study is designed to help make decisions today about tomorrow, not evaluate the results of yesterday's decisions today. You misread it as being a TCO study that took 5 years to compile.
Well, in all fairness, you didn't have to call Microsoft. You could have done what the OSS developers would have done at that time: scour the Internet for help from other people. Other Usenet groups have been around for a lot longer than the microsoft.public.* groups. And the MSDN subscription is totally optional. I've developed MSDN subscriptions for years and only in the last year received one from my employer. All of the MSDN documentation is available on their website. If the MSDN subscription is paid for, then the cost of development tools is eliminated, so it's one or the other for TCO. Without details, it's impossible to tell how they calculated it, but there are valid counterpoints to the equally valid points that you raised.
Especially in a corporate environment, you're more likely to see old operating systems running old systems because it costs money to migrate and upgrade the system rather than keep the old stuff running. I've seen large companies keep VAX machines up and running from 20 years ago. Sounds silly, but it performs a function and does it well. The cost of development time to migrate it to a spankin' new Windows or Linux machine far outweighs the cost of maintaining the old system once every solar eclipse.
The study didn't take five years to compile. The study analyzes the costs of what it takes to run Windows 2000 for five years -- big difference. Either the quote is poor or the quoter is reading what he wants to read, not what's written.
Sorry, but it doesn't work that way. Moore's law is for processing power, not for work achieved with that power. Otherwise, that will be like saying that computer-based productivity, weather forecasting accuracy, and entertainment value from computer games would also double. Spammer's resources might double in power, but their output might not. Bandwidth is still required (is bandwidth doubling every 18 months? If so, I'm sure that Worldcom has some fiber to sell you), policies change, filters get better, and all kinds of other mitigating factors.
That sounds like it's an excellent project -- good to see a practical implementation of a sound theory. However, nothing precludes a commercial solution from working just as well or being just as successful. There's a low barrier to entry, so ease of use or availabilty may be a greater deciding factor. This one particular one seems like it'll do very well, but it's not a "bright future" for OSS as a whole -- they will all face the challenges that OSS faces today.
Perhaps I misspoke, but what I meant was that looking at the prices of gasoline the consumer pays doesn't really yield all the information about why those prices are set that way and how much. You're right in that taxes can be used for all kinds of purposes -- it's just that I don't think that should be the case.
I'm a huge fan of cars, so I'd live in Europe just for the cars alone! I don't like SUVs, so they can all go! Just gimme some of those sportscars that you just can't get here! If only that gasoline was a li'l bit cheaper!
My comment wasn't wrong -- but it was admittedly incomplete. SUVs and trucks aren't included in CAFE because when CAFE was first legislated, SUVs didn't exist (they at least weren't named as such). However, now, with top pick-up trucks more than doubling the sales of top consumer vehicles, it's difficult situation. Should vehicles used for commercial traffic be included in CAFE? Should 18 wheelers and farm tractors? Or should it be just those limited to passenger vehicles? How do we count just those sold to consumers? How do states report those vehicles that are commercially registered vs. those that are privately registered? It's a sticky issue with no clear answers. So, yes, I'll concede that the US hasn't covered EVERY single option to completion, but you can't do everything at once. There's always something to criticize about everyone if you hunt long enough (such as why aren't catalytic converters mandatory in Europe?). BTW, road surfacing isn't really considered a viable option for the US because of its high cost per mile. With many more miles of roadway than any other country in the world, it's a daunting task. It's too easy to oversimplify situations in these posts, but it's suffice to say that everyone could be doing more.
Europe uses its fuel tax for all sorts of things, not just related to transportation. As much as 80% of the price per liter that consumers pay goes to the government. That's very high and makes it almost intolerable. Europe could make their fuel much more affordable if they wanted to, but they've chosen to structure their tax system in a manner which makes it difficult. I have admittedly limited understanding of Europe, but all I ever hear is complaints about the cost of driving. I just don't feel that taxing to encourage certain things (such as sin taxes for cigarettes -- apparently it's about $8/pack in NYC) are the proper way to deal with such things.
Yes, SUVs are exempt from CAFE, which is a huge loophole that many SIGs are interested in closing. There's a big controversey about 6,000 lb trucks and the exemptions that they get.
The US is actually very interested in fuel economy: Corporate Average Fuel Economy is very much responsible for pushing manufacturers to increase fuel economy. Note that this isn't just American manufacturers, but all auto manufacturers that sell in the US. In fact, this may be a primary reason that BMW launched the MINI brand: to boost the company's overall fuel economy in its largest single market. With talk of it being pushed over 40 mpg, it could be a real race for fuel economy very soon.
Comparing fuel costs in the US to those in Europe is just short-sighted. I agree that the US has interests in keeping oil petroleum prices down, just as every other country in the world, but Europe has chosen to tax their fuel very heavily, making non-gasoline options more attractive. It's not really an apples-to-apples comparison, as those taxes subsidize all kinds of other efforts and don't really reflect the true cost of driving on the consumer.
European taxes on fuels is extraordinary and it makes "us Americans" happy about having low cost fuel. However, the diesel isn't an alternative fuel ... most heavy machinery and trucks run on diesel fuel. The alternative fuel is biodiesel and ultra-low sulphur diesel, not regular diesel. Europe has many more diesel-fueled consumer cars and there are only a few offerings available here in the US with diesel, but that may change soon.
Are you sure about that? I've seen the same thing at other companies, but they often switch applications to something else for a different purpose. They will lookup prices in up to three different pricing systems, each accessed via a different method. Just minimize one window and open up another. Besides, there's always the question of is it worth maintaining two different kinds of systems vs keeping Windows up and running everywhere. I think HSBC they're really doing this, then it probably doesn't make sense, but hiring a group of guys to maintain a separate OS is a lot more money than it would cost to just use the same OS. 500 installations of a $100 OS is only $50,000 -- which is not enough money to pay for just one guy. With at least 1 guy in each of 4 shifts (rotating weekends) being paid $50k each (which is more like $75k cost to the company), you'll have to install at least 2,000 installations in order to exceed that amount. I understand your argument, but without knowing the details, it's very difficult to gauge whether or not it's more cost effective for them to do that.