That's a great argument in theory, but unfortunately the reality is that the older proprietary drivers that haven't been supported in half a dozen years still often perform better than the up-to-date open source alternatives (if there even are any non-generic drivers). If your motivation is pragmatism, then you're going to use proprietary in that case.
What exactly is different or more efficient about making the browser parse new HTML tags and incorporate new libraries instead of making the browser run JavaScript every time the page is loaded and relying on third parties to produce browser plug-ins for standard web content?
What did you use to format the partition? XP uses NTFS 5.1. Windows 7 uses NTFS 6.x. NTFS 6 adds transactional NTFS support and support for bitlocker encryption, among other things. If you used XP's setup utility or used a Linux-based utility to format the partition, it simply may not be completely compatible. Remember, XP is not a supported upgrade path for Windows 7.
It's just the the topic itself is a bit snarky. It's a little insulting, as a sysadmin who happily calls himself an IT person, to hear that a developer is insulted by being called an "IT guy". Developers often have the opinion that being a sysadmin is little more than being a janitor. It's very frustrating to be dismissed out-of-hand as "someone who was too stupid to program" by developers who ought to know better. I can forgive most users and most management for their ignorance, but I have a hard time forgiving developers.
That's not true at all. RAM draws a fixed supply of power, but flash certainly doesn't. That's why you're able to unplug a flash drive and have it keep the data!
The only parts constantly in need of power are the controller electronics. The power rating for Intel's enterprise SSDs, for example, is shown as 2.5 Watts active and 0.06 Watts idle. On the newer 34 nm drives, Intel (PDF warning) claims 0.15 W active, 0.075 W idle.
Chromium's power management may not suck. It seems that Chromium is not using EIST, so the processor is always running at 1600 MHz whereas the other distros could scale the processor down to 800 MHz to save power. Given that this system had a SSD, the CPU likely accounts for the vast majority of power consumption.
It's not an argument. It's an expression of the thin-thick cycle that myself and many others have seen undergone in the real world. It's a pendulum. One side is running thin clients and having everything done server-side. The other is thick clients and having as much as possible being client-side. The "prevailing" theory will oscillate about every two or three years or so, IMX.
It's absolutely possible to do it all correctly, but in general management is sold the idea that thin clients are cheaper than desktops -- or at least that's what they seem to remember the most. However, at some point you invariably reach capacity. Either you start running more apps, need to run a new OS or apps that require different operating systems or an OS that uses more resources, etc. You will need to periodically reinvest in your servers. And you'll need to maintain your infrastructure to a higher level of quality because basic client activity is now leveraged over the network. This is all on top of licensing for your thin client software. The problem with needing to invest in servers and networks is the high cost of investment. Buying desktops is really appealing if it means you don't need to upgrade all your switches and cable runs and you don't need to add half a dozen server racks to an already-cramped, power-hungry server room.
It's not that thin clients are bad. It's that "cheaper" doesn't mean cheap, and it certainly doesn't mean you get to neglect IT for half a dozen years. Unfortunately, that's precisely what tends to happen.
Maybe you hadn't noticed, but development of IE7 and IE8 have not been tied to a specific OS at all. IE7 was released before Vista and installs on XP, and IE8 well before Win 7 and that installs on Vista and XP. Microsoft has said that IE9 will be released in 2010, while Windows 8 is set for 2012. IE and Office are both on different development timetables than Windows -- although Office is almost always released 6 to 8 months after a desktop Windows release. Sure, they're linked in some senses because each product has a target platform, but otherwise there is no specific tie-in.
Microsoft's fiscal incentive is to maintain market dominance and some semblance of standards compliance. If they lose too much market share, developers may not create websites to handle IE quirks any longer. Then IE will falter, and MS will not be able to develop web apps only for IE, which is part of their strategy to lock-in users to Windows.
Yes, but your users probably hate them if they have to do any kind of real work on them. That is, anything that can't be done in a web browser at least.
Go out and take a walk and ask people if they miss having a real PC. I bet they do. If they don't now, they will when the capacity for your servers approaches 80%, and then management will be unwilling to invest in more infrastructure. Then it will all fall apart when you exceed capacity and the number of complaints by users forces management to reinvest... in new desktop PCs.
The problem with associations may be limited to the 64-bit version, but it's a problem you can see with a Google search that I've seen on both Vista x64 and Windows 7 x64 with UAC enabled. It makes me worry that non-admins will not be able to easily specify file associations. I'm also positive I had to turn the shell integration on when I installed 7 x64 last, but that may just have been a random bug or me misremembering as Google doesn't show any similar hits.
I use preserve full pathnames when copying folders from one system to another. I don't need to worry about specifying a path at all (other than root) if I preserve full pathnames. Less possibility of me making a typing error or unzipping incorrectly. WinZIP has it. WinRAR has it. GNU tar has it. 7-Zip does not.
As far as deleting after archiving, I do use a script to do this, too. Of course, you have to remember with a script that you need to first move the files to a new directory, then archive, then delete. Otherwise you risk deleting a file that wasn't archived because it was locked when 7z ran but not when delete did. Again, WinZIP has it. WinRAR has it. GNU tar has it. 7-Zip does not.
The GUI not being as user-friendly (meaning it doesn't obey the same conventions that the Windows Explorer model does) is a problem for average users, not power users. I still get occasional complaints from less technical users because 7-Zip is Norton Commander-like in many respects.
The installer not having options is a problem for corporate or enterprise deployment. What good is the tool if it isn't turned on and there's no easy way to configure it centrally? With most software I'd expect (that is, like) to be able to use an.mst to configure the installer to do typical tasks such as file associations and configuration so that it's easy to deploy.
As I said, I use 7-Zip for personal use. Average Windows users seem to have a lot better experience with WinZip's and WinRAR's conventions.
I've been hoping for a long time that there would be some kind of CPA/SPE/bar/board equivalent for both IT and software design and engineering. It would really help our industries so very much. I believe it's only a matter of time before such things are created, but I really wish it would happen sooner rather than later.
Perhaps some law will be passed requiring official certification or recognition... perhaps reducing the cost of insurance for an entity.
Eh, 7-Zip is a better compression algorithm than it is software. The GUI is not as nice as that in WinZIP or WinRAR (it's designed around Norton Commander instead of the more well-understood Windows Explorer model), the installer doesn't provide the option to associate the files, the association mechanism within the program itself wants admin rights but doesn't properly ask for them so it conflicts with Vista and 7's UAC, the archiver lacks "preserve full pathname" support (extraction support works fine), the context menu component defaults to off, etc. The command line version makes me miss tar, as it lacks the seemingly-basic feature of deleting files after archiving which would make archiving logs so easy.
It's a good program that's fine for personal and home use, but it requires a lot of post-install configuration that is difficult to deploy for a business to get the kind of uniform desktops that make IT life easier. About the only good thing business-wise is that the installer is an.msi file.
I use 7-Zip because it's free as in beer. If I had my choice, though, I'd use WinRAR.
I don't need to know how the chemistry of combustion works to be able to use it to make a fire to cook food, generate light and heat for my shelter, and so on. Fire was not well understood as a chemical process until relatively recently; certainly not until after the discovery of oxygen in the 1770s.
It frustrates nerds to no end, but "why" is often a pretty useless question to ask in the grand scheme of things: "The hard drive seems to have failed." "Why?" "I could answer that with a level 4 clean room and a team of investigators, or I could just replace it for $80."
I have the opposite experience. Windows XP and later work out-of-the-box on my system, but OpenSUSE never has. I started with 9, and also tried 10.1, 10.2, and 11. The system typically works fine from LiveCD, giving me false hope, but installation always fails to produce what I'd call a usable system. Anything from no sound or no network to X not starting. I tend to just get frustrated and go back to Windows. It's too bad. I want to run something other than XP or Ubuntu, but those are the only systems that work well with the PC.
However, if you're terrified of Ext4 eating your dog, you're welcome to run Fat32 if it makes you feel better.
Don't be obtuse. FAT is a horrible file system. He's going to stick with ext3.
A simple Google search shows that many people and many different distros have experienced data corruption and data loss and it has been attributed to anything from bugs in ext4 to the kernel not behaving as the ext4 developers were expecting due to specific configurations made for a given distro. Some people are very paranoid about data loss. Using ext3 instead is his choice, and choice is what Linux is all about.
There's only so much improvement you can make when the manufacturer insists on packing so much into the "kernel."
So in trying to bash Microsoft you're saying that Linux sucks?
Linux is a monolithic kernel. Windows is a hybrid kernel. Linux puts a lot more into kernel mode/real mode than Windows does. Many drivers in Windows are user mode drivers, for example, particularly printers. The only thing I can think of that runs in kernel mode in Windows and not in Linux is the graphics system -- which is why the screen flickers and changes resolutions slower in Linux and Windows tends to run full screen games and video better with DirectX, but it also rarely brings the system down... not that a system you can't get desired display output from is useful entirely.
If 99.9% of the time I don't need to configure Eclipse, how come whenever I try to use it I have to mess with the configuration for an hour or two and even then I'm still fighting with the damn thing to get it to behave in a manner I consider sane? Grandparent knows exactly what he's talking about when he says Eclipse doesn't correctly favor convention over configuration. Time was I thought I hated IDEs and swore by plain text editors. Turns out I just hate Eclipse. The IDE should not be more arcane than the language I'm trying to program with.
Hey, maybe things have changed in the last 5-6 years, but somehow I really doubt it.
Way to nuke any possibility of credibility, dude. Using currency symbols in company names just makes you look like a nutjob, regardless of how accurate your accusations might be. Nevermind that company of nVidia's, Intel's, Microsoft's, or indeed even ATI/AMD's size has "a very long history of dirty tricks, anti-competitive and anti-consumer behavior". Pick the card that works the best for your needs. Giving the name on the box more press -- even bad press -- simply makes the brand name that much more valuable than the hardware you're buying.
I worked in healthcare. I'd wager a guess that only the Microsoft apps were Win2k certified (although that was the OS we used). Many were intended for NT 4 or Win9x.
Not that all the MS apps were, but just that none of the non-MS apps weren't.
When you have little or no say in what software gets selected for use but are required to maintain local support for the same software as well as maintain the security of the network, it is not a waste of time at all. You do not give users Admin privileges. You give them the permissions they require to do their job and no more. That's basic best practice.
It's really not even that difficult to figure out. Nine times out of ten, the program either wants to write to HKLM\Software\$appname or wants to write to two or three configuration or log files in %programfiles%\$appname. About a quarter of the time (IMX) the documentation contains detailed information about what permissions are necessary. After that it's merely a case of using the various SysInternals monitors to figure out what's causing the problem. Between Xcacls and regini it's not difficult at all to script the changes. I typically maintain a single script which checks for the presence of each application and, if found, applies the necessary permissions changes.
Slashdot Mirror
That's a great argument in theory, but unfortunately the reality is that the older proprietary drivers that haven't been supported in half a dozen years still often perform better than the up-to-date open source alternatives (if there even are any non-generic drivers). If your motivation is pragmatism, then you're going to use proprietary in that case.
What exactly is different or more efficient about making the browser parse new HTML tags and incorporate new libraries instead of making the browser run JavaScript every time the page is loaded and relying on third parties to produce browser plug-ins for standard web content?
Yeah, these guys aren't so much as pirates as privateers.
What did you use to format the partition? XP uses NTFS 5.1. Windows 7 uses NTFS 6.x. NTFS 6 adds transactional NTFS support and support for bitlocker encryption, among other things. If you used XP's setup utility or used a Linux-based utility to format the partition, it simply may not be completely compatible. Remember, XP is not a supported upgrade path for Windows 7.
It's just the the topic itself is a bit snarky. It's a little insulting, as a sysadmin who happily calls himself an IT person, to hear that a developer is insulted by being called an "IT guy". Developers often have the opinion that being a sysadmin is little more than being a janitor. It's very frustrating to be dismissed out-of-hand as "someone who was too stupid to program" by developers who ought to know better. I can forgive most users and most management for their ignorance, but I have a hard time forgiving developers.
That's not true at all. RAM draws a fixed supply of power, but flash certainly doesn't. That's why you're able to unplug a flash drive and have it keep the data!
The only parts constantly in need of power are the controller electronics. The power rating for Intel's enterprise SSDs, for example, is shown as 2.5 Watts active and 0.06 Watts idle. On the newer 34 nm drives, Intel (PDF warning) claims 0.15 W active, 0.075 W idle.
Chromium's power management may not suck. It seems that Chromium is not using EIST, so the processor is always running at 1600 MHz whereas the other distros could scale the processor down to 800 MHz to save power. Given that this system had a SSD, the CPU likely accounts for the vast majority of power consumption.
But otherwise it was a pretty bland review.
So, all those people wanting to run Windows 7 on a SheevaPlug/NSLU2 or wireless router will be so upset?
Really, SlashDot? Really?
It's not an argument. It's an expression of the thin-thick cycle that myself and many others have seen undergone in the real world. It's a pendulum. One side is running thin clients and having everything done server-side. The other is thick clients and having as much as possible being client-side. The "prevailing" theory will oscillate about every two or three years or so, IMX.
It's absolutely possible to do it all correctly, but in general management is sold the idea that thin clients are cheaper than desktops -- or at least that's what they seem to remember the most. However, at some point you invariably reach capacity. Either you start running more apps, need to run a new OS or apps that require different operating systems or an OS that uses more resources, etc. You will need to periodically reinvest in your servers. And you'll need to maintain your infrastructure to a higher level of quality because basic client activity is now leveraged over the network. This is all on top of licensing for your thin client software. The problem with needing to invest in servers and networks is the high cost of investment. Buying desktops is really appealing if it means you don't need to upgrade all your switches and cable runs and you don't need to add half a dozen server racks to an already-cramped, power-hungry server room.
It's not that thin clients are bad. It's that "cheaper" doesn't mean cheap, and it certainly doesn't mean you get to neglect IT for half a dozen years. Unfortunately, that's precisely what tends to happen.
Maybe you hadn't noticed, but development of IE7 and IE8 have not been tied to a specific OS at all. IE7 was released before Vista and installs on XP, and IE8 well before Win 7 and that installs on Vista and XP. Microsoft has said that IE9 will be released in 2010, while Windows 8 is set for 2012. IE and Office are both on different development timetables than Windows -- although Office is almost always released 6 to 8 months after a desktop Windows release. Sure, they're linked in some senses because each product has a target platform, but otherwise there is no specific tie-in.
Microsoft's fiscal incentive is to maintain market dominance and some semblance of standards compliance. If they lose too much market share, developers may not create websites to handle IE quirks any longer. Then IE will falter, and MS will not be able to develop web apps only for IE, which is part of their strategy to lock-in users to Windows.
Yes, but your users probably hate them if they have to do any kind of real work on them. That is, anything that can't be done in a web browser at least.
Go out and take a walk and ask people if they miss having a real PC. I bet they do. If they don't now, they will when the capacity for your servers approaches 80%, and then management will be unwilling to invest in more infrastructure. Then it will all fall apart when you exceed capacity and the number of complaints by users forces management to reinvest... in new desktop PCs.
The problem with associations may be limited to the 64-bit version, but it's a problem you can see with a Google search that I've seen on both Vista x64 and Windows 7 x64 with UAC enabled. It makes me worry that non-admins will not be able to easily specify file associations. I'm also positive I had to turn the shell integration on when I installed 7 x64 last, but that may just have been a random bug or me misremembering as Google doesn't show any similar hits.
I use preserve full pathnames when copying folders from one system to another. I don't need to worry about specifying a path at all (other than root) if I preserve full pathnames. Less possibility of me making a typing error or unzipping incorrectly. WinZIP has it. WinRAR has it. GNU tar has it. 7-Zip does not.
As far as deleting after archiving, I do use a script to do this, too. Of course, you have to remember with a script that you need to first move the files to a new directory, then archive, then delete. Otherwise you risk deleting a file that wasn't archived because it was locked when 7z ran but not when delete did. Again, WinZIP has it. WinRAR has it. GNU tar has it. 7-Zip does not.
And yes, in a corporate setting, I'd absolutely prefer to use WinRAR.
The GUI not being as user-friendly (meaning it doesn't obey the same conventions that the Windows Explorer model does) is a problem for average users, not power users. I still get occasional complaints from less technical users because 7-Zip is Norton Commander-like in many respects.
The installer not having options is a problem for corporate or enterprise deployment. What good is the tool if it isn't turned on and there's no easy way to configure it centrally? With most software I'd expect (that is, like) to be able to use an .mst to configure the installer to do typical tasks such as file associations and configuration so that it's easy to deploy.
As I said, I use 7-Zip for personal use. Average Windows users seem to have a lot better experience with WinZip's and WinRAR's conventions.
I've been hoping for a long time that there would be some kind of CPA/SPE/bar/board equivalent for both IT and software design and engineering. It would really help our industries so very much. I believe it's only a matter of time before such things are created, but I really wish it would happen sooner rather than later.
Perhaps some law will be passed requiring official certification or recognition... perhaps reducing the cost of insurance for an entity.
Eh, 7-Zip is a better compression algorithm than it is software. The GUI is not as nice as that in WinZIP or WinRAR (it's designed around Norton Commander instead of the more well-understood Windows Explorer model), the installer doesn't provide the option to associate the files, the association mechanism within the program itself wants admin rights but doesn't properly ask for them so it conflicts with Vista and 7's UAC, the archiver lacks "preserve full pathname" support (extraction support works fine), the context menu component defaults to off, etc. The command line version makes me miss tar, as it lacks the seemingly-basic feature of deleting files after archiving which would make archiving logs so easy.
It's a good program that's fine for personal and home use, but it requires a lot of post-install configuration that is difficult to deploy for a business to get the kind of uniform desktops that make IT life easier. About the only good thing business-wise is that the installer is an .msi file.
I use 7-Zip because it's free as in beer. If I had my choice, though, I'd use WinRAR.
I don't need to know how the chemistry of combustion works to be able to use it to make a fire to cook food, generate light and heat for my shelter, and so on. Fire was not well understood as a chemical process until relatively recently; certainly not until after the discovery of oxygen in the 1770s.
It frustrates nerds to no end, but "why" is often a pretty useless question to ask in the grand scheme of things:
"The hard drive seems to have failed."
"Why?"
"I could answer that with a level 4 clean room and a team of investigators, or I could just replace it for $80."
I have the opposite experience. Windows XP and later work out-of-the-box on my system, but OpenSUSE never has. I started with 9, and also tried 10.1, 10.2, and 11. The system typically works fine from LiveCD, giving me false hope, but installation always fails to produce what I'd call a usable system. Anything from no sound or no network to X not starting. I tend to just get frustrated and go back to Windows. It's too bad. I want to run something other than XP or Ubuntu, but those are the only systems that work well with the PC.
Don't be obtuse. FAT is a horrible file system. He's going to stick with ext3.
A simple Google search shows that many people and many different distros have experienced data corruption and data loss and it has been attributed to anything from bugs in ext4 to the kernel not behaving as the ext4 developers were expecting due to specific configurations made for a given distro. Some people are very paranoid about data loss. Using ext3 instead is his choice, and choice is what Linux is all about.
So in trying to bash Microsoft you're saying that Linux sucks?
Linux is a monolithic kernel. Windows is a hybrid kernel. Linux puts a lot more into kernel mode/real mode than Windows does. Many drivers in Windows are user mode drivers, for example, particularly printers. The only thing I can think of that runs in kernel mode in Windows and not in Linux is the graphics system -- which is why the screen flickers and changes resolutions slower in Linux and Windows tends to run full screen games and video better with DirectX, but it also rarely brings the system down... not that a system you can't get desired display output from is useful entirely.
Because malice from Microsoft creates a lot more SlashDot nerd rage.
If 99.9% of the time I don't need to configure Eclipse, how come whenever I try to use it I have to mess with the configuration for an hour or two and even then I'm still fighting with the damn thing to get it to behave in a manner I consider sane? Grandparent knows exactly what he's talking about when he says Eclipse doesn't correctly favor convention over configuration. Time was I thought I hated IDEs and swore by plain text editors. Turns out I just hate Eclipse. The IDE should not be more arcane than the language I'm trying to program with.
Hey, maybe things have changed in the last 5-6 years, but somehow I really doubt it.
Because, unlike Congress, this nothing serves a purpose and it's designed to protect the citizenry.
Although it is true that they're both supported by little more than hot air.
Way to nuke any possibility of credibility, dude. Using currency symbols in company names just makes you look like a nutjob, regardless of how accurate your accusations might be. Nevermind that company of nVidia's, Intel's, Microsoft's, or indeed even ATI/AMD's size has "a very long history of dirty tricks, anti-competitive and anti-consumer behavior". Pick the card that works the best for your needs. Giving the name on the box more press -- even bad press -- simply makes the brand name that much more valuable than the hardware you're buying.
I worked in healthcare. I'd wager a guess that only the Microsoft apps were Win2k certified (although that was the OS we used). Many were intended for NT 4 or Win9x.
Not that all the MS apps were, but just that none of the non-MS apps weren't.
When you have little or no say in what software gets selected for use but are required to maintain local support for the same software as well as maintain the security of the network, it is not a waste of time at all. You do not give users Admin privileges. You give them the permissions they require to do their job and no more. That's basic best practice.
It's really not even that difficult to figure out. Nine times out of ten, the program either wants to write to HKLM\Software\$appname or wants to write to two or three configuration or log files in %programfiles%\$appname. About a quarter of the time (IMX) the documentation contains detailed information about what permissions are necessary. After that it's merely a case of using the various SysInternals monitors to figure out what's causing the problem. Between Xcacls and regini it's not difficult at all to script the changes. I typically maintain a single script which checks for the presence of each application and, if found, applies the necessary permissions changes.