I bought into the coolaid that "If you follow the standards your site will look good on most browsers, including IE." It should read "If you follow standards your site will look good on most browsers, EXCEPT IE."
Being a Linux user I made the mistake of initially designed a site using Opera, Konqueror and Firefox following the w3c recommendations. The differences between how these browsers render are relatively minor and its usually easy to find a good common denominator. Then there is IE, especially 6 but 7 still isn't in the same league as those other browsers. I had to restructure my HTML to get it to work good enough in IE. I wanted a tabular layout for screen rendering for one part but I thought divs would be better markup. This isn't a problem in all modern browsers except IE. And IE 6 had this really weird bug where part of the page background would turn grey when you scrolled the page down then up again. I never seen something that pathological in the other browsers. I shuffled the markup around a bit and it went away.
If you need to support IE, you should test with it from day 1 (it will shape how you implement a site). If you need to support Opera, Konqueror and Safari you can test with just Firefox (following w3c recommendations) and fix the minor problems at the end.
And using a w3c validator is not enough to verify if a page will render right on all browsers. There are many little gotchas in CSS that can vary between browsers, especially with IE.
I personally prefer Firefox for development (firebug...) and Konqueror for daily browsing.
Don't forget that the typical way to install software on a Linux distro is to use a package manager with a large selection of usefull software. The typical way to install software on windows is to buy box software or download it of some random site on the internet. It is much easier to get crapware on a users box if its typical for them to run unsigned software of the net rather than refer to a package manager with all signed packages. In fact Linux typically makes it somewhat difficult to just run whatever random shit. Vista just presents the user with yet another dialog.
I strongly suspect the best mechanism to attach a Vista box will be some variant of social engineering.
If the US military decides to bomb the building your family was in in Iraq, and they get killed, DON'T COMPLAIN. You should have become a US citizen and voted.
If you are declared a "terrorist sympathizer" and executed because you are not unequivocally "with us" DON'T COMPLAIN if you didn't vote. Because if you didn't vote its not murder, and you have no rights.
Can this be disabled? Not after the system boots but with some boot option, much as SELinux can be disabled. It is relatively trivial to provide a boot option to disable this that can not be circumvented once the system boots. In doing this, you would ensure the average user only uses signed quality drivers while enabling a developer to write drivers if they need to. If this can't be disabled it this would just be one more reason to use Linux/BSD's.
This use of cryptography is a good thing if the user is given the final say. Otherwise its downright tyrannical.
The primary value IE6 offers over the competition is familiarity, compatibility and convenience. A typical user doesn't have to do anything to install it (installed with windows) and it will work on almost every web site (because developers slave away to make sure it does). For many users it is also familiar, they are comfortable with it. Yet the default IE7 is a substantial departure from past web browser interface designs. It seems IE7 is ignoring one of IE's primary assets. Which traditionally seems out of character for Microsoft but reflects a more recent trend. Traditionally MS products have very few changes (but many additions) from previous versions. MS Office has many aspects that haven't changed for a decade, and in my opinion many of these aspect are exceptionally convoluted e.g. styles in Word, graphing in Excel. But now we are seeing MS change old habits and start to change existing interfaces, look at IE7, Vista or Office 12.
Do you expect these changes could cost market share considering an IE6 user will feel more at home with Firefox than IE7.
In theory and from my experience with reiser4 for nearly 3 years, it does not have this problem. But it is extremely lazy. By default reiser4 can take up to 10 minutes to sync data to disk. So you will get a perfectly consistent file system after a crash but it might be a bit out of date.
Also note the reiser4 is extreally dependent on data being written in a specific order (like softupdates). I had some fs corruption issues on my laptop with reiser4 until I disabled the write cache. I believe that hd wasn't respecting write barriers or whatever it is (if anything) for that drive. I have never had a problem like this with my SCSI Cheetahs.
I have experienced this individual file corruption with reiser3.
Hans vision (http://www.namesys.com/whitepaper.html) is about unifying namespaces (fs, database, email, config files, etc.). So I believe his primary objective was to build a file system that performs well under all situations especially those situations where current filesystems perform so bad nobody writes software to use the fs that way. Most notably lots of tiny files. Ext(2/3) use a traditional UNIX FS design and most software has been written to work acceptably on this type of file systems. In other words, nobody writes software that creates thousands to millions of tiny files because ext absolutely sucks at this (and ext4 doesn't help much) . So to get around this most of today's software create a new namespace in a single file./etc is exactly this. reiser(fs/4) also use extents which improves its handling of really large files which is becoming more common (movies, music, etc.). ext4 introduces this feature. So for today's common usage patterns and file sizes ext4 will perform reasonably well compared to reiser4.
But if you where to extend fs semantics enough to make it reasonable to use the fs to directly back something like gconf (http://en.wikipedia.org/wiki/GConf), reiser4 would be substantially (10x) faster than ext(2,3,4) could ever hope to be. In principal if you extend the filesystem to directly support gconf and similar you would be able use of many fs tools for free.
Consider the windows registry. It is a mistake because it obscures the information by its convoluted structure and just as significantly, one can't use all those usefull tools on the registry you use with your file system (well not to many in Windows). Gconf addresses the first problem but not the second. Overall, I would consider the/etc approach superior because it is better documented, more transparent and I can use cp and subversion on it. Nobody will ever write a tool as powerful as subversion for the windows registry because the value isn't great enough. But subversion was never designed specifically to help with/etc it was primarily for coding but because coding has much in common with/etc the same tool can be used for both.
Yet, the windows registry had the good idea of creating a consistent way to store the the same type of data. Saving programming time and building a single consistent source for all things configuration. I believe it is possible to build something that gets the best of/etc and the best of the windows registry or better yet gconf.
The more use one can get out of a tool, the more people will work to develop and improve it. By using the same namespace for as much as possible, the more value a tool designed to work with that namespace will have. In other words, the utility of a computer system is proportional to the number of ways the components can interact, not the number of components.
To achieve this objective, one needs a solid foundation, a system that can efficiently store whatever you ask of it, that is what reiser4 is trying to be. But, clearly much work remains.
Well, I am going to try being honest and see what happens.
I am more than 170k in debt, and Namesys is doing badly fiscally. A technical great success being stabilized now, but then there is my ongoing fiscal disaster. Once again, we are missing payroll. My wife is divorcing me in part because I keep going deeper into debt, and I thank her for divorcing me now rather than later. Unfortunately she is making the divorce messy enough to keep me from pulling Namesys out of the fiscal tailspin by consuming all my time with things like proving I am not making the fantastic amounts of money she claims I am. I hope next month is better."
Joe sixpack does not care if a driver is binary or open source, he wants it to work. So do we care about joe sixpack? I hate to say, I do. I have to help people with their Windows computers and I find it exceedingly frustrating that I actually have more years of Window experience but much less of an ability to actually fix problems on Windows than Linux, let alone get Windows to do anything outside of that small box Microsoft has prescribed. It would be great if the systems I need to support could run an operating system that doesn't make me hate IT work. But it takes a bit of time to learn a new OS and I can only expect that a user will only switch if the expected value of the switch is greater than the cost. Hardware that doesn't work will quickly make the cost appear high.
I would like to see linux distributions that target the average user and provide at least the same ease of use as Windows. I think Ubuntu, Xandros, Linspire, etc are seeking this end. Enabling proprietary drivers to just work in many cases will drasticly improve the users experience with a these linux distributions.
I understand the kernel developers desire to not be tied down to a fixed interface. But why not declare a kernel binary stable on a predictable time table (e.g. every year) and let binary driver developers target that kernel? That kernel will likely not get any updates short of something extremely critical. Distributions that would gain from binary compatibility by default would use these binary stable kernels. Might not get the latest and greatest features, but I expect a years worth of kernel features would be of less value to most people than hardware that actually works.
I also think for the most part binary drivers are assinine. But I don't see that anyone will convince companies like nVidia of this anytime soon. So why not work with them to make these drivers just work but clearly inform the user that they are using binary drivers and the problems with binary drivers.
Making more hardware just work with Linux can only encourage its use. Allowing binary drivers in the short term could encourage more open source drivers (and binary ones) in the long term.
If open source drivers really provide a substantial technical advantage (stability, working properly, lower cost) I would like to think with a little time those companies producing endorsing open source drivers would find a competitive advantage over their proprietary counterparts.
Is the goal to rid the world of binary drivers or to empower the users/developers?
Hans Reisers vision is about unifying namespaces (filesystem, relational database, XML, etc...) by providing the functionality in a filesystem to make this reasonable. In otherwords, making the file system better than current databases.
Do we evolve the file system into a database (Reiser approach) or evolve a database into a file system (Microsoft WinFS approach)?
If a binary XML file is semantically equivalent to its text counterpart and you have good tools to convert between the two, binary XML would be much like lossless (possibly minus beautification) compression for XML. Yet, if done right, it could speed up how long it takes to process XML files instead of slow it down as something like gzip would do.
Why do most so many people use Microsoft products? Is this because their products are functionally better? Or does a network effect play a large part in Microsoft's success. The more people that use Microsoft products and especially the OS, the more applications will work with them, thereby increasing the utility of their products. In addition, many users use Microsoft products because it appears to be easier to conform to the status quo.
I believe Palladium is an excellent means to extend the magnitude of this network effect. Microsoft knows that consumers will not be eager to purchase products that seem inferior to there predecessors. And Palladium will not look inferior. A Palladium system does not directly restrict what the user can do with the system. You will be able to run all the applications you run now and use them in the same way on a Palladium system. But, a content provider will be able to effectively only provide content to consumers running the software they deem appropriate. This software can be very restrictive. If you decide not to use the restrictive software, you give up your right to receive content from providers that require it.
The success of this technology as a DRM tool rests in not restricting the consumer to much. Lets assume the system is developed to a point where it can reliably authenticate an individual user via a smart card or something. This would allow a user to receive the content they licensed at any system that can authenticate a user and is trusted by the content provider. While this trusted systems will only allow users to access content they are authorized to access. If Microsoft could provide a reasonably high penetration of Palladium products, many consumers would find the restrictions of the system reasonable enough to justify purchasing protected content. The more consumers that purchase content, the greater the demand for Palladium products to utilize that content. The more Palladium products the greater the demand for the protected content. There is clearly a critical mass in which palladium would prosper or flounder.
Its important to note that in this scenario, Palladium didn't restrict the user from doing an explicit thing they could do before like playing there mp3's. It simply provides the consumer with access to more content. This is assuming this content isn't provided by means outside of Palladiums control. For this reason, I would expect software will be the first candidate for exclusive distribution within the palladium realm. Infact, Microsoft can add value to palladium by providing software that can only be acquired by a palladium system.
If the use of palladium becomes wide spread, a palladium enabled computer would offer a distinct added value in terms of available content over a non palladium counter part. Yet, to be an effective palladium system, the content providers must trust that system. But, establishing a system as trusted will be an expensive task. An individual would not be able to modify their palladium open source kernel (if such a thing will ever exist) and expect it to be trusted. If this where the case, palladium would be ineffective. This will prove to be a major challenge to open source development. It would inherently make working on many open source projects reduce the value of your computer.
If successful, the Microsoft palladium products will be better than the alternatives not because of technical merits but simple because they are trusted. Establishing software as trusted by the plethora of content providers could prove to be a task only the largest Corporations could afford.
This leads to the question. How will a content provider know what software to trust? Will each content provider need to explicitly define what software they trust? If an entity developed a palladium OS, would that entity need to get each content provider to trust it in order to compete with the Microsoft products that will undoubtedly be trusted by all?
Slashdot Mirror
I bought into the coolaid that "If you follow the standards your site will look good on most browsers, including IE."
...) and Konqueror for daily browsing.
It should read "If you follow standards your site will look good on most browsers, EXCEPT IE."
Being a Linux user I made the mistake of initially designed a site using Opera, Konqueror and Firefox following the w3c recommendations. The differences between how these browsers render are relatively minor and its usually easy to find a good common denominator. Then there is IE, especially 6 but 7 still isn't in the same league as those other browsers. I had to restructure my HTML to get it to work good enough in IE. I wanted a tabular layout for screen rendering for one part but I thought divs would be better markup. This isn't a problem in all modern browsers except IE. And IE 6 had this really weird bug where part of the page background would turn grey when you scrolled the page down then up again. I never seen something that pathological in the other browsers. I shuffled the markup around a bit and it went away.
If you need to support IE, you should test with it from day 1 (it will shape how you implement a site). If you need to support Opera, Konqueror and Safari you can test with just Firefox (following w3c recommendations) and fix the minor problems at the end.
And using a w3c validator is not enough to verify if a page will render right on all browsers. There are many little gotchas in CSS that can vary between browsers, especially with IE.
I personally prefer Firefox for development (firebug
Don't forget that the typical way to install software on a Linux distro is to use a package manager with a large selection of usefull software. The typical way to install software on windows is to buy box software or download it of some random site on the internet. It is much easier to get crapware on a users box if its typical for them to run unsigned software of the net rather than refer to a package manager with all signed packages. In fact Linux typically makes it somewhat difficult to just run whatever random shit. Vista just presents the user with yet another dialog.
I strongly suspect the best mechanism to attach a Vista box will be some variant of social engineering.
If the US military decides to bomb the building your family was in in Iraq, and they get killed, DON'T COMPLAIN. You should have become a US citizen and voted.
If you are declared a "terrorist sympathizer" and executed because you are not unequivocally "with us" DON'T COMPLAIN if you didn't vote. Because if you didn't vote its not murder, and you have no rights.
Can this be disabled? Not after the system boots but with some boot option, much as SELinux can be disabled. It is relatively trivial to provide a boot option to disable this that can not be circumvented once the system boots. In doing this, you would ensure the average user only uses signed quality drivers while enabling a developer to write drivers if they need to. If this can't be disabled it this would just be one more reason to use Linux/BSD's.
This use of cryptography is a good thing if the user is given the final say. Otherwise its downright tyrannical.
The primary value IE6 offers over the competition is familiarity, compatibility and convenience. A typical user doesn't have to do anything to install it (installed with windows) and it will work on almost every web site (because developers slave away to make sure it does). For many users it is also familiar, they are comfortable with it. Yet the default IE7 is a substantial departure from past web browser interface designs. It seems IE7 is ignoring one of IE's primary assets. Which traditionally seems out of character for Microsoft but reflects a more recent trend. Traditionally MS products have very few changes (but many additions) from previous versions. MS Office has many aspects that haven't changed for a decade, and in my opinion many of these aspect are exceptionally convoluted e.g. styles in Word, graphing in Excel. But now we are seeing MS change old habits and start to change existing interfaces, look at IE7, Vista or Office 12.
Do you expect these changes could cost market share considering an IE6 user will feel more at home with Firefox than IE7.
In theory and from my experience with reiser4 for nearly 3 years, it does not have this problem. But it is extremely lazy. By default reiser4 can take up to 10 minutes to sync data to disk. So you will get a perfectly consistent file system after a crash but it might be a bit out of date.
Also note the reiser4 is extreally dependent on data being written in a specific order (like softupdates). I had some fs corruption issues on my laptop with reiser4 until I disabled the write cache. I believe that hd wasn't respecting write barriers or whatever it is (if anything) for that drive. I have never had a problem like this with my SCSI Cheetahs.
I have experienced this individual file corruption with reiser3.
Hans vision (http://www.namesys.com/whitepaper.html) is about unifying namespaces (fs, database, email, config files, etc.). So I believe his primary objective was to build a file system that performs well under all situations especially those situations where current filesystems perform so bad nobody writes software to use the fs that way. Most notably lots of tiny files. /etc is exactly this.
/etc approach superior because it is better documented, more transparent and I can use cp and subversion on it. Nobody will ever write a tool as powerful as subversion for the windows registry because the value isn't great enough. But subversion was never designed specifically to help with /etc it was primarily for coding but because coding has much in common with /etc the same tool can be used for both.
/etc and the best of the windows registry or better yet gconf.
Ext(2/3) use a traditional UNIX FS design and most software has been written to work acceptably on this type of file systems. In other words, nobody writes software that creates thousands to millions of tiny files because ext absolutely sucks at this (and ext4 doesn't help much) . So to get around this most of today's software create a new namespace in a single file.
reiser(fs/4) also use extents which improves its handling of really large files which is becoming more common (movies, music, etc.). ext4 introduces this feature.
So for today's common usage patterns and file sizes ext4 will perform reasonably well compared to reiser4.
But if you where to extend fs semantics enough to make it reasonable to use the fs to directly back something like gconf (http://en.wikipedia.org/wiki/GConf), reiser4 would be substantially (10x) faster than ext(2,3,4) could ever hope to be. In principal if you extend the filesystem to directly support gconf and similar you would be able use of many fs tools for free.
Consider the windows registry. It is a mistake because it obscures the information by its convoluted structure and just as significantly, one can't use all those usefull tools on the registry you use with your file system (well not to many in Windows). Gconf addresses the first problem but not the second. Overall, I would consider the
Yet, the windows registry had the good idea of creating a consistent way to store the the same type of data. Saving programming time and building a single consistent source for all things configuration. I believe it is possible to build something that gets the best of
The more use one can get out of a tool, the more people will work to develop and improve it. By using the same namespace for as much as possible, the more value a tool designed to work with that namespace will have. In other words, the utility of a computer system is proportional to the number of ways the components can interact, not the number of components.
To achieve this objective, one needs a solid foundation, a system that can efficiently store whatever you ask of it, that is what reiser4 is trying to be. But, clearly much work remains.
http://marc.theaimsgroup.com/?l=reiserfs&m=1095355 06122706&w=2
1 78128079&w=27 5720520&w=2
Hans Reiser:
Well, I am going to try being honest and see what happens.
I am more than 170k in debt, and Namesys is doing badly fiscally. A
technical great success being stabilized now, but then there is my
ongoing fiscal disaster. Once again, we are missing payroll. My wife
is divorcing me in part because I keep going deeper into debt, and I
thank her for divorcing me now rather than later. Unfortunately she is
making the divorce messy enough to keep me from pulling Namesys out of
the fiscal tailspin by consuming all my time with things like proving I
am not making the fantastic amounts of money she claims I am. I hope
next month is better."
Others
http://marc.theaimsgroup.com/?l=reiserfs&m=108353
http://marc.theaimsgroup.com/?l=reiserfs&m=984246
Joe sixpack does not care if a driver is binary or open source, he wants it to work. So do we care about joe sixpack? I hate to say, I do. I have to help people with their Windows computers and I find it exceedingly frustrating that I actually have more years of Window experience but much less of an ability to actually fix problems on Windows than Linux, let alone get Windows to do anything outside of that small box Microsoft has prescribed. It would be great if the systems I need to support could run an operating system that doesn't make me hate IT work. But it takes a bit of time to learn a new OS and I can only expect that a user will only switch if the expected value of the switch is greater than the cost. Hardware that doesn't work will quickly make the cost appear high.
I would like to see linux distributions that target the average user and provide at least the same ease of use as Windows. I think Ubuntu, Xandros, Linspire, etc are seeking this end. Enabling proprietary drivers to just work in many cases will drasticly improve the users experience with a these linux distributions.
I understand the kernel developers desire to not be tied down to a fixed interface. But why not declare a kernel binary stable on a predictable time table (e.g. every year) and let binary driver developers target that kernel? That kernel will likely not get any updates short of something extremely critical. Distributions that would gain from binary compatibility by default would use these binary stable kernels. Might not get the latest and greatest features, but I expect a years worth of kernel features would be of less value to most people than hardware that actually works.
I also think for the most part binary drivers are assinine. But I don't see that anyone will convince companies like nVidia of this anytime soon. So why not work with them to make these drivers just work but clearly inform the user that they are using binary drivers and the problems with binary drivers.
Making more hardware just work with Linux can only encourage its use. Allowing binary drivers in the short term could encourage more open source drivers (and binary ones) in the long term.
If open source drivers really provide a substantial technical advantage (stability, working properly, lower cost) I would like to think with a little time those companies producing endorsing open source drivers would find a competitive advantage over their proprietary counterparts.
Is the goal to rid the world of binary drivers or to empower the users/developers?
Hans Reisers vision is about unifying namespaces (filesystem, relational database, XML, etc...) by providing the functionality in a filesystem to make this reasonable. In otherwords, making the file system better than current databases.
Do we evolve the file system into a database (Reiser approach) or evolve a database into a file system (Microsoft WinFS approach)?
If a binary XML file is semantically equivalent to its text counterpart and you have good tools to convert between the two, binary XML would be much like lossless (possibly minus beautification) compression for XML. Yet, if done right, it could speed up how long it takes to process XML files instead of slow it down as something like gzip would do.
Bill Gates said of Office 'it's too hard to find things in e-mail' and described some features of Word as 'clunky.'"
My respect for Microsoft has doubled, its now 0.
"Linux. The Future is Open. IBM."
should be
"GNU/Linux. The Future is Open. IBM."
Why do most so many people use Microsoft products? Is this because their products are functionally better? Or does a network effect play a large part in Microsoft's success. The more people that use Microsoft products and especially the OS, the more applications will work with them, thereby increasing the utility of their products. In addition, many users use Microsoft products because it appears to be easier to conform to the status quo.
I believe Palladium is an excellent means to extend the magnitude of this network effect. Microsoft knows that consumers will not be eager to purchase products that seem inferior to there predecessors. And Palladium will not look inferior. A Palladium system does not directly restrict what the user can do with the system. You will be able to run all the applications you run now and use them in the same way on a Palladium system. But, a content provider will be able to effectively only provide content to consumers running the software they deem appropriate. This software can be very restrictive. If you decide not to use the restrictive software, you give up your right to receive content from providers that require it.
The success of this technology as a DRM tool rests in not restricting the consumer to much. Lets assume the system is developed to a point where it can reliably authenticate an individual user via a smart card or something. This would allow a user to receive the content they licensed at any system that can authenticate a user and is trusted by the content provider. While this trusted systems will only allow users to access content they are authorized to access. If Microsoft could provide a reasonably high penetration of Palladium products, many consumers would find the restrictions of the system reasonable enough to justify purchasing protected content. The more consumers that purchase content, the greater the demand for Palladium products to utilize that content. The more Palladium products the greater the demand for the protected content. There is clearly a critical mass in which palladium would prosper or flounder.
Its important to note that in this scenario, Palladium didn't restrict the user from doing an explicit thing they could do before like playing there mp3's. It simply provides the consumer with access to more content. This is assuming this content isn't provided by means outside of Palladiums control. For this reason, I would expect software will be the first candidate for exclusive distribution within the palladium realm. Infact, Microsoft can add value to palladium by providing software that can only be acquired by a palladium system.
If the use of palladium becomes wide spread, a palladium enabled computer would offer a distinct added value in terms of available content over a non palladium counter part. Yet, to be an effective palladium system, the content providers must trust that system. But, establishing a system as trusted will be an expensive task. An individual would not be able to modify their palladium open source kernel (if such a thing will ever exist) and expect it to be trusted. If this where the case, palladium would be ineffective. This will prove to be a major challenge to open source development. It would inherently make working on many open source projects reduce the value of your computer.
If successful, the Microsoft palladium products will be better than the alternatives not because of technical merits but simple because they are trusted. Establishing software as trusted by the plethora of content providers could prove to be a task only the largest Corporations could afford.
This leads to the question. How will a content provider know what software to trust? Will each content provider need to explicitly define what software they trust? If an entity developed a palladium OS, would that entity need to get each content provider to trust it in order to compete with the Microsoft products that will undoubtedly be trusted by all?