When my generation was growing up... That was because your parents didn't have much money right? Wrong. It's because our parents didn't spend 98% of the salary on buying shit for us kids....
1. Java is far more reboust and atmoic than Parrot and Perl5. But its all preference. 2. There is a huge difference between loading swapped memory rather than reading data from disk and loading the classes in the VM. Loading from swap would take 1-2 seconds at the most. The first load of a JVM takes 5-6 seconds. 3. That _is_ poorly coded java. I believe this problem is not nearly as prevalent as it was 2-3 years ago but is still and always be an issue with all platforms.
In any even preloading a JVM should be an option, an user configurable adapter pattern. There are few applications these days that require native code. JIT environments are the future (In my opinion).
You completely missed the point. Making it an Operating System Component has nothing to do with a repository.
Essentially Operating Systems would startup a JRE when they startup. The actual memory footprint would be around 16MB for Java5. That way when apps loaded, they could reuse the existing JVM, sharing common classes with other applications. Think of dynamic libraries on steroids. It wouldnt just be a shared library but possibly instatiated objects.
As for a repository system for libraries, this uneeded. Applications ship with the libraries that they need rather than a perl application that forces the user to download them. Java applications can thus force exact version of libraries on the user without the user having to make decisions. Its a different way to do things, and in my opinion far better. For the desktop, Java has the ideal model. Let users run apps, without having to install libraries (and thus depend on an internet connection or other medium) just to get them to run.
"Reaganomics: Help the poor by giving to the rich!" rather than helping the poor directly."
I think you need to attend a more centrist univerisity. Reagan definitely had his good and bad imprints on US history. His economic decisions fueled our economy, promoted the technology boom and pulled us out of Carter's recession. Unfortunately he did not get to see his impact based on his illness. Bush Sr. and Clinton rode out his legacy for free (until the market collapsed).
That of course came at a price, and that was he was not very sympathetic to the real poor - people that cannot help themselves. And that is a shame. His overspending cut many programs and made many peoples lives harder. But even more benefitted. College enrollment exploded in the 80s fueling our technology/engineering foundation today. Children who otherwise had no change to get into (pay) for college could. I am not sure how any competent "economist" (and I used that term loosely) could possibly say reagonomics were summarily "bad".
I am a staunch liberal and even I can see his place in american history even though I disagreed strongly with many of his views and policies. I still can recognize his economic legacy that we enjoy today. The 80s could have turned out very differently if a different president just sat on the pot (like Bush Sr, Clinton, possibly Bush Jr). I look forward to the next time we elect a truely "Great" president and not just a sleezy politician.
This whole argument about GPLv3 is moot anyway. Why on earth would anyone want to relicense? Its like upgrading software that works just fine the way it is. I think Linus made the right call - There is no reason to promote a relicensing of the kernel. In fact there is no reason why anyone would.
Not many people license under the GPL anymore (compared to 3-5 years ago). Most use dual BSD/Apache style licensing now. This latest "update" to the GPL is pushing more companies away from OSS, not closer. I know my company wants nothing to do with GPL but instead dual licenses their OSS projects.
The linux kernel depends on corporate sponsership in a variety of ways. I think if Linus could reset the license, it too would have used something other than GPL.
Microsoft can do whatever they want with their products, they own them. Congress can regulate or compel any company that does business in America. Its called freedom, by the people. We can regulate any business entity in our own country, if they don't like it, they can leave =). As for the US forcing countries to use their currency? That is just ridiculous, lets see some proof =)
Sounds like you have a predisposition that you use to justify your ideals. Very strange way to form political foundations - "Building facts off of fiction".
Now i know the common response to this here is to say we live in a police state run by christian oil lovers and their cronies. But whats neat is, you can say whatever you want, its a free country =) I would just ask that you back up your statements and refrain from ad hominem attacks. It will make you more credible to the people in the middle and you will convince far more people instead of just fueling the extremists that believe everything they read.
You're missing the point. Of course vendors will pass on this cost to consumers, but they will not just "move on", they'll be pissed and they'll be more careful next time.
Negative. They wont be pissed because they dont care, its only 25 cents. Shipping portable drives unformatted will not happen. A few here and there might but all the big dogs will buck up. Its really not a big deal, this sort of process happens in many industries - all the time.
If this was say a $5.00 "tax", then they would explore alternate filesystems, but probably end up just shipping devices unformatted.
Its actually not in my opinion. kmail is lightweight and easy to use. Thunderbird takes a ridiculous amount of RAM to run on my machine - ~48m compared to ~17m with not just kmail but kontact. Although I have plenty of RAM on my desktop and I really dont notice the difference, it still makes you wonder about the moz platform. Firefox for instance is current sitting at 96m with slashdot loaded. konqueror right beside it is at 34m. Both have a single instance running.
I like firefox and thunderbird, but given the enourmous memory foorprint, you have to wonder what the future will be hold.
After this, electronics manufacturers are going to be reluctant to adopt any kind of Microsoft format.
At 25 cents? Vendors will just jack the price up and move on. No one wants to make a removable storage device that requires proprietary drivers and won't just work after being plugged in. Vendors don't want to pay for R&D costs to support a new format and then worry about users who have trouble using their devices. Its easier to just buck up. MS knows this and that is why they agreesively pursued this. Its easy money, they have nothing to lose.
Its also a nice sucker punch to Apple, which could have been the primary motivator;) I would prefer that their be an open/free filesystem for all removable devices, but there is no hope for this unless Apple gets behind it. Apple is turning into one of OSS's biggest allies.
They are our representatives, yet they don't represent us.
BS - They do represent you, just like the represent record companies. The difference is the record companies utilize superior tactics persuading representatives to their point of view. And no, this isnt always in the form of bribes or benefits, there are good politicians on both sides of the isle that are simply misinformed. Record companies put out literature, hold roundtables and most importantly - "convince". Our voices get trampled because we simply do not speak their langauge. Simply throwing up your arms saying congress and hollywood sucks lets all join a pity party does absolutely nothing.
Anyone on our side of the debate that raises their voice is usually a fanatic or at least considered one (FSF is a good example). If you we really want to change the future and secure what we consider our rights, you we have to form organizations that at least have the potential to gain respect in the house and senate. It is the _only_ way to get our voices heard. I am sorry, I like the FSF but they come off no better than southern baptists - "We are right because we know we are but we really can't prove it, just trust us". Believe it or not, not every OSS advocate is a GPL fan and they dont need to be. OSS and fair use are concepts that can exist in a capitalist society if we want them to.
Prove the idea of fair use, let congressman see what a future would be like that is completely DRM centric and employ intelligent and professional lobbyists. Get the message heard, comprehended and acted on.
I think you missed the point. Since gnome does not have a solid framework, the usability issues that linus pointed out exist. If the framework was solid the look would be consistent. The graphical usability problems that gnome exihibts are a direct result of the lack of a complete, extendable, reusable framework. Each project must tackle trivial HIG issues on their own. This is also tied into the fact that gnome uses C. When using the KDE framework (not just Qt) this is all done for you. Developers can extend the frameworks to fit their needs while being forced to use certain patterns. Thus KDE (the apps) has a unified look and feel and gnome (the apps) does not. It does not help that applications like firefox and evolution are not actual gnome apps, but rather just use gtk.
gnome-vfs is a horrible example since is it prime indicator of the lack of organization in gnome. When applications want to directly interface with a file, they are not forced to use gnome-vfs (like gthumb). This causes a huge problem since the apps somtimes take it upon themselves to utilize their own methodologies. Good application frameworks both empower and contain developers.
The fact that gnome utlizes GObject simply says, "they know its a problem are trying to work around it." it only adds complexity to developers thus increasing time to delivery.
The problems with gnome that Linus has pointed out are well known to everyone except gnome fanboys and developers. The framework GTK/Gnome frameworks are in shambles. They are not easily extendible and do not utilize sound OO principles to promote code reusability.
Gnome has made great strides in making a simplified DE for linux. Unfortunately they paid no attention on how to get there (see toolbar, io slave examples). Thus they skipped steps and now each project is responsible for writing code, that should have already existed in the framework. My message to gnome? Buy the GOF (Gang of Four) book, read it and live it. If you don't know what the GOF book is, you have no room to comment on the programming practices of either project. OO programming techniques are extremely important for GUIs. Gnome being written in C made it a challenge from the start. To rewrite or refactor the frameworks, would take an enourmous amount of time and effort (apps have to catch up).
KDE has made great strides in making a sound, extendible framework (See kparts, ioslaves, themes). However they have paid little attention to usability concerns from users. But by having a top notch infrastructure, they have ensure the project can adapt in the long term. The patterns and programming methodlogies the platform uses are modern and portable. If the KDE project stays true to its work, KDE4 will quite simply blow away gnome in the technical and user arenas. The primary goal of that release, is to revamp the usability of the DE.
Both projects actually have the same problem but with different results. They don't listen. Arguably, gnome suffers more and in a more critical area. KDE 3.3 was pretty miserable. 3.3 and is very stable and scalable.
Linus is a programmer, a very good one. He has simply pointed out the corner that gnome has painted themselves into by not utilizing true OO principles and modern design patterns. This a fact, not an opinion and is evident to any modern programmer. Gnome needs a paradigm shift to survive the long term. The KDE developers have put great effort into the KDE framework and it has paid off big time. Unfortunately, this meant the have ignored usability concerns. But usability is far easier to correct than poor frameworks and the lack of truely reusable code.
I use gnome on my desktop at home and KDE at work. But its common sense as to which platform has the better implementation.
The latest version of the connector does not function properly with Exchange 2003. I have observed this with SUSE10, gentoo and Ubuntu. Checking email takes 5 minutes, ldap/gal doesnt work at all and the calender app frequently loses appointments or changes the time on them. Exact same results on three different platforms. I used to like Evolution, but I must say its getting progressivly worse.
I dont want to have to run 5 apps that do not integrate with eachother to manage my business. I want a single app or better yet a pluggable infrastructure like kpart (see kontact) where I can run each component individually or in a single application.
Yeah I am not buying the using email offline is a power user function argument. That is insane. In a corporate environment you are constantly bombarded with attachments for various tasks. For people that travel frequently, webmail would not work. Once offline, they have no access their email or more importantly their attachments. I would hardly call this type of user a power user...
Webmail is simply a connector to an email provider. In the office that can work great. but it certainly is not a solution for all users, all the time.
egroupware is unfortunately very slow, even on a LAN. Its also not very modular. Plugging it into an existing authentication mechanism takes far too much effort. It has potential, but like openxchange, is a couple years off.
However, you have no right to complain unless you are giving something back to the project.
And that is exactly where I dissagree. If people are dissatisfied with a product they have every right to complain about it since they are using "public software". My point is that developers too frequently disregard complaints and give the typical response rather than catalog and create metrics. This is a severe problem with the OSS model. Complaints and Contrstuctive Criticism are often the same thing. There is usually no good way for actual users (people that only use software) to voice their concerns for many projects and even if there was, developers are normally too hard headed to listen with the two most common excuses - "Fix it yourself" or "give me money". Not everyone can code and no one likes to have money extorted from them. If people complain, its because they care in one way or another. Otherwise they would just not move on and not say a word. As a developer, the latter is absolutely the worst case senario - people not telling me when something isnt up to par.
In my opinion, OSS has never been able to break to mold due to this simple fact. Most of the truely successful OSS projects are commercialized (or at least heavy influence such as Qt, firefox, khtml, evolution). Notable exceptions include amarok, kopete, gthumb and gaim. Nautilus and Debian (2 years ago) are easy examples of what I am talking about.
So let me get this straight, if you are a user and do not plan to contribute you are not allowed to complain or point out faults. For after all you could fix them, even if you are an 80 year old grandma who can barely write emails. Everyone has the right to suggest changes, but they may not have the ability.
Sorry, I dont buy it. If people write software for the public domain than the public domain has every right to promote it or critisize it with absolutely no formal contributions. If OOS projects were smart (a few are), they would realize that these comments whether positive or negative could be used to better the project. The users that need to be polled are not the ones that know how to use IRC, forums or write their own code. Its the real users. KDE for example should have regular polling and provide a easy to use web based mechanism that allows regular users to see concepts and give feedback on what they want or need. Programmers do not make good usability analysts since they are be definition technical and have the skillset to avoid usability issues. Programmers should be implementors, not necessarily designers or dictators. Unfortunately, programmers are notoriously ego centric. This is no different in the open source community, however there are many notable exceptions.
By the way, I use an open source OS on my desktop every day.
Slashdot Mirror
When my generation was growing up
Word.
Yes there is.
a sterid=8112110/
http://ostg.pricegrabber.com/search_getprod.php/m
I always get a kick out of marketing hype surrounding wireless capabilities.
30 meters? I bet its more like 3 with marginal quality.
Think about this:
1. Java is far more reboust and atmoic than Parrot and Perl5. But its all preference.
2. There is a huge difference between loading swapped memory rather than reading data from disk and loading the classes in the VM. Loading from swap would take 1-2 seconds at the most. The first load of a JVM takes 5-6 seconds.
3. That _is_ poorly coded java. I believe this problem is not nearly as prevalent as it was 2-3 years ago but is still and always be an issue with all platforms.
In any even preloading a JVM should be an option, an user configurable adapter pattern. There are few applications these days that require native code. JIT environments are the future (In my opinion).
You completely missed the point. Making it an Operating System Component has nothing to do with a repository.
Essentially Operating Systems would startup a JRE when they startup. The actual memory footprint would be around 16MB for Java5. That way when apps loaded, they could reuse the existing JVM, sharing common classes with other applications. Think of dynamic libraries on steroids. It wouldnt just be a shared library but possibly instatiated objects.
As for a repository system for libraries, this uneeded. Applications ship with the libraries that they need rather than a perl application that forces the user to download them. Java applications can thus force exact version of libraries on the user without the user having to make decisions. Its a different way to do things, and in my opinion far better. For the desktop, Java has the ideal model. Let users run apps, without having to install libraries (and thus depend on an internet connection or other medium) just to get them to run.
"Reaganomics: Help the poor by giving to the rich!" rather than helping the poor directly."
I think you need to attend a more centrist univerisity. Reagan definitely had his good and bad imprints on US history. His economic decisions fueled our economy, promoted the technology boom and pulled us out of Carter's recession. Unfortunately he did not get to see his impact based on his illness. Bush Sr. and Clinton rode out his legacy for free (until the market collapsed).
That of course came at a price, and that was he was not very sympathetic to the real poor - people that cannot help themselves. And that is a shame. His overspending cut many programs and made many peoples lives harder. But even more benefitted. College enrollment exploded in the 80s fueling our technology/engineering foundation today. Children who otherwise had no change to get into (pay) for college could. I am not sure how any competent "economist" (and I used that term loosely) could possibly say reagonomics were summarily "bad".
I am a staunch liberal and even I can see his place in american history even though I disagreed strongly with many of his views and policies. I still can recognize his economic legacy that we enjoy today. The 80s could have turned out very differently if a different president just sat on the pot (like Bush Sr, Clinton, possibly Bush Jr). I look forward to the next time we elect a truely "Great" president and not just a sleezy politician.
This whole argument about GPLv3 is moot anyway. Why on earth would anyone want to relicense? Its like upgrading software that works just fine the way it is. I think Linus made the right call - There is no reason to promote a relicensing of the kernel. In fact there is no reason why anyone would.
Not many people license under the GPL anymore (compared to 3-5 years ago). Most use dual BSD/Apache style licensing now. This latest "update" to the GPL is pushing more companies away from OSS, not closer. I know my company wants nothing to do with GPL but instead dual licenses their OSS projects.
The linux kernel depends on corporate sponsership in a variety of ways. I think if Linus could reset the license, it too would have used something other than GPL.
Good call Linus, and keep up the good work!
Microsoft can do whatever they want with their products, they own them. Congress can regulate or compel any company that does business in America. Its called freedom, by the people. We can regulate any business entity in our own country, if they don't like it, they can leave =). As for the US forcing countries to use their currency? That is just ridiculous, lets see some proof =)
Sounds like you have a predisposition that you use to justify your ideals. Very strange way to form political foundations - "Building facts off of fiction".
Now i know the common response to this here is to say we live in a police state run by christian oil lovers and their cronies. But whats neat is, you can say whatever you want, its a free country =) I would just ask that you back up your statements and refrain from ad hominem attacks. It will make you more credible to the people in the middle and you will convince far more people instead of just fueling the extremists that believe everything they read.
You're missing the point. Of course vendors will pass on this cost to consumers, but they will not just "move on", they'll be pissed and they'll be more careful next time.
Negative. They wont be pissed because they dont care, its only 25 cents. Shipping portable drives unformatted will not happen. A few here and there might but all the big dogs will buck up. Its really not a big deal, this sort of process happens in many industries - all the time.
If this was say a $5.00 "tax", then they would explore alternate filesystems, but probably end up just shipping devices unformatted.
Its actually not in my opinion. kmail is lightweight and easy to use. Thunderbird takes a ridiculous amount of RAM to run on my machine - ~48m compared to ~17m with not just kmail but kontact. Although I have plenty of RAM on my desktop and I really dont notice the difference, it still makes you wonder about the moz platform. Firefox for instance is current sitting at 96m with slashdot loaded. konqueror right beside it is at 34m. Both have a single instance running.
I like firefox and thunderbird, but given the enourmous memory foorprint, you have to wonder what the future will be hold.
After this, electronics manufacturers are going to be reluctant to adopt any kind of Microsoft format.
;) I would prefer that their be an open/free filesystem for all removable devices, but there is no hope for this unless Apple gets behind it. Apple is turning into one of OSS's biggest allies.
At 25 cents? Vendors will just jack the price up and move on. No one wants to make a removable storage device that requires proprietary drivers and won't just work after being plugged in. Vendors don't want to pay for R&D costs to support a new format and then worry about users who have trouble using their devices. Its easier to just buck up. MS knows this and that is why they agreesively pursued this. Its easy money, they have nothing to lose.
Its also a nice sucker punch to Apple, which could have been the primary motivator
Okay so a cheap memory sticks runs for $15. So now its $15.25. Big deal. Consumer will not realize the difference and companies know this.
Nothing will change, they will stick with fat because its proven.
Pentium M single core 2.26GHz w/ 15.4" WUXGA LCD (This is going to be slower than a dual core 1.83, period.)
Yeah, I stopped reading right there.
I noticed this as well. I would recommend people completely disregard this test as its very impractical and inaccurate.
They are our representatives, yet they don't represent us.
BS - They do represent you, just like the represent record companies. The difference is the record companies utilize superior tactics persuading representatives to their point of view. And no, this isnt always in the form of bribes or benefits, there are good politicians on both sides of the isle that are simply misinformed. Record companies put out literature, hold roundtables and most importantly - "convince". Our voices get trampled because we simply do not speak their langauge. Simply throwing up your arms saying congress and hollywood sucks lets all join a pity party does absolutely nothing.
Anyone on our side of the debate that raises their voice is usually a fanatic or at least considered one (FSF is a good example). If you we really want to change the future and secure what we consider our rights, you we have to form organizations that at least have the potential to gain respect in the house and senate. It is the _only_ way to get our voices heard. I am sorry, I like the FSF but they come off no better than southern baptists - "We are right because we know we are but we really can't prove it, just trust us". Believe it or not, not every OSS advocate is a GPL fan and they dont need to be. OSS and fair use are concepts that can exist in a capitalist society if we want them to.
Prove the idea of fair use, let congressman see what a future would be like that is completely DRM centric and employ intelligent and professional lobbyists. Get the message heard, comprehended and acted on.
I think you missed the point. Since gnome does not have a solid framework, the usability issues that linus pointed out exist. If the framework was solid the look would be consistent. The graphical usability problems that gnome exihibts are a direct result of the lack of a complete, extendable, reusable framework. Each project must tackle trivial HIG issues on their own. This is also tied into the fact that gnome uses C. When using the KDE framework (not just Qt) this is all done for you. Developers can extend the frameworks to fit their needs while being forced to use certain patterns. Thus KDE (the apps) has a unified look and feel and gnome (the apps) does not. It does not help that applications like firefox and evolution are not actual gnome apps, but rather just use gtk.
gnome-vfs is a horrible example since is it prime indicator of the lack of organization in gnome. When applications want to directly interface with a file, they are not forced to use gnome-vfs (like gthumb). This causes a huge problem since the apps somtimes take it upon themselves to utilize their own methodologies. Good application frameworks both empower and contain developers.
The fact that gnome utlizes GObject simply says, "they know its a problem are trying to work around it." it only adds complexity to developers thus increasing time to delivery.
The problems with gnome that Linus has pointed out are well known to everyone except gnome fanboys and developers. The framework GTK/Gnome frameworks are in shambles. They are not easily extendible and do not utilize sound OO principles to promote code reusability.
Gnome has made great strides in making a simplified DE for linux. Unfortunately they paid no attention on how to get there (see toolbar, io slave examples). Thus they skipped steps and now each project is responsible for writing code, that should have already existed in the framework. My message to gnome? Buy the GOF (Gang of Four) book, read it and live it. If you don't know what the GOF book is, you have no room to comment on the programming practices of either project. OO programming techniques are extremely important for GUIs. Gnome being written in C made it a challenge from the start. To rewrite or refactor the frameworks, would take an enourmous amount of time and effort (apps have to catch up).
KDE has made great strides in making a sound, extendible framework (See kparts, ioslaves, themes). However they have paid little attention to usability concerns from users. But by having a top notch infrastructure, they have ensure the project can adapt in the long term. The patterns and programming methodlogies the platform uses are modern and portable. If the KDE project stays true to its work, KDE4 will quite simply blow away gnome in the technical and user arenas. The primary goal of that release, is to revamp the usability of the DE.
Both projects actually have the same problem but with different results. They don't listen. Arguably, gnome suffers more and in a more critical area. KDE 3.3 was pretty miserable. 3.3 and is very stable and scalable.
So let me guess, you are a gnome user =)
Linus is a programmer, a very good one. He has simply pointed out the corner that gnome has painted themselves into by not utilizing true OO principles and modern design patterns. This a fact, not an opinion and is evident to any modern programmer. Gnome needs a paradigm shift to survive the long term. The KDE developers have put great effort into the KDE framework and it has paid off big time. Unfortunately, this meant the have ignored usability concerns. But usability is far easier to correct than poor frameworks and the lack of truely reusable code.
I use gnome on my desktop at home and KDE at work. But its common sense as to which platform has the better implementation.
This just shows how much the US has fallen behind in the technology arena thanks to the Bush Administration (tm). Oh wait... Wrong thread, sorry.
The latest version of the connector does not function properly with Exchange 2003. I have observed this with SUSE10, gentoo and Ubuntu. Checking email takes 5 minutes, ldap/gal doesnt work at all and the calender app frequently loses appointments or changes the time on them. Exact same results on three different platforms. I used to like Evolution, but I must say its getting progressivly worse.
Its currently unusable.
Since its not a mistake, yes.
I dont want to have to run 5 apps that do not integrate with eachother to manage my business. I want a single app or better yet a pluggable infrastructure like kpart (see kontact) where I can run each component individually or in a single application.
Yeah I am not buying the using email offline is a power user function argument. That is insane. In a corporate environment you are constantly bombarded with attachments for various tasks. For people that travel frequently, webmail would not work. Once offline, they have no access their email or more importantly their attachments. I would hardly call this type of user a power user...
Webmail is simply a connector to an email provider. In the office that can work great. but it certainly is not a solution for all users, all the time.
egroupware is unfortunately very slow, even on a LAN. Its also not very modular. Plugging it into an existing authentication mechanism takes far too much effort. It has potential, but like openxchange, is a couple years off.
However, you have no right to complain unless you are giving something back to the project.
And that is exactly where I dissagree. If people are dissatisfied with a product they have every right to complain about it since they are using "public software". My point is that developers too frequently disregard complaints and give the typical response rather than catalog and create metrics. This is a severe problem with the OSS model. Complaints and Contrstuctive Criticism are often the same thing. There is usually no good way for actual users (people that only use software) to voice their concerns for many projects and even if there was, developers are normally too hard headed to listen with the two most common excuses - "Fix it yourself" or "give me money". Not everyone can code and no one likes to have money extorted from them. If people complain, its because they care in one way or another. Otherwise they would just not move on and not say a word. As a developer, the latter is absolutely the worst case senario - people not telling me when something isnt up to par.
In my opinion, OSS has never been able to break to mold due to this simple fact. Most of the truely successful OSS projects are commercialized (or at least heavy influence such as Qt, firefox, khtml, evolution). Notable exceptions include amarok, kopete, gthumb and gaim. Nautilus and Debian (2 years ago) are easy examples of what I am talking about.
So let me get this straight, if you are a user and do not plan to contribute you are not allowed to complain or point out faults. For after all you could fix them, even if you are an 80 year old grandma who can barely write emails. Everyone has the right to suggest changes, but they may not have the ability.
Sorry, I dont buy it. If people write software for the public domain than the public domain has every right to promote it or critisize it with absolutely no formal contributions. If OOS projects were smart (a few are), they would realize that these comments whether positive or negative could be used to better the project. The users that need to be polled are not the ones that know how to use IRC, forums or write their own code. Its the real users. KDE for example should have regular polling and provide a easy to use web based mechanism that allows regular users to see concepts and give feedback on what they want or need. Programmers do not make good usability analysts since they are be definition technical and have the skillset to avoid usability issues. Programmers should be implementors, not necessarily designers or dictators. Unfortunately, programmers are notoriously ego centric. This is no different in the open source community, however there are many notable exceptions.
By the way, I use an open source OS on my desktop every day.