It's very easy to reproduce these problems: 1. Download the Firefox installer for your platform from mozilla.org. This could be the latest 3.6 release, or the latest 4.0 beta. 2. Install it. Make sure you do not install any sort of plugins or extensions. We want a clean, default Firefox installation. 3. Browse the web for 30 minutes. Visit a variety of sites, including Slashdot, Facebook, and some of the popular news sites. 4. Use top or the Task Manager or whatever your system offers to see the memory usage of the Firefox-related processes. Notice that they'll be in the gigabytes.
I have had firefox running for over a week now. I've visited numerous sites, loads of tabs of Slashdot, had previously many open tabs on furaffunity and various sites but eventually closed them. There is one window with a god awful amount of tabs, infact all the monster.co.uk's jobs for Glasgow and that has been sitting there since Monday. So, you would think this would be a good candidate to observe you problem. Now, my system as a god awful amount of memory too, yet.. What is Firefox using?
According to http://dl.dropbox.com/u/58565/firefox-memory.jpg It's just over 300MB and I was expecting far more with the amount of tabs I have open. I doubt foxyproxy or firefox sync (the only addons I have) magically made the "memory leak" go away.
Can you explain why I am not seeing this issue please?
If you have to jump through a few hoops (as you tend to do with Linux/Unix), then it will remove a lot of chaff from the sysadmin field who don't want to bother.
I find it easier and faster to setup Linux servers in most scenarios since it's just yast -i package or sudo apt-get install package, then changing a few things via the configuration screen. This compared to screwing around with GUIs that don't have standardized installation procedures, unreliable update mechanisms etc.
You're making it sound like it's such an onerous requirement.
No, I'm saying that it's simply not "right off the box" as claimed. Other mp3 players I have used however are, I just drag and drop and go.
and there is a downloadable version for Windows.
Indeed, thus I don't see the "right off the box" as claimed.
perhaps installing iTunes on Windows is not much of a problem to a large percentage of users.
Again, you're missing the point, it's not about how difficult it to install something, it's the claim about it being "right off the box" for what you would call "a large percentage of users".
Maybe it's just me, but my computer came with iTunes pre-installed.
The only computers I'm aware of that have iTunes preinstalled don't have a majority market share. So, very likely a very small percentage of users, yes.
Maybe it's just me... But I don't consider having to install iTunes, maintain a music collection inside iTunes's library "right off the box" when compared to drag and dropping files onto a drive like other mp3 players.
I'm not SIP expert, but I suspect similar issues are going on there as well.
SIP is more of an addressbook server, not the target location you're connecting to. I don't really see this going away. SFTP works because the target machine is one and the same and you know exactly who and where you're connecting to. You can't do that with SIP, because you're asking SIP, "I want to call X" and SIP will reply, "You can call via the Internet Protocol addresses X, Y, Z on protocols P, Q using codecs T, G, D"
then people will wonder why they don't just do an extremely good IPv4 NAT solution and go IPv4->IPv4 and drop the entire IPv6 part.
They won't because then they run out of IPv4 addresses. The idea is that the local pool of IPv4 addresses can be entirely different from the IPv4 addresses on the other side.
That sounds nice, but in practice you probably wouldn't be able to connect at all. At least one side must have a public IP address for P2P to work (with TCP), or at least be able to open incoming ports with something like UPnP. What do you think the odds are of ISPs letting customers reserve incoming ports?
Probably will just use a form of hole punching like a lot of peer2peer stuff is already.
There is a difference between saying "I assume X is not possible" verses "X is not possible".
If you have a GPU that only supports DX9 then that is what it will use.
Which was stated by Microsoft already in channel 9 videos as I recall in which they mentioned slower routines are used for backwards compatibility with older hardware and certain shaders would not be available.
If anything, you giving that example serves to help my argument - as there is no reason why Shattered Horizon could not work on XP as clearly it happily works with DX9 cards.
Except for the fact that the software (Windows XP) doesn't do DX10, which is all I care about at this point and likely all the user of the game cares about too and before you suggest it, km-soft's poor DX10 implementation does not work with Shattered Horizon and has not worked with other DX10 games I own either.
If you want to talk about what's possible, yes, it's entirely possible to make any kind of software abstraction layer for anything API related for practically any mainstream platform out there. Doesn't mean it doesn't require a large amount of resources to produce it.
happily works with DX9 cards.
Not really, the FPS in the game wasn't that great until I upgraded to a later card model that had DX10 support (which amusingly had less video ram).
what dx10 brings over dx9, is only water shading in certain circumstances. (beaches, lakesides). the difference wouldnt be noticeable in many gaming environments unfortunately, and they are noticeable in crysis because it passes in an island. not to mention that hacking the game makes it play with dx10 standards, ironically, under dx9, with all the bells and whistles dx10 would bring...
I don't know if that's what it's used for. I just know that I have games that will not work on DX9, period. I don't know why games like Shattered Horizon would use it then since it's purely a space shooter, but I assume they had a legitimate reason of some kind. Either way, I can't play those games on XP.
Can we stop with the FUD & look at the *REALITY* of the situation?
I did, the reality is that many kinds of security related patches will not be made for XP any more, exploits such as the one I mentioned can be exploited through malware used on websites in various ways. One example is embedding windows media player component inside a webpage to play a DRMed media stream that needs to download managed code licensing, which uses the mentioned exploit to take control of the system.
As a home user of XP, there is nobody sat there on the big bad Internet waiting to pounce on me the moment I connect my PC up.
Of course not, you'll eventually come to them. That's how most malware works these days.
Big corporations are much bigger & juicier targets for malware
I disagree, credit card, bank details from individual users who are not experienced or even have any idea what is going on are generally a bigger and less risky target.
So, in other words, if you don't act stupid then XP is probably just as secure as Vista or 7.
I have three games right now that require DX10 or higher, and I'm not even that much of a gamer... From a gamer perspective, I suspect some may have moved to a newer operating system for that reason, which admittedly, even if all gamers moved to a higher version of Windows, is likely an insignificant amount of people in the percentage of the overall statistics.
Are you joking? It's tiny in comparison to Second life, it's not even 1% the size.
Media streaming tech and servers.
How is that meant to be impressive? Media streaming technology has existed for years, xbox has had it for years for their video services. Meanwhile, Home shows how great it is with streaming when it can't even load an area live on the spot, it has to download an entire area first unlike other virtual world platforms like Second life.
Or am I supposed to be impressed by the fact they have just two episodes of something from adult swim on repeat for months on end with trailers all the time?
Microsoft would have to spend years coming up with anything remotely similar.
Speaking of which, Second life has media streaming support built in, you just select an object's face and feed it a URL, be it a video link, streaming server or even a webpage. In comparison, PS3 home needs a lot work, it can't even handle dynamic updates to an area like Second life, everything has to be prerendered, prefetched and prebuilt.
This is Slashdot... Don't you check the sources? The source was Tizzers Foxchase's twitter account, a known griefer from the banned Woodbury University / Wrong Hands group on Second life. Microsoft aren't buying Second life.
UNIX compliance is decided by the Open Group. Posix compliance is an IEEE standard. Windows can have a POSIX layer without being POSIX compliance and Posix compliance doesn't imply UNIX(tm) compliance.
I don't see anything there that again, disputes the acronym that Apple gave to the kernel which is the question, which what I keep bringing up and you keep conveniently(?) ignoring.
As far as the fork without exec. Are you sure that's not a POSIX requirement?
Considering things like "The Single UNIX Specification", which includes POSIX, pretty sure.
Has the Open Group certified Windows as being Unix(tm)?
Microsoft's POSIX subsystem has been previously certified, yes. Then later a company called Interix evolved the subsystem later and implemented the later POSIX specifications on top of it along with a full userland, Microsoft ate them and they became "Windows Services for UNIX" which holds a UNIX certification from opengroup (it's not on the first page of Google when looking for 'Windows Services for UNIX certification' and I honestly cannot be assed to go further). The POSIX subsystem it self is shipped with Windows without the Windows Services for UNIX userland which uses it. Although there are the UNIX based applications still shipped with windows like FTP, Telnet etc. just like OS X.
Well you don't get to decide what is and is not Unix(tm).
I'm not making a decision here, it's obvious, it is not UNIX and it's no more based on UNIX than Windows is, they both hold certifications and they both contain UNIX code, which doesn't happen to be even close to the majority of code in either.
And to go further into the matter, while there is certification saying OS X is compliant with UNIX standards, their test case applications obviously did not take into account everything into the POSIX specification which is why OS X can fail at porting basic UNIX applications that are fully compliant with the specification (such as in my forking example where you refused to even address the point), you can't claim it's compliant to the specification when I just pointed an area where it is. Now, I could go and bitch at opengroup for not making enough test cases to enforce their specifications, but to be frank, it's hard enough trying to push bugs to get fixed from other developers without having to deal with these groups.
I wouldn't be opposed to saying there are pieces of OS X based on UNIX, but I am opposed to the implication that it is majority based on UNIX. There is very little of UNIX code, philosophy, structures even used in the entire operating system by comparison, that's not to say there aren't adaptations, but it isn't what people appear to be claiming in earlier threads.
But, the problem with the previous posts, is that they even go further and claim that because there is UNIX code in there, it's magically more secure. It's like saying that because I included a random.c file from UNIX in my application, it's been turned magically secure. Then to go again, even further and claim that a platform that had a history of many viruses has had few viruses is just ludicrous and obviously looking like a lot of this is marketing hype. You can write some of the worst most insecure services/daemon code in the world, the fact it runs on UNIX or Windows generally wouldn't make that much of a difference, the application will get penetrated either way.
But, I digress...
In your original reply, you said that "The Open Group, you know the people who certify a system as Unix, disagrees with you....." about "X is Not Unix" (you can see the definition on Apple's own website).
So, I looked at the URL given and don't see any protest about the acronym for the XNU kernel, which means "X is Not Unix" on the page. Perhaps you could point it out more clearly for me? I can't even find the word 'XNU' or 'acronym', or is Firefox find just not working for me because I'm not at a computer that "just works" at this moment?
http://en.wikipedia.org/wiki/Mac_OS_X: "Mac OS X (pronounced/mæk o s tn/ mak oh es ten)[6] is a series of Unix-based operating systems and graphical user interfaces...."
So, I went to the wikipedia article, went to the sources list and they verified what I mentioned. The BSD subsystem etc. I don't see your point? Just because Wikipedia and Apple make a bad summary that a subsystem is apparently UNIX based doesn't mean anything, the fact Apple passes certification but doesn't comply to UNIX specifications (as mentioned in my other post) doesn't make it anymore of a UNIX than Windows which by the way, is more compliant with the specifications in it's POSIX subsystem which also has UNIX certification, where Windows also is sharing BSD code in it's subsystem, in some small utilities included with the OS etc. just like OS X.
As of 2006, there are relatively few security exploits targeting Mac OS X (with a Unix-based file system and kernel)."
How is this relevant to "UNIX" ? Windows is UNIX too if you're going by the vague reasoning OS X is a "real UNIX" and that didn't magically protect it. How is that relevant to my previous comment where UNIX has in the past been a major attack vector for worms, unlike the previous claim that apparently "real UNIX that also has very few wild viruses", UNIX has had many viruses.
Slashdot Mirror
I have had firefox running for over a week now. I've visited numerous sites, loads of tabs of Slashdot, had previously many open tabs on furaffunity and various sites but eventually closed them. There is one window with a god awful amount of tabs, infact all the monster.co.uk's jobs for Glasgow and that has been sitting there since Monday. So, you would think this would be a good candidate to observe you problem. Now, my system as a god awful amount of memory too, yet.. What is Firefox using?
According to http://dl.dropbox.com/u/58565/firefox-memory.jpg It's just over 300MB and I was expecting far more with the amount of tabs I have open. I doubt foxyproxy or firefox sync (the only addons I have) magically made the "memory leak" go away.
Can you explain why I am not seeing this issue please?
Maybe apple should improve their x11 implementation instead since so much uses it.
Solaris of course.
I find it easier and faster to setup Linux servers in most scenarios since it's just yast -i package or sudo apt-get install package, then changing a few things via the configuration screen. This compared to screwing around with GUIs that don't have standardized installation procedures, unreliable update mechanisms etc.
Yes.
Often doesn't return the results I am interested in on the first few pages, Google does for me. So for my personal use, Google is better.
You have my condolences on all three.
No, I'm saying that it's simply not "right off the box" as claimed. Other mp3 players I have used however are, I just drag and drop and go.
Indeed, thus I don't see the "right off the box" as claimed.
Again, you're missing the point, it's not about how difficult it to install something, it's the claim about it being "right off the box" for what you would call "a large percentage of users".
I've been able to use Facebook and Steam on Linux for years. I don't get your point.
In other words, Microsoft's ads are lying to us D:
http://www.youtube.com/watch?v=MN28Ad3TNrQ
http://www.youtube.com/watch?v=H0f8X_SOVjA
http://www.youtube.com/watch?v=6jMt6saTqq4
http://www.youtube.com/watch?v=ZSkaTcjDIMk
http://www.youtube.com/watch?v=DfMfek7IRWA
http://www.youtube.com/watch?v=CffImVdKCVE
Why would Microsoft lie to us? Oh the humanity, it's a trap!
The only computers I'm aware of that have iTunes preinstalled don't have a majority market share. So, very likely a very small percentage of users, yes.
Note: I am not the grand father poster.
Maybe it's just me... But I don't consider having to install iTunes, maintain a music collection inside iTunes's library "right off the box" when compared to drag and dropping files onto a drive like other mp3 players.
Indeed.
SIP is more of an addressbook server, not the target location you're connecting to. I don't really see this going away. SFTP works because the target machine is one and the same and you know exactly who and where you're connecting to. You can't do that with SIP, because you're asking SIP, "I want to call X" and SIP will reply, "You can call via the Internet Protocol addresses X, Y, Z on protocols P, Q using codecs T, G, D"
They won't because then they run out of IPv4 addresses. The idea is that the local pool of IPv4 addresses can be entirely different from the IPv4 addresses on the other side.
Probably will just use a form of hole punching like a lot of peer2peer stuff is already.
There is a difference between saying "I assume X is not possible" verses "X is not possible".
Which was stated by Microsoft already in channel 9 videos as I recall in which they mentioned slower routines are used for backwards compatibility with older hardware and certain shaders would not be available.
Except for the fact that the software (Windows XP) doesn't do DX10, which is all I care about at this point and likely all the user of the game cares about too and before you suggest it, km-soft's poor DX10 implementation does not work with Shattered Horizon and has not worked with other DX10 games I own either.
If you want to talk about what's possible, yes, it's entirely possible to make any kind of software abstraction layer for anything API related for practically any mainstream platform out there. Doesn't mean it doesn't require a large amount of resources to produce it.
Not really, the FPS in the game wasn't that great until I upgraded to a later card model that had DX10 support (which amusingly had less video ram).
I don't know if that's what it's used for. I just know that I have games that will not work on DX9, period. I don't know why games like Shattered Horizon would use it then since it's purely a space shooter, but I assume they had a legitimate reason of some kind. Either way, I can't play those games on XP.
It's even an incomplete list too, FUEL, while not DX10 only, is a DX10 game and it's not on there.
Indeed, it's still missing games on the list though.
Actually, I was running Shattered Horizon, which is a DX10 only game on a DX9 graphics card on Windows Vista. Please stop spreading misinformation.
I did, the reality is that many kinds of security related patches will not be made for XP any more, exploits such as the one I mentioned can be exploited through malware used on websites in various ways. One example is embedding windows media player component inside a webpage to play a DRMed media stream that needs to download managed code licensing, which uses the mentioned exploit to take control of the system.
Of course not, you'll eventually come to them. That's how most malware works these days.
I disagree, credit card, bank details from individual users who are not experienced or even have any idea what is going on are generally a bigger and less risky target.
Not in the scenario I suggested.
I have three games right now that require DX10 or higher, and I'm not even that much of a gamer... From a gamer perspective, I suspect some may have moved to a newer operating system for that reason, which admittedly, even if all gamers moved to a higher version of Windows, is likely an insignificant amount of people in the percentage of the overall statistics.
I will clarify for your benefit what is going on with XP:
No more mainstream patches. The only patches that are done now are funded by large corporations. Don't expect things like http://secunia.com/community/advisories/windows_insecure_library_loading/ to get ever fixed properly in Windows XP's system software now.
Are you joking? It's tiny in comparison to Second life, it's not even 1% the size.
How is that meant to be impressive? Media streaming technology has existed for years, xbox has had it for years for their video services. Meanwhile, Home shows how great it is with streaming when it can't even load an area live on the spot, it has to download an entire area first unlike other virtual world platforms like Second life.
Or am I supposed to be impressed by the fact they have just two episodes of something from adult swim on repeat for months on end with trailers all the time?
Speaking of which, Second life has media streaming support built in, you just select an object's face and feed it a URL, be it a video link, streaming server or even a webpage. In comparison, PS3 home needs a lot work, it can't even handle dynamic updates to an area like Second life, everything has to be prerendered, prefetched and prebuilt.
This is Slashdot... Don't you check the sources? The source was Tizzers Foxchase's twitter account, a known griefer from the banned Woodbury University / Wrong Hands group on Second life. Microsoft aren't buying Second life.
If you say so...
Then we have nothing to discuss on this matter, because that's not what the previous comments were saying.
I don't see anything there that again, disputes the acronym that Apple gave to the kernel which is the question, which what I keep bringing up and you keep conveniently(?) ignoring.
Considering things like "The Single UNIX Specification", which includes POSIX, pretty sure.
My friend who owns "Caprica" on Second life would likely rent a plot to them if they asked.
Microsoft's POSIX subsystem has been previously certified, yes. Then later a company called Interix evolved the subsystem later and implemented the later POSIX specifications on top of it along with a full userland, Microsoft ate them and they became "Windows Services for UNIX" which holds a UNIX certification from opengroup (it's not on the first page of Google when looking for 'Windows Services for UNIX certification' and I honestly cannot be assed to go further). The POSIX subsystem it self is shipped with Windows without the Windows Services for UNIX userland which uses it. Although there are the UNIX based applications still shipped with windows like FTP, Telnet etc. just like OS X.
I'm not making a decision here, it's obvious, it is not UNIX and it's no more based on UNIX than Windows is, they both hold certifications and they both contain UNIX code, which doesn't happen to be even close to the majority of code in either.
And to go further into the matter, while there is certification saying OS X is compliant with UNIX standards, their test case applications obviously did not take into account everything into the POSIX specification which is why OS X can fail at porting basic UNIX applications that are fully compliant with the specification (such as in my forking example where you refused to even address the point), you can't claim it's compliant to the specification when I just pointed an area where it is. Now, I could go and bitch at opengroup for not making enough test cases to enforce their specifications, but to be frank, it's hard enough trying to push bugs to get fixed from other developers without having to deal with these groups.
I wouldn't be opposed to saying there are pieces of OS X based on UNIX, but I am opposed to the implication that it is majority based on UNIX. There is very little of UNIX code, philosophy, structures even used in the entire operating system by comparison, that's not to say there aren't adaptations, but it isn't what people appear to be claiming in earlier threads.
But, the problem with the previous posts, is that they even go further and claim that because there is UNIX code in there, it's magically more secure. It's like saying that because I included a random .c file from UNIX in my application, it's been turned magically secure. Then to go again, even further and claim that a platform that had a history of many viruses has had few viruses is just ludicrous and obviously looking like a lot of this is marketing hype. You can write some of the worst most insecure services/daemon code in the world, the fact it runs on UNIX or Windows generally wouldn't make that much of a difference, the application will get penetrated either way.
But, I digress...
In your original reply, you said that "The Open Group, you know the people who certify a system as Unix, disagrees with you....." about "X is Not Unix" (you can see the definition on Apple's own website).
So, I looked at the URL given and don't see any protest about the acronym for the XNU kernel, which means "X is Not Unix" on the page. Perhaps you could point it out more clearly for me? I can't even find the word 'XNU' or 'acronym', or is Firefox find just not working for me because I'm not at a computer that "just works" at this moment?
So, I went to the wikipedia article, went to the sources list and they verified what I mentioned. The BSD subsystem etc. I don't see your point? Just because Wikipedia and Apple make a bad summary that a subsystem is apparently UNIX based doesn't mean anything, the fact Apple passes certification but doesn't comply to UNIX specifications (as mentioned in my other post) doesn't make it anymore of a UNIX than Windows which by the way, is more compliant with the specifications in it's POSIX subsystem which also has UNIX certification, where Windows also is sharing BSD code in it's subsystem, in some small utilities included with the OS etc. just like OS X.
How is this relevant to "UNIX" ? Windows is UNIX too if you're going by the vague reasoning OS X is a "real UNIX" and that didn't magically protect it. How is that relevant to my previous comment where UNIX has in the past been a major attack vector for worms, unlike the previous claim that apparently "real UNIX that also has very few wild viruses", UNIX has had many viruses.