That's the article excerpt in a nutshell. It doesn't explain what it does or how it does it.
Yup, that's exactly how I feel. Aside from the part about using a new XML-based config file format (I can pretty much imagine how that will look), there aren't a lot of details.
For example, startup scripts are called that because they are scripts. Sometimes they do more than just launch a program (e.g. a firewall script which sets a bunch of ipfw rules). Somehow I do not associate XML config files with scriptability. Based on what I've read so far, I have the feeling that I'm going to end up just "launching" several of my old scripts.
Likewise, I wonder what happens when you add some weird new third-party service to your machine. How does launchd know that my distributed batch scheduler needs several NFS mounts to have completed successfully? If launchd can't auto-detect this sort of thing, then we're back to scripting again, either by me or by that third-party developer.
What it boils down to is that I need to see this thing in action.
RTFC! (the "C" is for Comments, or rather Ars' discussion section for this article)
Comment by Deffexor (Ars Audio/Visual Moderator): The problem with including the iPod Shuffle in the review was that we then would have had to include a bunch of other USB based Audio players in the review. Then to make matters worse, we would have had to benchmark/test the audio players, etc. This article is already 12 pages and I wasn't going to push it beyond that...
The good news is that I'm planning a portable audio player (flash based) round-up for May and will include the iShuffle.
As for the 20 year estimate, that sounds more the result of negotiations with the transit workers union than ability to get things switch over. You know City Hall, when it comes to a budget, they suddenly know the value of each penny and would switch the whole thing over in a couple years, tops.
As someone who actually lives in New York, I would say that it probably has more to do with the fact that MTA doesn't have the money to do any of this. They're just sort of hoping to have the cash to do this (and extend the 7 line to the West, and refurbish the south ferry stop on the 1/9, and build a 2nd avenue line, and repair any number of stations, and do all the other things they supposedly want to do) in 20 years.
Talking about video drivers shows how much Linus is not ready for the Laptop. If this is a problem, how much are audio, USB, FireWire, and WLAN are going to be a problem?
Extrapolation is bad. There's a known problem with video support for the latest 3d accelerated video cards (2d support is there), but that does not imply that other hardware is not supported.
Having just bought a new laptop and installed Linux on it (to replace an old laptop with Linux on it) I can tell you that audio, USB, and FireWire aren't a problem. There are only so many mobile chipsets and only so many integrated audio/USB/FireWire solutions which go with those. WLAN is a problem, most likely due to the lack of availability of hardware specs (as with video).
That if, say, you're an English speaker unfamiliar with the "i's" long "e" sound, you'll be likely to pronounce it "Man-dreyeva"
Yes, now that all of the English speakers have had their chance to say that the name of a French company sounds weird to them (gee, really?) do any French speakers have something to say about this? It's not supposed to be pronounced like "monde riva" or something is it?
the whole delete your cookies thing is silly. i run several web sites that use cookies to track logins, not for me to track them but for the site to track who is logged in.
Translation: I'm doing the right thing, so obviously the other 99.9999% of the world is as well and we are all "fools" for believing otherwise.
Please, most websites try to hit me with a doubleclick.net cookie or an advertising.com cookie right away. I'm no "fool" for deleting that sort of thing. Nor am I a fool for deleting all of the miscellaneous cookies I get, e.g. from misconfigured sites which leave Apache's mod_unique_id enabled for no reason.
As for user-tracking cookies, which may well be useful, there are two kinds: session cookies (which my browser does and should delete at the end of the session) and unreliable ones (e.g. ones which treat everyone on a public terminal or a family computer system as the same person). Ditching isn't such a bad idea (though I personally leave a few around from sites which I do use and trust).
Remember, the web is not a publication medium. It is designed to be interpreted by the user's web browser. If the user turns off images, they will see no images. If they turn off flash, there's no flash. If they use a screen-reader... well, you get the idea. That's the way the web was always intended to work. Turning off/deleting cookies is no different. The user controls the experience, plain and simple, and apparently 58% of people have decided to do that. Good for them, especially given the number of junk cookies out there.
The PSSU feature, though (as I mentioned in another post), that blocks incoming traffic on first boot and immediately directs the user to download updates is awesome. Why other companies haven't thought about this, I have no idea.
In the last several versions of Linux that I've installed, downloading the latest updates has been part of the install process. So I'm thinking that maybe other companies have thought of this, though they arrived at a slightly different solution.
Now, if Windows did that on _every_ boot (configurably, I would hope), that would be interesting. It might be useful for a server which have been offline for a while, e.g. after a power outage, fire, or other similar emergency.
In all fairness, this would probably have been the advice which the average slashdotter would have given if Apple weren't involved.
The problem is that it's not a particularly accurate complaint. There are some online music stores out there which sell mp3s, and the Shuffle can play those just fine. Not to mention other sources of mp3-based music, such as archive.org and, you know, ripping CDs which you own (remember those?) Not to mention that subscription-based services like Napster are actually worse in this regard.
On the other hand, with the wide variety of online music stores out there, the whole argument that iTunes is a reason why people should buy Apple products doesn't make much sense either. It's not as though every store has every song that anyone could ever want. _If_ iTunes' selection matches your tastes, _then_ it makes sense to go with an Apple product. (Then you get to hope that your taste in music doesn't change)
*Any* local exploit is *also* a potential remote exploit (just like the IRC conversation shows). I had someone nearly pwn a box of mine by using an exploit in a buggy PHP script, then trying to elevate privileges through a local exploit.
How can a fork bomb be used as a remote exploit? You've just eliminated all the meaning from the words "local" and "remote". I mean really, even in your own example it was a real remote exploit (buggy PHP script) which allowed access to the system in the first place.
Yes local exploits are dangerous (I don't think the grandparent post disagreed with that) but without remote access to the machine, in practice that system will be fine. So no, they're not the same thing. The remote access/remote exploit is by far the more dangerous part. Once an attacker has that, they can do all sorts of things even without privilege escalation (filling up the HD with junk files, sending out spam e-mails, etc.). Unless of course you very carefully protect against every one of those things, but that's awfully hard. Show me an OS which by default would prevent an attacker who got access to a non-root account from sending out spam, and I'll show you an OS with no TCP/IP stack.;)
There can be no distinction between uploading and downloading via P2P. There is no client-server relationship (thats what peer-to-peer means!).
Even if you don't make anything else available for sharing, chunks of the file you are currently downloading can and do get shared as you are downloading it.
The reason why nobody raised this issue is that it you are factually incorrect. You most certainly can download and not upload with any number of P2P applications. With some programs all you have to do is not put anything in your shared directory and not download into that directory. While other programs like BitTorrent do attempt to prevent this type of behavior, proper firewalling can deal with it.
What it boils down to is this: P2P is still TCP/IP. There's still a connection from a client to a server. With P2P the servers are individuals' systems rather than big webservers somewhere, and there is often one connection in each direction, but that's it. If one peer wants to connect to you but not allow the reverse, it can be done.
You may not get "faster CPU" or something of the sort, but you do get a great OS, tiny case made of adonized aluminium, a practically silent computer, great software-bundle and the like. Staring at just few specs (CPU-speed etc.) is pretty short-sighted IMO.
See, the problem is that I'm not convinced that you can't get all that for less than $825 _with_ a faster CPU and so on. Of course, that leads to the usual PC-vs-Mac pricing flamewars, which I'm trying to avoid.
Also, I could just as easily say that you shouldn't ignore the specs in favor of stuff like the material the case is made of. You may get a pretty computer, but will the experience of using it be pleasant (e.g. because of the slow processor, limited video paired with graphics-intensive OS, etc.)"?
What it boils down to is this: the specs aren't exactly impressive. Yes the Mac mini may have other strengths, but that doesn't change the numbers. It's not short-sighted to accurately note those numbers, either. As I pointed out above, they could reflect upon other deeper issues, like whether the user experience is good.
While I wouldn't call you a zealot, I do think you glossed over one thing: CPU speed. No, most people don't need half the speed their computers have, but $825 (not including monitor, keyboard, etc) isn't exactly my idea of a "cheap" system. For that money I would expect a faster CPU. More video memory would be nice too.
Of course, I assume that small size and running quietly are two of the Mac mini's big advantages, and a faster CPU might cause problems there. Still, I don't know, it just seems to me that the Mac mini's specs seem low end. Couldn't they have put a G5 in there? Yes, I'm sure that the build quality and being able to run OS X is worth something, but somehow it still doesn't seem right.
Not surprising. The doppler effect shouldn't come into play until more like 80,000mph. At that point your wireless connection will be unusable (the data will be corrupted). Also, you might notice increased latency as you move further away from Sprint's satellites and the Earth in general.
Whether this is patentable is another issue. But you can certainly patent a published idea -- it's the only way to protect it.
I assume you mean that you can patent _your_ published idea, right? At least, that's what the idea is supposed to be.;)
This patent sounds like comparing pointers, which has been doable forever. Pointers are ultimately just numbers (after taking type into account and so on), so basically this has been doable as long as there have been pointers. Now whether it's been doable in BASIC is a whole other question. I don't remember anything about pointers in BASIC 20 years ago. Still, a patent which boils down to "comparing two numbers" does seem a bit broad to me, even if it's specific to a certain programming language.
I'm sick and tired of hearing this bullcrap. For the past 20-30 years, there's been nothing but active encouragement for women to denounce their traditional gender roles and perform tasks normally associated with men.
Wrong. You're confusing the culture with the people in it. In other words, while the feminist movement has led to a general sense that women should denounce their traditional gender roles, that doesn't mean that many or most individuals have acted in a manner consistent with this (they just pay lip service). If the increase in sexual harrassment lawsuits has taught us nothing else, it's that some guys are still old-fashioned. And Lord knows that it should be easy enough to find religious types who think that women should stay at home. For a more computer-specific example, notice the strong connection between video games and scantily-dressed women (in the games, as models at trade shows, in advertisements, and so on). Few scantily-dressed men appear in video games, though of course the programmers are usually men. What kind of impression do you think this higly visible part of the computer industry leaves on people?
Well, slashdot, you can't have it both ways, and no, predatory business practices and high prices don't excuse people infringing copyrights.
Is it not possible that there's one group of posters and moderators who support not infringing the GPL, and a _totally different group_ (possibly a younger one:) which supports infringing copyrights? It this perspective so out there that people have a hard time figuring it out for themselves?
I mean, really, the parent post mentions that there are "many views here", but that information does not seem to be a part of the poster's thought process. Yes, "many views", or "two sets of people", does explain everything. Period. Yes, it is easier to think of SlashDot as some sort of unified intelligence than to think of a collection of people. But SlashDot is just a collection of people. Assigning opinions to SlashDot itself is sophistry. Who exactly is the "you" in "you can't have it both ways"?
I say that as someone who opposes both GPL infringment and file sharing. I exist, thank you!
There are many Linux programs based on libnjb which allow Creative mp3 players to be used with Linux. One of them is this project which causes the device to appear as a file system in KDE. What did the grandparent post say again? Something about "easy to use KDE software"?
As another example, I use my Dell DJ (manufactured by Creative) on Linux with gnomad2. So yes, Linux support for Creative-built mp3 players does exist. "Excellent" might be an exaggeration since you may have to install from source, but that's about it.
Needless to say, it is the parent of _this_ post which should be modded down (or at least not modded up so highly). The fact that a particular individual cannot do something rarely means that it cannot be done, but people tend to make that generalization anyway. And moderators tend to believe them.
Apple is doing this stuff (e.g. you can mount WebDAV servers), but Apple is doing it right by integrating network resources into the real VFS layer so that all applications can access them. KDE's I/O slaves are not real filesystems and are not accessible by all applications.
But I thought the very nature of OSS makes this sort of thing impossible. What did I miss?
The whole point.
OSS (supposedly) makes bugs easier to find -- that's the advantage it gives. I can go back to Linux kernel 1.0 (say) and probably find an unpatched bug. I could also go back and look at Windows 95, but I'd have a much harder time finding unpatched bugs, having no access to the source code. It's easier to find bugs in OSS code.
If I do find a bug in either, though, the bug is not likely to be fixed anytime soon in either case.:) The simple reason being that nobody is bothering to fix that code anymore (Okay, technically I could patch the Linux kernel code if I really wanted to, but why bother with such old code?) Unfortunately, the same can be said for the Mozilla project in this case. Nobody is bothering to fix these old bugs, and for code which is "current", that's a bad problem.
It's also an organizational problem, not an open vs. closed source problem. The issue is that nobody is writing the patch to fix a known problem. As advanced as modern software is, there is as of yet no program, OSS or otherwise, which I know of which writes patches for itself! Humans have to write the code, plain and simple, and I really doubt that any OSS advocate ever claimed otherwise.
Slashdot Mirror
I've been running it thru cron at rather short intervals for 6 months now, and dropped 340 ips into hosts.deny.
/etc/ssh/sshd_deny_list \
/etc/ssh/sshd_deny_list.
One possible improvement -- have your script run whenever there's an ssh login:
hosts.allow
-----------
sshd: ALL EXCEPT
: spawn (/etc/cron.d/check_for_ssh_scans) & \
: ALLOW
Where the script "check_for_ssh_scans" adds any offending IPs it finds to
For the undead crowd out there:
OpenBSD is affected, and was patched [openbsd.org] on the 6th of June
Oops... You meant 'July', not June.
Yes, apparently the undead have psychic powers, perhaps from eating all those delicious brains.
That's the article excerpt in a nutshell. It doesn't explain what it does or how it does it.
Yup, that's exactly how I feel. Aside from the part about using a new XML-based config file format (I can pretty much imagine how that will look), there aren't a lot of details.
For example, startup scripts are called that because they are scripts. Sometimes they do more than just launch a program (e.g. a firewall script which sets a bunch of ipfw rules). Somehow I do not associate XML config files with scriptability. Based on what I've read so far, I have the feeling that I'm going to end up just "launching" several of my old scripts.
Likewise, I wonder what happens when you add some weird new third-party service to your machine. How does launchd know that my distributed batch scheduler needs several NFS mounts to have completed successfully? If launchd can't auto-detect this sort of thing, then we're back to scripting again, either by me or by that third-party developer.
What it boils down to is that I need to see this thing in action.
RTFC! (the "C" is for Comments, or rather Ars' discussion section for this article)
Comment by Deffexor (Ars Audio/Visual Moderator):
The problem with including the iPod Shuffle in the review was that we then would have had to include a bunch of other USB based Audio players in the review. Then to make matters worse, we would have had to benchmark/test the audio players, etc. This article is already 12 pages and I wasn't going to push it beyond that...
The good news is that I'm planning a portable audio player (flash based) round-up for May and will include the iShuffle.
As for the 20 year estimate, that sounds more the result of negotiations with the transit workers union than ability to get things switch over. You know City Hall, when it comes to a budget, they suddenly know the value of each penny and would switch the whole thing over in a couple years, tops.
:)
As someone who actually lives in New York, I would say that it probably has more to do with the fact that MTA doesn't have the money to do any of this. They're just sort of hoping to have the cash to do this (and extend the 7 line to the West, and refurbish the south ferry stop on the 1/9, and build a 2nd avenue line, and repair any number of stations, and do all the other things they supposedly want to do) in 20 years.
It'll take 50.
Gamer : Regular Folk :: Military-Grade : Civilian
Mainly referring to the price of their equipment...
Talking about video drivers shows how much Linus is not ready for the Laptop. If this is a problem, how much are audio, USB, FireWire, and WLAN are going to be a problem?
Extrapolation is bad. There's a known problem with video support for the latest 3d accelerated video cards (2d support is there), but that does not imply that other hardware is not supported.
Having just bought a new laptop and installed Linux on it (to replace an old laptop with Linux on it) I can tell you that audio, USB, and FireWire aren't a problem. There are only so many mobile chipsets and only so many integrated audio/USB/FireWire solutions which go with those. WLAN is a problem, most likely due to the lack of availability of hardware specs (as with video).
That if, say, you're an English speaker unfamiliar with the "i's" long "e" sound, you'll be likely to pronounce it "Man-dreyeva"
Yes, now that all of the English speakers have had their chance to say that the name of a French company sounds weird to them (gee, really?) do any French speakers have something to say about this? It's not supposed to be pronounced like "monde riva" or something is it?
the whole delete your cookies thing is silly. i run several web sites that use cookies to track logins, not for me to track them but for the site to track who is logged in.
Translation: I'm doing the right thing, so obviously the other 99.9999% of the world is as well and we are all "fools" for believing otherwise.
Please, most websites try to hit me with a doubleclick.net cookie or an advertising.com cookie right away. I'm no "fool" for deleting that sort of thing. Nor am I a fool for deleting all of the miscellaneous cookies I get, e.g. from misconfigured sites which leave Apache's mod_unique_id enabled for no reason.
As for user-tracking cookies, which may well be useful, there are two kinds: session cookies (which my browser does and should delete at the end of the session) and unreliable ones (e.g. ones which treat everyone on a public terminal or a family computer system as the same person). Ditching isn't such a bad idea (though I personally leave a few around from sites which I do use and trust).
Remember, the web is not a publication medium. It is designed to be interpreted by the user's web browser. If the user turns off images, they will see no images. If they turn off flash, there's no flash. If they use a screen-reader... well, you get the idea. That's the way the web was always intended to work. Turning off/deleting cookies is no different. The user controls the experience, plain and simple, and apparently 58% of people have decided to do that. Good for them, especially given the number of junk cookies out there.
The PSSU feature, though (as I mentioned in another post), that blocks incoming traffic on first boot and immediately directs the user to download updates is awesome. Why other companies haven't thought about this, I have no idea.
In the last several versions of Linux that I've installed, downloading the latest updates has been part of the install process. So I'm thinking that maybe other companies have thought of this, though they arrived at a slightly different solution.
Now, if Windows did that on _every_ boot (configurably, I would hope), that would be interesting. It might be useful for a server which have been offline for a while, e.g. after a power outage, fire, or other similar emergency.
In all fairness, this would probably have been the advice which the average slashdotter would have given if Apple weren't involved.
The problem is that it's not a particularly accurate complaint. There are some online music stores out there which sell mp3s, and the Shuffle can play those just fine. Not to mention other sources of mp3-based music, such as archive.org and, you know, ripping CDs which you own (remember those?) Not to mention that subscription-based services like Napster are actually worse in this regard.
On the other hand, with the wide variety of online music stores out there, the whole argument that iTunes is a reason why people should buy Apple products doesn't make much sense either. It's not as though every store has every song that anyone could ever want. _If_ iTunes' selection matches your tastes, _then_ it makes sense to go with an Apple product. (Then you get to hope that your taste in music doesn't change)
*Any* local exploit is *also* a potential remote exploit (just like the IRC conversation shows). I had someone nearly pwn a box of mine by using an exploit in a buggy PHP script, then trying to elevate privileges through a local exploit.
;)
How can a fork bomb be used as a remote exploit? You've just eliminated all the meaning from the words "local" and "remote". I mean really, even in your own example it was a real remote exploit (buggy PHP script) which allowed access to the system in the first place.
Yes local exploits are dangerous (I don't think the grandparent post disagreed with that) but without remote access to the machine, in practice that system will be fine. So no, they're not the same thing. The remote access/remote exploit is by far the more dangerous part. Once an attacker has that, they can do all sorts of things even without privilege escalation (filling up the HD with junk files, sending out spam e-mails, etc.). Unless of course you very carefully protect against every one of those things, but that's awfully hard. Show me an OS which by default would prevent an attacker who got access to a non-root account from sending out spam, and I'll show you an OS with no TCP/IP stack.
There can be no distinction between uploading and downloading via P2P. There is no client-server relationship (thats what peer-to-peer means!).
Even if you don't make anything else available for sharing, chunks of the file you are currently downloading can and do get shared as you are downloading it.
The reason why nobody raised this issue is that it you are factually incorrect. You most certainly can download and not upload with any number of P2P applications. With some programs all you have to do is not put anything in your shared directory and not download into that directory. While other programs like BitTorrent do attempt to prevent this type of behavior, proper firewalling can deal with it.
What it boils down to is this: P2P is still TCP/IP. There's still a connection from a client to a server. With P2P the servers are individuals' systems rather than big webservers somewhere, and there is often one connection in each direction, but that's it. If one peer wants to connect to you but not allow the reverse, it can be done.
Isn't this what (ex)hackers have been telling the IT industry all along?
Yeah, but for some reason nobody ever believes them, and I think I know why.
boss: "So, you're a computer geek hacker-type, eh?"
ex-hacker: "Yes."
boss: "And what you want to teach us is..."
ex-hacker: "How to relate to people."
boss: *laugh* *chortle* *door slam in face*
You may not get "faster CPU" or something of the sort, but you do get a great OS, tiny case made of adonized aluminium, a practically silent computer, great software-bundle and the like. Staring at just few specs (CPU-speed etc.) is pretty short-sighted IMO.
See, the problem is that I'm not convinced that you can't get all that for less than $825 _with_ a faster CPU and so on. Of course, that leads to the usual PC-vs-Mac pricing flamewars, which I'm trying to avoid.
Also, I could just as easily say that you shouldn't ignore the specs in favor of stuff like the material the case is made of. You may get a pretty computer, but will the experience of using it be pleasant (e.g. because of the slow processor, limited video paired with graphics-intensive OS, etc.)"?
What it boils down to is this: the specs aren't exactly impressive. Yes the Mac mini may have other strengths, but that doesn't change the numbers. It's not short-sighted to accurately note those numbers, either. As I pointed out above, they could reflect upon other deeper issues, like whether the user experience is good.
While I wouldn't call you a zealot, I do think you glossed over one thing: CPU speed. No, most people don't need half the speed their computers have, but $825 (not including monitor, keyboard, etc) isn't exactly my idea of a "cheap" system. For that money I would expect a faster CPU. More video memory would be nice too.
Of course, I assume that small size and running quietly are two of the Mac mini's big advantages, and a faster CPU might cause problems there. Still, I don't know, it just seems to me that the Mac mini's specs seem low end. Couldn't they have put a G5 in there? Yes, I'm sure that the build quality and being able to run OS X is worth something, but somehow it still doesn't seem right.
people who use a businesses' air, light and even gravity?
Don't most customers bring their own gravity? Perhaps you just happen to frequent businesses where most of the customers are massless.
Time Warp? WTF?
;)
Yes, apparently customers are still demanding libc4 and the a.out binary format.
Those customers are obviously concerned about Xenix compatibility. So it's Microsoft's fault.
Speed is.. well, 115k at 80mph isn't bad. :)
Not surprising. The doppler effect shouldn't come into play until more like 80,000mph. At that point your wireless connection will be unusable (the data will be corrupted). Also, you might notice increased latency as you move further away from Sprint's satellites and the Earth in general.
Whether this is patentable is another issue. But you can certainly patent a published idea -- it's the only way to protect it.
;)
I assume you mean that you can patent _your_ published idea, right? At least, that's what the idea is supposed to be.
This patent sounds like comparing pointers, which has been doable forever. Pointers are ultimately just numbers (after taking type into account and so on), so basically this has been doable as long as there have been pointers. Now whether it's been doable in BASIC is a whole other question. I don't remember anything about pointers in BASIC 20 years ago. Still, a patent which boils down to "comparing two numbers" does seem a bit broad to me, even if it's specific to a certain programming language.
I'm sick and tired of hearing this bullcrap. For the past 20-30 years, there's been nothing but active encouragement for women to denounce their traditional gender roles and perform tasks normally associated with men.
Wrong. You're confusing the culture with the people in it. In other words, while the feminist movement has led to a general sense that women should denounce their traditional gender roles, that doesn't mean that many or most individuals have acted in a manner consistent with this (they just pay lip service). If the increase in sexual harrassment lawsuits has taught us nothing else, it's that some guys are still old-fashioned. And Lord knows that it should be easy enough to find religious types who think that women should stay at home. For a more computer-specific example, notice the strong connection between video games and scantily-dressed women (in the games, as models at trade shows, in advertisements, and so on). Few scantily-dressed men appear in video games, though of course the programmers are usually men. What kind of impression do you think this higly visible part of the computer industry leaves on people?
Well, slashdot, you can't have it both ways, and no, predatory business practices and high prices don't excuse people infringing copyrights.
:) which supports infringing copyrights? It this perspective so out there that people have a hard time figuring it out for themselves?
Is it not possible that there's one group of posters and moderators who support not infringing the GPL, and a _totally different group_ (possibly a younger one
I mean, really, the parent post mentions that there are "many views here", but that information does not seem to be a part of the poster's thought process. Yes, "many views", or "two sets of people", does explain everything. Period. Yes, it is easier to think of SlashDot as some sort of unified intelligence than to think of a collection of people. But SlashDot is just a collection of people. Assigning opinions to SlashDot itself is sophistry. Who exactly is the "you" in "you can't have it both ways"?
I say that as someone who opposes both GPL infringment and file sharing. I exist, thank you!
There are many Linux programs based on libnjb which allow Creative mp3 players to be used with Linux. One of them is this project which causes the device to appear as a file system in KDE. What did the grandparent post say again? Something about "easy to use KDE software"?
As another example, I use my Dell DJ (manufactured by Creative) on Linux with gnomad2. So yes, Linux support for Creative-built mp3 players does exist. "Excellent" might be an exaggeration since you may have to install from source, but that's about it.
Needless to say, it is the parent of _this_ post which should be modded down (or at least not modded up so highly). The fact that a particular individual cannot do something rarely means that it cannot be done, but people tend to make that generalization anyway. And moderators tend to believe them.
Apple is doing this stuff (e.g. you can mount WebDAV servers), but Apple is doing it right by integrating network resources into the real VFS layer so that all applications can access them. KDE's I/O slaves are not real filesystems and are not accessible by all applications.
So then what Apple's doing is like LUFS?
Remember folks, KDE is not a whole OS, it's just a frontend. Apples and oranges, and all that.
But I thought the very nature of OSS makes this sort of thing impossible. What did I miss?
:) The simple reason being that nobody is bothering to fix that code anymore (Okay, technically I could patch the Linux kernel code if I really wanted to, but why bother with such old code?) Unfortunately, the same can be said for the Mozilla project in this case. Nobody is bothering to fix these old bugs, and for code which is "current", that's a bad problem.
The whole point.
OSS (supposedly) makes bugs easier to find -- that's the advantage it gives. I can go back to Linux kernel 1.0 (say) and probably find an unpatched bug. I could also go back and look at Windows 95, but I'd have a much harder time finding unpatched bugs, having no access to the source code. It's easier to find bugs in OSS code.
If I do find a bug in either, though, the bug is not likely to be fixed anytime soon in either case.
It's also an organizational problem, not an open vs. closed source problem. The issue is that nobody is writing the patch to fix a known problem. As advanced as modern software is, there is as of yet no program, OSS or otherwise, which I know of which writes patches for itself! Humans have to write the code, plain and simple, and I really doubt that any OSS advocate ever claimed otherwise.