Linux NEEDS the killer apps. It needs Photoshop, Dreamweaver, Acrobat, Flash authoring tools, and tons of games. Most studios are totally based on Photoshop, GIMP has no chance there etc.
Also, it is an irony that XGL is being held as a shiny OSS example, but requires those much-hated closed source drivers because they are the only ones with decent OpenGL support.
Well, Adobe already ported their codebase. They are very conservative when it comes to porting their Nr.1 apps, which is quite understandable. The last time they commented on the situation they said that a port is still under consideration and the Linux platform being under evaluation.
As for Flash, I guess this is more a Macromedia issue. Yes, Macromedia still exists, not by this name, but the devs behind Flash etc. are still there.
Not everything is part of GNU. If a closed-source app is developed for Linux, this app is not part of GNU. Period. See Maya for example.
Also, nVidia's efforts are the only way of getting decent 3D on Linux. With decent I mean OpenGL 2.0 support. And as I said before, the binary blob is NOT illegal, the kernel module (which is opensourced) is, because its not GPLed. The binary blob just uses the kernel module's API. If nVidia GPLs its kernel module, their binary driver becomes perfectly legal. I hope they do so.
Well if you're sooo dependent on Microsoft products, and you admit it, then you should now understand WHY OSS is so important. We're seeking to empower the individuals, who in today's setup are at the mercy of software companies.
Does not change a thing. If there is no alternative to this app in the OSS world, then Windows stays. For example, I use Windows for WinUAE, Gens and other Emulators, because the Linux versions are pure crap compared to the Windows ones. The Windows ones run better even when started with Wine. Also, the Linux ones seem to be abandoned long ago. Photoshop exists for Windows only (and no, GIMP is no replacement, mainly not for technical, but for know-how reasons), 3D studio max is Windows only (Maya targets a different audience, 3DSmax is more aimed towards game development), many bookkeeping applications are Windows only, almost all PC games are Windows only (cedega is not a 100% replacement, and requires lots of finetuning, which is not exactly user friendly, but not cedega's fault), etc.
If you want to exclusively use Linux, then the first thing you need to learn to do is to COMPROMISE.
Hah! Often enough this is simply not an option, especially among the pragmatic Joe Sixpack users. Then again, if you want Linux to stay among the self-proclaimed pseudo-elitist "hackers" (most of them being more like script kiddies), then go on.
We're not here to emulate windows, we're here to offer a different desktop experience.
Now THIS line sounds like straight from the MS marketing department.
As for the lack of new interesting things in the OSS world, well I'll just say that you haven't been looking hard enough. Not all the interesting stuff comes in a.deb or.rpm..
1. I proposed killer apps with no Linux equivalent already. 2. Unless its in a.deb/.rpm, it is not an option. Compiling it manually is out of the question unless you are a programmer (and even then it is highly annoying if it doesnt compile because of yet another issue with different distros - do you really expect Joe Sixpack to fix this?)
Did *you* read my posting? I said exactly the same. The opensourced kernel module is *linked* to the kernel, and the binary driver then *uses* the module's open API, thereby just *using* the module. Using an open API of a GPLed system is no GPL violation, otherwise Linux could not be used at all (BIOS is closed-source for example).
It is not, as long as the source code is available. Otherwise, all non-source based distros would be illegal, including Ubuntu, Debian, Suse, Fedora,....
Sorry, my mistake. Obviously I misunderstood something in the nVidia sources. But still, the layer idea allows binary drivers without violating the GPL.
Using the GPL layer approach, they are not. But apparently, I was mistaken about the open-source nVidia kernel modules that act as the layer. They are not GPLed. If they are re-released under the GPL, then the nvidia binary drivers are perfectly legal.
The problem with #2 is that the drivers need constant support, because the kernel changes so fast. Of course it is logical that a driver written for Kernel 2.6.1 may not work with 2.6.15, but sometimes stuff even breaks because it was written for 2.6.14! This makes Linux driver maintenance expensive. If the maintainer does not have 24/7 devotion behind it, the driver is gone.
Contrast this with stuff like ext2fs for Windows. I can *still* use it, even after the updates and the switch from 2000 to XP. There should be a stable API for GPLed OSS drivers, which is changed ONLY between major releases.
Indeed, automounting still has issues. For example, you unmount it using the KDE menu, but after unplugging, its still present. Or, it simply does not automount etc. IMO automounting is not well integrated, even after Project Utopia. The Unix design is simply not made for automounting and removable media, and it shows. Only after the creation of pmount *useful* media mounting is possible (i.e. you dont have to be root).
Another thing I am not fond of is samba share mounting. It is really not trivial to mount a share being an user, and with the rights set so that ordinary users can access it. It takes tons of samba tutorials to actually get this to work. KDE and Gnome support mounting using their own VFS', which is bad for non-KDE/GNOME apps. Fix this!
It is illegal to distribute non-GPL binary kernel modules (ask any kernel developer), and Freespire should respect the GPL since they are a Linux company.
Well, the actual nVidia/ATI kernel module IS GPLed. It acts as layer between the kernel and the binary blob. This layer has an open API, which is used by the blob. Thus, the blob uses the kernel, it does not extend it. The extension comes from the GPLed layer.
The real issue is that nVidia & ATI use GPL code from other parts of the kernel in their drivers. This is the only issue, because the way I described above makes binary blobs perfectly legal.
This does not change anything. There were STILL more important issues back in the 15th century than one wacko eager to find India. Most space research is fundamental, and thus cannot and must not be classied into "useful" and "useless". You can do this with applied research, but not with fundamental one. Now, if you want some money for education, it is STILL wiser to scrap Yet Another Bomb. Unfortunately, it is more popular to target NASA.
t is an indication of how far wrong we are going with large science projects in space that we can define support for museums and college labs as "pork" and therefore waste. How can they honestly think that frog sex in space and the dynamics of a burning candle are more important than education here on Earth? How long are we going to accept that spewing our seed into barren space is the thing to do?
Flawed thinking. First, the argument "first we should solve all our problems" is age-old and still pointless. It is likely that we will NEVER solve all our problems, no matter how many resources we pour over them. A good analogy is Christopher Columbus' journey. Should he have waited for Europe to solve all its problems? Definitely not. So aim your morale club at something else, like the budget the DoD gets. You are aware that the money NASA gets is absolutely NOTHING compared to the trillions of $ the DoD receives? Now, tell me, what would be wiser to scrap? Space Research or Yet Another Bomb?
And if something goes wrong, the graphics card is broken, because the ROM is gone, and has to be flashed by some pros for a hefty amount of $$$. What a GREAT idea!
Autotools are the wrong solution, LSB is the right one. Autofoo is an awful build system (the nick "autohell" has its reasons, just see the results in google). They are horrible to use, the scripts are a nightmare to debug, and M4 is just plain from hell. SCons, CMake, Premake, Jam are better tools (btw, CMake has been created by autotools veterans and creators). The interest in better build systems is quite high, just look at the KDE guys (they moved everything from autofoo to CMake), or at Blender. Autotools may have been a good idea back then, but its time for some progress. Also, NO ONE ever uses autotools in Windows, there is just no reason to do that.
CS is bad bad bad? Oh my. CS is THE ONLY WAY Linux is ever going to have serious games (is Doom3 OS? UT2004? NWN?) and if Adobe decides to port Photoshop, be prepared for a binary blob. Maya is closed source. Baad programs we don't need? Dream on. Besides, the aforementioned games have their issues with incompatible libc's etc. which is the precise reason why LSB is a GOOD thing.
Intuition is being able to read other people's emotions from the signs they give off.
Empathy would be a logical step above intuition. If people's emotions emit signs we do not know of, and if these signs are just crude psychic projections or whatever, then it would be hard to distinguish it from intuition in the first place. Of course, once detected, an experiment would be easy: sit the supposed psychic in a sealed chamber, stick a totally random person whose mood is known (sad, amused, enraged...) in a second sealed chamber. If the psychic senses the emotions correctly, then its empathy.
This would predict that most psychics with empathic abilities may never become aware of their unique traits, simply seeing it as deep intuitiveness.
This is akin to saying the fairies are shy of people, and that's why you can't see them. It's an excuse to rationalise away the non-existence of the phenomena to begin with.
This wasnt mentioned as evidence, just some random thoughts. I am allowed to have random thoughts without some rationalist inquisition knocking on my door, am I not?
This stance also requires that any psychic who appears on tv, writes a book, sells tours, offers advice through a premium phone number, holds seances, helps the cops, takes part in studies etc. is by definition not a psychic. After these folks are clearly not shy or even afraid of lining their pockets, so what's their excuse other than they're frauds?
If one thinks about the implications of having ESP abilities, I really doubt one would be able to behave the way these psychics do. I would rather expect them to be mostly insane, very bizarre, loners, etc.
One thing I wonder: what if those paranormal phenomena are actually glitches in our perception of the world? Remember that the conscious mind is aware of only fractions of the actual input. The subconscious is full of wonders. This goes all the way down to the question whether the world actually behaves like our understanding of causality. If causality is actually a simplification constructed by our mind, then there might be a chance that there are situations where this approximation is just plain wrong.
Of course, this raises some serious question about how to verify this in an experiment.
Although I'm often called a strict spectic I find science and biology magical and exiting just because of all the things we don't know. However I'm dismayd by the lack of cricitical mind and imagination that people have. There are plenty of magic left even after you ignore all these weak speudo science ideas.
Sounds like a mental dictatorship.
"We decide what your scientific interests have to be! Deviating from them is heresy and will be punished by exclusion from the scientific community! Be a good scientist! You must not think what you are not allowed to think!"
Might explain why there is a decline in science students.
In other words: what YOU see as magic means nothing. Magic is subjective. You are in no position to force everyone what has to be seen as interesting.
What if telepathy is real, but the experiments are wrong? I wonder if telepathy actually works differently than one usually thinks. After sorting out the clear frauds and hoaxes, one can see that most psychics seem to be very intuitive. If telepathy is some sort of enhanced intuition, then maybe the ability depends heavily on the environment and situation. For example, being in a very familiar room triggering telepathic abilities. Unfortunately, this would render telepathy unprovable.
It goes further than that, though. This touches the question whether there are phenomena that cannot be described by current scientific practices or not. If true, then telepathy may well be unproved for a very long time.
Also, if someone is REALLY capable of telepathy, chances are high that this person keeps it a secret. Reading thoughts allow revealing true motives. If one reads the minds of ESP-interested people, one may well find some rather sinister motives (like, abusing it for stealing, blackmail, military applications..) Also, reading other's minds could be quite scary and disturbing, so it would not surprise me to find lots of insane people among the real psychics.....
OF COURSE it is a bad idea. This is the result when marketing is the only decision maker. GTA sold well because its FUN to play. Is it fun to play a game that feels like a marketing product and is filled to the top with cliche gansta-hiphop crap? No. Maybe by putting someone in charge with a REAL game development background (most likely game design) the CEOs/CMOs/ would give games that are actually fun a chance. Til then, expect cold, heartless products designed for making money ONLY. (Kinda like Eisner's Disney).
I'm not a php/perl guru, but
on
PHP Hacks
·
· Score: 4, Interesting
anyone can tell me if this is to be taken seriously, or rather seen as worthless php-bashing? http://tnx.nl/php
AFAIK this has been considered before, namely: version control repositories, bugtrackers, logs valid as evidence in court (just like firewall logs are valid evidence).
Slashdot Mirror
Linux NEEDS the killer apps. It needs Photoshop, Dreamweaver, Acrobat, Flash authoring tools, and tons of games. Most studios are totally based on Photoshop, GIMP has no chance there etc.
Also, it is an irony that XGL is being held as a shiny OSS example, but requires those much-hated closed source drivers because they are the only ones with decent OpenGL support.
Nonsense. Free Software is not automatically better. Compare Blender to Maya. The only difference is in the licensing and availability.
Well, Adobe already ported their codebase. They are very conservative when it comes to porting their Nr.1 apps, which is quite understandable. The last time they commented on the situation they said that a port is still under consideration and the Linux platform being under evaluation.
As for Flash, I guess this is more a Macromedia issue. Yes, Macromedia still exists, not by this name, but the devs behind Flash etc. are still there.
I don't know what bizarro world you're inhabiting, but it all sounds like FUD to me.
Welcome to Slashdot!
Not everything is part of GNU. If a closed-source app is developed for Linux, this app is not part of GNU. Period. See Maya for example.
Also, nVidia's efforts are the only way of getting decent 3D on Linux. With decent I mean OpenGL 2.0 support. And as I said before, the binary blob is NOT illegal, the kernel module (which is opensourced) is, because its not GPLed. The binary blob just uses the kernel module's API. If nVidia GPLs its kernel module, their binary driver becomes perfectly legal. I hope they do so.
Well if you're sooo dependent on Microsoft products, and you admit it, then you should now understand WHY OSS is so important. We're seeking to empower the individuals, who in today's setup are at the mercy of software companies.
.deb or .rpm ..
.deb/.rpm, it is not an option. Compiling it manually is out of the question unless you are a programmer (and even then it is highly annoying if it doesnt compile because of yet another issue with different distros - do you really expect Joe Sixpack to fix this?)
Does not change a thing. If there is no alternative to this app in the OSS world, then Windows stays. For example, I use Windows for WinUAE, Gens and other Emulators, because the Linux versions are pure crap compared to the Windows ones. The Windows ones run better even when started with Wine. Also, the Linux ones seem to be abandoned long ago. Photoshop exists for Windows only (and no, GIMP is no replacement, mainly not for technical, but for know-how reasons), 3D studio max is Windows only (Maya targets a different audience, 3DSmax is more aimed towards game development), many bookkeeping applications are Windows only, almost all PC games are Windows only (cedega is not a 100% replacement, and requires lots of finetuning, which is not exactly user friendly, but not cedega's fault), etc.
If you want to exclusively use Linux, then the first thing you need to learn to do is to COMPROMISE.
Hah! Often enough this is simply not an option, especially among the pragmatic Joe Sixpack users.
Then again, if you want Linux to stay among the self-proclaimed pseudo-elitist "hackers" (most of them being more like script kiddies), then go on.
We're not here to emulate windows, we're here to offer a different desktop experience.
Now THIS line sounds like straight from the MS marketing department.
As for the lack of new interesting things in the OSS world, well I'll just say that you haven't been looking hard enough. Not all the interesting stuff comes in a
1. I proposed killer apps with no Linux equivalent already.
2. Unless its in a
Did *you* read my posting? I said exactly the same. The opensourced kernel module is *linked* to the kernel, and the binary driver then *uses* the module's open API, thereby just *using* the module. Using an open API of a GPLed system is no GPL violation, otherwise Linux could not be used at all (BIOS is closed-source for example).
It is not, as long as the source code is available.
Otherwise, all non-source based distros would be illegal, including Ubuntu, Debian, Suse, Fedora,....
Sorry, my mistake. Obviously I misunderstood something in the nVidia sources.
But still, the layer idea allows binary drivers without violating the GPL.
Using the GPL layer approach, they are not.
But apparently, I was mistaken about the open-source nVidia kernel modules that act as the layer. They are not GPLed. If they are re-released under the GPL, then the nvidia binary drivers are perfectly legal.
The problem with #2 is that the drivers need constant support, because the kernel changes so fast. Of course it is logical that a driver written for Kernel 2.6.1 may not work with 2.6.15, but sometimes stuff even breaks because it was written for 2.6.14! This makes Linux driver maintenance expensive. If the maintainer does not have 24/7 devotion behind it, the driver is gone.
Contrast this with stuff like ext2fs for Windows. I can *still* use it, even after the updates and the switch from 2000 to XP. There should be a stable API for GPLed OSS drivers, which is changed ONLY between major releases.
Indeed, automounting still has issues. For example, you unmount it using the KDE menu, but after unplugging, its still present. Or, it simply does not automount etc.
IMO automounting is not well integrated, even after Project Utopia. The Unix design is simply not made for automounting and removable media, and it shows. Only after the creation of pmount *useful* media mounting is possible (i.e. you dont have to be root).
Another thing I am not fond of is samba share mounting. It is really not trivial to mount a share being an user, and with the rights set so that ordinary users can access it. It takes tons of samba tutorials to actually get this to work. KDE and Gnome support mounting using their own VFS', which is bad for non-KDE/GNOME apps. Fix this!
It is illegal to distribute non-GPL binary kernel modules (ask any kernel developer), and Freespire should respect the GPL since they are a Linux company.
Well, the actual nVidia/ATI kernel module IS GPLed. It acts as layer between the kernel and the binary blob. This layer has an open API, which is used by the blob. Thus, the blob uses the kernel, it does not extend it. The extension comes from the GPLed layer.
The real issue is that nVidia & ATI use GPL code from other parts of the kernel in their drivers. This is the only issue, because the way I described above makes binary blobs perfectly legal.
This does not change anything. There were STILL more important issues back in the 15th century than one wacko eager to find India. Most space research is fundamental, and thus cannot and must not be classied into "useful" and "useless". You can do this with applied research, but not with fundamental one. Now, if you want some money for education, it is STILL wiser to scrap Yet Another Bomb. Unfortunately, it is more popular to target NASA.
t is an indication of how far wrong we are going with large science projects in space that we can define support for museums and college labs as "pork" and therefore waste. How can they honestly think that frog sex in space and the dynamics of a burning candle are more important than education here on Earth? How long are we going to accept that spewing our seed into barren space is the thing to do?
Flawed thinking. First, the argument "first we should solve all our problems" is age-old and still pointless. It is likely that we will NEVER solve all our problems, no matter how many resources we pour over them. A good analogy is Christopher Columbus' journey. Should he have waited for Europe to solve all its problems? Definitely not. So aim your morale club at something else, like the budget the DoD gets. You are aware that the money NASA gets is absolutely NOTHING compared to the trillions of $ the DoD receives? Now, tell me, what would be wiser to scrap? Space Research or Yet Another Bomb?
And if something goes wrong, the graphics card is broken, because the ROM is gone, and has to be flashed by some pros for a hefty amount of $$$. What a GREAT idea!
Rubbish.
Autotools are the wrong solution, LSB is the right one. Autofoo is an awful build system (the nick "autohell" has its reasons, just see the results in google). They are horrible to use, the scripts are a nightmare to debug, and M4 is just plain from hell. SCons, CMake, Premake, Jam are better tools (btw, CMake has been created by autotools veterans and creators). The interest in better build systems is quite high, just look at the KDE guys (they moved everything from autofoo to CMake), or at Blender. Autotools may have been a good idea back then, but its time for some progress.
Also, NO ONE ever uses autotools in Windows, there is just no reason to do that.
CS is bad bad bad? Oh my. CS is THE ONLY WAY Linux is ever going to have serious games (is Doom3 OS? UT2004? NWN?) and if Adobe decides to port Photoshop, be prepared for a binary blob. Maya is closed source. Baad programs we don't need? Dream on. Besides, the aforementioned games have their issues with incompatible libc's etc. which is the precise reason why LSB is a GOOD thing.
Intuition is being able to read other people's emotions from the signs they give off.
Empathy would be a logical step above intuition. If people's emotions emit signs we do not know of, and if these signs are just crude psychic projections or whatever, then it would be hard to distinguish it from intuition in the first place. Of course, once detected, an experiment would be easy: sit the supposed psychic in a sealed chamber, stick a totally random person whose mood is known (sad, amused, enraged...) in a second sealed chamber. If the psychic senses the emotions correctly, then its empathy.
This would predict that most psychics with empathic abilities may never become aware of their unique traits, simply seeing it as deep intuitiveness.
This is akin to saying the fairies are shy of people, and that's why you can't see them. It's an excuse to rationalise away the non-existence of the phenomena to begin with.
This wasnt mentioned as evidence, just some random thoughts. I am allowed to have random thoughts without some rationalist inquisition knocking on my door, am I not?
This stance also requires that any psychic who appears on tv, writes a book, sells tours, offers advice through a premium phone number, holds seances, helps the cops, takes part in studies etc. is by definition not a psychic. After these folks are clearly not shy or even afraid of lining their pockets, so what's their excuse other than they're frauds?
If one thinks about the implications of having ESP abilities, I really doubt one would be able to behave the way these psychics do. I would rather expect them to be mostly insane, very bizarre, loners, etc.
One thing I wonder: what if those paranormal phenomena are actually glitches in our perception of the world? Remember that the conscious mind is aware of only fractions of the actual input. The subconscious is full of wonders. This goes all the way down to the question whether the world actually behaves like our understanding of causality. If causality is actually a simplification constructed by our mind, then there might be a chance that there are situations where this approximation is just plain wrong.
Of course, this raises some serious question about how to verify this in an experiment.
Although I'm often called a strict spectic I find science and biology magical and exiting just because of all the things we don't know. However I'm dismayd by the lack of cricitical mind and imagination that people have. There are plenty of magic left even after you ignore all these weak speudo science ideas.
Sounds like a mental dictatorship.
"We decide what your scientific interests have to be! Deviating from them is heresy and will be punished by exclusion from the scientific community! Be a good scientist! You must not think what you are not allowed to think!"
Might explain why there is a decline in science students.
In other words: what YOU see as magic means nothing. Magic is subjective. You are in no position to force everyone what has to be seen as interesting.
What if telepathy is real, but the experiments are wrong? I wonder if telepathy actually works differently than one usually thinks. After sorting out the clear frauds and hoaxes, one can see that most psychics seem to be very intuitive. If telepathy is some sort of enhanced intuition, then maybe the ability depends heavily on the environment and situation. For example, being in a very familiar room triggering telepathic abilities. Unfortunately, this would render telepathy unprovable.
It goes further than that, though. This touches the question whether there are phenomena that cannot be described by current scientific practices or not. If true, then telepathy may well be unproved for a very long time.
Also, if someone is REALLY capable of telepathy, chances are high that this person keeps it a secret. Reading thoughts allow revealing true motives. If one reads the minds of ESP-interested people, one may well find some rather sinister motives (like, abusing it for stealing, blackmail, military applications..) Also, reading other's minds could be quite scary and disturbing, so it would not surprise me to find lots of insane people among the real psychics.....
OF COURSE it is a bad idea. This is the result when marketing is the only decision maker. GTA sold well because its FUN to play. Is it fun to play a game that feels like a marketing product and is filled to the top with cliche gansta-hiphop crap? No.
Maybe by putting someone in charge with a REAL game development background (most likely game design) the CEOs/CMOs/ would give games that are actually fun a chance. Til then, expect cold, heartless products designed for making money ONLY. (Kinda like Eisner's Disney).
anyone can tell me if this is to be taken seriously, or rather seen as worthless php-bashing?
http://tnx.nl/php
AFAIK this has been considered before, namely: version control repositories, bugtrackers, logs valid as evidence in court (just like firewall logs are valid evidence).
In short:
Linus did not fall - Linux rose.