C'mon, how many people do you know who paid for Winzip?
No one. Well apparently my work copy is registered but I can't verify that they paid for it. I use a free command-line unzip I downloaded. I hear Stuffit for Windows comes free too. If piracy weren't so socially acceptable, people might be more amenable to seeking out these alternatives.
Everyone pirates, some to a greater degree than others, of course.
Unless you're a some kind of saint, sure. I'm no saint. I've got a couple of Ghost in the Shell mp3 tracks sitting on my hard drive somewhere. And probably some other crap I can't remember.
The point I failed to make was that (I believe) there is a time and a place for breaking the law; and that it should always be done with consideration for the consequences, and willingness to face said consequences. I just happen to think "everybody's doing it" and "it's just for fun" are both really crappy excuses.
I know this is not the point of your argument but I still feel the need to say "speak for yourself". There are several reasons one might not participate in even casual piracy.
1) You can actually afford it -- college grad perhaps?
2) You want to support the company that makes the product
3) You can use a free alternative which makes the piracy issue moot
Of course if none of the above applies to you (and I can think of several scenarios, thx) then I guess I'd have to say "go for it"... I just think there's a difference between stealing loaves of bread and random crime...
Maybe this warming is just part of a very long cycle in Martian atmosphere, taking decades of even centuries, that we haven't observed yet
Well AFAIK, current theory is that Mars used to have a thicker atmosphere but it dissipated over the millenia due to (basically) lack of gravity. So yeah, if that CO2 stays in the atmosphere eventually it should reach escape velocity and leave the planet. I suppose then the planet would cool back down.. at least at night anyway:)
I'm no movie critic but I found that movie horrible. The blind angle was the best part of it, but that's not saying much. Kilmer came off as *incredibly* whiny and the chemistry was exceedingly forced and shallow. Especially the crying on the massage table bit. Little character growth. Almost as bad as "Meet Joe Black". (So if you liked that movie, go see this one.)
I am *not* trying to belittle this condition, but if you want to know more about it instead of just watching the Beautiful People mumble their lines (..and that can be fun sometimes..) I would just pick up the book instead.
Think about how crippled the ports will be. How on earth can you issue ssh user@host -X and expect it to work under M$?
Well, methinks naively that it would take a user account and an X server. xfree runs under cygwin, I don't see what the problem is..
The average M$ box lacks basic security features such as users, PIDs and embeded file permisions.
Aaah, this is what you're getting at. XP (afaik), 2000, and NT have what you mention, and I don't see a problem with restricting the port to these platforms. If MS is successful at converting the heathens to XP your problem should eventually go away.
The path issues alone are enough to make me gag, can they be passed to new shells in M$ land?
Have you actually run cygwin? I have no problem passing environment vars (including PATH) under bash..
Heck, I can't even find the much vaunted (and sorely incomplete) kshell on my NT cripled work box.
That's why you want Debian/w32, so you can run pdksh or whatever they bundle..
Good luck to the folks at Cygwin. They are taking on a endless, difficult and thankless task
Yup
Loss32 would be a good name for Win32 TM.
I like that, maybe you should suggest it to the debian list:)
The difference here is, the windows bug reports go to/dev/null.
Technically that would be "NUL":)
("type nul" is the windows equiv of "cat/dev/null". The interesting bit about windows' special file names is that they're case insensitive and you can add arbitrary extensions, which bit me once when I wanted to save off a console log and wondered why saving to con.log just hung..)
So you're agreeing that it can't open non-trivial word documents?
No; I could prove that false if I could get StarOffice to "open" a single non-trivial word doc. However to answer what I think is your point: yes, if you need 100% MS Office compatibility linux is not and likely will never be for you (unless you run vmware)
From my standpoint as a user, I would say that they both have the same vision---and that is to create a desktop environment.
I was using "vision" in the context of who calls the shots, rather than the general end goal, but I can't disagree w/what you say.
What does this have to do with anything? I don't know any windows users that submit bug reports.
Oops.. at that point I was merely belittling the parent post, flamebait style. Good point, and apologies all around for dredging that up.
Why are you so small-minded and petty that you cry troll at the slightest provocation?
Fine, I'll bite. That's not slightest provocation. You offer one paragraph of barely on-topicness immediately sequeing into three paragraphs of rehashed rant.
Linux falls on its face for mundane day-to-day productivity work.
That depends on what you use it for doesn't it?
it can't even open a simple Word document without formatting errors.
StarOffice (you did say "simple" Word documents..)
Linux stills falls short of Windows when playing Quake.
Client benchmarks that I have seen are dominated by graphics card/driver combinations and as such are dependent on the card vendor. Indeed in the case of the matrox g200 I could actually *play* qIII linux while matrox sat on their asses and refused to release a working windows opengl driver for the longest time. People generally favor linux for quake servers.
If all the man-hours poured into KDE and GNOME were combined into a common vision, we would have one perfect end-user desktop, instead of two poor imitations of Windows.
My guess is we'd just have one "poor imitation of Windows" as you call it. The practical weakness in your argument is deciding whose vision should be followed.
Don't give me the old "competition" argument either. There is only one Linux kernel, which seems to progress just fine without another competing project nipping at its feet and instigating flamewars.
*BSD. To a lesser extent, Windows itself. And to a still-lesser extent, GNU/Hurd, which if everybody had swallowed their ego (and worked on it instead of that upstart Linux) probably still wouldn't be ready for prime time.
The endless KDE vs. GNOME, Applix vs. StarOffice, and other feuds have wasted more productivity than would be gained by and competitive drive.
These feuds are generally propagated by users such as yourself rather than the actual developers. As such, little productivity is lost.
I, for one, am somewhat miffed that while my operating system powers Hollywood blockbusters and NASA supercomputers, it still can't fully replace Windows on my office desktop.
Show me one instance where you have properly submitted a bug report/feature request for any of these office programs you need and I'll reconsider dismissing you as an opinionated parasite.
Hardware autodetection is great, if you don't mind having it simply not work and freezing a user's system every once in a while.
Which is why autodetection should be an install option, and not mandatory. The hangs you describe mostly happen with legacy ISA devices that don't support probing. Barring sloppy programming, your freezes should occur less and less.
It's not hard to just have a person enter in their hardware.
That's lazy programmer syndrome (I should know, I've got a bad case of it).. it can be hard for the user. For instance, it's been a couple of years since I got my oem matrox g200 and now it's kindof hard to remember if it came with 4 or 8 megs of ram...
Perhaps this just reflects that the implementation of IPC in Linux, while complete, is not as fast or optimized as it should be.
Be careful when you throw that term around...
I can't tell whether you're talking about sysV IPC ('man ipc'), since it's the only other interface that provides shared memory similar to mmap(), or the posix threading mechanisms.
They have nothing to do with each other; and that's a good thing, since sysV IPC is a legacy, poorly designed POS (I can give examples on demand).
but there is little reason why the most calls can't be done purely in user space.
True; the fact they don't under linux is an artifact of LinuxThreads. As Xavier Leroy notes in his
FAQ, a one-to-one (every thread maps to a kernel thread) thread implementation implies that every context switch must be at the kernel level, which is more expensive than a pure user-space context switch. It's the price you pay for simplicity. This is somewhat mitigated by linux's fast context switching.
Processor instructions such as "test and set" don't typically need supervisor priveleges.
Correct, so you can do atomic variables in user space:)
Actually, in the 6.2 days, it would install the SMP kernel if your motherboard had more than one processor slot, regardless of how many procs were actually installed. I'm not sure if that's true now, tho.
Slashdot Mirror
No one. Well apparently my work copy is registered but I can't verify that they paid for it. I use a free command-line unzip I downloaded. I hear Stuffit for Windows comes free too. If piracy weren't so socially acceptable, people might be more amenable to seeking out these alternatives.
Everyone pirates, some to a greater degree than others, of course.
Unless you're a some kind of saint, sure. I'm no saint. I've got a couple of Ghost in the Shell mp3 tracks sitting on my hard drive somewhere. And probably some other crap I can't remember.
The point I failed to make was that (I believe) there is a time and a place for breaking the law; and that it should always be done with consideration for the consequences, and willingness to face said consequences. I just happen to think "everybody's doing it" and "it's just for fun" are both really crappy excuses.
Unless of course you are a COBOL programmer.. who wants to admit nowadays they even know black arts like that?
I know this is not the point of your argument but I still feel the need to say "speak for yourself". There are several reasons one might not participate in even casual piracy.
1) You can actually afford it -- college grad perhaps?
2) You want to support the company that makes the product
3) You can use a free alternative which makes the piracy issue moot
4) Risk of getting busted (while depressingly low) possibly outweighs benefits
5) (Most important) You don't really need it
Of course if none of the above applies to you (and I can think of several scenarios, thx) then I guess I'd have to say "go for it"... I just think there's a difference between stealing loaves of bread and random crime...
(/pulpit)
Well AFAIK, current theory is that Mars used to have a thicker atmosphere but it dissipated over the millenia due to (basically) lack of gravity. So yeah, if that CO2 stays in the atmosphere eventually it should reach escape velocity and leave the planet. I suppose then the planet would cool back down.. at least at night anyway :)
Somebody refute this, I'm no climatologist..
I am *not* trying to belittle this condition, but if you want to know more about it instead of just watching the Beautiful People mumble their lines (..and that can be fun sometimes..) I would just pick up the book instead.
Your entire chord reportoire is going to be eaten up for emacs key combos anyway :)
God, I thought carpal tunnel was bad before..
Well, methinks naively that it would take a user account and an X server. xfree runs under cygwin, I don't see what the problem is..
The average M$ box lacks basic security features such as users, PIDs and embeded file permisions.
Aaah, this is what you're getting at. XP (afaik), 2000, and NT have what you mention, and I don't see a problem with restricting the port to these platforms. If MS is successful at converting the heathens to XP your problem should eventually go away.
The path issues alone are enough to make me gag, can they be passed to new shells in M$ land?
Have you actually run cygwin? I have no problem passing environment vars (including PATH) under bash..
Heck, I can't even find the much vaunted (and sorely incomplete) kshell on my NT cripled work box.
That's why you want Debian/w32, so you can run pdksh or whatever they bundle..
Good luck to the folks at Cygwin. They are taking on a endless, difficult and thankless task
Yup
Loss32 would be a good name for Win32 TM.
I like that, maybe you should suggest it to the debian list :)
I would have preferred a giant megalomaniac pulsating brain.
Yes. Although to me it seems as though this mechanism would fail at the worst possible time. i.e. One way to make the thing fall over is just to
1) lean some way until you are going the vehicle's top speed
2) continue leaning that way, at which point IT can't compensate, and you lose your balance and splatter
I'm afraid this is something I don't want to race friends with, or be late in, for that matter.
Technically that would be "NUL" :)
("type nul" is the windows equiv of "cat /dev/null". The interesting bit about windows' special file names is that they're case insensitive and you can add arbitrary extensions, which bit me once when I wanted to save off a console log and wondered why saving to con.log just hung..)
No; I could prove that false if I could get StarOffice to "open" a single non-trivial word doc. However to answer what I think is your point: yes, if you need 100% MS Office compatibility linux is not and likely will never be for you (unless you run vmware)
From my standpoint as a user, I would say that they both have the same vision---and that is to create a desktop environment.
I was using "vision" in the context of who calls the shots, rather than the general end goal, but I can't disagree w/what you say.
What does this have to do with anything? I don't know any windows users that submit bug reports.
Oops.. at that point I was merely belittling the parent post, flamebait style. Good point, and apologies all around for dredging that up.
Fine, I'll bite. That's not slightest provocation. You offer one paragraph of barely on-topicness immediately sequeing into three paragraphs of rehashed rant.
Linux falls on its face for mundane day-to-day productivity work.
That depends on what you use it for doesn't it?
it can't even open a simple Word document without formatting errors.
StarOffice (you did say "simple" Word documents..)
Linux stills falls short of Windows when playing Quake.
Client benchmarks that I have seen are dominated by graphics card/driver combinations and as such are dependent on the card vendor. Indeed in the case of the matrox g200 I could actually *play* qIII linux while matrox sat on their asses and refused to release a working windows opengl driver for the longest time. People generally favor linux for quake servers.
If all the man-hours poured into KDE and GNOME were combined into a common vision, we would have one perfect end-user desktop, instead of two poor imitations of Windows.
My guess is we'd just have one "poor imitation of Windows" as you call it. The practical weakness in your argument is deciding whose vision should be followed.
Don't give me the old "competition" argument either. There is only one Linux kernel, which seems to progress just fine without another competing project nipping at its feet and instigating flamewars.
*BSD. To a lesser extent, Windows itself. And to a still-lesser extent, GNU/Hurd, which if everybody had swallowed their ego (and worked on it instead of that upstart Linux) probably still wouldn't be ready for prime time.
The endless KDE vs. GNOME, Applix vs. StarOffice, and other feuds have wasted more productivity than would be gained by and competitive drive.
These feuds are generally propagated by users such as yourself rather than the actual developers. As such, little productivity is lost.
I, for one, am somewhat miffed that while my operating system powers Hollywood blockbusters and NASA supercomputers, it still can't fully replace Windows on my office desktop.
Show me one instance where you have properly submitted a bug report/feature request for any of these office programs you need and I'll reconsider dismissing you as an opinionated parasite.
Sincerely, a fellow opinionated parasite
Agree with your general sentiment, but I also object to using +1 bonus to post offtopic rants like you just did, so I guess it balances out.
Where's the sourceforge page?
A suprising statement given the number of people here who post w/out appearing to have read the freakin' articles.
(Like me.)
Thanks dude! I had no add-on modules..
of course the fuckers never came out with an OpenGL driver for the longest time
True, but I'm linux-only and was happy with my John Carmack Special (aka Utah GLX... incidentally that was why I bought the g200 *and* quake3/linux)
Which is why autodetection should be an install option, and not mandatory. The hangs you describe mostly happen with legacy ISA devices that don't support probing. Barring sloppy programming, your freezes should occur less and less.
It's not hard to just have a person enter in their hardware.
That's lazy programmer syndrome (I should know, I've got a bad case of it) .. it can be hard for the user. For instance, it's been a couple of years since I got my oem matrox g200 and now it's kindof hard to remember if it came with 4 or 8 megs of ram...
Impressive.. a screenshots page w/out the screenshots.. Nice first post, welcome :)
Actually (correct me if I'm wrong) I think it used to be, but some people thought newbies would find it "confusing". Or something.
Yeah, but Konqy rocks :)
I don't think originality was a design goal.. and when windows came out, Apple sued because it looked like the Mac.. it all comes around.
Maybe they should call it fink/OpenOSX..
Be careful when you throw that term around...
I can't tell whether you're talking about sysV IPC ('man ipc'), since it's the only other interface that provides shared memory similar to mmap(), or the posix threading mechanisms.
They have nothing to do with each other; and that's a good thing, since sysV IPC is a legacy, poorly designed POS (I can give examples on demand).
True; the fact they don't under linux is an artifact of LinuxThreads. As Xavier Leroy notes in his FAQ, a one-to-one (every thread maps to a kernel thread) thread implementation implies that every context switch must be at the kernel level, which is more expensive than a pure user-space context switch. It's the price you pay for simplicity. This is somewhat mitigated by linux's fast context switching.
Processor instructions such as "test and set" don't typically need supervisor priveleges.
Correct, so you can do atomic variables in user space :)
Actually, in the 6.2 days, it would install the SMP kernel if your motherboard had more than one processor slot, regardless of how many procs were actually installed. I'm not sure if that's true now, tho.