> Yes it is. In the best case, OS/2 ran Windows
> apps exactly as well as Windows did (==not
> well), because it was by-in-large the same
> codebase. The advantage was that you didn't
> have to reboot after one of the frequent
> crashes,
Well, yes. You are right. I suppose I should've mentioned how running Win16 applications under OS/2 was better.
> Your suggestion would leave people like me who
> don't want icons on the desktop SOL.
Well, people like you don't matter. People like you are people who "think think they are |-r4d 3|33+ for using a non-Windows OS, but still don't know jack shit about said OS thus making them no better than your typical Windows luser".
The kind of people that do matter (in this area anyway) are the people who are like your typical Windows luser except they don't give a damn about what OS they are using, they just want to get some work done. These kind of people represent the majority of the desktop PC users out there. And these are the people that the various Linux desktop projects are trying to reach.
I'll never understand why people think that just because theres a standard Linux desktop, that automatically means you are restricted to use that desktop and that desktop only. Shit, you're not even restricted to only use the Windows UI in Windows!
> I used to say that I thought it unlikely that
> Windows was truly any easier to learn or use
> than KDE/Gnome for the truly computer illiterate
I don't think the issue is whether or not the Windows interface is truly easy to use or not. Because no matter how easy you make it, there will always be some idiot out there who won't be able to figure it out.
The issue is whether or not having multiple desktops is really a good idea. If Linux had one standard desktop, then theres only one thing to learn. It almost doesn't matter if said desktop is unintuitive and hard to use (but of course, you should endeavor to make it as easy to use as possible), because you only learn it once. If theres multiple desktops available, then that means you have to learn how to use multiple desktops. And most people can barely manage to learn one.
And the argument that standardizing on one desktop removes freedom of choice doesn't hold any water, imho. Any Linux user with a clue will know how to change his desktop to look and feel any way he or she wants. And any clueless Linux user most likely doesn't even give a damn that you can change the look and feel of your UI; they just want to get their work done.
> Would Linux users really be willing to give up
> their freedom of choice to gain broad
> acceptance?
With Linux, you never lose your freedom of choice so that point is irrelevant. Even in Windows its possible to radically change the look and feel of the UI; its just not as easy to do as it is on Linux.
The idea here is to standardize on one easy-to-use GUI so that the non-geek can sit down at a Linux machine and get to work immediately (assuming said non-geek has already been briefed on how to use the GUI). The problem with having multiple GUIs is that you need to learn how to use multiple GUIs. And for most non-geeks, learning just one GUI is hard enough.
> And about OS/2, GEM (no, not old enough to
> remember, but I'll assume...) - they weren't
> Windows-compatible.
OS/2 often times ran Win16 apps better than Windows itself. And thats not just typical anti-MS FUD! At the time of Windows 95's release, the majority of Windows applications on the market were Win16 applications, so OS/2 was in a pretty good position to take over the PC desktop.
> When he says "on thr fly" he means tht a
> filesystem can be grown WHILE it is still
> mounted. You can also do a mirror WHILE it is
> still mounted and being used by users. I started
> out on linux, and like the lvm, but there is
> nothing worse than realizing you
> need more space in a filesystem
FWIW, Linux LVM can do this too. You can resize a logical volume online. You just need to use a filesystem which supports resizing while still mounted. While ext2 doesn't support this (not out of the box anyway... theres a patch floating around tho that lets you do this), both Reiser and XFS do.
Linux LVM also lets you do that "mirroring" thing you're talking about. They are called "snapshots". Its pretty nifty!
> posting a vague inflammatory and judgemental
> post like this is ridiculous.
There was a time when calling a kernel release "stable" actually meant something. Now, some things don't even compile in the 2.4.x series. And this is not even in things marked "experimental". For example, up until 2.4.5 the OV511 USB driver wouldn't compile. Why? They simply forgot to include a declaration for 'version'. It's simple shit like this which proves that they don't even actually test to see if the thing even compiles before they release it as "stable". To me, this is mind boggling. This is the sort of thing you used to see in the development series, not in the stable series. Either make sure it at least compiles first before releasing it as stable, or stop calling it stable. Calling it "stable" when it doesn't even compile is just plain embarassing.
> I like the fact that growing an XFS volume to
> take up more space is simple, and does not
> require unmounting the volume (in fact, you
> CANNOT grow an unmounted XFS volume, you MUST
> mount it first).
FWIW, ReiserFS can do this too. That is, allow you to grow the filesystem while it is still mounted.
ReiserFS even goes a step further by allowing you to shrink the fileystem too. But you have to unmount said filesystem first to shrink it.
However, XFS is a full fledged journalling filesystem while Reiser just journals metadata.
> C++ is definetly not a low level language. Where
> did you get that idea? They're both (C++ and
> Java) in the same ballpark for abstraction-
> C is only slightly more low level.
ANSI C is for the most part a subset of ANSI C++. Therefore, ANSI C++ can be as low level as ANSI C.
In general, any valid ANSI C program is also a valid ANSI C++ program.
> constuctors: one that takes no arguments
> (default) and one that takes an int argument.
> The int argument one can't call the default
> constructor (this creates a temporary, contructs
> it, and then deletes the temporary). D allows
> you to do that. Maybe the next C++ specification
> will fix that.
Probably not, because all you have to do is put the common code in a private helper member function and have both of those constructors call it. Make said private helper function inline if you don't want the overhead of a function call.
> There are other differences, too. Most of them
> aren't common, but the above are a few I can
> remember off the top of my head.
Yes, which is why I qualified my statement with "a few exceptions". They are all nicely enumerated in the back of The C++ Programming Language, 3rd edition. But for the most part, ANSI C is a subset of ANSI C++.
What I'd really like a response to is dmelomed's statement that "C++ lets the programmer go lazy on many things at the expense of bloat and speed". I suspect he can't as most anti-C++ zealots tend to talk out of their ass.
> here's much more cruft in C++ than there's in C.
> C++ lets the programmer go lazy on many things
> at the expense of bloat and execution speed.
Such as?
> Even "Hello world" examples are much larger in
> size than C equivalents.
The following *C++* program compiles to a 3368 byte dynamic binary on my Linux box:
#include
int main() { printf("Hello, world!\n"): }
Oh whats this? You say thats not a C++ program but a C program? Hate to break it to you, but (with a few exceptions) any valid ANSI C program is also a valid ANSI C++ program.
> DNA/Human genome proves evolution? How? Sure,
> living things adapt to their surroundings,
> that's been proven on a small scale, but where's
> the proof of adaptation into new species?
There are many documented instances of speciation both in a controlled setting (i.e. experimentation in a laboratory) and in nature. Please visit http://www.talkorigins.org for more information.
Your statement that evolution theory grows weaker by the day is nothing but pure bullshit and thats what probably enraged the previous poster. If anything evolution theory has the most evidence to back it up than any other scientific theory to date.
> Did you say impossible????? Isn't that what
> they use to say about Win95 password lists, 40
> bit encryption and when pigs fly(I use FedEx
> for this)?
Bone head. Win95 passwords were encrypted. The information in WPA is hashed. Just like UNIX passwords. Once hashed, you can't get the original back.
> Nothing is impossible. Some things are
> difficult, some things are improbable and some
> things still elude us but, NOTHING is
> impossible!
No matter how hard I flap my arms, I'll never be able to fly.
> What planet do you live on? Who do you hang out
> with? What real people really told you they had
> their fillings removed by the government? I
> noticed you used the plural form, so you must
> have met more than one person like this.
Never underestimate the level of stupidity that is out there. Even now, in the 21st century, you have people here in the U.S. who still believe in ghosts and demons. So having people who take thier fillings out so the government can't track them wouldn't suprise me not one bit.
> We're talking about computers here, people, not
> a toaster or curling iron. There are people
> running RedHat who have VCRs flashing 12:00! If
> you want to turn a PC into an internet
> appliance, that's just fine and dandy with me.
> Those machines SHOULD be
Perhaps you should try reading the post I was responding to. You know.. the one by atheos. Here, let me repeat what he said so as to not make it too difficult for you ("you.. you mean you were responding to someone else's post???"):
atheos> So, in order to push Linux, you think we
atheos> need "Windoes" looking clones? Or gay RPM
atheos> packages? If you want to push Linux, you
atheos> should introduce people to an Operating
atheos> System that requires people to think for
atheos> themselves.
What I think atheos is trying to say here is that Linux ought to remain the difficult-to-use OS it started out as. Do you really believe Linux would be where it is today if it remained the difficult-to-use OS it started out as? If so, then I have a bridge I'd like to sell you. Or perhaps you're one of those 3l33+ h4xx0rz d00dz who draw self-esteem from using obscure OSes that no one has ever heard about? Which is why you'd object to Linux going mainstream because then you can't brag about being the 1 out of 4 people in the entire world that knows how to use it.
> What's GAY about source? Where do you think
> your RPMs come from, you idiot? Again, it all
> comes down to getting what you need to do what
> you need to do, and doing it your way. If you
> need to feel homophobic about source code,
> that's your problem.
YOU DUMB FUCK. Now I KNOW you didn't read atheos's post (the one I was replying to in case you forgot already). Again, I'll repeat what he said here just to help you out:
atheos> they don't rank anywhere near Slackware.
atheos> Shit, I can't even get anything to
atheos> compile on RH7.1 without some gay RPM to
atheos> throw at it.
Here, atheos is bitching about how he can't get shit to compile because obviously he neglected to install some dependent RPM (for example, he might be trying to compile a program which uses Gtk but he didn't install the gtk-devel rpm). Now, the problem here isn't due to RPM or the idea of package management in general. The problem is he's missing some dependencies. Now he can fulfill these dependencies by either installing the necessary RPMs (i.e. the Easy Way) or go and download the source, compile it, and install it to/usr/local or something (i.e. the Hard Way). In the presence of a package management system you have the choice of either doing it the Easy Way or doing it the Hard Way. In the absence of one though you have no choice but to do it the Hard Way.
All clear? Probably not. But you can't say I didn't try.
And as for the alleged homophobic remarks.. well, thats what I get for trying to use sarcasm in a forum where the bulk of the target audience is made up of people who aren't sharp enough to tell when it is being used. Again, read what atheos said. Then re-read what I said. Repeat until it starts to sink in.
> The only options I have are to try forcing the
> install (which might work and which *will*
> result in broken dependencies in the RPM
> database) or to compile from source. Both
> options lead to inconsistencies in the RPM data
> relative to what's actually on my disk.
Thats the whole purpose of the/usr/local tree though. Its for locally installed software, which in the presence of a vendor specified package manager, means software which doesn't come prepackaged in a rpm/deb (imho).
Your "leads to inconsistencies in the RPM data" with regard to installing software to/usr/local statement is a non-issue because no package thats been installed outside of the/usr/local tree should depend on any thing inside the/usr/local tree.
As far as the package manager is concerned, anything installed to/usr/local doesn't exist. Its up to you to manage the software in/usr/local. Let the package manager handle the rest of the system.
> No good. No good. Every time I've used Debian
> or an RPM-based system, I end up with EVERY
> LAST DAMN THING in the system in the/usr/local
> tree because I've compiled it myself as way to
> get around the versioning and dependencies in
> package management.
Hmm... wouldn't you be doing this ANYWAY in the absence of a package management system?? That is, you grab the source for a package and install it yourself somewhere under/usr/local or where ever it is you like to stick local software packages?
The only people who I see bitch about package managers are the ones who like to fight them rather than work with them. If you're in need of a particular software package, you first check to see if it already exists in rpm/deb form. If it does, you use that. If it does, but you happen to not like the compiled-in defaults you grab the source rpm/deb, rebuild it, then install it. Failing all that, you then grab the source and go ahead and compile/install to/usr/local.
> Package management gets in the way of a
> competent system administrator's day.
Only if said "competent" system administrator is a lazy fuck who doesn't want to learn how to use the particular system he or she is administering.
The package management system is for managing the various components of the system. You would maintain the/usr/local tree (or where ever it is you install local packages) independent of the rest of the system. Some people opt to use third party package management systems (such as epkg or stow) to mangage this local software tree. Personally, I make my own debs for my/usr/local tree for my Debian systems.
> If you want to push Linux, you should introduce
> people to an Operating System that requires
> people to think for themselves
Oh yes. Computers shouldn't be made easier to use. What a stupid idea. I long for the days when "programming a computer" actually meant rewiring the thing..... not.
> Shit, I can't even get anything to compile on
> RH7.1 without some gay RPM to throw at it.
Yes, instead you'll need to go download some gay source distribution of a dependent package, compile it in true gay fashion, and then install it in some gay location such as "/usr/local".
That is assuming of course that Slackware doesnt already have a Slackware package for that dependent package. And if it did, and you decide to install that instead of compiling the source, how would that be any different from using an RPM or DEB?
Slashdot Mirror
> Yes it is. In the best case, OS/2 ran Windows
> apps exactly as well as Windows did (==not
> well), because it was by-in-large the same
> codebase. The advantage was that you didn't
> have to reboot after one of the frequent
> crashes,
Well, yes. You are right. I suppose I should've mentioned how running Win16 applications under OS/2 was better.
- Arcadio
> Your suggestion would leave people like me who
> don't want icons on the desktop SOL.
Well, people like you don't matter. People like you are people who "think think they are |-r4d 3|33+ for using a non-Windows OS, but still don't know jack shit about said OS thus making them no better than your typical Windows luser".
The kind of people that do matter (in this area anyway) are the people who are like your typical Windows luser except they don't give a damn about what OS they are using, they just want to get some work done. These kind of people represent the majority of the desktop PC users out there. And these are the people that the various Linux desktop projects are trying to reach.
- Arcadio
> I'd rather have choices.
I'll never understand why people think that just because theres a standard Linux desktop, that automatically means you are restricted to use that desktop and that desktop only. Shit, you're not even restricted to only use the Windows UI in Windows!
WAKE THE FiretrUCK UP!
- Arcadio
> I used to say that I thought it unlikely that
> Windows was truly any easier to learn or use
> than KDE/Gnome for the truly computer illiterate
I don't think the issue is whether or not the Windows interface is truly easy to use or not. Because no matter how easy you make it, there will always be some idiot out there who won't be able to figure it out.
The issue is whether or not having multiple desktops is really a good idea. If Linux had one standard desktop, then theres only one thing to learn. It almost doesn't matter if said desktop is unintuitive and hard to use (but of course, you should endeavor to make it as easy to use as possible), because you only learn it once. If theres multiple desktops available, then that means you have to learn how to use multiple desktops. And most people can barely manage to learn one.
And the argument that standardizing on one desktop removes freedom of choice doesn't hold any water, imho. Any Linux user with a clue will know how to change his desktop to look and feel any way he or she wants. And any clueless Linux user most likely doesn't even give a damn that you can change the look and feel of your UI; they just want to get their work done.
- Arcadio
> Would Linux users really be willing to give up
> their freedom of choice to gain broad
> acceptance?
With Linux, you never lose your freedom of choice so that point is irrelevant. Even in Windows its possible to radically change the look and feel of the UI; its just not as easy to do as it is on Linux.
The idea here is to standardize on one easy-to-use GUI so that the non-geek can sit down at a Linux machine and get to work immediately (assuming said non-geek has already been briefed on how to use the GUI). The problem with having multiple GUIs is that you need to learn how to use multiple GUIs. And for most non-geeks, learning just one GUI is hard enough.
- Arcadio
> And about OS/2, GEM (no, not old enough to
> remember, but I'll assume...) - they weren't
> Windows-compatible.
OS/2 often times ran Win16 apps better than Windows itself. And thats not just typical anti-MS FUD! At the time of Windows 95's release, the majority of Windows applications on the market were Win16 applications, so OS/2 was in a pretty good position to take over the PC desktop.
- Arcadio
> When he says "on thr fly" he means tht a
... theres a patch floating around tho that lets you do this), both Reiser and XFS do.
> filesystem can be grown WHILE it is still
> mounted. You can also do a mirror WHILE it is
> still mounted and being used by users. I started
> out on linux, and like the lvm, but there is
> nothing worse than realizing you
> need more space in a filesystem
FWIW, Linux LVM can do this too. You can resize a logical volume online. You just need to use a filesystem which supports resizing while still mounted. While ext2 doesn't support this (not out of the box anyway
Linux LVM also lets you do that "mirroring" thing you're talking about. They are called "snapshots". Its pretty nifty!
- Arcadio
> posting a vague inflammatory and judgemental
> post like this is ridiculous.
There was a time when calling a kernel release "stable" actually meant something. Now, some things don't even compile in the 2.4.x series. And this is not even in things marked "experimental". For example, up until 2.4.5 the OV511 USB driver wouldn't compile. Why? They simply forgot to include a declaration for 'version'. It's simple shit like this which proves that they don't even actually test to see if the thing even compiles before they release it as "stable". To me, this is mind boggling. This is the sort of thing you used to see in the development series, not in the stable series. Either make sure it at least compiles first before releasing it as stable, or stop calling it stable. Calling it "stable" when it doesn't even compile is just plain embarassing.
- Arcadio
> I like the fact that growing an XFS volume to
> take up more space is simple, and does not
> require unmounting the volume (in fact, you
> CANNOT grow an unmounted XFS volume, you MUST
> mount it first).
FWIW, ReiserFS can do this too. That is, allow you to grow the filesystem while it is still mounted.
ReiserFS even goes a step further by allowing you to shrink the fileystem too. But you have to unmount said filesystem first to shrink it.
However, XFS is a full fledged journalling filesystem while Reiser just journals metadata.
- Arcadio
> C++ is definetly not a low level language. Where
> did you get that idea? They're both (C++ and
> Java) in the same ballpark for abstraction-
> C is only slightly more low level.
ANSI C is for the most part a subset of ANSI C++. Therefore, ANSI C++ can be as low level as ANSI C.
In general, any valid ANSI C program is also a valid ANSI C++ program.
- Arcadio
> constuctors: one that takes no arguments
> (default) and one that takes an int argument.
> The int argument one can't call the default
> constructor (this creates a temporary, contructs
> it, and then deletes the temporary). D allows
> you to do that. Maybe the next C++ specification
> will fix that.
Probably not, because all you have to do is put the common code in a private helper member function and have both of those constructors call it. Make said private helper function inline if you don't want the overhead of a function call.
- Arcadio
> There are other differences, too. Most of them
> aren't common, but the above are a few I can
> remember off the top of my head.
Yes, which is why I qualified my statement with "a few exceptions". They are all nicely enumerated in the back of The C++ Programming Language, 3rd edition. But for the most part, ANSI C is a subset of ANSI C++.
What I'd really like a response to is dmelomed's statement that "C++ lets the programmer go lazy on many things at the expense of bloat and speed". I suspect he can't as most anti-C++ zealots tend to talk out of their ass.
- Arcadio
> here's much more cruft in C++ than there's in C.
> C++ lets the programmer go lazy on many things
> at the expense of bloat and execution speed.
Such as?
> Even "Hello world" examples are much larger in
> size than C equivalents.
The following *C++* program compiles to a 3368 byte dynamic binary on my Linux box:
#include
int main() { printf("Hello, world!\n"): }
Oh whats this? You say thats not a C++ program but a C program? Hate to break it to you, but (with a few exceptions) any valid ANSI C program is also a valid ANSI C++ program.
- Arcadio
> DNA/Human genome proves evolution? How? Sure,
> living things adapt to their surroundings,
> that's been proven on a small scale, but where's
> the proof of adaptation into new species?
There are many documented instances of speciation both in a controlled setting (i.e. experimentation in a laboratory) and in nature. Please visit http://www.talkorigins.org for more information.
Your statement that evolution theory grows weaker by the day is nothing but pure bullshit and thats what probably enraged the previous poster. If anything evolution theory has the most evidence to back it up than any other scientific theory to date.
- Arcadio
> The internet would be worthless if only the US
> was allowed to use it.
Worthless? I wouldn't go that far. Not as cool, yes. But worthless, no.
Off the top of my head, I can't think of any non-US site that I goto with any regularity.
> I rest my case.
... I'm a little slow. But what case would this be? All you've managed to show was that you're just another anti-MS Slashdot sheep.
Forgive me
I'm a cheap freeloading scumfuck too! And damn proud of it! PH33R!
> Did you say impossible????? Isn't that what
> they use to say about Win95 password lists, 40
> bit encryption and when pigs fly(I use FedEx
> for this)?
Bone head. Win95 passwords were encrypted. The information in WPA is hashed. Just like UNIX passwords. Once hashed, you can't get the original back.
> Nothing is impossible. Some things are
> difficult, some things are improbable and some
> things still elude us but, NOTHING is
> impossible!
No matter how hard I flap my arms, I'll never be able to fly.
> What planet do you live on? Who do you hang out
> with? What real people really told you they had
> their fillings removed by the government? I
> noticed you used the plural form, so you must
> have met more than one person like this.
Never underestimate the level of stupidity that is out there. Even now, in the 21st century, you have people here in the U.S. who still believe in ghosts and demons. So having people who take thier fillings out so the government can't track them wouldn't suprise me not one bit.
> We're talking about computers here, people, not
.. the one by atheos. Here, let me repeat what he said so as to not make it too difficult for you ("you .. you mean you were responding to someone else's post???"):
/usr/local or something (i.e. the Hard Way). In the presence of a package management system you have the choice of either doing it the Easy Way or doing it the Hard Way. In the absence of one though you have no choice but to do it the Hard Way.
.. well, thats what I get for trying to use sarcasm in a forum where the bulk of the target audience is made up of people who aren't sharp enough to tell when it is being used. Again, read what atheos said. Then re-read what I said. Repeat until it starts to sink in.
> a toaster or curling iron. There are people
> running RedHat who have VCRs flashing 12:00! If
> you want to turn a PC into an internet
> appliance, that's just fine and dandy with me.
> Those machines SHOULD be
Perhaps you should try reading the post I was responding to. You know
atheos> So, in order to push Linux, you think we
atheos> need "Windoes" looking clones? Or gay RPM
atheos> packages? If you want to push Linux, you
atheos> should introduce people to an Operating
atheos> System that requires people to think for
atheos> themselves.
What I think atheos is trying to say here is that Linux ought to remain the difficult-to-use OS it started out as. Do you really believe Linux would be where it is today if it remained the difficult-to-use OS it started out as? If so, then I have a bridge I'd like to sell you. Or perhaps you're one of those 3l33+ h4xx0rz d00dz who draw self-esteem from using obscure OSes that no one has ever heard about? Which is why you'd object to Linux going mainstream because then you can't brag about being the 1 out of 4 people in the entire world that knows how to use it.
> What's GAY about source? Where do you think
> your RPMs come from, you idiot? Again, it all
> comes down to getting what you need to do what
> you need to do, and doing it your way. If you
> need to feel homophobic about source code,
> that's your problem.
YOU DUMB FUCK. Now I KNOW you didn't read atheos's post (the one I was replying to in case you forgot already). Again, I'll repeat what he said here just to help you out:
atheos> they don't rank anywhere near Slackware.
atheos> Shit, I can't even get anything to
atheos> compile on RH7.1 without some gay RPM to
atheos> throw at it.
Here, atheos is bitching about how he can't get shit to compile because obviously he neglected to install some dependent RPM (for example, he might be trying to compile a program which uses Gtk but he didn't install the gtk-devel rpm). Now, the problem here isn't due to RPM or the idea of package management in general. The problem is he's missing some dependencies. Now he can fulfill these dependencies by either installing the necessary RPMs (i.e. the Easy Way) or go and download the source, compile it, and install it to
All clear? Probably not. But you can't say I didn't try.
And as for the alleged homophobic remarks
> The only options I have are to try forcing the
/usr/local tree though. Its for locally installed software, which in the presence of a vendor specified package manager, means software which doesn't come prepackaged in a rpm/deb (imho).
/usr/local statement is a non-issue because no package thats been installed outside of the /usr/local tree should depend on any thing inside the /usr/local tree.
/usr/local doesn't exist. Its up to you to manage the software in /usr/local. Let the package manager handle the rest of the system.
> install (which might work and which *will*
> result in broken dependencies in the RPM
> database) or to compile from source. Both
> options lead to inconsistencies in the RPM data
> relative to what's actually on my disk.
Thats the whole purpose of the
Your "leads to inconsistencies in the RPM data" with regard to installing software to
As far as the package manager is concerned, anything installed to
thats just great ...
> No good. No good. Every time I've used Debian > or an RPM-based system, I end up with EVERY > LAST DAMN THING in the system in the /usr/local
> tree because I've compiled it myself as way to
> get around the versioning and dependencies in
> package management.
Hmm ... wouldn't you be doing this ANYWAY in the absence of a package management system?? That is, you grab the source for a package and install it yourself somewhere under /usr/local or where ever it is you like to stick local software packages?
The only people who I see bitch about package managers are the ones who like to fight them rather than work with them. If you're in need of a particular software package, you first check to see if it already exists in rpm/deb form. If it does, you use that. If it does, but you happen to not like the compiled-in defaults you grab the source rpm/deb, rebuild it, then install it. Failing all that, you then grab the source and go ahead and compile/install to /usr/local.
> Package management gets in the way of a
> competent system administrator's day.
Only if said "competent" system administrator is a lazy fuck who doesn't want to learn how to use the particular system he or she is administering.
The package management system is for managing the various components of the system. You would maintain the /usr/local tree (or where ever it is you install local packages) independent of the rest of the system. Some people opt to use third party package management systems (such as epkg or stow) to mangage this local software tree. Personally, I make my own debs for my /usr/local tree for my Debian systems.
And no, you don't break anything.
> If you want to push Linux, you should introduce
..... not.
> people to an Operating System that requires
> people to think for themselves
Oh yes. Computers shouldn't be made easier to use. What a stupid idea. I long for the days when "programming a computer" actually meant rewiring the thing
> Shit, I can't even get anything to compile on
> RH7.1 without some gay RPM to throw at it.
Yes, instead you'll need to go download some gay source distribution of a dependent package, compile it in true gay fashion, and then install it in some gay location such as "/usr/local".
That is assuming of course that Slackware doesnt already have a Slackware package for that dependent package. And if it did, and you decide to install that instead of compiling the source, how would that be any different from using an RPM or DEB?