CollegeLinux Released to the Public

YOU ARE SO FIRED! writes "It seems that the Swiss Robert Kennedy College (with the aptly named website) has released CollegeLinux, a Linux distribution based off of Slackware, to the public. If only my employees could've used this in school - I wouldn't have to fire them so much! See the interview with the dean of the school for more information."

Comments

Will it ever stop?

[mschoolbus]

Linux is great and all but I feel half the reason it isn't doing as well as it could is because there are just so many distros in general. I know the nature of Linux is about choice and open software but this hurts Linux in itself. Why don't hardware companies put out Linux drivers as much as they do Mac drivers? Because they expect certain things within the Mac OS, not everything is different from system to system, which makes it easier and more attractive to companies to write Linux code and drivers...

Re:Will it ever stop?

[Anonymous Coward]

Not writing drivers because of the number of Linux distributions would make a crappy excuse. The Linux kernel doesn't vary that much between distro's (and in the ways it does, it shouldn't matter much). The userland varies a small amount (just enough to be annoying for some low level tasks :P), but that doesn't affect the majority of drivers.

Re:Will it ever stop?

[jchristopher]

Yep, you'll probably get slammed with "-1" for saying it, but you're absolutely right.

Having so much effort wasted on many different distributions is stupid. Can you imagine what type of improvements could have been made to Linux in general with the programming time invested in maintaining many different distributions?

I mean absolutely no offense to the developers working on College Linux. But would you trade College Linux for a working way to change resolution on the fly, reliable working sound interfaces, simple dual monitor setup, and other enhancements that users have come to expect from other operating systems? I would.

Re:Will it ever stop?

[Anonymous Coward]

They're called kernel modules. If you use them, you might have to compile a module but not the whole thing.

Re:Will it ever stop?

[jchristopher]

Look into MAS if the currently 'common' interfaces don't work, I think you'll like it (MAS should become more common as time passes and it replaces the other sound servers).

LOL, I think that's sort of the point - the end user shouldn't have to think about interfaces, sound servers, and replacing one with the other!

Re:Will it ever stop?

[Billly Gates]

Why not more distro's ?

If you want to follow the herd install redhat or suse. Linux is about choice. I think Microsoft has bullied the industry so long that people have trouble with more then one way to think or have more then one way to do something. They want consistency and familiarity. The people I see complaining most about both gnome and kde for example are newbies and Windows users who are afraid to switch. People who use Linux/Unix understand that having more then one desktop environment is not bad.

How does this hurt Linux? You mean commercial support? They all target Suse, Redhat, and Debian. Companies like Oracle for example are specifically targeting RedHat advanced server since it changes the least. Also if you go to nvidia's website you will see drivers for quite alot of distro's.

Their are not numerous Linux versions, just distributions. They only look different because of different scripts and WindowManager themes running. Some of the package versions may differ but they are %95 the same from other distro's that are out at the same time. Mandrake 9, Redhat 8, Suse 8.1 all have the same versions of gcc, perl,kde, and apache for example. Its not like they are totally different beasts. Its easy to port.

Also I do not understand about your rat with hardware manufactures supporting mac users over linux ones. I find the situation opposite since Linux has more users and they use different kinds of hardware. Only ATI is behind and they are already at work with a unified driver model for Linux and Windows to compete against nvidia. Isn't competition great?

Companies and certain users want only one way of thinking are already in Microsoft's camp and will not change. A one and only one version of Linux will not convince them to switch. They are the ones already in .NET development, which once it's in is impossible to get out. They know it and do not care since an all Microsoft solution in their minds will lower support costs and its what everyone else is doing so they must stay with ms. They will stick with Windows as long as companies like PeopleSoft make Windows only products and mcse's are a dime a dozen and plentifull.

Re:Will it ever stop?

[$$$exy Gwen Stefani]

It will be unified to some extent, but it will take a bit more time.

There are already unified driver support projects, and there's a huge project at http://www.linuxbase.org/ in which the goal is "to develop and promote a set of standards that will increase compatability among Linux distributions and enable software applications to run on any compliant system."

We will do it, just give us a couple more years. Windows was written over several decades, and Linux is very new still!

Re:Will it ever stop?

[NanoGator]

"Erm, pretty much all driver code should be in the kernel anyways..."

Perhaps, but there needs to be an easy way to install them too. What's the guarantee that a given distro's not going to break something?

Question: Is there a driver installation standard that includes how to do it within KDE/Gnome? If there's not, then there's a nasty weakness there that'll pretty much guarantee most companies won't support it. If it is there, how often does it change? Does it change?

*Note: In case it's not obvious, I'm not knowledgable about Linux. From what little I've used it, the thought of installing a driver makes me go ewwwww. I'm spoiled by my 'Setup.exe, [Ok], [Ok], [Ok] *bam* you're up and running' expectation I've developed from using Windows 2000.

Re:Will it ever stop?

[Anonymous Coward]

*Note: In case it's not obvious, I'm not knowledgable about Linux. From what little I've used it, the thought of installing a driver makes me go ewwwww. I'm spoiled by my 'Setup.exe, [Ok], [Ok], [Ok] *bam* you're up and running' expectation I've developed from using Windows 2000.

And we (Linux users) are spoiled by our "it's in the kernel already as a module, don't worry about it" expectation we've developed from using Linux. So, on the (rare, for the most part) instances when a driver doesn't already exist and isn't already setup, some of us get kind of upset ;).

Re:Will it ever stop?

[vvikram]

huh....How does this hurt ? Of course it does. This is not economics, this is human society. When you want to provide an alternative to Microsoft then it is imperative to provide a unified face.

Actually its ironical that the most intelligent people in one dimension are so brainless [sorry for the harsh term] in another dimension. I have been like that when I was shouting up and down about linux. Let me give you an example: linux in itself came about due to a "critical mass" of organized people saying "hey lets get something out of this malleable chunk of cool code. There WAS before that the *BSD's and unix variants, what was definitely lacking was a concerted effort bounded together by the GPL and also the timing of the internet boom. But linux success HAS been in "rolling out" standard server based software - quickly roll out apache, php, mysql and get running....its SIMPLE isnt it ? its repeatable and its easy to do. quickly do a configure,make, make install - it works across MOST if NOT ALL of the software. WHY ? because the authors want to provide a uniform way of doing things. Extending this to a slightly larger scale doesnt seem to cross the minds of the ners?

While choice might be good for innovation and anti-monopolistic checks too much of choice does hurt. So a EE person asks me what is linux ? I say its an OS with a bunch of utilities. They ask "where can i download it"? and I say "huh....its like soap. you can get many flavours. The most popular is RedHat but you know this driver is supported better in the other distro but you know the security is best in Bastille but you know debian is the best in stability so you have to decide what you want to do" and the people go "huh-uh. thanks for the info.....later". This is for the end user side and believe me it does present a confused picture. For the developer side, thats us, it fractures a LOT of the effort. KDE reinvents the wheel,GNOME cannot *gasp* do what KDE has
done and so reinvents it in a slightly different form and so on and on. Imagine the number of install work, the number of packages, the number of hacks, the effort going into each of these distros - if they were to be combined into a select few then I can bet those distros will be awesome.

I cant believe I typed so much. Very sorry for the length. I just really dont agree that too much choice is great. There is a balance just like in real life for most things.

And, by the way, my univ has a linux distro too : SULinux

Thanks for reading.
vv

Re:Will it ever stop?

[NanoGator]

"Linux is great...as long as you have VMware "

Flamebait? He's got a respectable point. Most Windows users (including myself) are afraid of what we lose when moving to Linux. Having VM-Ware (at least initially) would make the transition much more bearable. If VM-Ware came with a distro of Linux for a reasonable price (I think VM-Ware is around $300, if memory serves) then I probably would attempt a switch because I'd have something to fallback on.

Honestly, I wish moderators wouldn't just assume one-liners are automatically insulting. I've gotten burned a couple of times like that over the last few days.

Re:Will it ever stop?

[bigberk]

all driver code should be in the kernel anyways

And it works beautifully, through loadable kernel modules. In this respect I think linux is superior to Windows. I spent hours this weekend trying to reinstall a Windows system that insisted on guessing and providing "the best" driver for my hardware. With linux I can use modprobe to load the exact appropriate driver I need, and these drivers are standard across all distros (kernel version specific however).

Re:Will it ever stop?

[fucksl4shd0t]

huh....its like soap. you can get many flavours. The most popular is RedHat but you know this driver is supported better in the other distro but you know the security is best in Bastille but you know debian is the best in stability so you have to decide what you want to do

Looks to me like you failed to provide a unified front. People ask me about Linux and I say "It's a free OS, you can download several for free." Then they say "Where?". I just tell them "Well, you should check out Mandrake first. It's the one I use, so I definitely recommend it. There's a lot of choice in distributions, but they're all built on the same stuff, so they interoperate well. Mandrake's a great one to start with, and you may never need another one. The beauty of choice, though, is that if you like Linux but you're dissatisfied with Mandrake, you can check out the others."

It's every bit in how you present it. Just make sure you're telling them the truth, and you're doing everything you can to help them (and charging an appropriate fee if you prefer).

From a university's standpoint, say they want their students to use Linux, but they don't want to support a few different distributions? Fine, roll their own and tell the kids "use this one or find someone else to support it. You're tuition covers this one." Anything unreasonable about that? They can give their kids something that installs out of the box and works on their network. Can they do that with Windows? Nope. They can achieve a close approximation, but not the real deal. Linux is the way...

Re:Will it ever stop?

[cascadefx]

Quote:

LOL, I think that's sort of the point - the end user shouldn't have to think about interfaces, sound servers, and replacing one with the other!

Sure, finding out which hardware works with your system is always a pain. A modern Operating System should just work, after all. I have found these resources to be invaluable when trying to find out whether something will work with my OS or not.

Of course, I am kidding. But realize all OSes have hardware issues and many times those issues force the user to step outside "the experience" (or pay someone else to) in order to fix the problem that they are having.

To give Microsoft credit, the end user has to do this less and less often if they are working with generic setups and applications. However, once you specialize (on any platform) you run the risk of forcing the user to "think about interfaces."

We forget that Microsoft has the same issues as Apple and Linux, but that is because there is a huge industry that has built up around the support of the OS that is used on 90% of the desktops. That "after market" industry is in many ways an extension of the "ease-of-use" on the user end.

I work in end-user support and the same problems that I hear that Linux must overcome are consistently present in Windows as well. The "average" user is no more adept at using Windows or dealing with its problems. They may be better at it than on a Mac and BeOS, but they still need lots of help.

Just my .02

Wireless

[ShadowDrake]

Colleges are ahead of the curve regarding wireless. GNU/Linux is nasty to prepare wireless on. This comes from experience. I had to pull the packets to my Thinkpad by hand!

Really valuable for a college environment would be a completely idiot-proofed wireless network setup utility. This utility or package should:

-Have all the driver modules compiled, and the configuration files kept up to date about different manufacturers' model identifications.

-Have a convinent popup tool, ideally triggered at the card-insertion time, and iconifying shortly after, that provides helpful stats and diagnostics. How hard would it be to convert 700 lines of iwconfig, ifconfig, and driver messages to:

"Discovered SSID "foo"."
"No IP number available after 20 seconds. Respawning DHCPCD." (to make up for some setups that seem to make DHCP have a fit if you pop the card and suspend, then pick up later."
"DHCP results: IP number is 127.0.0.43"
"Current situation: Signal/noise = 54/40. 353 bad sends, 107 bad recieves"

Absolutely wrong

[gorjusborg]

Having so much effort wasted on many different distributions is stupid. Can you imagine what type of improvements could have been made to Linux in general with the programming time invested in maintaining many different distributions?

You mean you would already have a free clone of YOUR favorite OS NOW if everyone would just team up and agree that your favorite GUI and OS's philosophy is the best?

The problem is, there are people who actually work on projects, and those who criticize other's projects. The reason that there are so many projects is that people disagree on what the 'correct' way of doing things is. If you want a windows clone, use windows. Otherwise, choose the distribution that fits your computing style the best.

Oh, and by the way, ctl-alt-+ and ctl-alt-- changes your resolutions in X on-the-fly, if you compile your quality sound driver and insert it into the kernel, you'll have reliable sound, and xfreee86 +xinerama works great for MULTI-monitor setups, not just dual-monitor setups.

Furthermore, the developers that work on making different distributions have totally different skill sets than driver developers and applications developers. In other words, you can't assume that if these people weren't developing their distribution they would be fixing problems with sound, video, etc.. In fact, if they weren't developing their distribution, they would probably be posting nonsense on slashdot, and complaining about how all the current distributions are crap.

Sorry if I come off as harsh, but I hate this type of thinking. Some people are so lazy, but expect the world of others. Parasites.

Re:And that is why OS X will ultimately beat Linux

[Anonymous Coward]

umm I take offense,
"us crackheads" are smart enough to add/edit stuff ourselves witch in reality if you know what you are doing is 10-20x faster than waiting on a stupid os wizard.

Add new drive:
echo /dev/sda1 /fire vfat \ rw,users,auto,showexec,umask=000,quiet 0 0 >> /etc/fstab
time: 2 seconds

sit and wait for gui
time: 2-5 minutes

Re:And that is why OS X will ultimately beat Linux

[Anonymous Coward]

I don't know what world you live in. Most people are still using Office 97.

And woulnd't notice the difference if you sat them down in front of Office 2000, Office XP, Star Office, Open Office, Abiword + Gnumeric, or MS Works. Except their will be a few frustrating oddities, but nothing worse than they're used to trying to format an outline, or left align numbers on a spreadsheet.

Slashdot jumped the shark on this one

[fartmaster]

No other slashdot submission has pissed me off more then this one.

YOU ARE SO FIRED! writes "It seems that the Swiss Robert Kennedy College (with the aptly named website) has released CollegeLinux, a Linux distribution based off of Slackware, to the public. If only my employees could've used this in school - I wouldn't have to fire them so much! See the interview with the dean of the school for more information."

Let's recap it -
YOU ARE SO FIRED!!. If only my employees could've used this in school.

I can just say that this bullshit additude towards different people pisses me off. Don't fire them. Teach them. Just because someone learns something different does not mean that they can not learn something different. Don't be such a total fucking asshole towards people and slashdot should be higher then posting this crap.

Re:Slackware again?

[ananke]

slackware is a good starting point. it's clean, simple, and easy to build upon. that's what i use for our linux workstations [with the addition of systemimager].

Because its very BSDish

[Lawrence_Bird]

and probably the most so of the vast universe of linux distros. Does this make it better? Thats for you to decide, but to me consistency and building upon what was already a good foundation seems like a good thing.

Re:And that is why OS X will ultimately beat Linux

[TheRaven64]

Trying to make Linux a desktop OS is like trying to make an octapus by nailing more legs onto a dog.

Actually, Linux isn't the problem. The Linux kernel is not a bad piece of software. The problem is X. X is a truly ancient solution to the problem of drawing windows on the screen. It was designed beck in the days when a graphical terminal was just that; a machine connected to a mainframe which had a graphical display instead of a textual one. X advocates claim that it has great network transparency. In a way, they are right, but the transparency is at too low a level. Compare X and MS Remote Desktop over a low-bandwidth link for an example. Modern features, like alpha blending are not supported by the X11 protocol, and adding 3d acceleration has to be done by hacking in an kludge of OpenGL, which destroys the network transparency.

There are some open source projects, like Fresco that aim at providing an alternative to X, but Apple have actually created one with Quartz Extreme. A lot of the problem with Linux is that it is living in the past, trying to recreate X/UNIX from decades ago. When they find a bit that's too antiquated to be useful, they hack it a bit until it looks kind-of modern.

Microsoft threw out DOS with NT (although they supported it until Windows Me), Apple threw out the old Mac OS with OS X. Th *nix crowd are still trying to adapt legacy ideas to modern computing. They will probably be able to for years to come, but eventually they will discover that you have to break backwards compatibility, or end up with a horrible kludge of aging ideas.

An aside: Have you noticed how many Linux users claim x86 is good, in spite of being a hideous architecture, because it's popular, but refuse to accept a parallel claim about Windows?

Re:Ack! Drivers in the kernel???

[Angry White Guy]

Actually, you should use the modules to test the hardware, then compile the kernel with only the drivers you need, and one with LKM on, and one with LKM off (Default). When you add new hardware, you reboot into the lkm-enabled kernel, load the proper modules, recompile the kernel for the new hardware, one with LKM, one without, and schedule a reboot.

By the same token, you should wear a seatbelt when riding in a car, but not everybody does.

Re:And that is why OS X will ultimately beat Linux

[0x0d0a]

The problem is X.

Joy. More unwarranted X bashing.

X advocates claim that it has great network transparency. In a way, they are right, but the transparency is at too low a level.

Why?

Modern features, like alpha blending are not supported by the X11 protocol, and adding 3d acceleration has to be done by hacking in an kludge of OpenGL, which destroys the network transparency.

What, you mean the RENDER extension? And I've coded in OpenGL under X11, and I have no idea what you mean by "destroying the network transparency" -- OpenGL is quite transparent over the network. In some ways more so than X11 itself.

Apple have actually created one with Quartz Extreme.

Which isn't network transparent and uses an insane amount of resources. Less features, more resource usage...why is it good again?

When they find a bit that's too antiquated to be useful, they hack it a bit until it looks kind-of modern.

Oh, for Chrissake. OS X doesn't look like anything but a large collection of not particularly usable eye candy to me. So I guess it's all in the taste.

They will probably be able to for years to come, but eventually they will discover that you have to break backwards compatibility, or end up with a horrible kludge of aging ideas.

I've found that the merit of a design is in how long it can run before it has to go. Windows 9X, for instance, lasted a little over five years. The classic Mac OS lasted a good fifteen. UNIX is at about 30 and shows no signs of slowing down.

An aside: Have you noticed how many Linux users claim x86 is good, in spite of being a hideous architecture, because it's popular, but refuse to accept a parallel claim about Windows?

What, that there's benefits to using something popular? Sure there is. That doesn't mean that the benefits outweigh the drawbacks in all cases. Windows is popular, so it's easy to find software for, etc, etc, but it's expensive, somewhat buggy, and a pain to know what's going on in the internals. x86 is popular, and while the instruction set is old, it's the only real contender out there. Alpha's dead, SPARC doesn't kick ass any more, the (desktop) PowerPC may one day become important again but lost a huge amount of ground when Apple trusted Motorola instead of IBM to do development.

I don't think anyone loves the x86 instruction set. It's just that the best bang/buck processors currently out there happen to use it.