Slashdot Mirror
Ubuntu Linux Review
JimLynch writes "Pardon me while I pimp one of my own stories. We've got a review of Ubuntu Linux up on ExtremeTech. Check it out. Overall we had quite a positive experience with it, we think it's going to be a good distro as it matures. If you're looking for an easy-to-install debian distro, give it a download." Update: 09/27 23:25 GMT by T :
Eugenia writes with another review from USALug, and a 6-page comprehensive Ubuntu preview at OSNews, writing "Gnome's & Ubuntu's release manager Jeff Waugh also had an interesting interview detailing lots of interesting tidbits. The final version of Ubuntu is expected mid-October."
Choices.
What does that even mean? If you need your computer to do 'X' you don't change your OS, you get an app or two that does the job.
Finally, we were disappointed (but not surprised) that Ubuntu did not detect or configure the wireless card in our laptop. We've come to expect this, unfortunately. But it would sure be nice at some point if we could connect wirelessly right after installing a Linux distro, with no extra effort required.
I don't know what brand of wireless card it was, but if it was one with a Broadcom chip inside, well your SOL on that one. If they would give out the specs, we'd have drivers for them.
Swapping a distro isn't swapping the OS at all, it's all Gnu/Linux after all. It's sorta like putting it in different wrapping paper.
Be Safe! Sleep with a Marine. Semper Fi!
Some people want a server, some people want a desktop, some want to run an FTP server.
Umm HELL NO. Look at what troube we have when a Web Browser is build into the OS. Keep it in userland people...
"It's not like your minds are as open as the source you love..." - Me to the majority of Slashdot.
The main thing, it seems, is that this disto provides a spoonful of sugar to make the Debian medicine go down. But this sugar may not be enough for laptop users. Quoting from the article, we were disappointed (but not surprised) that Ubuntu did not detect or configure the wireless card in our laptop. So that spoonful of sugar may be deceptive ... some real skill may be required after the pointy-clicky stage. Is it a good thing to mix the difficult and the simple?
What benefit, exactly, do Linux users get from the proliferation of distros?
No-one gets any benefit directly from the fact that there are a large number of distros.
However, for each specific distro, there is apparently at least one person who likes that distro better than the alternatives. Which is enough.
If someone decides he wants to make AbominationDistro, which is existing distro X but with the meaning of /etc and /usr switched around, and he creates it - more power to him, that doesn't influence me at all - and he has the distro he wants.
I believe posters are recognized by their sig. So I made one.
To me this is "unnecessary choice". The choices only evolve because the existing procedures fail at particular tasks. Why not develop, for example, a clean, comprehendable, human as well as machine usable startup system to handle services so that one distribution could act as a server or desktop.
Hell, isn't this the purpose of the SysV init system, to provide a separate runlevel for X as well as a stripped down network-and-servers-only runlevel?
Choice is bad where it limits interoperability. You don't see browsers speaking Debian-HTTP, Redhat-HTTP or SuSE-HTTP, do you?
People seem to confuse choice with not-a-standard sometimes.
IMHO
Of course it runs NetBSD. BTC: 1NT7QvbetmANwaMzhpVL6
"Some people have even told me that have a low usser number.."
Well, that's all depending on the perspectve...
On Topic: Your complaint on choice boils down to the same question as when people decry the existence of mutliple desktop environment projects - who is supposed to enforce anything?
Ok, assue that we have too many distros, and that it is hurting the community in some unsepcified way. How do you suggest reducing the number? You can assume that people building on, or using, any given distro aren't willing to volontarily dump their work just because someone says so. So how, then?
And once you've accomplished your purge, how do you propose avoiding a couple hundred new distro projects popping up like mushrooms after rain?
Trust the Computer. The Computer is your friend.
You're right the review sucks. For instance, under the heading 'Installation', it says,
"At the end of the installation, we were asked if we wanted to use APT (Advanced Packaging Tool) to update our system. We said yes and our system was updated over the Internet before we even booted into our Gnome desktop."
This totally glosses over the connection to the internet. Was it by broadband, satellite link or modem? Did the installer correctly identify the modem if there was one and did it create a connection to the user's ISP?
Mandrake 10 installation fails utterly in this task, particularly if the user has the misfortune to be in North America.
In times of trouble, the smell of frying onions usually gives confidence and comfort.
What benefit, exactly, do Linux users get from the proliferation of distros?
Very simple answer for you. You like Debian, I like Slackware, my work uses RedHat, but we are switching to SuSe soon. Some people like Mandrake, other's swear by Gentoo. Every Distro fills a niche, every person has a niche. The more distro's the better. If you have an issue with an array of choices, you can use Windows or Solaris. Some of us really like to have a distro that provides EXACTLY what we want.
You can not please all the people all the time, but giving people more choices increases your chances greatly.
If we don't make light of everything, we are just stumbling in the dark - Blank
To translate Jeff's market speak: you get commercial support with Ubuntu.
How we know is more important than what we know.
No, you're less likely to find one that does what you want, having to research 50 distros all claiming to be everything to everybody.
"Specialized" distros that are made for a specific purpose are great. Knoppix, Smoothwall, and Damn Small are examples of distros with specific reasons for existence. The problem comes when someone wants a desktop Linux distro, and there's an army of them claiming to be exactly what the user wants. After the user has tried 7 of them and found none of them were quite what they claimed, the user is forced to ask himself why he should bother with this whole "Linux thing".
What kind of VPN?
A Cisco VPN client? A MS style PPTP VPN client? Freeswan? That web based client for Nortel Convtivity?
It drives me nuts, and I'm the Network Admin at my company!
So the core of your argument is this:
Person A: Hey, I don't like the way X and Y work, why can't they work like Z
Person B: Do it yourself or stop complaining
And when they find a security vulnerability in zlib, you're prepared to download every single app using it again? That's assuming the project is still even being maintained.
Having each app statically linked or with its own libraries means you:
1) waste memory
2) waste developer effort
3) waste bandwidth
4) waste hard disk space
5) make maintaining systems a nightmare (DLL-hell)
6) Open yourself up to security problems - look at Microsoft's problems with their jpeg lib spread all over the system.
6) lose the ability to upgrade a system - say the zlib developers release a point upgrade that doubles the speed of decompressing. If I'm using shared libs, I can download the 100kb package and *every app* using zlib is now twice as fast at decompressing. In a static system, I have to wait for the developer of every app to take the time upgrade their app to the new version and then I have to download it.
How is keeping twenty or more copies of the same library any better than keeping one?
It reduces to the same problem of keeping libraries and the applications using them compatible, except now you have twenty or thirty as many files to keep track of.
In which case, you might as well just use system-wide shared libraries, with a few compatibility libs installed for those apps that need it.
You can do per-application testing with a chroot, if you feel the need.