Slashdot Mirror
YaST to Become Open Source
Space_Soldier writes "According to News.com, YaST is going open source: 'For years, SUSE has considered its YaST (Yet Another Setup Tool) software for installing, configuring and managing Linux an advantage over its competitors and forbade them from incorporating it into the products they sold. But with the new plan, to be announced Monday at Novell's Brainshare conference, the company will release YAST under the GPL, sources familiar with the plan said.'" Several years ago, when I first used YaST, I found it to be superior to the rest of the all-in-one administation tools around at the time. It was generally regarded as a great program, save for the licensing. Today, that's no longer a concern.
Hopefully we can get other large companies putting as much support into open source as Novell is.
-------
Chunky Bacon
The problem is that YAST has been going downhill ever since Rolf Schilling left the project. Now they have to GPL it to get development going again. It was a great AI1 tool once but it has languished for at least a couple years now.
I've been using Yast for the last few months. I've been really happy with it. Updates are a breeze. Installing new software couldn't be easier. It hasn't missed a dependency yet. Usability is pretty good and fairly intuitive.
Clearly Novell is taking the hint. They're aware of the fact that the world is going Open Source, and they're willing to deal with it. If they ensure a good relationship with the open source community now, they'll be rewarded with success for years to come. If they distance themselves from the open source community, like SCO, then they will make more money in the short term but be ousted in the long term.
Novell is a good organization that has been around since the beginning (or, at least, for a long time). I, for one, hope they continue to be around and keep up the good work.
Many distributions have open sourced their installers and administration tools, but for the most part, many have been useless for other distros. Many expect the EXACT filesystem hierchy of the original distro, exact package tools, etc. etc. So if I want to make a distribution and base it on someone else's installer or admin tools, I either have to dig into their source and do it myself, or make another distribution. There are some notable exceptions (webmin being one of them). While this is mostly good news, what I question is, will I be able to use it on linux from scratch without heavy porting? If not, it's not much use to many people.
Thing is, I think Novell's got the idea. Once we can develop good, solid, working ways to install the operating system, supporting it should be a lot easier. And Novell knows that there's no reason to NOT tap the millions of people online willing to help code this platform. I personally believe Novell's trying to secure itself as the second large Linux supporting company. By buying Ximian, they gave themselves a very viable desktop, by buying SuSE, they gave themselves a stable platform. Now they just need to do the middle work such as getting it to work on all hardware, and making it easy to support. And IMO, open source is a hell of a lot easier to support, especially since the people with the problems, usually know how to go about fixing them, and will send patches.
Don't discredit the selling power either. This probably won't hurt the sells of SuSE at all, in fact, it very well might augment sales, due to the people without fast internet connections wanting to get a taste of the YaST code. Don't count on it, but the potential's definitely there. Novell's making a good move here, I commend them.
"Victory means exit strategy, and it's important for the President to explain to us what the exit strategy is." G.W.Bush
Is this really such a good thing, in the long run?
It is if it increases SuSE's penetration as a distro. Before Novell (reasonably deep pockets) bought SuSE (pretty small pockets), the distro had to be a profit centre. Now Novell can afford to allow the entire distro to be free (a la Red Hat), so that more people use it and use Novell/SuSE's server and service offerings as a result.
Novell/SuSE will want as many people to try their software as possible: making their entire distro GPL-friendly will accomplish this, along with Red Hat's official abandonment of desktop Linux. Sure, short-term this may hurt them (I was planning on purchasing 9.1 soon, I may not now). It is *because* of the long-term benefits that this makes sense.
Look at the tomato! Isn't it sad? He can't dance! Poor tomato!
For years, SUSE has considered its YaST (Yet Another Setup Tool) software for installing, configuring and managing Linux an advantage over its competitors
Then they probably should not have named it with the "Yet Another" schema. It does not really give the idea to the user that the setup tool is an advantage or in any way innovative. Serious lack of confidence there.
To add to the previous three posters who all made excelent points - RedHat GPL'd most everything and provided ISO's for download, and still managed to make money off of Red Hat Linux. This will just increase install base, not decrease sales.
No, not at all.
You need to realise that Novell's product is not a Linux distro - that was never their reasoning behind the purchase of SuSE. Rather Novell purchased SuSE to give them a strong, established Linux distro on which to base their directory service offering.
Prior to purchasing SuSE, Novell evaulated its position in the market. What they found was that while they had a kick-ass directory service product, they were being kicked in the pants when it came to new deployments - primarily by MS Windows and Active Directory.
Rather then attempt to re-build and re-position the NetWare brand among IT decision makers, Novell realised they could do much better by taking an existing base Operating System with widespread appeal, and integrating NDS with that.
Essentially Novell's cut NetWare* and tied its future to NDS on Linux.
Enter Linux. It had everything Novell needed: stability; maturity; widespread developer support; GPL (why write a new base when you can modify an existing one?); a wicked reputation among IT techs and, best of all, an increasingly bright future with the potential to topple all challengers.
Announcing NDS on Linux and then subsequently purchasing a well established Linux distro was, not to put to fine a point on it, absolute genious. NDS gets the best possible base, loss of market share to Active Directory is significantly slowed or halted (and eventually reversed if all goes to plan) and Novell regains the reputation it had among techs back in the days when MS' best offering was WfW.
GPLing YaST isn't a loss for Novell, it's a gain for Linux. Which makes it a gain for the base OS Novell will see increasing use of NDS on. Which makes it a win for Novell.
*Yes, Novell will continue to support and even offer NetWare-based NDS installations. But the fact remains that if all goes to plan, Novell will see its new business increasingly tied to NDS+Linux rather than the old bundle of NDS+NetWare
Janie took my gun...
I think a better rule would be not to make excuses for badly written tools.
GUI config tools should follow three simple rules:
Otherwise they are useless.
eh, I'm not gonna go into the desktop wars, saying why Novell did something one way or another, I'm not on the Board of Directors and I wouldn't know these things. But Ximian is doing a damned good job at making a desktop along with the rest of the GNOME and KDE world. By buying SuSE, they gave themselves a floor, walls, electricity, the basic framework of what they'd need if they want to move all their products to linux. By buying Ximian, they gave something to skin the house, make it look pretty, make people want to buy it. SuSE is a KDE distro, but there's no reason you couldn't install Ximian over it, and that's one reason they may have bought the two companies, to give themselves a more diverse enviroment to attack RedHat at their own game.
Could be another reason behind opensourcing YaST: give it a GTK2 interface and wala, you've got a complete, working corporate desktop platform, which of course, they can then use to sell their own eDirectory software, and others as well... It's all about building a platform. Microsoft understood this too, how do you think they became so powerful? They built two seperate houses, both very shady but they got the job done. Then they skinned one house when they realized it was about to collaspe. Moved the skin from the first, to the second, and poof, a solid platform. Now they can sell Active Directory, Office, and other software for it, and not look like bumbling rejects.
It's all about process, format, proceedure.
"Victory means exit strategy, and it's important for the President to explain to us what the exit strategy is." G.W.Bush
The problem is SuSE and distros like it are NOT suited for you. You sound like someone who wants to do everything themself. Try slack, gentoo or arx, not SuSE. Use the CLI, it sounds like what you want and need. Or, as an alternative, a version of YaST that has various levels of interaction, like simple, moderate or expert.
*Fortitudo, aequitas, fidelitas.*
They are no longer redhat-config-* but system-config-* (as of Fedora).
Remember the roots of Novell. They are a network administration company and this is what they want to do. By making a decent, freely(or cheaply) available linux distro more popular in the business and home world they are causing a threat to the MS monoculture model and will have an advantage with dealing with all the different types of systems out there. They'll be the ones calling the shots of the network, where all really happens.
If you ever want Linux to become a real destop choice to Joe User, it must get more 'newbieized'. Or, to be more specific, 'windows-ized' (*shudder*). I'm not trolling, but speaking out of experience. Lots of it. I've tried to switch to Linux from Windows many times, but noticed that all too often, if I want to make the OS work the way I want, I have to dive into the dangerous world of config files. YaST is a step into the right direction, and with Novell's decision to set it free I'm sure it's going to get even better.
When you get a new system, please learn how to use it. If something annoys you, find a way to switch it off, or change its behavior.
/etc/sysconfig files directly, you'll love how it streamlines things. That way you will have more time for other things.
/etc/modules.conf for instance, just to watch TV on my Linux box?
Actually, if you change a file directly SuSEconfig can tell that has happened and will not touch it in virtually every case that I did that. But, the best approach is to edit
Why do people insist things are great for newbs when it makes ones life easier, and makes it quicker to get to the point where you want to be. Why should I spend two hours setting up a TV card manually in
I don't think YaST's appeal comes from its installer aspects, but rather from its system management aspects after you get installed and running.
Comment removed based on user account deletion
The registry?
.config file between the 2.4 & 2.6 kernel if necessary, it just ignores what it doesn't know.
1. It is way too complex. There is no way you can understand it all or hand edit it if required.
2. If it is corrupted, your whole OS won't even boot.
3. Its huge! 45MB of my fairly clean XP box.(although it is in a domain and has policies applied to it, etc, etc, but not much software)
4. You can't move the registry between machines, let alone between different versions of Windows. I can move my
Several smaller independent registiries might work better. e.g. one for linux conf, one for X, one for KDE, etc. So each one has a small well definied file for all configs.
Why don't you just use the ftp_conn_track module for ip tables? Then you don't need to leave large ranges of ports open, just the standard ftp port. Once a connection is established the connection tracker will manage opening and closing of ephemeral ports.
This is what package managers are for. Any program that overwrites the config file from another program is broken and should be fixed.
The whole idea of having the GUI config tools work "on top" of independtly developed applications, written with no thought of GUI, or even non-nerd, configurations, is a loss-loss situation.
What we need is for a standardized way for the application developers to communicate the possible configuration choices and their legal values to the config tools, and for the tools to communicate these choices to the applications.
The interface must be extremely simple to use and light weight in order to be acccepted by the application developers. And it must be stand alone, not depend on any particular framwork or other libraries. The primary interface should be to the application developers, because it is their accept we need first. Our ultimate goal, to serve the users, will have to come next. We won't serve users by having a cool interface that no applications support.
I believe it can be done, though. I got such an interface accepted among Emacs developers, and I suspect similar tools are accepted in the limited domain of KDE and Gnome. That such a tool can exist in the whole domain of free software, is shown by the acceptance of the gettext interface. Those free software projects that do localization, tend to use the gettext interface. Because it is so simple, non-intrusive, and toolkit independent.
Somehow you're supposed to know that
HKEY_LOCAL_MACHINE -> SOFTWARE ->Autodesk ->
AutoCAD -> R14.0 -> ACAD-12:409 ->
Applications -> AecBase -> LOADCTRLS
is supposed to be set to 0x0000000d (13)
(as opposed to say 5)
I suggest you take that complaint up with Autodesk; MS can hardly be held responsible for how other companies store their apps' configuration settings, and the documentation they may or may not provide.
True, MS are just as bad in this respect, but surely you could have picked a better example? That's like saying that text config fles are bad, because of sendmail's one.
It's official. Most of you are morons.
Sendmail's config file has documentation. It may be long and unfriendly, but it is possible to understand it.
Most of these "issues" wouldn't even exist if the documentation was clear, concise, and available. Then we would be saying:
Of course it should be 13, 13 means "load last file upon launching".
Whether it's Autodesk's fault for not including some sort of documentation, or MS's fault for not requiring descriptive strings for elements in the registry is up for debate. The prevailing opinion that the MS registry sucks (as it exists today) is hardly every questioned.
For so many companies (including Microsoft) to be using the MS registry so badly, I'd shudder to think that best pratices concerning the registry are being followed (or even published).
But they've been able to do this all along - the restrictions in the Yast license solely applied to commercial distribution. Giving it away for free, modifying the source etc - that's all been allowed already. I can see your point though, that someone selling the SuSE ISOs could reduce their sales.
> GUI config tools should follow three simple rules:
..
Interestingly, those are the same rules that I followed when developing Webmin, yet another administration GUI. Other programs that keep their own databases of settings from which the actual config files are built annoy me, as they make it hard to interoperate with other tools. Some of Redhat's control panels and Linuxconf are guilty of this
because mandrake drak* tools are the most buggy thingies from the whole distro, for ages.
There you are, staring at me again.
That's a programmer problem, not a design problem. Not to mention that many config files are way too complex as well. One thing that's nice about config files, however, is that you can include comments. While you could do this with the registry (with the EXPAND_SZ, expand string, type) it's not optimal as it increases the size. And nobody does it. [aside]If programmer's don't want you to change values or the values are meaningless...why make it changeable? Why not hard code it?
A redesign of the registry with a seperate table for comments would be interesting, I think. That way, when using editing tools, the comment table could be referenced. But, when loading or executing software, the comments would not hinder performance.
While I somewhat agree on this point, I have to note that corrupted config files will also prevent Linux from booting. I don't know the format that Windows uses for the registry tables, but it should be recoverable. Also note that I've yet to see any registry corruption on Win2000+, except with HW failures. I think the inclusion of something similar to BartPE or ERD Commander would also be a worthwhile replacement to MSFT's extremely limited Recovery Console. And frequent, automated, timed backups of the registry (at least OS configuration) should be done.
My Win2003 server, excluding registry backups and the user.dat portion, is only 23MB. 17MB of that is in HKLM\SOFTWARE (I have a lot of software installed). Perhaps someone handier than me in Linux could tell us what size all of the config files for a normal desktop come to (actual space on disk, ot just data size).
While true that you can't move some parts of the registry between machines (parts dealing with hardware and the like), software configuration is easily moved. I don't recommend moving the entire hive, as it would no doubt cause problems, but .REG files can be imported/exported with no problem. And .REG files are pretty portable (and text based), though it does require some editing and checking of data types to move from NT based to 9x based machines. With NT becoming the standard, though, that concern should go away.
Perhaps a DBA could chime in with better info, but I think that you would then be duplicating database structure overhead on each of those files. While I see the concern of a single point of failure for all software in the machine, automated backups and sensible defaults should mitigate that somewhat.
I think the main advantage of the registry is a central location for configurable values. By using a database, you should also have the advantage of database reliability and performance. Of course, the real problem with it would be getting everyone to use it.
The difference is that the files that will keep Linux from booting if corrupted are mostly static. Rarely, for example, is /etc/inittab edited. So, the likelihood that these files will suddenly be corrupted is fairly low.
The Registry is anything but static. Apps write to it all the time. That increases the likelihood that one wrong write will mess up the whole thing.
As a SuSe home user for 3 years, I say it's about time. This tool has made SuSe the perfect distro for a Linux newb such as myself. Had it not been for the YAST module I believe I would have spent countless hours, trying to figure out configurations.
Though the documentation was lacking in the SuSe distro, YAST made my transition from a strictly Windows user to a multi OS user. I now use Windows strictly for playing those games that refuse to work proerly under WINE. PSSST MESSAGE TO THE OPEN SOURCE COMMUNITY. How about improving the video acceleration of Linux! We need better games. And yes I did read the GAMES FOR LINUX ARTICLE
At any rate the article made no mention of YAST2 though. Is this to remain outside the GPL?
I am Bennett Haselton! I am Bennett Haselton!
Is that why Suse Personal 9.1 only has KDE on it?
(Please browse at -1 to read this comment.)