A New OS

[nko]

Welp, this just might turn in to a flame-fest, but if it does, it wont be for a good reason. I know several projects like this are probably out there, but I wanna do my own :)

I'm going to make a new GNU/Linux-based OS. It'll pretty much just be an LFS with decent HIGs. Here's the basic idea, so far:

Redesign the file heirarchy system

Write HIG's

Port applications

The idea being to have more standardization, be more user-friendly, don't sacrifice any of the power Linux already has, and to focus on more mature, less hacked interoperability.

Anyone interested? I know a lot of people will probably try and shoot me down. It's happened on many another board :)

[st0rmrd]

The only reason to flame you, I think, would be just the fear of forking. We all know what happened to UNIX... (they ended up copying Linux code :p)

But, if Linux is so good of an OS, it should survive any such things, and worse. So, if Linux is worth it, it shouldn't fear of any thing, at least not such things. If it's not, then the sooner we find out, the better (you know *BSD, right :p)

So, no, I won't flame you.

Actually, I'll encourage you to go on. Because Open Source and GPL code in general, has this good attribute, that you can share code, so if you write something good, it might end up in the main stream. Good luck :D

PS. You should be aware that any project that aims too high, may end up failing easier than others.

[nko]

Awesome! Thanks for the reception. It would definitely be a fork, but I'd like to think of it as a whole other thing that just uses the same thing. Totally no disrespect to Linux or the current distro situation, which I think is very healthy, if only a little immature. I'd rather it compliment the current Linux / OSS stuff.

And as far as aiming too high, I was afraid of that for a long time, and so didn't move on this idea for a while. Then I just realized that all it'd take is to basically do a somewhat customized LFS, and then some brainstorming. The only part that could bring the project down would be porting of applications.

On that topic, what I'd like to do is basically take the code from current projects, bend it to interoperate decently with the rest of the system, change the name to a generic name, and then make sure that it'll be very well known which projects' stuff we use, so it's not like we're ingrates, or worse, lawbreakers. I'd definitely like an easily-accessible credits page, creditting all the projects we'd "tear apart" :)

[MorphiusFaydal]

theres already a new linux os in progress that got started on these boards... Dux Linux...

their website is dux.sunsite.dk (http://dux.sunsite.dk), and they dont have many people working on the project, u should probably go and join them, instead of starting your own distro.

New Linux OS - by Oubipaws (http://www.justlinux.com/forum/showthread.php?s=&threadid=92261) they've pretty much stopped posting here, the forums on the Dux website are up, and thats where the main stuff is going on, they've even got a version up that you can d/l and install, its kinda hard to install, and has no graphics, but hey, its a start.

so, my recomendation to you is, go check out Dux, and if that doesnt fit the bill, then go and start your own distro.

still, you are advancing the linux cause, so, best of luck to ya, and hope se you around!!

chris

[El_Cu_Guy]

theres already a new linux os in progress that got started on these boards... Dux Linux...

Oh yeah, wasn't that the one with the 3D GUI that was to natively run Windows and Apple binaries too? It was a project that seemed to have been started by a cart before the horse marketing team. You know, the "this is what it will do/have, but we don't know how to make it do that anyone out there with any programming skills whatsoever?"

They've got screenshot but they don't seem to be related to Dux in anyway shape of form.

I'm going to make a new GNU/Linux-based OS. It'll pretty much just be an LFS with decent HIGs.

I agree with MorphiusFaydal. Why does everyone think they need to start their own distribution? Why not build on new projects? End fragmentation!

If you reall wanna go after some wild ideas that might appeal to you, why not check out another unix-like OS? Perhaps AtheOS? It's under the GNU GPL and LGPL. Take it and run with it or join teams already doing so. The lead developer is a dick and apparently doesn't seem to like letting others play in his sandbox.

[nko]

Cool! I can't figure out quite what it is that Dux Linux is doing, though. Their web site is still a work in progress (though it does seem to be a freshly started work in progress). Can anyone tell me what it is they're trying to do? I think I'm in a different direction.

[MorphiusFaydal]

The Dux team is trying to make a new linux distribution, they have gotten off the ground, and there is actually a working version available, command line only, root-access only, no installers, bare-bones...

yeah.

for the best synopsis of Dux, read the thread here on justlinux or cruise the forums on their site.

from what i can gather from your posts, you want to create an entirely new operating system based off of linux, not just a new linux distro. but, unless you have extensive coding skills, you probably wont get much done, other than a rehashed linux kernal with your name on it.

as for the Dux website, there website is/was based on the PostNuke content management system, and they decided that that wasn't going to work out, so one of the guys, who is also webmaster for the team, is building a new site. no idea when it'll be ready, but hes working on it.

El_Cu_Guy: you may want to check out the Dux website, they have decided to drop 'native' support for windoze and MacOS, because it is nearly impossible to build the binarys into the main kernal. as it is, they will be including Wine, WineX, and Darwin as some of the basic packages in the Dux install. you probably should read the Dux sites forums a bit, that'll really fill you in on the project.

peace out

chris

[glussier]

It was a project that seemed to have been started by a cart before the horse marketing team


Wasn't-it Lindows that marketing team?

[oubipaws]

The Dux Linux project was put on hold due to some issues with people having finals in college and needing to be able to study, but as of right now the project is back on and I would love to have your help... Here are some of the things I would like to be able to do:

~ 1st of all.. I would like to be able to change the way the system is layed out in the root directory. For example we could change the /home to /Users or whatever, just an easier system for people to understand so when they open up the / directory for the first time they don't get freaked out and never want to mess with it.

~ 2... I want to make Dux as user friendly as posssible, and to do this, we are using a Gentoo base so that we have EMERGE built into Dux to make installing things more easily.

~ 3... A 3D GUI, yes I said a 3D GUI, the same subject that has been downplayed by several people on this forum. The GUI would be something similar to 3DNA (3dna.net)

~ 4... A highly customized *Box GUI (openbox, fluxbox, blackbox, etc.) as the primary GUI instead pf having something like a specialized version of KDE or GNOME which can be quite clunky and buggy at times.

~ 5... Hardware support... I want Dux to be able to support as many possible types of hardware as it possibly can support.

~ 6... User Mode Linux... I want this to be included with Dux so that people can play around and learn how to tweak the kernel, GUI, etc without having to worry about destroying their system. This was discussed on the Dux forums and will not be included at first, but will be put in the OS.

~ 7... Games... To please so many of us gamers out there, I would like to be able to somehow include a way to play at least a few windows games, and if not that, I would like to include plenty of Linux games and I don't just mean those goofy arcade games, I mean somesort of serious games... something with a lengthy play time :-P

Ok, thats all I'm going to type, anyway, I just wanted to give you a small overview of the project to see if we could possibly sway someone of your talent (I am assuming you know what your talking about since you are wanting to achieve such major things) to our team for Dux.

By the way, you mentioned the Dux website being kinda sad looking, I am working on a new design as of right now for the site, so keep checking back.

[nko]

I didn't mean it's sad looking, just that it looks like it recently started redoing some things :) Looking at dates, things really don't seem too shabby.

Has there been much discission about directory layout? Dux is sounding more interesting. I just might have to zip back down to their forums.

[nko]

So, a 3D *box GUI?

I've always thought that anyone who says something is a waste of time is just too closed minded, and can't see the big picture. Any idea of an API? I cringe at the thought of X libs only slightly less than OpenGL ones. So much so that I've never tried coding in either :)

[bwkaz]

Hmm... changing the directory layout is probably not such a good idea. Yeah, it's crappy, but there's so much software out there whose build scripts absolutely depend on a certain directory structure, that I'd be afraid to change it.

For example, NEVER, NEVER move /usr to something else. X's default build system, for one, will probably fail, unless you reset ProjectRoot in the X configuration. /home might be acceptable, but I think I'd put a symlink called /home that points to /Users anyway, just in case.

But hey, whatever. You don't have to do any of what I'm saying with Dux, or with this other distro (you're doing it in the end, not me ;)), but just so you know the kinds of things that might break.

[nko]

Oh, I do dare to touch the directory layout ;)

I think that's my main hangup, that the directory structure is so violently bent to legacy's will. For all it matters, if I make such a fork from regular distros, it'll be devoid of symlinks to the old structure. Until an app was ported, I'd just leave it out. The idea, at least at first, not being functionality, but to try something new. See if it makes a big enough splash. If, for example, someone were to change the whole thing, and it went over so well that someone went through the trouble to port X build scripts to the new layout, I think it'd really say something.

It wouldn't take that much effort to do to a really basic LFS, would it?

[oubipaws]

2 things before I head off to bed (I will be on here tomorrow to talk to you about this some more):

1) The directory structure, I know this is dangerous, but there are work arounds, I will explain tomorrow.

2) oh well, I forgot what number 2 was, I've been up since 3am and its now 11pm so I'm off to bed... night guys

[El_Cu_Guy]

El_Cu_Guy: you may want to check out the Dux website, they have decided to drop 'native' support for windoze and MacOS, ............................................. you probably should read the Dux sites forums a bit, that'll really fill you in on the project.

I was discussing what is was, not what it is "supposed" to maybe be.

Wasn't-it Lindows that marketing team?

Didn't someone make that comparison in the presious thread?

I don't see Dux going anywhere. I mean I see so many goals and not a clear path to getting there. *box GUI yet it's supposed to be a "revolutionary" 3D UI. It seems more to me that this is more of an afterthought. I think you'll run into problems moving from one to the other this way.

As for changing the directory structure, why not? It can be done. Look at BeOS. It share commonality with Unix but it's different.

[CMonster]

YALD? Yet Another Linux Distro...?

Applications... applications... applications!!!

Give me software for Linux like:

Jumpstart (http://www.jumpstart.com)

Other Educational (http://www.broderbund.com)

Landscape (http://www.gardencomposer.com)

Home Improvement (http://www.punchsoftware.com/index.htm)

Pharmacy (http://www.rxitequick.com/software.php3)

Medical Billing Software (http://www.medicalbilling-direct.com)

Printing Photos (http://www.kodak.com/eknec/PageQuerier.jhtml?pq-locale=en_US&pq-path=130)


For a few examples, but the list goes on and on and on....

You see, there are hundreds of these kind of applications (and innumerable others) for Windows, but next to none for Linux. A few exist, but more are desperately needed if Linux is to succeed as a Desktop OS.... and the worst part is that few companies are willing to port a "for pay" applicaiton to an OS that will require a patch every 4-6 months as the new iterations are released with updated kernels and glibs and such...

[oubipaws]

Kernel redesign is not an easy task, but with time and effort it can be accomplished and if nothing else we could always design a way so that a newbie can use a script to create symlinks that show what each folder really has in it.

As for how much this has been discussed in the Dux forums, very very little, as in one post, but I'm going to post a detailed page of all of our options that we could be able to do this with.

As for the 3d gui, this is just an idea, if it fails, then it fails, but hey, atleast we tried, so just as every projects, some things will succeed and others will fail...

NKO: I'm not sure of how the GUI is going to built yet, we haven't gotten that far yet, first we have to get out a good stable base version of Dux...

CMonster: most of the applications you have mentioned below run fine using VMware which can be installed on any computer for small fees. WINE might possibly run them also.

NKO: LFS is a no go due to the fact that we are including Portage. Thats why we are using gentoo as our base design. IF you know of a better system thats as easy as portage which we can install on a LFS system, please tell me?

Alot of things are going to starting to happen with dux now that most everybody is done with finals and such and now has time to work on it...

NKO: another thing, I didn't mean that you said the website looked sad for Dux, I was meaning that I thought it was starting to look kind of sad
:p

[nko]

I meant LFS if I was gonna redesign the directory structure. Portage.... yeah, I'm gonna concent to Gentoo on that one without a problem :)

I think a directory structure redesign would be easy with LFS. Putting applications on there would be a whole different taco. KDE? Forget it. X? Not likely. PINE? We'll see :)

[oubipaws]

I don't really think that redesigning the / directory would be that difficult, I just think that the after parts would be really hard. Unless you want to go in manually and alter the X source code so that it sees the new files, but if you did that, you would have to do the same for every program out on the net if you wanted people to use it....

[bwkaz]

You should be able to compile portage on LFS...

Although I don't see a hint for it anywhere, hmm...

I can't think of any reason it wouldn't compile, personally... is there one?

[oubipaws]

I had completely forgotten, but I believe Mooktaking got Portage to work on the first Dux Base

[oblio]

This Dux Linux thing needs a bullet put in it. I've been following this "project" since it started here, tried to move to some other linux site where it got flamed into submission and regularly check the dux site dux.sunsite.dk for updates. I wanted to be wrong. I wanted to watch this thing happen that I thought was impossible and be proven wrong but dear God it doesn't end!
I have no problem with people dreaming, speculating and trying things but all the Dux team is talk-talk-talk blah-blah-blah. Oubipaws pretty quickly went from the visionary with the big idea to the administrator who seemed more concerned with assigning people tasks than about anything getting done. Mooktaking seems to be the only one who is actually working on a distro.
The whole project went very quickly from being about creating a distro to creating websites about creating a distro. In the several months that this "project" has not been happening the various people involved have acquired sourceforge space for a 3d-desktop that does not exist, sunsite space for a distro that appears to be little more than LFS on one member's machine, been in contact with VMWare regarding including their software in this hypothetical distro, and more talking-talking-talking.
I know a distro takes time but after following this as long as I have I am convinced that Dux Linux is little more than a sparkle in the eye of a few web-site designers, some wanna-be administrators and a few people who are too naive to tell the difference.
To anyone that is thinking of joining the Dux Linux team I plead with you to take any programming abilities you have and USE them on something, anything, that would actually benefit the linux community. Linux needs apps, linux needs gui's for programs, linux needs help on ease of use, drivers, and about a million other things.
I tried to bite my tounge and held off as long as I could but I just can't take it anymore.
http://www.justlinux.com/forum/showthread.php?s=&threadid=92261 This is the thread that started it all. If you read through it it pretty much ends up devolving into what would be cool in the website and who should work on the website and what things the website needs. Then, recently all the activity on their sunsite dried up, ostensibly, because they were all in the middle of exams. Fair enough. After pretty much no posts for 2 months. They get back together and what's about the first news I see? "Hey were gonna redesign the website!!" and of cousre much talking about what everyone likes and doesn't like about the current website. It's jsut too much. I gotta stop or I could go on and on even more but alas perhaps I have already said to much. Maybe you can get the gist of how long this has been coming...

[hop-frog]

Originally posted by oblio
This Dux Linux thing needs a bullet put in it. I've been following this "project" since it started here, tried to move to some other linux site where it got flamed into submission and regularly check the dux site dux.sunsite.dk for updates. I wanted to be wrong. I wanted to watch this thing happen that I thought was impossible and be proven wrong but dear God it doesn't end!
I have no problem with people dreaming, speculating and trying things but all the Dux team is talk-talk-talk blah-blah-blah. Oubipaws pretty quickly went from the visionary with the big idea to the administrator who seemed more concerned with assigning people tasks than about anything getting done. Mooktaking seems to be the only one who is actually working on a distro.
The whole project went very quickly from being about creating a distro to creating websites about creating a distro. In the several months that this "project" has not been happening the various people involved have acquired sourceforge space for a 3d-desktop that does not exist, sunsite space for a distro that appears to be little more than LFS on one member's machine, been in contact with VMWare regarding including their software in this hypothetical distro, and more talking-talking-talking.
I know a distro takes time but after following this as long as I have I am convinced that Dux Linux is little more than a sparkle in the eye of a few web-site designers, some wanna-be administrators and a few people who are too naive to tell the difference.
To anyone that is thinking of joining the Dux Linux team I plead with you to take any programming abilities you have and USE them on something, anything, that would actually benefit the linux community. Linux needs apps, linux needs gui's for programs, linux needs help on ease of use, drivers, and about a million other things.
I tried to bite my tounge and held off as long as I could but I just can't take it anymore.
http://www.justlinux.com/forum/showthread.php?s=&threadid=92261 This is the thread that started it all. If you read through it it pretty much ends up devolving into what would be cool in the website and who should work on the website and what things the website needs. Then, recently all the activity on their sunsite dried up, ostensibly, because they were all in the middle of exams. Fair enough. After pretty much no posts for 2 months. They get back together and what's about the first news I see? "Hey were gonna redesign the website!!" and of cousre much talking about what everyone likes and doesn't like about the current website. It's jsut too much. I gotta stop or I could go on and on even more but alas perhaps I have already said to much. Maybe you can get the gist of how long this has been coming... Sounds a lot like the Dilbert.

[MMA]

Ok let me introduce my self. My names MooktaKiNG, MMA, Mook, Mooktakim. And whatever else they call me.

I'm one of the dveloppers on Dux Linux.

Ok reading through this thread i found some good points, bad points and so on.

But the idiotic post ^^ up there has really annoyd me.

Making a new dsitro is not about putting a coin in a vending (spell?) machine and popping a new distro out.

It takes time. It takes LOTS of time. And lots of discussions.

yes, we have been using Time and having lots of discussions.

We have also beeing trying different ideas, like LFS. but that didn't workout.

this is a team effort OK.

this isn't about WHO makes the best OS or anything like that. This is about a linux distro that we think (our own opinions) is better and that we beleive is a way of contributing something to the opensource comunity.

OK Mr oblio. I would like to know what you are doing to contribute to this community.

Instead of just taking and taking. We feel we need to give something. This is our chance.


OK, the idea about new directory structure is a good one. But i do belive that its not a good idea to remove the compatibilty with other programs.

This is the best solution, i beleive:

We create the directory we like. Lets say here: /usr/share/mycomputer or maybe something closer to root /mycomputer.

Then we symlink all the programs that a user installs in there. I think it should programs that only the user installs becuase there's a lot of BS in /bin that the user has no idea what todo with and would clutter everything up.

then we have a choice:

1] We symlink this to the user's /home/user/Desktop/My Computer

and let the user have the control and get him away from the / root.

2] we somehow mask everything in root and stuff and only show /mycomputer (or /usr/share/mycomputer). We can do this by modufying all the GUI's that were gonna use.
But we leave the user the command line. So they can still access it using command line. This is how Machintosh works.

OK if you want to know all the things we've discused then goto the dux website and read some of the threads. Its better then here, people who go there are mostly interested.

[oblio]

BLAH-BLAH-BLAH:rolleyes:

[nko]

Alright, alright, maybe they're going somewhere, maybe they're not. Same situation as any OSS project. Can we stop flaming them based on conjecture and opinion?

On that topic, may anyone touting the cliche, "if you don't like it, start hacking" heretofor be stricken from trying to help. The community just needs to be more understanding than that. I'm no mad hacker. I can do stuff, but I don't have the skills it takes to fix ANY problem I can identify. Most people are the same way (including SOME people who CAN be otherwise described as a "mad hacker").

Johnny Torch, please flame off.