Unununium, and interesting project which aims at producing an ‘organic’ operating system where all components are interchangable at run-time, recently put a screenshot online.
They have been making a lot of progress lately, with the website becoming increasingly active and filled with content. This project is a bookmark must-have for any assembly buffs out there as it is entirely written in assembly.
That sucks, guess I won’t be able to check it out.
When all else fails, view the source.
Being a OS screenshot junky I jumped right on the screenshot link. Well seeing as how the site doesn’t support IE… or more correctly, I suppose, IE doesn’t support the website, I desided to try Mozilla. I’ve read that the firebird project was one of top distribution of it so I d/led it…. Wow, is all I have to say. The while the project seems neat the screen shot was… um well what was that supposed to show? Either way, what is really immpressive is my new default browser… It is really really good.
Lets hope this OS becomes self-aware in the near future and decides to help the developers by programming itself
I just finished reading the documentation for UUU (on my new favorite browser)and I think this is some really immpressive stuff. While I don’t like the way the faq explains the concept of UUU’s “cells” by compairing them to living cells, the idea behind it is very cool.
The project site claims that the entirety of the code is written in assembly and that this will result in a single platform OS. Fair enough but given that we are moving towards 64-bit desktop computing and that Intel may not support 32-bit x86 instructions is this wise?
If the jump to 64-bit had already been made this would result in a lean and mean OS for sure. My ASM is rusty but there’s no doubting its performance. Hope these guys would need to go through hell to port it in 1-2 years.
… you seem to assume they are going to get something running. From all the buzzwords I have read from their website’s documentation, I spect a reality check in …. 4,3,2,1.
Wow…we made it on OSNews without lifting a finger this time.
The screenshot is actually VERY old, it dates back to pre-Uuu actually, when the project was called 4th Axis. But it’s the only screenshot worth showing, cause most people dont like looking at text consoles.
For those of you that may already know the project, think again. We’ve changed most of the goals since it’s re-emergence. We’re still doing 100% assembly, but not 100% x86 Assembly. Some x86 Assembly, but mostly Bismuth Virtual Machine Assembly. Bismuth is our very own VM, built with performance in mind. The OS will be able to handle CPU native code, aswell as the VM code, but for the time being we’re confident that most of the stuff will be in the VM. So that means that porting Uuu to any other CPU architecture would only take as long as it takes to port stuff like the memory manager, threader, and virtual machine (a few thousand lines each, not bad actually).
What the project really needs is coders. We’re very few coders (3-4), and none of us have much time, so we need help. If anyone has questions, post them up, i’ll do my best to answer them, and i’m sure Dave Poirier will also once he gets the chance. Or you can email me, whichever.
“get something running”.. I suppose by that you mean something you can try? Have you checked out the download section?
Why the decision to go with the SAS? Is it because you’re using a software-based process switcher? I’m curios how you are handling address space separation for user vs cell memory locations.
This one would be better answered by Dave, but i’ll give it a shot… We chose SAS because when you dont have segments, you can do all calls as ‘near calls’, which are a fair bit faster than far calls, so we gain speed that way. Plus it simplifies things a lot as a whole. You never need to play with segment registers, which can give a (very minimal) speed increase in the applications, but at the very least, it simplifies their design. But at this time we dont hvea support for Virtual Memory, we only support as much memory as the machine has, but hat has yet to be an issue. We’ve run Uuu on a box with 4 megs of RAM, without any problems, and plenty of room left. Currently everything can be seen as “kernel space”, we dont do security like that with rings. When we do implement security like that, chances are it not be done like most OSes out there. Hope that helps.
I’ve been interested in learning assembly for a while now, I was wondering if there are any good books, sites . . . anything to get started really. If I can get the hang of it I’d be more than happy to do what I can, and I’ve got tons of time on my hands (which is why I want to learn assembly in the first place), anyway, thanks for any response.
I find the best way to learn assembly is just to dive in and start coding. You can do things with very little knowledge with assembly. If you can find a tutorial that can teach you what a register is, and instructions, and the nitty gritty stuff like that, then the rest is mainly just learning how to use an API to make stuff happen. If you need help understanding stuff, #uuu on irc.oftc.net has quite a few people who could give you a hand. It’s a wonderful language, and really a joy to write if you ask me. A good OS to learn assembly on is DOS, if you want something cooler, there’s Uuu, and V2_OS, and i suppose you could learn to code for linux too.
thanks for the help
when unununium actually gets a name? (The element, that is)
Could the UUU project simply be called what is more commonly known as an “exo-kernel” OS? Various universities have been working on projects using this concept for some time now, so it is not entirely out of the blue. Still, UUU looks quite interesting if it can gain enough coders to contribute to it. Best of luck to you.
Before big changes and redesign there was UUU demo which loaded Sokoban game So UUU could do more than what’s on screenshot.
Now they’re redesigning all, so we’ll have to wait a little to get new Sokoban
I cannot imagine a faster way to kill a project… Denying 95% of your potential viewers to see what you are all about..
Well I’ve known these guys for awhile, and it’s good to know they most commonly have problems with renaming their projects rather than coding. These guys are brilliant and talented, if I had more time I’d finish my study of ASM and start coding beside them. Best of luck Richard see you in IRC.
why asm, isn’t pure C fast enough(?!?) on platforms UUU would be running when it’s done(4Ghz+ 32bit machines or 2Ghz 64bit machines)? considering pure C ANSI code could be compiled on miriad of platforms(once the gcc re-design would be done even on IA-64 or using intel compiler even now) and with optimization human rarely could make especially when project rises to be big.
and as to low level if you really really need _asm_ you can inline at&t syntax that would be also portable
Looks cool. Will check out the technical details later. One thing set alarm bells ringing:
On X11 licensing: “The license is compatible with the GPL but allows you to make money out of your creation, while still allowing other people to freely obtain and modify the sources.”
This statement is contradictory. By definition, the only way you can make money out of BSD/X11 code (that doesn’t apply to GPLd code) is by closing the source and placing it under a more restrictive/proprietary license, which then excludes other people from freely obtaining and modifying the sources.
You can’t have it both ways – if you want to make money by selling code, you shouldn’t be using a free software license (as otherwise it becomes meaningless when the first person forks to proprietary). If you want to make money from free software then services is the only way to go. It’s certainly doable, even for research projects like this one. The maintainer of PicoGui has done some consulting work for instance, that gave a nice rate of return.
Ah nope. If you are interested in doing OS work in assembly, you are almost by definition not using Internet Explorer. In fact even on this site, there are more Mozilla users than IE users. So really if it doesn’t work in IE, that’s just tough on IE users. If they’re at work, try again at home. If they’re at home, get a better browser.
they’re actually increasing their audience. the kind of people who are most likely to appreciate UUU won’t be caught dead near IE. 🙂 i’m still working through tanenbaum’s Modern OSes but i’ll have a lookie at it when i understand enough of it 😉
This project has some good ideas but since it is not written by machines but rather humans, than the motivation for assembly is hard to understand. Modern C++ offers multi-paradigms (action based, OOP, generic) and you can apply traceable and recognizable design patterns. I value the organic architecture in concept, that would certainly be nice, but I think I’d want it to be organic so that more people could get involved as maintainers of their own platform, so that the architecture would be easier to understand, and the source code components more approachable. In terms of performance, that would be realized by the new level of complexity that organic systems would be able to take on, as the system grows…bigger than anything we currently know.
Thats just moronic.
HTML (or any ) programming isnt just mindlessly supporting whatever browser you like but as many USERS as possible.
Is he trying to get his information out or just piss off 80% of us who MUST use IE at work.
Nice work pal.
Microsoft doesn’t typically follow internet standards, and now you know how everyone who doesn’t use Microsoft feels like:
I think they based their website code on a standard.
so.. uuu is still alive? i thought that died off ages ago..
i thought uuu was just a spinoff of v2? its been too long since i checked it out.
I don’t think Uuu is an exokernel. AFAIK the idea behind an exokernel is seperation of security and management – the kernel enforces security, e.g. access to devices and memory, and management (such as filesystems, drivers, etc.) is done at the application level by an application or a LibOS (library operating system, an abstraction normally provided by the kernel in traditional OSes). It’s a little bit like a microkernel taken to the extreme. Uuu looks like it has no security.
Yes we brought the project back from the dead, we missed working on it too much. We were never a V2_OS spinoff, what happened is that most of us met while developing for V2_OS, and Dave Poirier was one of the major coders for it. When V2 started going downhill, we all jumped ship, and started our own OS. There is very little in common between the 2 projects except that they’re both in assembly.
For the people bitching about IE… make it comply to CSS standards, or stop your bitching. Our HTML is 100% standard compliant, and so works with damn near any browser BUT IE. We dont care about IE users not being able to see the site. If someone REALLY wants to use the site, they’ll install a browser that is actually worth something.
Uuu is not an exo-kernel OS, there are differences. Even the exo-kernel relies on the fact that there is a certain amount of code that remains statically in place. We dont have that.
As for not showing people what they want to see… why not just download the boot floppy and try it out? I dont know about you guys, but i much rather try an OS out than look at screenshots. The ‘sad’ thing is that there is a lot of drivers written for it that we cant really use, cause we dont have anything using them.
These guys are likely writing HTML code for what browsers they can get their hands on. I’m willing to bet they’re not even able to run IE.
Don’t complain to me or anyone else about not being able to view a web page on a platform dependent browser. Go get one that can view it.
As long as I can’t get IE/ActiveBLOAT for my platform I have every right to bitch about IE/ActiveBLOAT only sites.
>>> “Could the UUU project simply be called what is more commonly known as an “exo-kernel” OS? Various universities have been working on projects using this concept for some time now, so it is not entirely out of the blue. Still, UUU looks quite interesting if it can gain enough coders to contribute to it. Best of luck to you.”
Not really. The concepts behind and exokernel and Uuu are quire different. An Exokernel has a very minimalistic static piece of code loaded in memory responsible for sharing the underlying hardware between various OS Libraries.
In Uuu, we have a single “OS” that is built out of various totally separate reloadable dynamically linked components with no static piece of code anywhere to be found.
A lot of people asked why we selected assembly rather than a higher level language, which technically, would make more sense.
The reason is rather simple, it’s because we prefer writing assembly, we feel closer to the hardware, more in control. We do not write this project with the hopes of conquering the world, or writing the next desktop OS, or the next server OS.
We are experimenting and developing new concepts, trying out new ideas and enjoying developing together. For these very reasons, we selected a language we like working in, and that is assembly (for x86, alpha and more!)
“We dont care about IE users not being able to see the site. If someone REALLY wants to use the site, they’ll install a browser that is actually worth something.”
Well, fuck you too. Have you considered there’s people who are not so “politically-minded” as you, who are not bitching about your site not working, but actually likes to use Internet Explorer. I don’t remember pissing on your favorite browser…
I’m fed up with people who constantly talks about open software and choise, and don’t accept that some people CHOOSE to use closed software.
To me it seems that some moron deliberately choose to use some w3c standard code structure that IE, because of some error don’t know how to interpret, just to show off.
Thomas: That wasnt dave that said that, it was I.
I’m not trying to flaunt OSS, there are good closed source browsers too. Opera comes to mind. We did not create the site with the goal of not letting IE users view it. We created it, we liked it, and then IE users told us it did not work. We investigated the problem, and found out that it was because IE did not support all of CSS. So we tried to fix it, but we couldnt fix it and have it look exactly like we wanted. If someone were to make us a design which looks and functions the exact same way, and is IE-workable, then we’d switch, no doubt about it. But personally, i dont see it as a problem. I’ve managed to navigate the site on IE, not fun, but doable.
I’ve had CSS problems with IE on every site i’ve used CSS with. I realize that IE has a huge chunk of the market, and i put effort in to make sites work properly with IE, often it makes the mozilla version slightly broken. But for this site, one geared towards more tech-oriented people, we did not feel the need to bring our standards down because MS doesnt want to make a standard-compliant browser.
IE does not properly support “fixed” positions, it cannot either have a right-determined width, you can adjust the right margin but you cannot control the actual “border” of a DIV tag. There are also various other issues.
In all our developers and fans, we had only 8 individuals out of a couple hundreds who actually reported having difficulties with the site, most of them installed a minimal 8.5MB Phoenix/Firebird browser and were able to work from there. IE users for us are a very small minority and this problem will not be present in the future version of IE, since those bugs have been fixed (currently in beta-testing). I’m not degrading IE or showing off, I just believe in standards.
“8< … we had only 8 individuals out of a couple hundreds who actually reported having difficulties with the site … 8< … IE users for us are a very small minority … 8< … I’m not degrading IE or showing off, I just believe in standards.”
I’ll bet there was a small smile dancing across your lips as you wrote that though
I only wish I could get away with it. I’ve been wanting to use “position: fixed;” ever since CSS2 officially launched and have been constantly hampered by Microsoft’s continued resistance to full standards compliance (Alpha transparency in PNG’s as well but that’s another story).
Still, nice to see IE users get all stroppy when a site doesn’t work for them rather than do what most “alternative browser” users do – ie: a polite private email to the webmaster explaining the problem and asking them to try a little harder then working your way around it.
Back On-Topic – interesting project you have there!
That browser elitism is why Linux will never amount to anything on the Desktop.
Joost, VP, …
That browser elitism is why Linux will never amount to anything on the Desktop.
hmm.. i thought it’s Microsoft’s browser elitism because IE ISN’T STANDARDS COMPATIBLE ;P They think they own the world so they don’t care about standards, but times change – more and more users jump to alternative browsers (because they want or because they use computer/OS with default browser based on different engine).
So You IE people stop complaining about sites – complain to Microsoft to properly update their browser.
Once again you guys just dont get it.
The people you are snubbing your noses at are the very people you WANT to use Linux/BSD/whatever.