Post a Comment
eyeOS is amazing great work guys!
if you haven't seent he demo on the site yet it is a must.
http://www.eyeos.org/about
Edited 2007-06-07 23:05
It is a neat idea, but nobody seems to remember WebOS.
I used it back in 2000 and it was way ahead of anything I've seen 7 years later. I just found this from one of the progammers.
http://talkback.zdnet.com/5208-11423-0.html?forumID=1&threadID=2020...
Some highlights:
"The WebOS(.com) API was basically a JavaScript conversion of Sun's Swing API used in Java applications.... We even had a server communication channel running on a Java Servlet backend. We were doing AJAX two years before Jesse James Garrett wrote his famous article.
I laugh whenever I hear people talk about how AJAX and the WebOS Space is changing computing. We were there... six whole years ago we were doing what is only now begining to come around again."
Sadly though:
"In late 2004 Shervin and I made plans to release the WebOS API under an open source license, but it never materialized."
I just installed EyeOS and though it's not as fully featured yet, it's works nicely - very attractive, solid and easy to use. Being open-source and free to install on your own server, I think it'll be more useful to people than WebOS was.
Edited 2007-06-08 01:47
Sorry to be off-topic again, but I found the WebOS.com homepage from mid-2000 on archive.org. There's s tiny screenshot showing an email client and some other apps. Incredible, really.
http://web.archive.org/web/20000815233247/www.webos.com/webos/index...
I looked over their site.
I didn't see even a basic overview.
No idea what architecture this runs on.
No idea what language is their base.
No idea if this is an OS or just another linux distro
It would be nice if they at least had some basic bullet items about what the heck their OS id.
There's a link for this?
Edited 2007-06-08 05:02 UTC
As it's already mentioned, you're wrong: a compiled code will never be as speed efficient as a native one. Except for the cases which are about programmer mistakes, of course.
Maintainability? Why do you think it's any harder than elsewhere? Portability? This is a new platform. Robustness? They're maturing the new platform - it will be in time, if they can stand up for their plans.
And... here's a note, ope you won't take this personal. You've heard something somewhere and you just belived that one without understanding. Be more careful next time, and try to be less offensive - at least think before you do claims like "infintely stupid".
In the purest form you are correct, native code => compiled code in speed; However, in practice it is going to be just the opposite. First there is the fact that few coders are skilled enough, and for that matter patient enough to create large optimized programs in assembly. But lets say someone did spend the time and created an awsome O.S. in the i386 instruction set. What about other platforms, or even updates to the intel platform? Less well-written compiled code could very possibly reach higher speeds just becuase it can more fully take advantage of a platform with just a recompile.
You are right, in a big part the issue is about programmer qualities and the quality of the OS basis. The goal of the project, of course, is to provide a high-quality package at the end.
But at this point I see no difference: the same applies to other platforms. Going simple: bad programmers will produce bad code. No meter what platform they build on. On the other hand: good OS dev management and hight quality distribution building will let the talented developers to do their best in writing apps.
The thing they can't help is the small number of assembly developers out there. But a project like this definitly will increase the attractivity of the platform.
As per the last two comments:
jello, there is no reason. I worked on ORCA over a year ago, and the market for a web-based windowing environment is slim to none. The only place it would even be slightly viable is perhaps in a school, running on thin-client systems. The main benefit is instant deployment of software to client systems, as the code is constantly being-reloaded over the LAN.
But as far as providing a web-based "operating system" - this alone pisses me off - it's completely pointless.
----------
Luposian, no, it's a bunch of HTML and JS files. It's not an operating system, it's not even an operating environment. "WebOS" is just a buzzword, it doesn't provide a meaningful description of the project.
First time I saw something like that was YouOS.com, now eyeOS, and then as someone posted WebOS.
I certainly don't see the point. It's hack, on top of a hack, on top of hack, I guess Richard Gabriel was right: "Worse is Better" - http://www.dreamsongs.com/WorseIsBetter.html
Please, take a look at the source code in eyeOS - plainly most of the stuff written there is not reusable. How would one write an application for it? Where are the tools?
And what is the purpose?
As far as KolibriOS goes, there could be markets where this OS would be highly needed. Coin-Op operators, certain embedded devices, etc - it's rather small, and seeing FAR running under it, made me excited 
EyeOS or any web environment is not an OS due to the meaning of the words of Operating System. But they do provide the user with an experience that reminds them to something that they call their OSs on their PCs, but here they mean a graphical user environment actually.
And what can be the point in having such environment? Well, that's something that is avaialable from anywhere, with the same look and functionality without any extra requirement (like download, install etc.). As it requires only a browser, it will be the same on vistually all OSs, it may be a powerful and easy-to-manage software environemnt for many.



