Linked by Kroc Camen on Thu 5th Nov 2009 21:05 UTC
There's no right way to do it, only ideas that are better than others in certain situations. But if you had the opportunity to head up the design of a new OS, one to Put Things Right, one that could be radical enough to varnish out those UI/X bumps that have clung on for years, but practical enough to be used every day, what would you design? How would you handle application management? What about file types and compatibility? Where would you cherry pick the best bits from other OSes and where would you throw away tradition? I've tackled this challenge for myself and present (an unfinished idea): KrocOS (warning: HTML5 site, will display without CSS in IE/older browsers). OSnews Asks: What would make your perfect OS?
Thread beginning with comment 393461
To view parent comment, click here.
To read all comments associated with this story, please click here.
To view parent comment, click here.
To read all comments associated with this story, please click here.
RE[5]: Why was he modded down?
by sorpigal on Mon 9th Nov 2009 01:01
in reply to "RE[4]: Why was he modded down? "
Member since:2005-11-02
The evidence is staring you in the face, but you don't want to hear it. Why can't the vast majority of distros do something as simple as play DVDs or Blu-Rays without being sued? This is 2009, for chrissakes. How can you possibly call that success?
What has any of that got to do with a stable or changing kernel ABI? Oh yes, nothing. What conversation do you think we're having here?
The primary reason for this failure is the insistence of Linux zealots on having source code for anything they ship. Constantly shifting the ABIs isn't rational, from a development perspective. People claim they want "flexibility", but that's a red herring.
I find this laughable. Firstly, I have a pretty good idea why Linux is failing on the desktop and it has nothing to do with an 'insistence' on Free software. Distribution vendors routinely ship non-Free software when they have a good reason to do it... and a workable agreement with the rights holder.
Second, if shifting ABIs isn't rational why is it done? Yes, I know what you're about to say:
What these same folks really want is to create a binary incompatibility such that you would HAVE to have the source code and compile the drivers with that specific version of the kernel in order to have a prayer of working.
So now it's a conspiracy! The fact is that kernel devs do want the source code to drivers written for their kernel. The GPL requires this without any shifting of ABIs. This is, in fact, the rational argument against my argument that a stable ABI helps proprietization of the kernel. It's amazing no one has attempted it.
So, given that we get the source code anyway, why do you think there isn't a stable ABI? Could it be that there is some rational reason not to have one?
Hardware OEMs aren't going to disclose all of their secret sauce to open source developers. Just as they won't share all of their IP with Microsoft. It's not a trust or fairness issue. It's a trade secret and patent issue.
This is simply not true. They will disclose that information if that's what it takes to sell hardware. Their business wont die if they do, it may if they don't.
some stuff on opensolaris
I for one welcome our new OpenSolaris overlords. If what you say is true it should gradually take over from Linux. If it does I will owe you a caffeinated drink of your choice.
RE[6]: Why was he modded down?
by tomcat on Tue 10th Nov 2009 02:12
in reply to "RE[5]: Why was he modded down? "
Member since:2006-01-06
What has any of that got to do with a stable or changing kernel ABI? Oh yes, nothing. What conversation do you think we're having here?
Source code. FOSS zealots insist on having the source code for everything that they distribute -- drivers, media players, etc -- and having an unstable ABI is a perfect illustration of how they force that to happen. Content owners don't want to make media playback possible without DRM--and they're not going to allow the DRM-related code checked into the Linux tree. Result: Epic fail for Liux distros. They can't even play a damned DVD or Blu-Ray with a default install.
I find this laughable. Firstly, I have a pretty good idea why Linux is failing on the desktop and it has nothing to do with an 'insistence' on Free software.
See above. Tell me how that is a better experience for users. I'd love to hear your response.
Distribution vendors routinely ship non-Free software when they have a good reason to do it... and a workable agreement with the rights holder.
They ship commercial software because it includes things that people want -- like basic DVD playback, etc -- that can't be included in Linux without a license. And since the FOSS zealots have an ideological problem with DRM -- thus causing them to ignore basic & reasonable customer requirements -- the result is epic fail.
So now it's a conspiracy!
More like a confederacy of dunces.
The fact is that kernel devs do want the source code to drivers written for their kernel.
Duh. Ignoring market realities.
The GPL requires this without any shifting of ABIs.
The GPL does NOT prevent establishing a binary interface contract for the kernel ABI. Any suggestion to the contrary is ridiculous and wrong.
This is, in fact, the rational argument against my argument that a stable ABI helps proprietization of the kernel. It's amazing no one has attempted it.
So far, all that you've argued is that kernel devs like to have source code for their drivers. That's a pretty pathetic defense.
So, given that we get the source code anyway, why do you think there isn't a stable ABI? Could it be that there is some rational reason not to have one?
Continued idiocy is not a rationale for idiocy.
This is simply not true. They will disclose that information if that's what it takes to sell hardware. Their business wont die if they do, it may if they don't.
LMAO. Wrong. You guys keep prodding the hardware manufacturers to release specs about their hardware, and then you whine, wring your hands, and complain when they don't. You're like a dog chasing its tail.
I for one welcome our new OpenSolaris overlords. If what you say is true it should gradually take over from Linux. If it does I will owe you a caffeinated drink of your choice.
Apparently, you missed the part where I mentioned that Linux has had a decade headstart on OpenSolaris, and its forebear Solaris was primarily run on expensive workstation and server hardware, not widely available x86 desktop PCs. So, it's unlikely that it will ever catch up to Linux, for reasons wholly unrelated to the kernel ABI. But, by all means, feel free to erect another strawman and flog it, if it makes you feel better.
Recent Original Stories
Recent Comments
Headlines
Random Comments
Random Stories
Random OS Link
Member since:
2006-01-06
The evidence is staring you in the face, but you don't want to hear it. Why can't the vast majority of distros do something as simple as play DVDs or Blu-Rays without being sued? This is 2009, for chrissakes. How can you possibly call that success?
The primary reason for this failure is the insistence of Linux zealots on having source code for anything they ship. Constantly shifting the ABIs isn't rational, from a development perspective. People claim they want "flexibility", but that's a red herring. What these same folks really want is to create a binary incompatibility such that you would HAVE to have the source code and compile the drivers with that specific version of the kernel in order to have a prayer of working.
Hardware OEMs aren't going to disclose all of their secret sauce to open source developers. Just as they won't share all of their IP with Microsoft. It's not a trust or fairness issue. It's a trade secret and patent issue. If you give away your technological advantages to your competitors, you cease to offer differentiation -- and you die as a commercial entity. Your competitors can leverage and replicate your research without putting in the same level of investment. You know, I'm surprised that so many people in the FOSS community are either blissfully unaware of the realities of this situation -- or simply don't care. Either way, you're shooting yourselves in the foot. It's like you're afraid of success or something, even when people explain step-by-step how you can win.
As for me, I don't care, one way or another. I write commercial software for a living, and the fact of the matter is that, if you guys fail to solve this problem, it just means more business for me. So, by all means continue to ignore what I'm saying. Enrich me, please.
First, OpenSolaris hasn't been open very long. Linux has had more than a decade headstart, and being first means a lot. Some people refer to this as First Mover Advantage. It takes time to write drivers, and there has to be sufficient interest and motivation to target any OS. Second, no slight to OpenSolaris (it's a good OS), but it was derived from Sun's relatively expensive workstation and server OS codebase which didn't use horizontal market hardware. Somebody has to write those drivers. Comparing one form of failure against another is counterproductive.
Edited 2009-11-08 04:46 UTC