- Site menu:
- Latest Discussions
- Latest Newswire
- Archives
- Submit Story
- Story HOWTO
- Syndicate
- E-mail Alerts
Top Five Linux Contributor: Microsoft
|
|
|
Comments (58)
38 earlier comments in the full thread.
Quoting:That's not merely propaganda, it's a fairly clumsy attempt to revise history. As a "peace offering" it is distinctly barbed.Could also be a non-historical, non-computer Trojan horse. This definition of a Trojan horse is from http://www.thefreedictionary.com/Trojan+horse :
Quoting:Trojan horse n. 1. A subversive group or device placed within enemy ranks.With a slight spelling change, here:
Quoting:Trojan Horse n 2. a trap intended to undermine an enemy
-----
Quoting:I can only assume that the video is actually aimed at some other audience than the Linux community.OTOH, the video could conceivably be aimed at the "Linux community" to put them on notice concerning a possible Trojan horse given in the above non-historical, non-computer meaning.
Windows not a failure? Who actually LIKES to use it?
People like to use MacOS. People like using Unix and Linux. People tolerate Windows, only because other people tolerate Windows. One only has to contrast the presentation styles of Ballmer and Jobs to see what I mean.
People like to use MacOS. People like using Unix and Linux. People tolerate Windows, only because other people tolerate Windows. One only has to contrast the presentation styles of Ballmer and Jobs to see what I mean.
@fewt Microsoft has significant opportunity costs if people wanting to run linux and Windows end up running Windows over Linux. They will see Linux run well in comparison to Windows. The alternative allows Windows to slow down, make unstable, corrupt, data-analyze and intercept, etc, Linux as much as suits them for competitive advantage. This is not tin-foil hat territory. It's rather easy to have apps on your platform (and Linux would be an "app") have issues and lack integration in contrast to your "apps", which in this case would be any competing software feature Microsoft offers over Windows that a Linux virtualization would also be running.
For not recognizing this and perhaps past comments, I think many people view you as a shill, etc. Do you recognize that Microsoft has clear advantages in having customers pick to run Linux over Windows rather than vice-versa?
As for the degree of self-interest of corporations like IBM and Oracle, Microsoft competes aggressively directly against Linux. IBM and most others leverage Linux to a greater extent because they have significant income streams on hardware, services, and other areas that complement Linux. Then you have firms like Red Hat which leave all of these firms behind in terms of contributions to Linux and support of open Linux platforms. Canonical contributes at a different level than the kernel; however, as has been stated here already, Microsoft contributing drivers can certainly be seen as a negative to the Linux ecosystem (especially given what Linux has to gain potentially), regardless of the existence, generally, of a call for Linux drivers. If Microsoft were to be bringing a net negative to Linux, surely we it's possible to understand how Canonical and others who contribute less quantity of code to the kernel would still end up potentially very far ahead. The main question is, to compare Canonical and Microsoft, if Microsoft were not actively paying attention at all to Linux, would Linux be better or worse off? And we'd ask the same question of Canonical, Red Hat, etc.
For not recognizing this and perhaps past comments, I think many people view you as a shill, etc. Do you recognize that Microsoft has clear advantages in having customers pick to run Linux over Windows rather than vice-versa?
As for the degree of self-interest of corporations like IBM and Oracle, Microsoft competes aggressively directly against Linux. IBM and most others leverage Linux to a greater extent because they have significant income streams on hardware, services, and other areas that complement Linux. Then you have firms like Red Hat which leave all of these firms behind in terms of contributions to Linux and support of open Linux platforms. Canonical contributes at a different level than the kernel; however, as has been stated here already, Microsoft contributing drivers can certainly be seen as a negative to the Linux ecosystem (especially given what Linux has to gain potentially), regardless of the existence, generally, of a call for Linux drivers. If Microsoft were to be bringing a net negative to Linux, surely we it's possible to understand how Canonical and others who contribute less quantity of code to the kernel would still end up potentially very far ahead. The main question is, to compare Canonical and Microsoft, if Microsoft were not actively paying attention at all to Linux, would Linux be better or worse off? And we'd ask the same question of Canonical, Red Hat, etc.
...I have criticism of IBM and others who I believe have proprietary products and platforms they try to push onto those using Linux, attacking the value of Linux and openness. Society rewards this kind of behavior in business because we have rules of the game (including tax policy) that empower and encourage those who want to place money above most everything else. We give increasingly more privilege to those who are already wealthy and excel at trading and other business skills. At the same time, we are rather harsh on a fair number who aren't particularly aggressive and leveraged in these areas. The result being that many talented people find it hard to have their own modest business or interesting job under non-profit and free sharing conditions and instead are pulled into the sphere of influence of potentially very greedy and already powerful individuals. FOSS (for software) has made significant change on this landscape from how it was and where it might have been headed, but much more can be done, for software and for other areas as well.
Quoting:Microsoft has significant opportunity costs if people wanting to run linux and Windows end up running Windows over Linux. They will see Linux run well in comparison to Windows.
Windows and Linux are typically used to solve different business problems, and don't often conflict which is why I would argue there isn't a significant opportunity cost.
Linux and Windows both perform well in their respective functions.
Quoting:The alternative allows Windows to slow down, make unstable, corrupt, data-analyze and intercept, etc, Linux as much as suits them for competitive advantage.
This is pure speculation. In the desktop sense, you may have a point in that Windows can and sometimes will slow down and become unstable over time. On the other hand we have a completely different problem set with Linux in that often times we will find software or hardware work for a while and then no longer function the same or work at all after an upgrade.
There really is no competitive advantage. For example, in a datacenter, RHEL is more expensive than Windows.
Quoting:This is not tin-foil hat territory. It's rather easy to have apps on your platform (and Linux would be an "app") have issues and lack integration in contrast to your "apps", which in this case would be any competing software feature Microsoft offers over Windows that a Linux virtualization would also be running.
The exact same could be said of Windows as an app on Linux (KVM or Xen).
Quoting:For not recognizing this and perhaps past comments, I think many people view you as a shill, etc. Do you recognize that Microsoft has clear advantages in having customers pick to run Linux over Windows rather than vice-versa?
I don't recognize nonsense, no. If that makes me a shill, then I guess people can believe that I'm a shill. To be honest, I don't really care what other people think, it isn't my problem that the people making these baseless claims are simple minded, and ignorant. :)
fewt, fewt, fewt,...
You're spitting out "facts" like they came out of the M$ playbook...
The only "problem" that running Windoze rather than Linux in the data center solves is when you have applications that only run on Windoze. As far as the jobs that an infrastructure type of server is asked to do, file serving, database serving, print serving, web serving, etc., I can think of no reason to pick Windoze over Linux, short of a proprietary database format, or a particular proprietary communications protocol. In many instances there replacements for those, as well.
As far as losing hardware functionality when upgrading a Linux distribution, it is almost always a regression that is fixed inside a week, and almost always involves hardware that's appropriate to a desktop workstation, and inappropriate on a server. I don't know why most servers need to be updated immediately when running Linux, anyway. The majority of Linux security patches are for patching local privilege escalation threats, not to prevent outside attack. And there are things that can be done to configure around most of those to make them irrelevant. That provides the time to study the upgrade plan and evaluate it for the kind of regressions that break hardware.
You're spitting out "facts" like they came out of the M$ playbook...
The only "problem" that running Windoze rather than Linux in the data center solves is when you have applications that only run on Windoze. As far as the jobs that an infrastructure type of server is asked to do, file serving, database serving, print serving, web serving, etc., I can think of no reason to pick Windoze over Linux, short of a proprietary database format, or a particular proprietary communications protocol. In many instances there replacements for those, as well.
As far as losing hardware functionality when upgrading a Linux distribution, it is almost always a regression that is fixed inside a week, and almost always involves hardware that's appropriate to a desktop workstation, and inappropriate on a server. I don't know why most servers need to be updated immediately when running Linux, anyway. The majority of Linux security patches are for patching local privilege escalation threats, not to prevent outside attack. And there are things that can be done to configure around most of those to make them irrelevant. That provides the time to study the upgrade plan and evaluate it for the kind of regressions that break hardware.
JaseP said:
Why don't you tell us a little bit more about what makes you qualified to imply that I'm wrong when bugzillas and Launchpads everywhere seem to drastically conflict with your opinion.
Then, take the immature childish comments out of your reply, and maybe you'll be worth the time of day.
Until then..
Quoting:fewt, fewt, fewt,...
You're spitting out "facts" like they came out of the M$ playbook...
The only "problem" that running Windoze rather than Linux in the data center solves is when you have applications that only run on Windoze. As far as the jobs that an infrastructure type of server is asked to do, file serving, database serving, print serving, web serving, etc., I can think of no reason to pick Windoze over Linux, short of a proprietary database format, or a particular proprietary communications protocol. In many instances there replacements for those, as well.
As far as losing hardware functionality when upgrading a Linux distribution, it is almost always a regression that is fixed inside a week, and almost always involves hardware that's appropriate to a desktop workstation, and inappropriate on a server. I don't know why most servers need to be updated immediately when running Linux, anyway. The majority of Linux security patches are for patching local privilege escalation threats, not to prevent outside attack. And there are things that can be done to configure around most of those to make them irrelevant. That provides the time to study the upgrade plan and evaluate it for the kind of regressions that break hardware.
Why don't you tell us a little bit more about what makes you qualified to imply that I'm wrong when bugzillas and Launchpads everywhere seem to drastically conflict with your opinion.
Then, take the immature childish comments out of your reply, and maybe you'll be worth the time of day.
Until then..
Take the insults out of your reply, and maybe I will...
For the record I have used microcomputers and Unix mainframes since the early to mid 1970s, since before the IBM PC was even developed. My first use of computers was with Unix CLI commands, on an ADM-3A crt terminal, connected through an acoustic coupler modem... that well before DOS came out as a poor-man's CP/M "clone." I'm no Linux script kiddie in my teens, twenties or even thirties...
About the only thing I have done nearly as long as computers is Martial Arts,... and I am the senior ranking practitioner and instructor of my particular martial arts style in my geographic region (and it's not an obscure style, it was Chuck Norris's first style).
Your "experience" doesn't meet up with real world practice. What are these particular Bugzillas and PPAs of which you speak??? I would like to see if they are obscure hardware bug reports that shills use to spout meaningless statistics, or whether they have any basis as a serious real world complaint.
For the record I have used microcomputers and Unix mainframes since the early to mid 1970s, since before the IBM PC was even developed. My first use of computers was with Unix CLI commands, on an ADM-3A crt terminal, connected through an acoustic coupler modem... that well before DOS came out as a poor-man's CP/M "clone." I'm no Linux script kiddie in my teens, twenties or even thirties...
About the only thing I have done nearly as long as computers is Martial Arts,... and I am the senior ranking practitioner and instructor of my particular martial arts style in my geographic region (and it's not an obscure style, it was Chuck Norris's first style).
Your "experience" doesn't meet up with real world practice. What are these particular Bugzillas and PPAs of which you speak??? I would like to see if they are obscure hardware bug reports that shills use to spout meaningless statistics, or whether they have any basis as a serious real world complaint.
Quoting:Take the insults out of your reply, and maybe I will...
Sorry, you first.
Quoting:For the record I have used microcomputers and Unix mainframes since the early to mid 1970s, since before the IBM PC was even developed. My first use of computers was with Unix CLI commands, on an ADM-3A crt terminal, connected through an acoustic coupler modem... that well before DOS came out as a poor-man's CP/M "clone." I'm no Linux script kiddie in my teens, twenties or even thirties...
No no, I want a resume.
My experience and credentials are easy to find, but you come up empty. Experience talks, bullsh#t walks.
Quoting:About the only thing I have done nearly as long as computers is Martial Arts,... and I am the senior ranking practitioner and instructor of my particular martial arts style in my geographic region (and it's not an obscure style, it was Chuck Norris's first style).
Good for you, but unrelated to the thread.
Quoting:Your "experience" doesn't meet up with real world practice. What are these particular Bugzillas and PPAs of which you speak??? I would like to see if they are obscure hardware bug reports that shills use to spout meaningless statistics, or whether they have any basis as a serious real world complaint
I didn't say PPAs. There's the silly baseless S word again.
The problems are really simple to find if you try.
For example:
[url=https://bugs.launchpad.net/ubuntu/natty/ source/linux/ bug/712082]https://bugs.launchpad.net/ubuntu/natty/ source/linux/ bug/7...[/url] [url=https://bugs.launchpad.net/ubuntu/natty/ source/linux/ bug/755066]https://bugs.launchpad.net/ubuntu/natty/ source/linux/ bug/7...[/url] [url=https://bugs.launchpad.net/ubuntu/natty/ source/linux/ bug/751689]https://bugs.launchpad.net/ubuntu/natty/ source/linux/ bug/7...[/url] [url=https://bugs.launchpad.net/ubuntu/natty/ source/linux/ bug/774563]https://bugs.launchpad.net/ubuntu/natty/ source/linux/ bug/7...[/url]
Those are just some recent bugs, from Ubuntu. There are more, many many more.
My experience paints a very real world picture. Take off your blinders, and put away your Linux Youth t-shirt.
From google: Fedora Kernel Panic - 735,000 results, Ubuntu Kernel Panic - 1,310,000 results.
You can pretend that I'm a shill, but who was it that spent 12 hours yesterday building Kernel 3.0.0, and bringing it and it's associated dependencies into Fuduntu?
Oh right, that was me.
I'll let you guess how much I was paid for the time spent doing it (hint, it's a number less than 1)....
>> Windows and Linux are typically used to solve different business problems, and don't often conflict which is why I would argue there isn't a significant opportunity cost.
There is definitely overlap and this probably is increasing (eg, with Microsoft discovering the "open source" on Windows religion and striking deals with numerous FOSS projects and with hw vendors supporting Linux more and more).
There is also the impression of which system is better. That one that runs on top gets the short end of the stick. People might virtualize to save on hardware for example and have to decide which to have right above the hardware.
>> On the other hand we have a completely different problem set with Linux in that often times we will find software or hardware work for a while and then no longer function the same or work at all after an upgrade.
Not sure what you mean since you are contrasting with Windows. Do you mean upgrade of.. OS? app? hw drivers?
>> There really is no competitive advantage.
No competitive advantage for running native and managing the other platform through virtualization?
>> This is pure speculation.
Each layer of software you add extra has a very good chance of degrading performance, stability, and fidelity. This is even more so when (a) the software are managed by different entities, (b) especially if these compete, (c) one is closed source, and (d) there is much on the line for the firm in consideration (Microsoft sw as host to Linux).
>> The exact same could be said of Windows as an app on Linux (KVM or Xen).
Absolutely. This composes part of the opportunity cost. ..why the Linux drivers are a gain to Microsoft.
There is definitely overlap and this probably is increasing (eg, with Microsoft discovering the "open source" on Windows religion and striking deals with numerous FOSS projects and with hw vendors supporting Linux more and more).
There is also the impression of which system is better. That one that runs on top gets the short end of the stick. People might virtualize to save on hardware for example and have to decide which to have right above the hardware.
>> On the other hand we have a completely different problem set with Linux in that often times we will find software or hardware work for a while and then no longer function the same or work at all after an upgrade.
Not sure what you mean since you are contrasting with Windows. Do you mean upgrade of.. OS? app? hw drivers?
>> There really is no competitive advantage.
No competitive advantage for running native and managing the other platform through virtualization?
>> This is pure speculation.
Each layer of software you add extra has a very good chance of degrading performance, stability, and fidelity. This is even more so when (a) the software are managed by different entities, (b) especially if these compete, (c) one is closed source, and (d) there is much on the line for the firm in consideration (Microsoft sw as host to Linux).
>> The exact same could be said of Windows as an app on Linux (KVM or Xen).
Absolutely. This composes part of the opportunity cost. ..why the Linux drivers are a gain to Microsoft.
Quoting:There is definitely overlap and this probably is increasing (eg, with Microsoft discovering the "open source" on Windows religion and striking deals with numerous FOSS projects and with hw vendors supporting Linux more and more).
I agree with the increase in overlap as Microsoft discovers and "embraces" Open Source (however you decide to read that).
I don't think Microsoft is very successful in creating deals to keep vendors off of the Linux bandwagon anymore, at least it isn't as bad as it used to be. More and more vendors are releasing drivers daily which is fantastic, if it could only have happened 10 years ago...
Quoting:There is also the impression of which system is better. That one that runs on top gets the short end of the stick. People might virtualize to save on hardware for example and have to decide which to have right above the hardware.
I don't think you can really count a Hyper-V license like a Windows license, even though to be fair it is (a service that runs on Windows).
In my mind, people virtualize Windows and Linux on a hypervisor which could be Hyper-V, Xen, KVM, or VMWare. The most important decision in choosing which hypervisor to use really boils down to supportability.
Can your existing resources support VMWare, Xen, KVM, or MS Hyper-V.
Today, it's arguable that most can support VMWare, with few running KVM or Xen and even fewer running Hyper-V.
It isn't really a matter of which is "better', but one of which "solves the problem we have best".
Quoting:Not sure what you mean since you are contrasting with Windows. Do you mean upgrade of.. OS? app? hw drivers?
OS which normally includes app and hw drivers. Specifically I was referring to hardware drivers in the kernel which sometimes see regressions that can cause problems like kernel panics.
Also true of Windows of course, but less often (which is a sharp contrast to historical Windows).
Quoting:No competitive advantage for running native and managing the other platform through virtualization?
Not really, when in nearly every case you would run both platforms on top of that hypervisor.
Remember, a Windows Hyper-V instance wouldn't normally also do anything else.
You wouldn't see a Hyper-V domain controller / Exchange server. It is a platform solution like VSphere.
Quoting:Each layer of software you add extra has a very good chance of degrading performance, stability, and fidelity. This is even more so when (a) the software are managed by different entities, (b) especially if these compete, (c) one is closed source, and (d) there is much on the line for the firm in consideration (Microsoft sw as host to Linux).
I agree with everything except the implication that closed source reduces stability and performance, because there isn't any evidence to prove this to be true.
Quoting:Absolutely. This composes part of the opportunity cost. ..why the Linux drivers are a gain to Microsoft.
Fair enough.
Why on Earth are you trying to replace a working kernel on your distro with one that is less than 96 hours old??? I can appreciate trying to be cutting edge, but let the Red Hats and Canonicals of the world work out the kinks. By the way,... Looked at the Fuduntu screen shots, and it looks remarkably like the modified Ubuntu that I use. Looks decent,... Although I prefer to hide the panel, and relocate it to the bottom left. It frees screen real estate. Oh, and I prefer Cairo-dock to Avant (or whatever that dock is). Gets more "Ooos" and "Ahhs."
Oh, and Googling "Kernel Panic" is not how you get serious stats on bugs. I bet I'd get worse stats by googling Microsoft and Virus...
Oh, and I wasn't calling you a schill, I was just implying that you sounded like one... As in you're "smoking" too much of what M$ brought over.
Oh, and Googling "Kernel Panic" is not how you get serious stats on bugs. I bet I'd get worse stats by googling Microsoft and Virus...
Oh, and I wasn't calling you a schill, I was just implying that you sounded like one... As in you're "smoking" too much of what M$ brought over.
Quoting:Why on Earth are you trying to replace a working kernel on your distro with one that is less than 96 hours old???
I'm not replacing my stable kernel, but if it doesn't get tested it can never become stable. The work was done to make it ready for testing. This morning it was released to testing.
Once it's sufficiently tested we'll move it to stable for everyone else.
Quoting:I can appreciate trying to be cutting edge, but let the Red Hats and Canonicals of the world work out the kinks.
No thank you. I have a great amount of respect for RedHat, but one issue I find consistently is that problems in released kernels remain in released kernels across distribution subreleases (Fedora).
RedHat backports a lot security fixes, (and hardware support in the enterprise product) but many problems remain through the lifecycle of a distribution release.
I trust Linus, and I have found the best way to resolve regressions in the kernel is to follow the latest (stable) kernel.
I had more problems with Canonical's "release" kernels than any other kernel. They make it painfully obvious that they lack kernel maintainers (evidenced by their lack of upstream contribution).
Quoting:By the way,... Looked at the Fuduntu screen shots, and it looks remarkably like the modified Ubuntu that I use. Looks decent,... Although I prefer to hide the panel, and relocate it to the bottom left. It frees screen real estate. Oh, and I prefer Cairo-dock to Avant (or whatever that dock is). Gets more "Ooos" and "Ahhs."
Glad you like it.
Quoting:Oh, and Googling "Kernel Panic" is not how you get serious stats on bugs. I bet I'd get worse stats by googling Microsoft and Virus...
No doubt, but it does make the case that it's not all blue skies and pink ponies which was my point.
Quoting:Oh, and I wasn't calling you a schill, I was just implying that you sounded like one... As in you're "smoking" too much of what M$ brought over.
It certainly sounded like it, my apologies for misinterpreting your wording and reading it with a hostile tone. I'm not smoking anything though, Linux has been my primary desktop since the mid 1990s, and I certainly don't drink Microsoft kool-aid. That said though, I don't find fault in everything they do either. I give credit where it's due.
Quoting:You can pretend that I'm a shill, but who was it that spent 12 hours yesterday building Kernel 3.0.0, and bringing it and it's associated dependencies into Fuduntu?
Well, in this case you are not a shill, but rather a mole and a Trojan Horse. We now know that Florian Mueller is not alone and has good company.
Quoting:FUDuntu
With a name that starts with FUD and a Distro uglier than its name, there is nothing that could give or restore credibility.
Clearly, some background information is in order.
1st, in reverse order, is info concerning the naming of Wyatt's distro.
~~~~
2nd is identifying Florian Mueller.
Florian Mueller is the author of 'FOSS Patents'; a blog that "covers software patent news and issues with a particular focus on the competitiveness of Free and Open Source Software (FOSS)"
Source: http://fosspatents.blogspot.com/
~~~~ (Both near-identical variants of "Trojan Horse" are defined above.) ~~~~
3rd is the definition of a mole.
~~~~
4th is the definition of a shill.
Source: http://www.urbandictionary.com/define.php?term=shill
~~~
From the same Urban Dictionary, there is yet another appropriate definition; that of the troll.
~~~~
1st, in reverse order, is info concerning the naming of Wyatt's distro.
Quoting:Fuduntu - Behind the name Fuduntu Pronounciation: [fuhduhntoo]Source: http://www.fuduntu.org/name.html
Fuduntu is a fusion of the names Fedora and Ubuntu.
Fedora - F and D from Fedora Ubuntu - U and UNTU from Ubuntu
The name is a pun, intended to be fun, and funny, while implying that the distribution fits in-between Fedora and Ubuntu. The fun uses of the name mean that it is successful, so go ahead, have fun with it!
~~~~
2nd is identifying Florian Mueller.
Florian Mueller is the author of 'FOSS Patents'; a blog that "covers software patent news and issues with a particular focus on the competitiveness of Free and Open Source Software (FOSS)"
Source: http://fosspatents.blogspot.com/
~~~~ (Both near-identical variants of "Trojan Horse" are defined above.) ~~~~
3rd is the definition of a mole.
Quoting:mole n.Source: http://www.thefreedictionary.com/mole
3. A spy who operates from within an organization, especially a double agent operating against his or her own government from within its intelligence establishment.
~~~~
4th is the definition of a shill.
Quoting:1. shill
A person engaged in covert advertising. The shill attempts to spread buzz by personally endorsing the product in public forums with the pretense of sincerity, when in fact he is being paid for his services.
Source: http://www.urbandictionary.com/define.php?term=shill
~~~
From the same Urban Dictionary, there is yet another appropriate definition; that of the troll.
Quoting: 1. trollSource: http://www.urbandictionary.com/define.php?term=troll
One who posts a deliberately provocative message to a newsgroup or message board with the intention of causing maximum disruption and argument
~~~~
Quoting: No thank you. ... I trust Linus, and I have found the best way to resolve regressions in the kernel is to follow the latest (stable) kernel.
Too much G__ D___ work, if you ask me... Recompiling the kernel is the very reason I don't have a 100% functional Viliv S5 UMPC, ... yet. But, if that's how you roll, more power to you.
On a related note, ... How is Fuduntu with Marvel SD8686 SDIO interfaced Wifi chipsets, ps/2 port interfaced touch screens and the Intel GMA500 graphics chipset???
Quoting:Too much G__ D___ work, if you ask me... Recompiling the kernel is the very reason I don't have a 100% functional Viliv S5 UMPC, ... yet. But, if that's how you roll, more power to you.
Even more G__ D___ work when you distribute one. lol.
Quoting:On a related note, ... How is Fuduntu with Marvel SD8686 SDIO interfaced Wifi chipsets, ps/2 port interfaced touch screens and the Intel GMA500 graphics chipset???
I don't know anything about the SD8686, so YMMV but your GMA500 is not currently supported. I believe you will have OK 2D support, but on my list of things to do is to backport xorg from Fedora 15 or rawhide at which time we will be able to provide a working Poulsbo driver.
I tried bringing the newer Intel driver down and building out Poulsbo support but it was pretty ugly so I pulled it until I can build all of Xorg to correct the issues back porting the driver causes.
>> In my mind, people virtualize Windows and Linux on a hypervisor
But if that is Windows proprietary code, then that is Windows proprietary code, the impression you, I, or customer X has in their mind notwithstanding.
Microsoft is not engineering a black box towards their own OS. I've heard of no such verifiable firewall as concerns Hyper-V (not that I've really checked). Microsoft has always touted the benefits of integration (eg, you don't get a browser, you get an extension of the OS). We don't have to assume or speculate.
If you want clear separation, then you get that in writing from Microsoft (and the accompanying license to reverse engineer as necessary) or you take the safer chance by going with a third party for the virt host (eg, open source version of Xen or even vmware).
>> Remember, a Windows Hyper-V instance wouldn't normally also do anything else.
I actually don't remember (never got the source code), nor know about the "abnormal" conditions.
I don't know for example (and certainly suspect) that a cooperating Windows on top can get extra privileges and access that any other OS would not be able to achieve.
If you have source code that proves otherwise, I might consider looking at it. And it isn't difficult to write software that favors any "app" you run and know intimately.
Without source code we can all just speculate. This is why I try to avoid as much proprietary code as I can. I trust independent third parties and myself with source code on hand, not vendor vague marketing brochures and then hostile EULAs.
>> I agree with everything except the implication that closed source reduces stability and performance
Well, in terms of probability, the size of the door to problems increases if the software must be treated as black boxes best case v. if you have the extra option to study the internal details. All extra information helps avoid problems and resolve odd bugs. This is one reason why open source exists as open and why mega proprietary software houses exist (where lots and lots of software functions are aggregated under one roof, with source code kept closed to outside world largely just because of this help that would otherwise be afforded to the competition).
>> Fair enough.
So that is an important implicit point that goes hand in hand with what I have tried to cover on this thread, that, not only has the door been opened for superior integration when you can control the virtualization software in addition to your own OS software, but there is an opportunity cost of not having that integration potential while the competition does.
But if that is Windows proprietary code, then that is Windows proprietary code, the impression you, I, or customer X has in their mind notwithstanding.
Microsoft is not engineering a black box towards their own OS. I've heard of no such verifiable firewall as concerns Hyper-V (not that I've really checked). Microsoft has always touted the benefits of integration (eg, you don't get a browser, you get an extension of the OS). We don't have to assume or speculate.
If you want clear separation, then you get that in writing from Microsoft (and the accompanying license to reverse engineer as necessary) or you take the safer chance by going with a third party for the virt host (eg, open source version of Xen or even vmware).
>> Remember, a Windows Hyper-V instance wouldn't normally also do anything else.
I actually don't remember (never got the source code), nor know about the "abnormal" conditions.
I don't know for example (and certainly suspect) that a cooperating Windows on top can get extra privileges and access that any other OS would not be able to achieve.
If you have source code that proves otherwise, I might consider looking at it. And it isn't difficult to write software that favors any "app" you run and know intimately.
Without source code we can all just speculate. This is why I try to avoid as much proprietary code as I can. I trust independent third parties and myself with source code on hand, not vendor vague marketing brochures and then hostile EULAs.
>> I agree with everything except the implication that closed source reduces stability and performance
Well, in terms of probability, the size of the door to problems increases if the software must be treated as black boxes best case v. if you have the extra option to study the internal details. All extra information helps avoid problems and resolve odd bugs. This is one reason why open source exists as open and why mega proprietary software houses exist (where lots and lots of software functions are aggregated under one roof, with source code kept closed to outside world largely just because of this help that would otherwise be afforded to the competition).
>> Fair enough.
So that is an important implicit point that goes hand in hand with what I have tried to cover on this thread, that, not only has the door been opened for superior integration when you can control the virtualization software in addition to your own OS software, but there is an opportunity cost of not having that integration potential while the competition does.
Quoting: I don't know anything about the SD8686, so YMMV but your GMA500 is not currently supported. I believe you will have OK 2D support, but on my list of things to do is to backport xorg from Fedora 15 or rawhide at which time we will be able to provide a working Poulsbo driver.
Thanks. The SD8686 is supported in the kernel, but there are regressions from patches made to the SDIO code for it that break support without patching. The chipset is used on a variety of handhelds for its low power wifi support. Some of those handhelds, like the now defunct Viliv models are x86 machines (Intel Atom processors). Support for that and Poulsbo would elevate Fuduntu to "must have" status on these machines. The other issue is touch screen support for ps/2 port connected touch screens. The touch screen issue has to do with udev rules for the ps/2 ports which ends up giving random event numbers to the devices, such that the drivers cannot consistently find or work with them...
Post a comment
A free account is needed to post. Write your comment here first: your text is kept in this browser while you sign up or log in, and is waiting for you when you come back.
Your message:Markdown: **bold**, *italic*, [text](https://link), - lists, > quotes, $x^2$ for math. Formatting help
Sign up and post (it is free) or Log in and post
I can only assume that the video is actually aimed at some other audience than the Linux community.