Linux on the Desktop? A long way off...



  • @morbiuswilters said:

    @bnt said:

    In English, please.

    Have I been using any other language?

     

    @bnt said:

    So, why does saying M$ make me a douchebag?

    Because it's a lame Slashdot meme. 

    Okay, now explain it again, but this time in Swahili.



  •  you keep defending microsoft and saying it's the hardware vendor's problem there isn't drivers.

     

    Maybe a lot of those hardware manufacturer's don't want to pay microsoft the ransom money to be allowed to write drivers for vista that can be loaded.  Maybe they don't think that it is a very good business decision to pay into such demands because microsoft will just keep raising the ransom demand every iteration of windows.



  • @bnt said:

    In English, please. 
     

    Ha!



  • @Kazan said:

     you keep defending microsoft and saying it's the hardware vendor's problem there isn't drivers.
     

    Who does? Who are you replying to?
    @Kazan said:

     

    Maybe a lot of those hardware manufacturer's don't want to pay microsoft the ransom money to be allowed to write drivers for vista that can be loaded. 


    FUD.

    Try and get a clue what you are talking about please.

    @Kazan said:

    Maybe they don't think that it is a very good business decision to pay into such demands because microsoft will just keep raising the ransom demand every iteration of windows.

    Really? Funny, you would think they would be FLOCKING to Linux then!



  • @Kazan said:

     you keep defending microsoft and saying it's the hardware vendor's problem there isn't drivers.

     

    Maybe a lot of those hardware manufacturer's don't want to pay microsoft the ransom money to be allowed to write drivers for vista that can be loaded.  Maybe they don't think that it is a very good business decision to pay into such demands because microsoft will just keep raising the ransom demand every iteration of windows.

    Well then perhaps they should develop for Linux and it's whopping 7% market share. That sounds like a good business plan.



  • @bnt said:

    I bet you're seeing the low post count and thinking "n00b".
     

    Not really, I don't pay attention to post count like you apparently do. 

    I see your posts and think 'retard' though.



  • @morbiuswilters said:

    Because it's a lame Slashdot meme. 

     

    Ah - thank you. I wouldn't know that, since I don't hang out there. I saw it years ago, somewhere else, and it stuck in my head. There's a reason why assumption always starts with an ass.

    This is off-topic and boring.



  • @bstorer said:

    Well then perhaps they should develop for Linux and it's whopping 7% market share. That sounds like a good business plan.
     

    Protecting users who have been scared by the FUD these /. trolls spread everywhere is a better business plan if you ask me.

    Anyone who complains about driver signing clearly doesn't understand it all.



  • @bnt said:

    I saw it years ago, somewhere else, and it stuck in my head.
     

    Is that your "personal, non-technical reason"?



  • @MasterPlanSoftware said:

    @bnt said:

    I saw it years ago, somewhere else, and it stuck in my head.
     

    Is that your "personal, non-technical reason"?

    I u$e "M$", too, but that'$ becau$e my key between 'a' and 'd' on the keyboard i$ broken.



  • @bstorer said:

    it's whopping 7% market share
     

    We passed the 5 percent? WOHOOOO! We need FIREWORKS!

    *yells a bit more*

    What's that you say? Oh, oh, right!

    We need CHEM KITS! We'll make our own fireworks! GPL'd!

     

    Seriously though, wouldn't it be cool to have the entire GPL 3 license written in fire in the sky?  Then, you can say that God just licensed the Universe under GPL and we all get a free ride to Saturn, provided that we build ourselves a space shuttle.

     



  • @archivator said:

    @bstorer said:

    it's whopping 7% market share
     

    We passed the 5 percent? WOHOOOO! We need FIREWORKS!

    Probably not. I picked the highest figure a quick Google search turned up.



  • @Kazan said:

    Maybe a lot of those hardware manufacturer's don't want to pay microsoft the ransom money to be allowed to write drivers for vista that can be loaded.  Maybe they don't think that it is a very good business decision to pay into such demands because microsoft will just keep raising the ransom demand every iteration of windows.
    I find it interesting that several free and open-source projects can get signed Vista drivers, while some hardware manufacturers still can't.



  • @bstorer said:

    Well then perhaps they should develop for Linux and it's whopping 7% market share. That sounds like a good business plan.

     

    Of course they should not develop for Linux only, but IMO it makes sense to get 7% of the market that the competition can't get. For example, Epson all-in-one printers like the DX4050 work well with Linux (both printing and scanning), while comparable devices from Canon do not. If other vendors (HP, Lexmark etc.) ignore the Linux market too, it means Epson gets 7% of the market for sure, while it has to fight for the remaining 93% against several competitors.

    BTW, offering enough information so the "community" can write their own drivers is enough in many cases.



  • @bstorer said:

    @archivator said:

    @bstorer said:

    it's whopping 7% market share
     

    We passed the 5 percent? WOHOOOO! We need FIREWORKS!

    Probably not. I picked the highest figure a quick Google search turned up.

    That is the most absurd number imaginable.  Linux has less than 0.5% penetration on the desktop.  Apple is around 6-8%, for Christ's sake and you know Linux is nowhere near that. 



  • @ammoQ said:

    Of course they should not develop for Linux only, but IMO it makes sense to get 7% of the market that the competition can't get. For example, Epson all-in-one printers like the DX4050 work well with Linux (both printing and scanning), while comparable devices from Canon do not. If other vendors (HP, Lexmark etc.) ignore the Linux market too, it means Epson gets 7% of the market for sure, while it has to fight for the remaining 93% against several competitors.
     

    Obviously this is a business decision, and should be based on their perceived ROI.

    I cannot speak for every company and say this is worth it for everyone.

    @ammoQ said:

    BTW, offering enough information so the "community" can write their own drivers is enough in many cases.

    Sure, it works for many people, but they don't have to. And that is just a fact everyone has to accept.



  • @morbiuswilters said:

    That is the most absurd number imaginable.  Linux has less than 0.5% penetration on the desktop.  Apple is around 6-8%, for Christ's sake and you know Linux is nowhere near that. 

     

    http://www.w3counter.com/globalstats.php claims something in the 2% range.



  • @MasterPlanSoftware said:

    @Kazan said:

     you keep defending microsoft and saying it's the hardware vendor's problem there isn't drivers.
     

    Who does? Who are you replying to?


     considering your reply I would say you knew exactly what troll I was speaking to.

     

     @MasterPlanSoftware said:

    @Kazan said:

     

    Maybe a lot of those hardware manufacturer's don't want to pay microsoft the ransom money to be allowed to write drivers for vista that can be loaded. 


    FUD.

    Try and get a clue what you are talking about please.

     

    That term doesn't mean what it thinks you mean.  

    Can you defend their practice of making companies fork out money to be allowed to write drivers?

     

    http://en.wikipedia.org/wiki/Fear,_uncertainty_and_doubt

    Pointing out a SPECIFIC and concerete issue cannot qualify as FUD by any reasonable definition.  It is a REAL issue that microsoft requires companies to buy very expensive licenses to write drivers for windows, which denies market entry to small developers and non-comercial developers.  Real applications such as ATITray have been harmed by this.  A company wrote a driver that would allow you to chain load other drivers with it so you could use your software - it took a knowledgeable user to use, would be nearly impossible to stealth install, etc.  Fortunately for this company they bought a second certification signature for this driver - because microsoft labeled it malware, revoked the cert, and put it's signature into windows defender.

     If you would care to explain how specific examples creating clear problems is FUD then be my guest.   However know that you are blatantly misusing the term to change the subject away from the actual point mentioned because you cannot criticize the point.

     

    Stop trying to change the subject and address the criticism

     

     

     @MasterPlanSoftware said:

    @Kazan said:

    Maybe they don't think that it is a very good business decision to pay into such demands because microsoft will just keep raising the ransom demand every iteration of windows.

    Really? Funny, you would think they would be FLOCKING to Linux then!

     

    wow because obviously i mentioned linux AT ALL in my post.   You know it should be possible to post a criticism of windows without this become a windows vs linux argument - especially not where I post salient points and you just try to dismiss them as FUD and then make some attack on linux that isn't related to my point in the slightest.

     



  •  @ender said:

    @Kazan said:
    Maybe a lot of those hardware manufacturer's don't want to pay microsoft the ransom money to be allowed to write drivers for vista that can be loaded.  Maybe they don't think that it is a very good business decision to pay into such demands because microsoft will just keep raising the ransom demand every iteration of windows.
    I find it interesting that several free and open-source projects can get signed Vista drivers, while some hardware manufacturers still can't.

     

    as do i.. which makes me think it's more of a policy decision for those companies than a financial one (well.. a policy one based on financial)

     

    but creates a barrier to entry, not all small projects - closed or open source - can raise the money



  • @ammoQ said:

    @bstorer said:

    Well then perhaps they should develop for Linux and it's whopping 7% market share. That sounds like a good business plan.

     

    Of course they should not develop for Linux only, but IMO it makes sense to get 7% of the market that the competition can't get. For example, Epson all-in-one printers like the DX4050 work well with Linux (both printing and scanning), while comparable devices from Canon do not. If other vendors (HP, Lexmark etc.) ignore the Linux market too, it means Epson gets 7% of the market for sure, while it has to fight for the remaining 93% against several competitors.

    BTW, offering enough information so the "community" can write their own drivers is enough in many cases.

     

     

    yup.. Hauppauge has a long reputation for providing all the information needed to get their gear working in linux.  they have so many people purchase their hardware to use with mythtv that they hired on someone closely related to mythtv to maintain the linux drivers.

     

     



  • @MasterPlanSoftware said:

    @ammoQ said:

    Of course they should not develop for Linux only, but IMO it makes sense to get 7% of the market that the competition can't get. (..)
     

    Obviously this is a business decision, and should be based on their perceived ROI.

    I cannot speak for every company and say this is worth it for everyone.

    @ammoQ said:

    BTW, offering enough information so the "community" can write their own drivers is enough in many cases.

    Sure, it works for many people, but they don't have to. And that is just a fact everyone has to accept.

     

    Personally, as a Linux user, I'm very satisfied with the current level of hardware support. Of course I have to be a bit careful when purchasing desktop hardware (no hw vendor can ignore the Linux server market), but it's not difficult to find something that works in every category. In theory, having invested in the "wrong" hardware might be a barrier for some users, but I don't expect too many people to considering an OS switch anyway. If Linux ever gets past the 10% mark, it will be because of pre-installed devices like the EeePC.



  • @Kazan said:

    as do i.. which makes me think it's more of a policy decision for those companies than a financial one (well.. a policy one based on financial)

     

    but creates a barrier to entry, not all small projects - closed or open source - can raise the money


    I don't know about the rest of you, but I make it a habit to buy all of my hardware from companies so small and poor that they can't even raise as much money as an open source project.  That's why I'm opposed to the added security of signed drivers!


  • @Kazan said:

    troll
     

    I don't think that word means what you think it does.

    @Kazan said:

    Can you defend their practice of making companies fork out money to be allowed to write drivers?

    Yes. I already have. I know I will never satisfy your criticism and I don't care, your opinion of MS's practices means next to nothing to me or anyone else here.

    @Kazan said:

    Pointing out a SPECIFIC and concerete issue cannot qualify as FUD by any reasonable definition. 

    It is when you try and turn it into and issue by not giving the whole story and trying to make a problem where there isn't one.

    @Kazan said:

    It is a REAL issue that microsoft requires companies to buy very expensive licenses to write drivers for windows, which denies market entry to small developers and non-comercial developers.  Real applications such as ATITray have been harmed by this.  A company wrote a driver that would allow you to chain load other drivers with it so you could use your software - it took a knowledgeable user to use, would be nearly impossible to stealth install, etc.  Fortunately for this company they bought a second certification signature for this driver - because microsoft labeled it malware, revoked the cert, and put it's signature into windows defender.

    Citation needed here. FOSS applications are obtaining them. There is no conspiracy beyond your /. FUD. We have seen it all before, it is getting old and boring.

     

    @Kazan said:

    wow because obviously i mentioned linux AT ALL in my post.

    You don't have to. We can all see where it is going. You seem to think you are something special, but your type is a dime a dozen. You are one of the most boring of that type too. 



  • @morbiuswilters said:

    @Kazan said:

    as do i.. which makes me think it's more of a policy decision for those companies than a financial one (well.. a policy one based on financial)

     

    but creates a barrier to entry, not all small projects - closed or open source - can raise the money


    I don't know about the rest of you, but I make it a habit to buy all of my hardware from companies so small and poor that they can't even raise as much money as an open source project.  That's why I'm opposed to the added security of signed drivers!
     

     

    troll much?

     A) I never said signed drivers were bad, i said REQUIRING them is.  I should be able to install them and tell windows "yes shut the fuck up i know what i'm doing"

    B)  Do you understand the definition of a policy decision?

    C) do you understand that that primary non-comercial products  such as ATI Tray are what is harmed by this

    D) Do you ever stop trolling for 20 seconds and engage your brain?



  • @morbiuswilters said:

    I don't know about the rest of you, but I make it a habit to buy all of my hardware from companies so small and poor that they can't even raise as much money as an open source project.  That's why I'm opposed to the added security of signed drivers!
     

    Right, I can't even bring myself to argue with this jackass. He makes two conflicting arguments, backs them up with no fact, and doesn't understand a single thing he is talking about.



  •  @MasterPlanSoftware said:

    @Kazan said:
    Can you defend their practice of making companies fork out money to be allowed to write drivers?

    Yes. I already have. I know I will never satisfy your criticism and I don't care, your opinion of MS's practices means next to nothing to me or anyone else here.

    where exactly is this defense?

     

     @MasterPlanSoftware said:

    @Kazan said:

    Pointing out a SPECIFIC and concerete issue cannot qualify as FUD by any reasonable definition. 

    It is when you try and turn it into and issue by not giving the whole story and trying to make a problem where there isn't one.

    exactly what part of the story did i leave out?

     

     @MasterPlanSoftware said:

     @Kazan said:

    It is a REAL issue that microsoft requires companies to buy very expensive licenses to write drivers for windows, which denies market entry to small developers and non-comercial developers.  Real applications such as ATITray have been harmed by this.  A company wrote a driver that would allow you to chain load other drivers with it so you could use your software - it took a knowledgeable user to use, would be nearly impossible to stealth install, etc.  Fortunately for this company they bought a second certification signature for this driver - because microsoft labeled it malware, revoked the cert, and put it's signature into windows defender.

    Citation needed here. FOSS applications are obtaining them. There is no conspiracy beyond your /. FUD. We have seen it all before, it is getting old and boring.

     

    http://arstechnica.com/journals/microsoft.ars/2007/08/07/microsoft-revokes-companys-digital-certificate-prevents-unsigned-driver-loading?rel

    http://www.linchpinlabs.com/resources/atsiv/usage-design.htm

    didn't expect me to have a citation did you troll?

     throw around the term FUD all you want - it's NOT fud when it's a concrete solid issue.  Just because some OSS projects can raise the funds required doesn't mean it's not a barrier to entry.  You seem to not understand what that term means, take a basic college intro to economics class.

     

     @MasterPlanSoftware said:

     


    @Kazan said:

    wow because obviously i mentioned linux AT ALL in my post.

    You don't have to. We can all see where it is going. You seem to think you are something special, but your type is a dime a dozen. You are one of the most boring of that type too. 

    I didn't mention linux, I didn't have any intention to either.  Perhaps when you stop making assumptions you won't be such an ass.  You're pathetic, you're like a scientologist defending their nutjobbery.



  • @morbiuswilters said:

    @bstorer said:

    @archivator said:

    @bstorer said:

    it's whopping 7% market share
     

    We passed the 5 percent? WOHOOOO! We need FIREWORKS!

    Probably not. I picked the highest figure a quick Google search turned up.

    That is the most absurd number imaginable.  Linux has less than 0.5% penetration on the desktop.  Apple is around 6-8%, for Christ's sake and you know Linux is nowhere near that. 

    I think the number was combined server + desktop market share. But that's still way too high, as most estimates give Linux about 12% server market share, and I sincerely doubt that servers outnumber desktops.



  • @MasterPlanSoftware said:

    @Kazan said:
    It is a REAL issue that microsoft requires companies to buy very expensive licenses to write drivers for windows, which denies market entry to small developers and non-comercial developers.  Real applications such as ATITray have been harmed by this.  A company wrote a driver that would allow you to chain load other drivers with it so you could use your software - it took a knowledgeable user to use, would be nearly impossible to stealth install, etc.  Fortunately for this company they bought a second certification signature for this driver - because microsoft labeled it malware, revoked the cert, and put it's signature into windows defender.
    Citation needed here. FOSS applications are obtaining them. There is no conspiracy beyond your /. FUD. We have seen it all before, it is getting old and boring.
    I actually read about this one, too (don't remember where, it was almost a year ago). Of course, the real reason for requiring signed drivers on Vista x64 isn't stability, but DRM - MS doesn't want somebody to write a kernel driver that would be able to get around the built-in protection. You can still load unsigned drivers, if you choose a specific bootup menu setting, but then the protected media path will not be activated (and you can't make that the default choice - you have to do it on every boot).



  • An appeal to good sense and reason in the face of FUD

     

    I am not going to reply to Kazan, I think we can all see it is a waste of everyone’s time to even read his posts. Also, I don’t delude myself into thinking he will ever get past the FUD he has been fed.

     

    For anyone who would like to actually know what is being discussed and form a reasonable conclusion about this subject, instead of just spouting ignorant nonsense, I will do a little explanation.

     

    First of all, the pricing is not prohibitive. (Disclaimer: I have not done this myself, since I don’t develop drivers or hardware for commercial use. There could be other costs involved than I mention.) However a little research (http://download.microsoft.com/download/d/e/1/de1e0c8f-a222-47bc-b78b-1656d4cf3cf7/WLP-Policies_03-21-08.pdf) has turned up a price of $250 (USD) for each OS family you want to target. This is to get your software tested by the MS QC team and digitally signed. Seems pretty reasonable to me. I doubt MS is getting rich off of this, and I doubt that anyone who seriously wants to sell hardware/drivers is going to be scared away by an investment under $1000 that makes their software guaranteed a little more reliable, and helps it fall neatly into the Windows security environment. AFAIK most of this cost is for the Verisign key that needs to be generated. Again, never been through it, so I cannot say definitively. More info: http://www.microsoft.com/whdc/winlogo/drvsign/drvsign.mspx

     

    Another point that needs to be made is that 32 bit Vista does not enforce driver signatures. 64 bit does. You cannot turn it off in 64 bit Vista permanently, but you can tell it to disable it with the F8 menu on startup. Major catastrophe? I think not.

     

    Reasons for the driver signing? I think most of that is pretty obvious, I doubt I need to explain this to most people, but here goes:

     

    From: ”http://msdn.microsoft.com/en-us/library/aa906341.aspx”

    ·  To ensure the integrity of driver packages. Windows uses digital signatures to verify the identity of the publisher and to verify that the driver has not been altered since it was published.

    ·  To provide the best user experience by facilitating automatic driver installation. If a driver is not signed, the Plug and Play (PnP) driver installation policy requires that a system administrator manually authorize the installation of an unsigned driver, adding an extra step to the installation process. This extra step can be potentially confusing and bothersome to the average user.

    ·  To run kernel-mode drivers on 64-bit versions of Windows Server 2008 and Windows Vista. The kernel-mode code signing policy for 64-bit versions of Windows Server 2008 and Windows Vista requires that kernel-mode drivers be signed in order for the operating system to load the driver.



     

    I think this is a good step in the right direction for Windows security, and I have seen no issues with it yet.

    Please don’t listen to the FUD that is being spread by people like Kazan here, read the references and decide for yourself if you are concerned. 



  • @bstorer said:

    @morbiuswilters said:

    @bstorer said:

    @archivator said:

    @bstorer said:

    it's whopping 7% market share
     

    We passed the 5 percent? WOHOOOO! We need FIREWORKS!

    Probably not. I picked the highest figure a quick Google search turned up.

    That is the most absurd number imaginable.  Linux has less than 0.5% penetration on the desktop.  Apple is around 6-8%, for Christ's sake and you know Linux is nowhere near that. 

    I think the number was combined server + desktop market share. But that's still way too high, as most estimates give Linux about 12% server market share, and I sincerely doubt that servers outnumber desktops.
     

     

    you're numbers are pathetically small - last i saw their server market share was around 25%

    http://www.oreillynet.com/manila/tim/stories/storyReader$56

     that 7% number comes from here and is % of entire market

    http://blogs.zdnet.com/ITFacts/?p=68

    apache has half the httpd market according to netcraft.. and you know less than half of those machines are running windows

    http://news.netcraft.com/archives/2008/06/22/june_2008_web_server_survey.html

    several major vendors (HP, etc) are selling linux on laptops.

     

    Let's try to be accurate with the numbers people - no matter how many technical flaws windows has i wouldn't say it has half the market penetration it has  let's try to show some intellectual honesty



  •  MPS you truly are pathetic - keep ignoring the legitimate issues I raised with "it's fud" repeated to infinity.  $250? they must have lowered the price recently - it used to be like $600

     either way it's a barrier to entry - look up the definition for that - what a barrier for entry is (something that makes it harder to do what you want to do)

     can you give a sound reason for microsoft to disallow power users from bypassing driver signing on 64-bit without having to disable other subsystems?

     

    Of course, i don't expect you to actually reply to my concerns - you're just going to scream that it's FUD again and go off on a separate tangent.  You are guilty of what you keep accusing me of being.  Fuck dude I'm considering working for microsoft's business divison.  Stop making assumptions about me and answer the criticism or SHUT THE HELL UP



  • @Kazan said:

    you're numbers are pathetically small - last i saw their server market share was around 25%

    Posted by Tim O'Reilly, 8/9/00 at 1:58:51 PM.
    The last time you checked was 8 years ago? To be fair, this article from last October claims Linux has a 20% server market share.

    @Kazan said:

     that 7% number comes from here and is % of entire market

    http://blogs.zdnet.com/ITFacts/?p=68

    Yeah, that's probably it. But that's a 3.5-year-old projection, so my mistake there. This article pegs desktop share at about 1.3% last September, but I don't seem capable of finding any recent report of overall market share.



  • @ammoQ said:

    http://www.w3counter.com/globalstats.php claims something in the 2% range.

    That site only claims to track 25M unique visitors across 13K sites.  That's a pitifully small sampling by web standards.  Also keep in mind the following things points:

    • Linux users are more likely to be technically proficient meaning they will do more surfing than your average Windows users.  This will drive up counts.
    • Linux users tend to be more conscious of cookies and Javascript and tend to avoid them.  As these are the primary methods of tracking unique users they are more likely to be counter multiple times.
    • Linux users are less likely to obscure their UA string than Windows users.  Linux users want people to know they are Linux users and want the Linux numbers to increase.

    I've been doing web software for a long time.  I've had several large samples to draw from and have had access to many different for-fee reports that show at-large trends in browser and OS usage.  I can tell you quite confidently that Linux accounts for less than 0.5% of desktop users.  In fact, I'd go so far as to say less than 0.1% based on what I have seen.


  • @Kazan said:

    $250? they must have lowered the price recently - it used to be like $600

    Oh my god, what a big difference!  If you can't afford $10,000 to get your driver certified by MS I wouldn't want the filthy thing on my computer anyway.

     

    @Kazan said:

    can you give a sound reason for microsoft to disallow power users from bypassing driver signing on 64-bit without having to disable other subsystems?

    Yes, it's how the driver system works.  They want to maintain the integrity of the kernel.  Why would they make it easy to disable?  That's essentially like complaining because Toyota doesn't make it easy to remove the airbags from your car.  Meanwhile, what other option is there in the OS space for driver compatiblity?  Microsoft is by far the most compatible and most willing to interoperate with hardware vendors.  They disabled the key those asshats were using because it specifically violated the ToS and circumvented the security system.  I would have gone further and sued them for breach of contract if that were legally possible.  I think Microsoft acted very restrained given the circumstances.

     

    @Kazan said:

    Of course, i don't expect you to actually reply to my concerns - you're just going to scream that it's FUD again and go off on a separate tangent.

    You've been replied to many times.  There is a reasonable explanation for the system and nobody said you had to like it.  However, accusing Microsoft of anything but trying to improve the stability and security of the world computing ecosystem is pure bullshit.  It might be true but you have absolutely no evidence to support such a claim and it simply defies explanation.  That is why MPS is telling you that you are spouting FUD.  You're essentially just inventing a conspiracy theory and turning your lack of actual information into a series of leading questions and accusations.  "Oh, I don't know of a reason for this happening so obviously the intent is to suppress competition and extort money from hardware vendors.  Yeah, the costs are much now, but they will just keep raising the price and squeezing every last penny.  Just you wait and see!"  That's textbook, conspiracy nut FUD.



  • @morbiuswilters said:

    I've been doing web software for a long time.  I've had several large samples to draw from and have had access to many different for-fee reports that show at-large trends in browser and OS usage.  I can tell you quite confidently that Linux accounts for less than 0.5% of desktop users.  In fact, I'd go so far as to say less than 0.1% based on what I have seen.
    Are there any sort of biases for the audience of those websites?  Particularly in the technology domain, any sort of statistical sampling will be subject to the biases caused by the content of those sites.  I would be interested in knowing what sort of statistics can be found from non-technical websites like cnn.com, foxnews.com, but still have broad appeal.

    Also, I don't think enough users know how to obscure their UA string to make a significant difference. 



  • @Kazan said:

    troll much?

    It's called sarcasm.  I was mocking your ridiculous claims.

     

    @Kazan said:

    A) I never said signed drivers were bad, i said REQUIRING them is.  I should be able to install them and tell windows "yes shut the fuck up i know what i'm doing"

    This would defeat the purpose of signed drivers as soon every vendor would just include instructions telling people to ignore the "unsigned driver" dialog.

     

    @Kazan said:

    B)  Do you understand the definition of a policy decision?

    What the fuck does this have to do with anything.

     

    @Kazan said:

    C) do you understand that that primary non-comercial products  such as ATI Tray are what is harmed by this

    Well, if they're non-commercial it's pretty hard to claim any kind of actual damage, now isn't it?  I guess you could say it made the ATI Tray developer really sad and made him take up drinking again.  Hey, getting your driver signed is just another requirement to get it to work.  The guy has probably spend hundreds of hours pouring over ATI hardware documentation, working around various inconsistencies and errata.  He's spent a long time educating himself how the Windows API and kernel work and how you write a plugin for it.  He's probably had to buy several development machines and ATI cards as well.  But if you ask him to fork over $1000 to get his driver signed, well that's just a fucking travesty, now isn't it?  The guy has a donation page so if people care about his little app they will donate enough for him to buy a cert.  If they don't then he can just FOADIAF.  Seriously, this is your big "barrier to entry" that is holding this guy back?  What is wrong with you?  Seriously, I'd like to know.



  • @morbiuswilters said:

    • Linux users are more likely to be technically proficient meaning they will do more surfing than your average Windows users.  This will drive up counts.
    • Linux users tend to be more conscious of cookies and Javascript and tend to avoid them.  As these are the primary methods of tracking unique users they are more likely to be counter multiple times.
    • Linux users are less likely to obscure their UA string than Windows users.  Linux users want people to know they are Linux users and want the Linux numbers to increase.

    Oh com'on, it's not 1997 anymore. Everybody and his grandma surfs a lot. Nobody who wants anything to work turns of cookies and Javascript anymore. And why the hell should a Windows user obsure the UA string? The only reason to do that at all is to get through a browser check.


    I've been doing web software for a long time.  I've had several large samples to draw from and have had access to many different for-fee reports that show at-large trends in browser and OS usage.  I can tell you quite confidently that Linux accounts for less than 0.5% of desktop users.  In fact, I'd go so far as to say less than 0.1% based on what I have seen.

    I fail to find a single statistic which shows less than 0.8% market share for Linux (except one from 1999 ;-). Therefore: Pics or it didn't happen.



  • @ammoQ said:

    Oh com'on, it's not 1997 anymore. Everybody and his grandma surfs a lot.
    Yes, but of the two desktop markets: Windows and Linux, which do you think has the larger percentage surfing the web? I'd go with Linux, because you have to be fairly web-savvy to even know what Linux is.



  • @bstorer said:

    Yes, but of the two desktop markets: Windows and Linux, which do you think has the larger percentage surfing the web? I'd go with Linux, because you have to be fairly web-savvy to even know what Linux is.

    Linux users do not necessarily have to know what Linux is. Some of them probably bought an ultra-cheap PC at Walmart's and never wondered what that "Linspire Windows" really is.



  • @ammoQ said:

    Linux users do not necessarily have to know what Linux is. Some of them probably bought an ultra-cheap PC at Walmart's and never wondered what that "Linspire Windows" really is.
    Do you think the percentage of Linux users unclear that they're running Linux is larger than that of the percentage of Windows users who couldn't think Windows is the computer itself? Linux users trend far more technologically-aware.



  •  re: bstorer

     

    was having a hard time finding recent articles on it on google -I seriously couldn't care less about market share beyond the fact that some people here feel the need to cut the numbers by orders of magnitude



  • @bstorer said:

    Do you think the percentage of Linux users unclear that they're running Linux is larger than that of the percentage of Windows users who couldn't think Windows is the computer itself? Linux users trend far more technologically-aware.
     

    Considering the millions of linux-powered EeePCs Asus has sold during the last year, it's really hard to tell. And that's not even taking into account the corporate Linux users (yes, they exist too).



  •  A few points about driver signing/testing (and don't get on my case about quoting - I'm not quoting because there was nothing in that post worth quoting, this is for general information):

    • Microsoft revoked certain keys because the vendors were flagrantly violating the ToS (as has already been mentioned).
    • The cost to have a driver tested and certified is $250 (as has also already been mentioned), which is a tiny, tiny fraction of the cost to develop the hardware and even to pay the programmers who write the drivers.  Several people at Microsoft have been quite explicit about how they came to this number:
      • Microsoft has to pay people to do the actual testing, not to mention the customer hand-holding.
      • 15 years of experience have proven that the majority of manufacturers, given the opportunity, will simply dump their testing and compatibility burden on Microsoft.  The $250 figure is a completely reasonable one-time cost, but not something that manufacturers will want to pay 12 times over because they can't get their shitty drivers to work.
    • When a driver screws up (i.e. BSOD), users always blame Microsoft even if the OS had nothing to do with it.  Microsoft has to pay employees to handle the support calls and sort through these issues.
    • Microsoft has good reasons (aside from the above) to want to crack down on unsigned/unverified drivers.  Throughout the lifetime of XP, most manufacturers didn't even bother to go through the verification process, and just gave users instructions to ignore the warning without any explanation as to what it meant.  Some very well-known manufacturers even went so far as to put ugly hacks in their install programs to try to circumvent or automatically dismiss the warning.  Of course these hacks were never localized, or even tested, so they failed all the time, but when users called in for support, the worms blamed Microsoft anyway.

    The fact of that matter is, most hardware vendors put the absolute minimum amount of effort they think they can get away with into their drivers.  They didn't put out unsigned drivers because they couldn't afford the minuscule "ransom", they did it because they were too lazy and cheap to actually test their junk drivers and fix the bugs.  And then they had the audacity to spread FUD to their customers, using the expense as an excuse and claiming that the warnings were just another annoying Windows quirk that Microsoft never quite "fixed".

    So if some of you are wondering why these same companies don't put out Linux drivers, try to keep in mind that they probably wouldn't write any drivers at all if they could get away with it.  They're plagued by the same IT problems as any other company - clueless penny-pinching management, incompetent programmers, no QA.

    I'm glad that MS has been moving toward stricter requirements for drivers.  They're doing a major service to their customers.  Drivers tend to be both the most critical and poorest quality programs on any computer, and that needs to change.

    Incidentally, if you as a user want to go "shut up, I know what I'm doing" in Vista x64, you actually do have that ability.  It takes all of 5 seconds to find out how on Google.  Power users can do what they please, but ordinary users are now better protected against aggressively sloppy or even hostile 3rd-party drivers.



  • @ammoQ said:

    Considering the millions of linux-powered EeePCs Asus has sold during the last year, it's really hard to tell. And that's not even taking into account the corporate Linux users (yes, they exist too).
    There were over 70 million PCs sold in the second quarter of 2008. Dell alone sold 5.25 million. Asus shipped 1.7 million EeePCs in the first half of 2008. That's a drop in the bucket.



  • @bstorer said:

    There were over 70 million PCs sold in the second quarter of 2008. Dell alone sold 5.25 million. Asus shipped 1.7 million EeePCs in the first half of 2008. That's a drop in the bucket.
     

    When we are talking about a market share that could be 0.1% or 1% or 5%, those 1.7 million EeePCs definitely matter!



  • @ammoQ said:

    Considering the millions of linux-powered EeePCs Asus has sold during the last year, it's really hard to tell.

    You know eeePCs come with Windows XP too, right?  You also know that the vast majority of people buying the Linux versions are just trying to save $50 and will install a pirated copy of XP.  Yes, seriously. 



  • @ammoQ said:

    When we are talking about a market share that could be 0.1% or 1% or 5%, those 1.7 million EeePCs definitely matter!
    Sure they do. But what makes you think they aren't being counted. The EeePCs are called "netbooks" for a reason, and it ain't because they're useful in catching tuna.



  • @ammoQ said:

    When we are talking about a market share that could be 0.1% or 1% or 5%, those 1.7 million EeePCs definitely matter!
     

    Shipping and selling are two different things. I would be interested to see how many USERS there are of linux based eeepcs. 

    Like others have said, buying them with linux and installing windows would negate any advantage you perceive here.



  • @Kazan said:

    I seriously couldn't care less about market share beyond the fact that some people here feel the need to cut the numbers by orders of magnitude

    Bullshit.  Here's a report for the last 24 months: http://marketshare.hitslink.com/report.aspx?qprid=11&qpdt=1&qpct=4&qptimeframe=M&qpsp=90&qpnp=25#  They gather stats on 160 million unique visitors per month and have several mainstream sites like NYTimes, WSJ and TheStreet.  So even optimistically Linux only holds a 0.8% market share of average-to-heavy Internet users.  This neglects all of the Mac and Windows users who surf the web very little, which does artificially skew the numbers for Linux a bit higher.  So 0.5% certainly was correct. 



  • @morbiuswilters said:

    You know eeePCs come with Windows XP too, right?

    Yes, but the first version was Linux-only.


    You also know that the vast majority of people buying the Linux versions are just trying to save $50 and will install a pirated copy of XP.  Yes, seriously. 

    IMO not in this case (EeePC 701). This thing doesn't even have a CD-ROM drive. It's difficult to install and gains nothing, given the low capacity of the SSD.

     


Log in to reply