5 ms·

Microsoft has historically been the single greatest enemy of Linux and Open Source in general. The only reason they now changed face and hopped on the Linux tra

Microsoft has historically been the single greatest enemy of Linux and Open Source in general. The only reason they now changed face and hopped on the Linux train is that they realized they lost or cannot possibly win the battle when it comes to servers, embedded devices, a large section of developers, mobile, etc. > I can tell you unequivocally the thing holding back the average enterprise from moving their entire business to Linux on the Desktop is NOT Microsoft Office It may not be office specifically, but vendor lock-in with various MS products is definitely a huge factor. Also, every single "quoted" piece of your comment is cheap strawman.

Someone1234 7y ago
People don't use Linux Desktop because it is bad. Constantly looking for others to blame (Microsoft, DRM, Google, Valve, Apple, etc) doesn't change that fact. The fact that Google has been able to take the Linux kernel twice and produce a better and more popular desktop experience (with Android and then Chromebook) should be a good indication of that. Linux Desktop is exclusively designed for Linux developers, it is user hostile, highly inconsistent, difficult to centrally manage, not accessible (and thus could leave employers liable to lawsuits in some countries), and buggy. Incidentally Microsoft does have a version of Office that runs on Linux: Office Online. Yet somehow nobody is quitting Windows, MacOS, and Chromebook to go Linux Desktop even for completely free. If you want Linux Desktop to be a success, maybe spend less time finding the latest scapegoat and more time helping make it a good user experience. Personally I don't think it can be fixed not because the software is beyond saving, but the community surrounding it doesn't actually want it to be an experience for regular non-technical people. Try going to a mailing list and suggesting that editing text files in /etc isn't a practical end-user experience and see what response you get.
- pytester 7y ago
  >People don't use Linux Desktop because it is bad. There is a reason why it is not offered as an option to be preinstalled on laptops except in a few instances. That reason is not "because it is bad". The windows license is a tax on laptops. End of story. I have a dell xps 13 and even I paid it (dell's developer edition laptop is often hard to acquire for some reason).
  - Wowfunhappy 7y ago
    > That reason is not "because it is bad". I think it quite literally is, or at least that's a part of it (I won't say Microsoft strong-arming has zero effect). Linux would be an exceedingly poor experience for most customers—worse even than Windows with all the preinstalled crapware some manufacturers add. The first netbooks ran Linux. The choice seemed to make sense. Cost was paramount, and manufacturers needed something exceedingly lightweight. But I remember reviews from the time explicitly recommending that people ignore any netbooks that don't come with Windows. It didn't take very long before they almost all switched over to XP.
- excursionist 7y ago
  > People don't use Linux Desktop because it is bad. People do you Linux Desktop because it is great. I do. > The fact that Google has been able to take the Linux kernel twice and produce a better and more popular desktop experience (with Android and then Chromebook) should be a good indication of that. In what way is Android or Chromebook better than, say, Debian or Ubuntu on desktop? > Linux Desktop is exclusively designed for Linux developers.. blah blah blah Just false bullshit. > If you want Linux Desktop to be a success, maybe spend less time finding the latest scapegoat and more time helping make it a good user experience. If you want to debate the quality of Linux as a deskop OS maybe try using it, learn a little about it, and make statements that have even a little bit of substance and basis in reality. > Try going to a mailing list and suggesting that editing text files in /etc isn't a practical end-user experience and see what response you get. First of all, it is perfectly practical, especially given the fact that you can find step-by-step instruction to do just about anything in a matter of seconds. Second, you rarely have to do these things and graphical overlays are often available and/or come pre-installed on distros such as Ubuntu. Third, you sometimes have to do similar things on Windows, whether it be editing the hosts file or changing obscure registry settings - the difference is that the Linux versions of these things have much better documentation, and much more info and community support online.
  - dlivingston 7y ago
    Your response is little more than, "no, you're wrong, and I'm right." I tend to agree with parent that the general Linux UI/UX experience (even the flagship Linux OS, Ubuntu) is subpar compared to commercial operating systems. You only make a serious argument in the last bullet: > First of all, it is perfectly practical, especially given the fact that you can find step-by-step instruction to do just about anything in a matter of seconds. Second, you rarely have to do these things and graphical overlays are often available and/or come pre-installed on distros such as Ubuntu. Third, you sometimes have to do similar things on Windows, whether it be editing the hosts file or changing obscure registry settings - the difference is that the Linux versions of these things have much better documentation, and much more info and community support online. The vast majority of Windows and macOS users are not technical people, and that is fine. The vast majority of Windows users do not need to edit registry settings, just as the vast majority of macOS users don't need to edit `defaults write com.apple.[star]`. More importantly, they don't want to. My grandma is deathly afraid that any wrong action she takes will brick her PC. You're telling me that she'll be fine running sudo in a terminal, even if there's step by step directions?
- caspper69 7y ago
  > Personally I don't think it can be fixed not because the software is beyond saving, but the community surrounding it doesn't actually want it to be an experience for regular non-technical people. Try going to a mailing list and suggesting that editing text files in /etc isn't a practical end-user experience and see what response you get. Thank you. This is the truth, and your attitude on the issue will tell you which dog you have in the fight; The non-technical among us, the 80-90% if you will, are not going to use a desktop Linux distribution the way they are currently setup. A home user is not going to use Google to fix a problem. They're not going to call their technically-inclined family member. They're not going to edit text files nor understand the minutiae of their hardware (and it WILL happen, eventually, 100% guaranteed). They are either going to throw "that piece of shit away" or they're going to call their buddy Steve who can put Windows on it. Unfortunately, the community is still "the community," and much like those kids who wouldn't get out of their parents' basements, they're now in their 50s, crusty and pedantic as hell, sitting in their own ivory tower, incidentally not much different than the one they rebelled against in the 90s. Anyway... blah, blah, blah.
  - excursionist 7y ago
    > The non-technical among us, the 80-90% if you will, are not going to use a desktop Linux distribution the way they are currently setup. A home user is not going to use Google to fix a problem. They're not going to call their technically-inclined family member. They're not going to edit text files nor understand the minutiae of their hardware (and it WILL happen, eventually, 100% guaranteed). They are either going to throw "that piece of shit away" or they're going to call their buddy Steve who can put Windows on it. So what are they going to do when they encounter a problem on windows? Pretty sure MS hasn't yet figured out how to directly install a Window manual into their userss brains, and the thousands of unanswered windows troubleshooting questions online can attest to that. > Unfortunately, the community is still "the community," and much like those kids who wouldn't get out of their parents' basements, they're now in their 50s, crusty and pedantic as hell, sitting in their own ivory tower, incidentally not much different than the one they rebelled against in the 90s. Anyway... blah, blah, blah. Is this basement-dweller fantasy something you have personal experience with or is it just a story you like to tell to make yourself feel better about your own technical ignorance and laziness?
- AnIdiotOnTheNet 7y ago
  I agree with everything you said, but most especially this: > Personally I don't think it can be fixed not because the software is beyond saving, but the community surrounding it doesn't actually want it to be an experience for regular non-technical people The community behind Linux Desktop is, by far, the biggest reason it is as unpopular as it is.
- K0SM0S 7y ago
  There are distros targetting more specifically mainstream end-users, historically Mint and more recently Elementary (which does a fantastic albeit opinionated job with UI/UX, or so I hear). The problem of general adoption however is not related to that. It's really not the best software that wins the wallets of consumers, but the best marketing — same for music, movies, vacation destinations, virtually any product: marketing 'bypasses' quality (it just needs to be 'good enough') by 'brute-forcing' awareness and familiarity (sheer repetition, seduction, etc). It's true there's a lasting power to great UX (ask Apple), but Linux desktop never even entered that phase for the mainstream (first you have to exist before we even ask if you're going to last). Making the best Linux experience is thus a vain enterprise. It just won't sell more or less (if we agree that even free has a cost), it just won't bootstrap massive adoption. It's at best preparing the product for its encounter with the public, but it's not making said encounter more likely. A strategy like Microsoft's with Windows shows where the battle of OS's is fought and won: with OEMs. Microsoft correctly identified (just like Intel) in the 1990's that people would buy computers (thus OS, thus CPU) through middlemen, OEMs, the Dell's and HP's and ASUS's of the world. So that's who you need to sell on your product. Apple is its own OEM, as they integrate vertically, and indeed the 'integration' between hard- and soft-wares is what sold most of us who ever bought into OS X or iOS et. al. The interesting take is that Apple for the longest time oscillates between 5 and 15% market share on average on all their products worldwide (much closer to 50% on star products in the US, nigh-zero in the poorest markets), and this is the nonetheless massive (9 figures) user base that guarantees adoption of their OS, which are designed entirely to sell the hardware by making it shine (what MS is now trying to do with Surface lead by Panos Panay, or Google with Pixels and other things). This whole 'integration' space is where Linux never even tried itself, it seems, because there's never been a "Linux Marketing Department" nor a "Linux Sales Person" — when Dell does Linux, it's because a bunch of fans inside the company made it their project; when Red Hat sells Linux it's closer to the goal with OEMs but certainly not mainstream/consumer in that case. As for Ubuntu, I don't know who they sell to; they're certainly big on the servers, but absent of consumer laptops despite being totally usable these days (and most consumers certainly don't need the advanced features of paid O365, so that debate is moot in this case). There are Linux brands (notably System76 with their awesome PopOS), but they also don't target the mainstream it seems, or not in a 'big' way likely to turn market figures around. The truth is that I don't think anyone with big-tech decision power really cares now that we've moved to IaaS and serving apps on containers (alternatively 'minified' through some CD pipeline from a common base) is going to increasingly become the 'easy' way to have apps 'just work'; and by that point who cares if it's Windows, Linux or even iOS in the kernel underneath. By 2030, I anticipate that OS/kernel won't make much more difference than the CPU you're using (tech people and nerds may care, most people just want apps X Y Z to run fine and they'll get just that). Whether server or client side, we'll all live on top of some common or at least expected abstraction and most of us will be none the wiser. I'm not sure BSD will be much around in this scenario, except legacy and of course as darwin for Apple. Not saying I predict the future or this is what must or should happen, but it definitely can as I see it, it's a 'valid scenario' with a high degree of probability IMHO.
- jcranberry 7y ago
  With Linux + Gnome 3, you don't really ever have to go change text files in /etc. If you're only using native software/browsing/etc, the experience is fine for a non-technical person. People don't use Linux just because it only has downsides for the average person. Not being able to play games, use MS office, other domain specific software, get used to a new UI...plus the only cultural associations with it is that it's highly technical. So there's no incentives for people to switch.
tracker1 7y ago
MS has shown almost no effort to lock things into windows only for the better part of a decade at this point. I'm not saying they never did, just not seeing that recently. I am only slightly disappointed it took so long to get MS Teams working properly in Linux. The changes in the past few months have the unofficial clients working much better... A/V calling/meetings actually works now (though I haven't tried sharing my desktop) and an official client, if not already out, should be eminent. Office 365's online versions are definitely usable and leaps and bounds better than Google Docs. I have mixed feelings on Azure AD and a few other pieces. Right now, Windows feels like the odd man out, it's what I'm using on my work Desktop, I've phased out my Mac usage, and mostly use Linux outside work and at work, my flows are mostly Docker. I use msys bash (via git) on windows and there are a couple gotchas. I remember the crap they pulled in the late 90's very well... The MS I see today is not the same... the management, culture and interactions have dramatically changed... the people have changed. Companies aren't people, but made up of them.
gowld 7y ago
Whatever proprietary company is most sucessful will be the greatest enemy. If not MS, then Apple, or Google, or whoever. If you believe in Open Source, you should focus on building and promoting (and fundraising for) open source. Until we get the same $5T-equivalent investment in Open Source that FAANGMwhatever have, Open Source will will trail the rest in many quality and network-effect dimensions that will hinder the virtuous sycle of growth.
- excursionist 7y ago
  > Whatever proprietary company is most sucessful will be the greatest enemy. If not MS, then Apple, or Google, or whoever. That's not true. Apple and Google have not done even 1% of the damage to Linux and Open Source that MS has. > Until we get the same $5T-equivalent investment in Open Source that FAANGMwhatever have, Open Source will will trail the rest in many quality and network-effect dimensions that will hinder the virtuous sycle of growth. Except for the fact that most successful modern tech company stacks are mostly FOSS and Linux. Microsoft was essentially forced to offer Linux on Azure because the overwhelming consensus is MS Server cannot hold a candle to Linux.

Someone1234 7y ago

People don't use Linux Desktop because it is bad. Constantly looking for others to blame (Microsoft, DRM, Google, Valve, Apple, etc) doesn't change that fact. The fact that Google has been able to take the Linux kernel twice and produce a better and more popular desktop experience (with Android and then Chromebook) should be a good indication of that. Linux Desktop is exclusively designed for Linux developers, it is user hostile, highly inconsistent, difficult to centrally manage, not accessible (and thus could leave employers liable to lawsuits in some countries), and buggy. Incidentally Microsoft does have a version of Office that runs on Linux: Office Online. Yet somehow nobody is quitting Windows, MacOS, and Chromebook to go Linux Desktop even for completely free. If you want Linux Desktop to be a success, maybe spend less time finding the latest scapegoat and more time helping make it a good user experience. Personally I don't think it can be fixed not because the software is beyond saving, but the community surrounding it doesn't actually want it to be an experience for regular non-technical people. Try going to a mailing list and suggesting that editing text files in /etc isn't a practical end-user experience and see what response you get.

pytester 7y ago
>People don't use Linux Desktop because it is bad. There is a reason why it is not offered as an option to be preinstalled on laptops except in a few instances. That reason is not "because it is bad". The windows license is a tax on laptops. End of story. I have a dell xps 13 and even I paid it (dell's developer edition laptop is often hard to acquire for some reason).
- Wowfunhappy 7y ago
  > That reason is not "because it is bad". I think it quite literally is, or at least that's a part of it (I won't say Microsoft strong-arming has zero effect). Linux would be an exceedingly poor experience for most customers—worse even than Windows with all the preinstalled crapware some manufacturers add. The first netbooks ran Linux. The choice seemed to make sense. Cost was paramount, and manufacturers needed something exceedingly lightweight. But I remember reviews from the time explicitly recommending that people ignore any netbooks that don't come with Windows. It didn't take very long before they almost all switched over to XP.
excursionist 7y ago
> People don't use Linux Desktop because it is bad. People do you Linux Desktop because it is great. I do. > The fact that Google has been able to take the Linux kernel twice and produce a better and more popular desktop experience (with Android and then Chromebook) should be a good indication of that. In what way is Android or Chromebook better than, say, Debian or Ubuntu on desktop? > Linux Desktop is exclusively designed for Linux developers.. blah blah blah Just false bullshit. > If you want Linux Desktop to be a success, maybe spend less time finding the latest scapegoat and more time helping make it a good user experience. If you want to debate the quality of Linux as a deskop OS maybe try using it, learn a little about it, and make statements that have even a little bit of substance and basis in reality. > Try going to a mailing list and suggesting that editing text files in /etc isn't a practical end-user experience and see what response you get. First of all, it is perfectly practical, especially given the fact that you can find step-by-step instruction to do just about anything in a matter of seconds. Second, you rarely have to do these things and graphical overlays are often available and/or come pre-installed on distros such as Ubuntu. Third, you sometimes have to do similar things on Windows, whether it be editing the hosts file or changing obscure registry settings - the difference is that the Linux versions of these things have much better documentation, and much more info and community support online.
- dlivingston 7y ago
  Your response is little more than, "no, you're wrong, and I'm right." I tend to agree with parent that the general Linux UI/UX experience (even the flagship Linux OS, Ubuntu) is subpar compared to commercial operating systems. You only make a serious argument in the last bullet: > First of all, it is perfectly practical, especially given the fact that you can find step-by-step instruction to do just about anything in a matter of seconds. Second, you rarely have to do these things and graphical overlays are often available and/or come pre-installed on distros such as Ubuntu. Third, you sometimes have to do similar things on Windows, whether it be editing the hosts file or changing obscure registry settings - the difference is that the Linux versions of these things have much better documentation, and much more info and community support online. The vast majority of Windows and macOS users are not technical people, and that is fine. The vast majority of Windows users do not need to edit registry settings, just as the vast majority of macOS users don't need to edit `defaults write com.apple.[star]`. More importantly, they don't want to. My grandma is deathly afraid that any wrong action she takes will brick her PC. You're telling me that she'll be fine running sudo in a terminal, even if there's step by step directions?
caspper69 7y ago
> Personally I don't think it can be fixed not because the software is beyond saving, but the community surrounding it doesn't actually want it to be an experience for regular non-technical people. Try going to a mailing list and suggesting that editing text files in /etc isn't a practical end-user experience and see what response you get. Thank you. This is the truth, and your attitude on the issue will tell you which dog you have in the fight; The non-technical among us, the 80-90% if you will, are not going to use a desktop Linux distribution the way they are currently setup. A home user is not going to use Google to fix a problem. They're not going to call their technically-inclined family member. They're not going to edit text files nor understand the minutiae of their hardware (and it WILL happen, eventually, 100% guaranteed). They are either going to throw "that piece of shit away" or they're going to call their buddy Steve who can put Windows on it. Unfortunately, the community is still "the community," and much like those kids who wouldn't get out of their parents' basements, they're now in their 50s, crusty and pedantic as hell, sitting in their own ivory tower, incidentally not much different than the one they rebelled against in the 90s. Anyway... blah, blah, blah.
- excursionist 7y ago
  > The non-technical among us, the 80-90% if you will, are not going to use a desktop Linux distribution the way they are currently setup. A home user is not going to use Google to fix a problem. They're not going to call their technically-inclined family member. They're not going to edit text files nor understand the minutiae of their hardware (and it WILL happen, eventually, 100% guaranteed). They are either going to throw "that piece of shit away" or they're going to call their buddy Steve who can put Windows on it. So what are they going to do when they encounter a problem on windows? Pretty sure MS hasn't yet figured out how to directly install a Window manual into their userss brains, and the thousands of unanswered windows troubleshooting questions online can attest to that. > Unfortunately, the community is still "the community," and much like those kids who wouldn't get out of their parents' basements, they're now in their 50s, crusty and pedantic as hell, sitting in their own ivory tower, incidentally not much different than the one they rebelled against in the 90s. Anyway... blah, blah, blah. Is this basement-dweller fantasy something you have personal experience with or is it just a story you like to tell to make yourself feel better about your own technical ignorance and laziness?
AnIdiotOnTheNet 7y ago
I agree with everything you said, but most especially this: > Personally I don't think it can be fixed not because the software is beyond saving, but the community surrounding it doesn't actually want it to be an experience for regular non-technical people The community behind Linux Desktop is, by far, the biggest reason it is as unpopular as it is.
K0SM0S 7y ago
There are distros targetting more specifically mainstream end-users, historically Mint and more recently Elementary (which does a fantastic albeit opinionated job with UI/UX, or so I hear). The problem of general adoption however is not related to that. It's really not the best software that wins the wallets of consumers, but the best marketing — same for music, movies, vacation destinations, virtually any product: marketing 'bypasses' quality (it just needs to be 'good enough') by 'brute-forcing' awareness and familiarity (sheer repetition, seduction, etc). It's true there's a lasting power to great UX (ask Apple), but Linux desktop never even entered that phase for the mainstream (first you have to exist before we even ask if you're going to last). Making the best Linux experience is thus a vain enterprise. It just won't sell more or less (if we agree that even free has a cost), it just won't bootstrap massive adoption. It's at best preparing the product for its encounter with the public, but it's not making said encounter more likely. A strategy like Microsoft's with Windows shows where the battle of OS's is fought and won: with OEMs. Microsoft correctly identified (just like Intel) in the 1990's that people would buy computers (thus OS, thus CPU) through middlemen, OEMs, the Dell's and HP's and ASUS's of the world. So that's who you need to sell on your product. Apple is its own OEM, as they integrate vertically, and indeed the 'integration' between hard- and soft-wares is what sold most of us who ever bought into OS X or iOS et. al. The interesting take is that Apple for the longest time oscillates between 5 and 15% market share on average on all their products worldwide (much closer to 50% on star products in the US, nigh-zero in the poorest markets), and this is the nonetheless massive (9 figures) user base that guarantees adoption of their OS, which are designed entirely to sell the hardware by making it shine (what MS is now trying to do with Surface lead by Panos Panay, or Google with Pixels and other things). This whole 'integration' space is where Linux never even tried itself, it seems, because there's never been a "Linux Marketing Department" nor a "Linux Sales Person" — when Dell does Linux, it's because a bunch of fans inside the company made it their project; when Red Hat sells Linux it's closer to the goal with OEMs but certainly not mainstream/consumer in that case. As for Ubuntu, I don't know who they sell to; they're certainly big on the servers, but absent of consumer laptops despite being totally usable these days (and most consumers certainly don't need the advanced features of paid O365, so that debate is moot in this case). There are Linux brands (notably System76 with their awesome PopOS), but they also don't target the mainstream it seems, or not in a 'big' way likely to turn market figures around. The truth is that I don't think anyone with big-tech decision power really cares now that we've moved to IaaS and serving apps on containers (alternatively 'minified' through some CD pipeline from a common base) is going to increasingly become the 'easy' way to have apps 'just work'; and by that point who cares if it's Windows, Linux or even iOS in the kernel underneath. By 2030, I anticipate that OS/kernel won't make much more difference than the CPU you're using (tech people and nerds may care, most people just want apps X Y Z to run fine and they'll get just that). Whether server or client side, we'll all live on top of some common or at least expected abstraction and most of us will be none the wiser. I'm not sure BSD will be much around in this scenario, except legacy and of course as darwin for Apple. Not saying I predict the future or this is what must or should happen, but it definitely can as I see it, it's a 'valid scenario' with a high degree of probability IMHO.
jcranberry 7y ago
With Linux + Gnome 3, you don't really ever have to go change text files in /etc. If you're only using native software/browsing/etc, the experience is fine for a non-technical person. People don't use Linux just because it only has downsides for the average person. Not being able to play games, use MS office, other domain specific software, get used to a new UI...plus the only cultural associations with it is that it's highly technical. So there's no incentives for people to switch.

tracker1 7y ago

MS has shown almost no effort to lock things into windows only for the better part of a decade at this point. I'm not saying they never did, just not seeing that recently. I am only slightly disappointed it took so long to get MS Teams working properly in Linux. The changes in the past few months have the unofficial clients working much better... A/V calling/meetings actually works now (though I haven't tried sharing my desktop) and an official client, if not already out, should be eminent. Office 365's online versions are definitely usable and leaps and bounds better than Google Docs. I have mixed feelings on Azure AD and a few other pieces. Right now, Windows feels like the odd man out, it's what I'm using on my work Desktop, I've phased out my Mac usage, and mostly use Linux outside work and at work, my flows are mostly Docker. I use msys bash (via git) on windows and there are a couple gotchas. I remember the crap they pulled in the late 90's very well... The MS I see today is not the same... the management, culture and interactions have dramatically changed... the people have changed. Companies aren't people, but made up of them.

gowld 7y ago

Whatever proprietary company is most sucessful will be the greatest enemy. If not MS, then Apple, or Google, or whoever. If you believe in Open Source, you should focus on building and promoting (and fundraising for) open source. Until we get the same $5T-equivalent investment in Open Source that FAANGMwhatever have, Open Source will will trail the rest in many quality and network-effect dimensions that will hinder the virtuous sycle of growth.

excursionist 7y ago
> Whatever proprietary company is most sucessful will be the greatest enemy. If not MS, then Apple, or Google, or whoever. That's not true. Apple and Google have not done even 1% of the damage to Linux and Open Source that MS has. > Until we get the same $5T-equivalent investment in Open Source that FAANGMwhatever have, Open Source will will trail the rest in many quality and network-effect dimensions that will hinder the virtuous sycle of growth. Except for the fact that most successful modern tech company stacks are mostly FOSS and Linux. Microsoft was essentially forced to offer Linux on Azure because the overwhelming consensus is MS Server cannot hold a candle to Linux.