Windows Subsystem for Linux

[Solar]

You forget that IBM - a much larger force than Microsoft - tried to take over the desktop market. They didn't succeed because their operating system - although technically superior to Windows - didn't do what the public wanted. It was nothing to do with Microsoft having a bigger market presence or (obviously) because Microsoft had the backing of IBM; it was simply that Microsoft made the product that was more in line with what people needed. And, like Linux, OS/2 was too complicated for the ordinary person to configure.

At almost no point throughout computing history has Microsoft presented the technologically superior product.

What they did was playing the business side of things perfectly. And their tactics haven't even changed that much over the decades: Making sure that their software was preinstalled. Making it hard and / or inconvenient to install or use alternatives. Providing "support", and marketing heavily, for businesses, where you had to convince only one or two manager types to have another couple of hundred wage drones educated into mentally equating "text document" with "Word .doc" and "spreadsheet" with "Excel", making any competition a second-grade choice and forcing everyone to play catch-up in compatibility instead of differenciating themselves in innovation. Playing the software patent game to perfection ("sign this agreement, sell out your IP, or be sued into bancruptcy").

Microsoft Windows 95 was basically just catching up, technology wise, with what competition already offered for a decade. Win3.11 was a JOKE by comparison. But by that time, marketing budget, business customer leverage etc. were already off the scale, and drove any commercial competition out of the market, with the exception of Apple who had decided to dance with the devil instead of fighting him (being, at that time, basically a nicer skin to run Word and Excel on).

Everything after that was just a continuation of the steamrolling they had done before.

Microsoft was never good at producing what the customer wanted. What they were undeniably good at was making the customer believe it was Microsoft he wanted, which is not the same thing.

Too bad it looks as if the mobile revolution might be putting an end to it, as Microsoft has so far been unable to shoulder aside iOS and Android.

[onlyonemac]

You forget that IBM - a much larger force than Microsoft - tried to take over the desktop market. They didn't succeed because their operating system - although technically superior to Windows - didn't do what the public wanted. It was nothing to do with Microsoft having a bigger market presence or (obviously) because Microsoft had the backing of IBM; it was simply that Microsoft made the product that was more in line with what people needed. And, like Linux, OS/2 was too complicated for the ordinary person to configure.

At almost no point throughout computing history has Microsoft presented the technologically superior product.

What they did was playing the business side of things perfectly. And their tactics haven't even changed that much over the decades: Making sure that their software was preinstalled. Making it hard and / or inconvenient to install alternatives. Providing "support", and marketing heavily, for businesses, so that the not-so-geeky populace came to equate "text document" with "Word .doc" and "spreadsheet" with "Excel", making any competition a second-grade choice and forcing everyone to play catch-up in compatibility instead of differenciating themselves in innovation. Playing the software patent game to perfection ("sign this agreement, sell out your IP, or be sued into bancruptcy").

Microsoft Windows 95 was basically just catching up, technology wise, with what competition already offered for a decade. Win3.11 was a JOKE by comparison. But by that time, marketing budget, business customer leverage etc. were already off the scale, and drove any commercial competition out of the market, with the exception of Apple who had decided to dance with the devil instead of fighting him (being, at that time, basically a nicer skin to run Word and Excel on).

Microsoft was never good at producing what the customer wanted. What they were undeniably good at was making the customer believe it was Microsoft he wanted, which is not the same thing.

Exactly, I agree with everything that you say and this is exactly what I've been trying to say myself. And we don't want this happening with Linux, because the whole Linux industry is built on principles that are exactly the opposite to this approach.

[onlyonemac]

They didn't succeed because their operating system - although technically superior to Windows - didn't do what the public wanted.

Saying that OS/2 "didn't do what the public wanted" is a bit broad. Specifically, I have used OS/2 in the past I believe that its main shortcoming was the inefficiency of the interface. I don't really know how to describe it, but the interface just felt really slow and cumbersome to use, even though the underlying operating system had some interesting concepts (which frankly I think were too ahead of their time to be implemented successfully).

[Solar]

And we don't want this happening with Linux, because the whole Linux industry is built on principles that are exactly the opposite to this approach.

Err, except that it isn't. There's a reason why there never was any effort to make Linux device drivers useable by anyone else but Linux.

Also, GNU/Linux and the GPL are inseperable, which is basically the software patent game played the other way around. You dabble in MS' playing field, they get you via the patents. You dabble in Linux' playing field, they just assimilate you via the GPL.

GNU/Linux is playing for market share(1), with the same hard bandages as Microsoft. Neither of the two is interested in opening up the field to actually compete on technological merit.

----

(1) The ones at the rudder are; not each individual developer actually buying into their happy "free software" mantras. They are just the zombies, just like the wage drones are Microsoft's zombies.

[onlyonemac]

Re: the static vs dynamic linking discussion:

Dynamic linking is better for most situations, as having duplicated object code is a waste of disk space and there are usually no significant issues with having to load object code at runtime. Static linking is better really only in situations where one cannot rely on being able to locate the dynamically-linked libraries and read them from disk, for example if one requires high fault tolerance or if one is working in a bootstrap or recovery shell and the filesystem holding the dynamically-linked libraries may not be mounted or readable. Static linking is also sometimes used for performance reasons, as dynamic linking requires more files to be read from disk and introduces extra overhead in the executable loader, but in reality the total amount of object code that needs to be loaded is often similar and the performance benefits of static linking are questionable.

My comment about glibc wasn't claiming that dynamic linking is always better. My point was that glibc is designed for dynamic linking, so if you statically link with it you shouldn't be too surprised if the results aren't that good because that isn't the way that it's meant to be used (one might argue that libraries should be developed with both dynamic linking and static linking in mind, but considering that glibc was and is developed primarily for systems that use dynamic linking as the main way of linking with libraries, and that getting the library to work well when static linking is likely to require a lot of work for something that's outside of the intended use cases, one can see why glibc still works better when linked dynamically). Furthermore, there are other C libraries that are more suited to static linking and if that's your use case then you should be using one of those alternative libraries.

/subthread

[onlyonemac]

Also, GNU/Linux and the GPL are inseperable, which is basically the software patent game played the other way around. You dabble in MS' playing field, they get you via the patents. You dabble in Linux' playing field, they just assimilate you via the GPL.

Except that

1. I'm not talking exclusively about the Linux kernel and common software that are GNU-licensed, but about many of the other projects that are not GNU-licensed and don't have restrictions that require them to remain open source
2. Microsoft can still include proprietary kernel modules and system components alongside the GNU-licensed ones, and they can still package proprietary software (e.g. Microsoft Office, Internet Explorer, Windows Media Player) alongside or in place of traditional open source alternatives (such as LibreOffice, Firefox, and VLC), and by getting enough users they can pull people away from the latter projects until they die off

[Rusky]

Let's start with the difference between code written by Microsoft and code written by most other people (especially true open-source developers), as evidenced by the end product

Now you've got a no-true-Scotsman fallacy going on- not only does Microsoft have to open source their code for them not to be evil, they have to conform to your particular idea of a release cycle, which the Linux/FOSS ecosystem does just as (or more!) awful a job of. Open source software is no stranger to changing and adding things too quickly and for no reason, breaking people's workflows, releasing buggy code, and failing to fix bugs users care about. It's so bad, in fact, that many distros wind up staying several versions behind by default for "stability" reasons, and heavily patching upstream source as well.

Microsoft, on the other hand, is nowhere near as bad as you claim. As you've seen already, they've gotten very thorough about security, and provide good tools for administrating large numbers of desktop machines. They also (for even longer) take backwards compatibility very seriously, because their income depends on it. I would even say (disagreeing with Solar here) that they do provide some of the highest-quality tools and applications in some areas- DirectX is a far better API than OpenGL (and I say this as someone paid to write graphics code), Visual Studio is far better than Eclipse at editing and debugging C++ (feature and interface-wise), Office (especially Excel) is top-notch.

Now none of this is to say that Linux is horrible (I use it as my main OS at home) or that Windows is perfect (I have plenty of complaints). The point is that your narrative of "everything Microsoft does is garbage, they're only widely used because they force themselves on everyone, and Linux is perfect" is utter hogwash. Microsoft does some great work, and Linux has some deep flaws.

So what do you suggest is their intention with "embracing" direct competitors? How does helping competing projects, such as Linux and Eclipse, bring Microsoft any financial gain unless they plan to eventually either take over or kill off those projects? (Also note that the open source .NET is not an "embracing" of Mono so I don't know why that's in your list.)

Helping people develop for Linux (via Visual Studio support, WSL, an official .NET port, as well as improving the tools people already use like Eclipse and Mono) helps people who are developing things to run on Azure, which brings them revenue no matter which OS they're using. It also helps them port user-space applications like Office (or, to speculate, maybe ActiveDirectory, etc.) to other platforms like Android, which also brings them revenue. They're diversifying, not planning to kill off or take over Linux or Eclipse or Mono.

Which question did I apparently dodge?

Let me ask you something, onlyonemac: how many years and products does Microsoft have not to embrace-extend-extinguish for you to accept anything they do as less than the devil's handiwork? Or are you just a religious zealot who will shift the goalposts anytime Microsoft changes, hating Microsoft just for the sake of hating Microsoft?

The "answer" you gave was 'Perhaps they need to actually start not "embrace extend extinguish"ing products for a start,' but you gave no evidence or even compelling argument that they are still doing so. So, since they aren't doing that (or any of your other conspiracy nonsense claims like "releasing too much code" lol), how much longer do they need to continue before you'll stop demonizing them for both open-source and closed-source releases?

[Solar]

FWIW, my take on the above is that Microsoft became acutely aware that, between Linux becoming an on-par desktop OS, Microsoft failing in the mobile market, and actual OS becoming increasingly secondary to things like web technologies and portals, they started shifting their aim. The long time that 2001-vintage WinXP was the latest on the OS market, and the clear lack of revolution in the former cash cows Office and Studio since pretty much the same time, is a clear testimony to that.

The number of utter failures Microsoft has produced since, both in the application and the web technology sector, underlines how much they had lost orientation.

That my employer is still hellbent on pushing such a turd as Sharepoint despite nobody at the lower echolons thinking that is a good idea goes to prove how much weight the name Microsoft still has.

But it's dwindling.

Make no mistake, the new big crushing monsters are already out there. (Hello Google.)

[jojo]

So my takeaway here is that everyone loves WSL and wants to make sweet love to it every night? Is that an accurate assessment?

Okay, sounds good

/thread

[alexfru]

So my takeaway here is that everyone loves WSL and wants to make sweet love to it every night? Is that an accurate assessment?

Perhaps, not every night, but it may be quite useful for certain things, no need to run Linux VMs. However, there are a few things that I'm not sure of...

• Unicode support in the console (it was problematic up to Windows 7, not sure if there have been any improvements since, in general or for this feature specifically (there was no programmatic way to choose a Unicode font, only manual or undocumented ways))
• graphics (X, whatever), whether it's gonna be supported at all
• direct access to some select hardware (e.g. USB ports / SD cards)
• whether this whole thing is gonna last (e.g. XP mode was available only in Windows 7, DOS support worsened with every release after Windows XP and is unavailable in 64-bit Windows)

The last is probably the most important.

[zdz]

What they did was playing the business side of things perfectly. And their tactics haven't even changed that much over the decades: Making sure that their software was preinstalled. Making it hard and / or inconvenient to install or use alternatives. Providing "support", and marketing heavily, for businesses, where you had to convince only one or two manager types to have another couple of hundred wage drones educated into mentally equating "text document" with "Word .doc" and "spreadsheet" with "Excel", making any competition a second-grade choice and forcing everyone to play catch-up in compatibility instead of differenciating themselves in innovation. Playing the software patent game to perfection ("sign this agreement, sell out your IP, or be sued into bancruptcy").

Show me one alternative for Office that can do what Office can do. There is no suitable replacement for Word or Excel. You can even show me one that doen't try to be a Word clone, because Word clones are in essence playing catch-up by default.

Let's not even touch on Visual Studio. There is no IDE comparable to it.

I think it is going to last and they are going to really improve on it. The main idea is that they don't really care what OS you are using as long as you are developing with / for one of their system / platforms / whatever. And this is one of the things that will make it easier for people to use Windows instead of something else.

[onlyonemac]

Microsoft, on the other hand, is nowhere near as bad as you claim. As you've seen already, they've gotten very thorough about security, and provide good tools for administrating large numbers of desktop machines. They also (for even longer) take backwards compatibility very seriously, because their income depends on it. I would even say (disagreeing with Solar here) that they do provide some of the highest-quality tools and applications in some areas- DirectX is a far better API than OpenGL (and I say this as someone paid to write graphics code), Visual Studio is far better than Eclipse at editing and debugging C++ (feature and interface-wise), Office (especially Excel) is top-notch.

And yet just earlier today I was struggling my way through a Microsoft knowledge base article on fixing the "duplicate emails in Outlook 2016" which told me to use IMAP instead of POP3 instead of providing me with an actual bug fix.

Let me ask you something, onlyonemac: how many years and products does Microsoft have not to embrace-extend-extinguish for you to accept anything they do as less than the devil's handiwork? Or are you just a religious zealot who will shift the goalposts anytime Microsoft changes, hating Microsoft just for the sake of hating Microsoft?

There is no exact number in terms of years and products, because I haven't defined one and that's not how I assess a company's motives. Instead I look at the actual evidence, and the evidence suggests to me that Microsoft are continuing their "embrace-extend-extinguish" approach (or some variation thereof) to either eliminate or take control of the competition.

[onlyonemac]

Show me one alternative for Office that can do what Office can do. There is no suitable replacement for Word or Excel. You can even show me one that doen't try to be a Word clone, because Word clones are in essence playing catch-up by default.

Sadly that's not possible. While LibreOffice, for example, would be a perfectly acceptable application on its own, Microsoft have already got control of the office software industry and unless anyone can produce a fully-compatible re-implementation of their proprietary "office open XML" file format (which is impossible because it's a proprietary format that's moving faster than anyone can keep up) then their office software is "not as good as Microsoft Office", even if it's got a better interface (although some users have been brainwashed into thinking that a ribbon interface is more efficient to use, even though I've observed them using both interfaces and noticed that they perform the same tasks quicker with a menubar and toolbar interface) and fewer bugs.

[Rusky]

And yet just earlier today I was struggling my way through a Microsoft knowledge base article on fixing the "duplicate emails in Outlook 2016" which told me to use IMAP instead of POP3 instead of providing me with an actual bug fix.

And yet just the other day GNOME broke everyone's themes for the umpteenth time, while simultaneously removing fundamental features of its file browser; I have to lock Freetype to a release three versions old because they disabled a feature I use with no way to enable it again; for the longest time LibreOffice Calc's tabs were unreadably small because they wrote their own crappy UI toolkit and forgot not all horizontal scrollbars are thick enough to put text next to; Linux refuses to disable the touch screen on my laptop; etc. See, I can cite random bugs too- doesn't mean Microsoft doesn't make overall quality products. (Or that free software doesn't have quality applications either- you're the only one being an absolutist here!)

There is no exact number in terms of years and products, because I haven't defined one and that's not how I assess a company's motives. Instead I look at the actual evidence, and the evidence suggests to me that Microsoft are continuing their "embrace-extend-extinguish" approach (or some variation thereof) to either eliminate or take control of the competition.

In other words, "I'll know it when I see it and I've dug myself too deep in this hole to see anything." You've given no evidence that Microsoft is continuing EEE, only that you assume their next step is in that direction. Meanwhile even Solar agrees that they're "shifting their aim."

Microsoft have already got control of the office software industry and unless anyone can produce a fully-compatible re-implementation of their proprietary "office open XML" file format (which is impossible because it's a proprietary format that's moving faster than anyone can keep up) then their office software is "not as good as Microsoft Office"

The question explicitly excluded the problem of .docx compatibility, so your whining is irrelevant:

You can even show me one that doen't try to be a Word clone, because Word clones are in essence playing catch-up by default.

LibreOffice, for example, would be a perfectly acceptable application on its own, ... even if it's got a better interface (although some users have been brainwashed into thinking that a ribbon interface is more efficient to use, even though I've observed them using both interfaces and noticed that they perform the same tasks quicker with a menubar and toolbar interface) and fewer bugs.

So Microsoft is evil for creating the Ribbon interface (which, incidentally, I prefer and am more efficient with, without ever having been brainwashed), while GNOME is good for completely annihilating all the features and customizability you praise Linux OSes for?

[zdz]

Show me one alternative for Office that can do what Office can do. There is no suitable replacement for Word or Excel. You can even show me one that doen't try to be a Word clone, because Word clones are in essence playing catch-up by default.

Sadly that's not possible. While LibreOffice, for example, would be a perfectly acceptable application on its own, Microsoft have already got control of the office software industry and unless anyone can produce a fully-compatible re-implementation of their proprietary "office open XML" file format (which is impossible because it's a proprietary format that's moving faster than anyone can keep up) then their office software is "not as good as Microsoft Office", even if it's got a better interface (although some users have been brainwashed into thinking that a ribbon interface is more efficient to use, even though I've observed them using both interfaces and noticed that they perform the same tasks quicker with a menubar and toolbar interface) and fewer bugs.

Here you can't say that Microsoft is dictating what people need. People need x feature of Office because they actually need it. The interface thing is purely subjective as I can pretty much customize it the way I like it and for the features I use often I use the keyboard shortcuts + I tend to find that one obscure thing that I'll never use again pretty fast. The problem here is that no one can offer a superior product. Should Microsoft care about them when taking care of Office? No. Libre Office can't do half the things Office does, and most of those it can do are poorly implemented. I rarely came across a document edited in Libre Office that looks how it should look - why should I give it a chance? Isn't "screwing up all the paragraphs and lists and references" also a bug? Why is Libre Office installed by default on Ubuntu? Is this fair?!

I hope they don't destroy Windows scheduler by implementing Ubuntu on Windows