How Microsoft went from “Linux is a cancer” to “Microsoft loves Linux”
idginsiderpro.com
idginsiderpro.com
Some years back, one criticism I heard against Microsoft was that they threatened to assert patent infringement against Linux, but refused to tell the alleged offenders which particular patents Microsoft considered to be violated. And the end result was Microsoft (imho) legally extorting some Linux vendors for licensing fees.
If that story is true, then is Microsoft still engaging in that behavior or similar?
This report[2] from October 2018 seems to indicate that Microsoft no longer does this:
> Ten to 15 years ago we were very focused on the competition between Linux and Windows,” says Microsoft’s chief IP counsel, Erich Andersen, in an interview with Yahoo Finance. “But today Microsoft is a very different company. With the rise of our Azure [cloud services] business, Microsoft has become one of the largest contributors to open source in the world. . . . We’re in a position where, as [CEO] Satya [Nadella] says, we love Linux. We love open source. That’s a big part of the future of our business.”
[1] https://google.com/search?q=microsoft+android+patent+fees
[2] https://finance.yahoo.com/news/microsoft-may-relinquishing-b...
Anyone with first-hand knowledge care to comment?
Said the fox to the chickens. Come real close and let me tell you a story.
Is Microsoft still a 800lbs gorilla that does 800lbs gorilla things? Yes. But let's not pretend they employ the same strategies today that made them the bad guys at the turn of the millennium.
6 years and a single CEO does not constitute a rebirth. When the next CEO is still doing the same thing, you can talk about what Microsoft is.
Reminds me of when Visual Studio Code came out and some people denied themselves of what would end up being one of the best editors to come out in a while just because they refused to use something with such a Microsoft-ey name...
They aren't just a large company that people have an old grudge against. They're very specifically the people who kept us from having the option to not consider a particular company and still have any reasonable options at all.
Because this is all tangential to the fact that in a world that now moves one financial quarter at a time, six years is a lifetime.
MS's direction has been consistent and it's been good. If you're waiting for another six years to let MS prove themselves, it's on you, but again, plenty of other people will reap the benefits of not doing so
Microsoft is a company run by people, and those people have changed (the "they" you mentioned). Gates & Balmer are now nothing more than shareholders (Gates stepped down from the board in March this year). Believe it or not, it's people that determine how a company is run and the people running Microsoft today are not the people that were running it in the 80s & 90s.
Large business units with billions in revenue take years to change direction, and we've seen that clearly happen under Nadella.
I will admit that it says something that Satya was able to survive inside of Microsoft during the worst parts of this, but I don't know if that says more about how amazing Satya is, or that the internal toxicity is over-reported. We all know multiple people who have reported horrible things about Amazon too, and yet there are still people who enjoy working there.
Everyone has their own version of the truth, but once bitten twice shy, and I've lost count of the bite marks. Have your truth, just keep it outside of arm's reach of my nose, thank you.
And I'm sure it had zero benefit from exaggerating stories to increase readership. I've worked at companies that have been the ire of the press, and know first hand it's many degrees separated from reality.
> It wasn't just a despot at the wheel
Except it was. We've seen this consistently in leadership, including just recently WeWork, Uber and the U.S. government. When a nasty person is at top, it legitimizes nasty people everywhere else.
> We all know multiple people who have reported horrible things about Amazon too,
No, we don't actually know them. We read some news stories specifically centered around warehousing and distribution as Amazon grew from 34,000 employees in 2010 to over 1,000,000 in 2020.
This is rhetoric. Microsoft wasn't committing genocide, it just abused capitalism.
Yeah, but did they decide to stop shooting, or did they just run out of ammunition?
(I guess the news that a bookseller was the world #1 in cloud computing would be even more of a shock.)
In their quarterly earnings reports you can see that about 1/3 of their revenue is from "more personal computing."[0] We know that this category includes Windows licensing, devices (Surface, etc), gaming, and search. About $11 billion per quarter, on average.
They don't neatly break this down in their report, as far as I can see, but some other internet sleuths have made educated guesses based on things that are directly reported. As far as anyone can guess, Windows is no more than 10-15% of Microsoft revenue.
It seems fairly obvious to me that Windows is no longer the backbone of Microsoft. 10-15% still represents billions of dollars, but I think Microsoft would be perfectly content to replace all of that Windows revenue with O365 licenses or Azure billings.
[0]: https://www.microsoft.com/en-us/Investor/earnings/FY-2020-Q1...
But I think Microsoft is much less paranoid, simply because desktop is a much less attractive target. It's good owning it today, but why would any potential competitor go to war to establish a beachhead in a slowly dying market? Certainly the other big tech companies are much more potentially threatening to MS today. Nobody in 1999 could hit as hard as Google, Amazon, Apple or even Facebook could today.
But why would they even bother? Defending Windows is like Britain worrying about defending Gibraltar. It's still somewhat strategically important, but much less so than in the past. Because of that legacy, it's already defended way past the point of its value to any rational adversary.
Microsoft has more or less abolished Windows as a separate business unit.
https://www.ft.com/content/5bf8b952-3396-11e8-b5bf-23cb17fd1...
Both public and private companies can have business models that are driven by the morality of their founders and key employees but it’s still a business model.
Cygnus wasn’t more moral than say Oracle, they just had developed a business model around FOSS services that was unique, since they didn’t held any IP they didn’t had the tools that some people may consider immoral such as protecting their patents through law suits or worse rent seeking based on IP they hold.
Same goes for other companies today that primarily rely on FOSS software I don’t consider Ubuntu, Red Hat (even before the acquisition) or even Purism to be “moral” companies just because they can’t or don’t hammer someone with law suits or rake in royalties.
I also don’t consider opposition to companies that do to be a simple moral issue. I can and do have my own opinions about companies that I feel practice more rent seeking than actual innovation or any sort of productive activity but I don’t see it to be a particularly moral stance to want them to go into financial trouble because in the end the losers will be their employees and shareholders.
I much rather see companies like Oracle for example be successful based on true productivity than go into financial trouble.
But companies are run by people. I think it's easier to think that companies are more like profit robots to get past our qualms of what we expect of humans, but in the end they are made of and run by humans. So much ink of US history has been used to write about corporate scandals and the profiteering that we just come to expect it -- notably the 2008 recession was caused by pure unadulterated greed on a vast scale.
JPMC plead guilty to three felonies and basically got a handslap. Talk about moral hazard!
https://www.huffpost.com/entry/time-for-three-strikes-an_b_7...
PS. You might think that $35B is a large fine, but they have $3T under management.
https://www.google.com/search?q=JPMC+assets+under+management...
What most morals are is boils down to what behavior is acceptable and beneficial or to be more exact the set of incentives and repercussions that society enforces.
No one can really argue that downloading a movie or a songs is an objectively moral act, however most people can easily justify it as a victimless crime, some even justify it by claiming that the entertainment industry itself is immoral.
What you essentially have is people with targets and an incentive structure that is misaligned with what you or society at large consider to be moral.
If a business can make a large sum of money by doing something against the rules and the punishment is smaller than the profits it will essentially always lead to “amoral” behavior.
Until we can actually deprive bad acting companies of the ability to influence the political landscape, nothing is going to stand a significant chance of actually making a misalignment of moral/legal compass hurt.
You are distinguishing between Microsoft the corporation and persons who might rehabilitate, although how I read your sentence, I infer that you consider Microsoft a person, only a different kind of person to human beings.
I don't know if I'm right about how I read you, but I think I can comment usefully on a different distinction which I think is the most important one necessary to make in these kinds of comparisons:
A company is a legal person.
But it is unhelpful to think about a corporate person in terms with which you might regard a human being.
I find it very much clearer to think about the body corporate as taking on the character and nature bestowed upon it by its directing executive management / board.
Certainly it is by this means that any company has ever acquired a psychopathy.
AS several comments also say in different ways, the caveat is that once you replace the human animus that directs the behaviour of the company, you can completely change the behaviour of a corporate being.
What I think is being conflated here frequently is the organisational behavioural properties of any large organisations that have vestigial and inhibiting effect on corporate body character redefining changes.
These things are always something that I have to carefully unpick whenever you want to avoid confusing myself and that's after 35 years of analysing corporate beings looking for long term bets. It's not easy or at least it never got easier for me. I wish I could weigh in on the issue which I think everyone cares about most : the sincerity of Microsoft and even the question of whether Microsoft is capable of sincerity. So far I can only agree with a tangential comment made last week, how the old evil Microsoft at least sure documented everything far better. If they documented like the Technet I remember, for the whole WSL game, now that might convince me they mean well towards the open world. But somehow I think Microsoft embraced and only extended the documentation shame...
Usually you try to stay as far away from that person as you can. If they want redemption they should find it with a new set of people. They set that bridge on fire while we asked them not to. They watched it burn. They gloated the entire time. The ashes have washed away but many of us remember.
Microsoft Office for Mac was first released 31 years ago. It’s nothing new and hence is orthogonal to anyone’s subjective impression of Microsoft’s engagement in EEE.
Microsoft has been the largest ISV in the Apple ecosystem by far, for decades.
I have to imagine that that was an innocent typo. Do you think so? Or do you think that it wasn’t, and so pointing it out helps to discredit the author?
Habitual abusers of relationship always promise that they will be better the next time and for a few days they are indeed better behaved. After a while they fall back to the same patterns
If Microsoft bankrolling the SCO lawsuits is the stand you want to take then be my guest. It was almost 20 years ago, the entire leadership of the company has changed, their culture has changed, and they are pretty much unrecognizable at this point.
So, I'm already "boycotting" almost every company you listed, without even needing any ideological goal to chase after. The fact is these giant evil mega-conglomerates aren't that great at making things. The more you avoid them the better off you will be.
Now, everything is connected to everything else, so I think the following rules are reasonable for avoiding evil companies:
- If it's a commodity, and it's mixed in with all the other commodities, it's probably not practical to separate the tainted stuff from the cleanly produced stuff.
- If it's a specific company with a specific product that they have specifically done evil stuff to push, that's an example of something to avoid.
- If the same people are in power as when the company was evil, it's probably still evil. If everything but the name has changed maybe they can get another chance. For degrees of change in-between those two, use your judgement.
He joined the company in 1992, when it was as bad as it got and did not see any reason to go work somewhere else. He was totally free to pack up and go but choose to hold on and continued to work there all the way up to today. That makes him at a minimum a silent supporter and at worst complicit. Neither of these speak well for his ethics.
Nadella gave up his Greencard to switch to the H-1B program in 1994 to bring his wife to the US (and surely this unprecedented switch was some time in the making). Subsequently he relied on Microsoft to sponsor him once again for permanent residency.
In that situation no immigrants switches employers regardless of their ethics. It often isn't even possible to make such a switch.
You're certainly entitled to your opinion, but I think it's a little silly to write them off over old stuff that they're no longer engaged in when there are bigger fish to fry. They are making pretty good products these days and, by all appearances, are one of the better players in big tech these days.
I hated Microsoft back then too. They were about as bad as I thought a company could be. At least until I lived long enough to see Amazon, Google and Facebook - or learned about the horrible things other household brands have done.
I don't know about Microsoft though - the CEO changed, but what about the rest?
The leadership (both the board and the executives) have changed since that time. I assume you also refuse to buy BMW and Volkswagen cars and actively petition governments worldwide to boycott them because "the company" supported the Nazi's during WWII?
Same with Toyota, Mitsubishi, Hitachi, etc?
Actually, they are going after Android handset makers. Who are paying Microsoft for each handset sold didn't you know Microsoft owned Android?.
https://www.theregister.com/2014/10/07/how_much_is_that_micr...
We don't know the exact specifics of any of the patent deals, though, because neither Microsoft nor its new Android-flogging partners are telling. For the most part, we don't even know which patents Redmond is actually licensing.
Few things are as scummy as legal extortion, with contract clauses that prevent that extortion from becoming public. At the very least, refusal to reveal which patents something allegedly infringes, should result in any subsequent patent claims being invalidated. It's like being accused of trespassing, but the property owner refuses to tell you what part of the land is theirs.
How would this work in court?
"We're charging you with breaking some laws, but we're not going to tell you which ones. How do you plead?"
Obviously, Microsoft disclosed what patents they believed were being infringed when they decided to actually take people to court. They decided not to disclose that when negotiating licenses, which is actually legal. Contracts can include non-disclosure agreements that would prohibit you from telling Linus what parts of Linus to change, and you're allowed to vaguely threaten people as a negotiating tactic, as long as there actually exists some underlying infringement which Microsoft could reasonably have believed to have occurred. (Otherwise, it would be extortion.)
Patents are public information, including what patents Microsoft has been issued, so the law would assign device manufacturers the burden of determining which of Microsoft's patents have or have not been infringed by their products. Microsoft is not legally required to do their work for them and tell them what they're infringing. All of the parties involved could have conducted patent reviews of the Linux kernel or AOSP to find and remove the infringing bits Microsoft was suing over. Instead, they decided to individually settle because the engineering and legal resources required to fix Android's apparent patent problem were less than just paying Microsoft for a license.
People tend to anthropomorphize companies. But companies are not people, they're organizations. They're driven by a mix of individuals who come and go and whose values, vision, motivation and influence shape the direction and actions of the organization at time T.
Because of this uncertainty, never trust an organization with anything forever and learn to be comfortable with the idea that it will change. The promises of one CEO have no guarantee to be upheld by the next.
A consequence of this is that an unintentional mis-step by someone of relatively low seniority can be seen as intentional strategy from the overall entity.
https://en.m.wikipedia.org/wiki/Fear,_uncertainty,_and_doubt
The first is regarding MSFT patents in Linux itself, and the SCO vs IBM/Linux courtcase of the early 2000s involving Linux. That was basically killed off by a 'put up or shut up' action by IBM and the Linux supporters against SCO/MSFT and their supporters.
The second is regarding MSFT patents in Android which is based on Linux. In that case MSFT threatened to sue manufacturers who were using Android. They found it cheaper to pay royalties to MSFT than having to battle through the courts. https://www.howtogeek.com/183766/why-microsoft-makes-5-to-15...
Or "embraces" (with the intention to extend)?
The Microsoft from 20 years ago does not look anything like the Microsoft of today, and trying to pretend it is is lazy and dishonest.
VSCode, Edge, .NET Core, WSL, DX12, deliberately crippling cross-platform Vulkan drivers, ...
And then you have the stereotypically anti-user bullshit, like the telemetry and enforced upgrades.
All things generally loved by the community (except Edge).
> deliberately crippling cross-platform Vulkan drivers
I find nothing doing a quick search for this. Do you have any evidence of this?
> like the telemetry and enforced upgrades
Neither of which have proven harmful to their users, other than the inconvenience of a forced restart. I'd rather have forced upgrades than the mayhem that existed before.
How do these have ANYTHING to do with Microsoft from 20 years ago?
I don't think it's a good faith argument to redefine proven harm to exclude all of the harms that people have complained about.
Where are the harms? No, really. It's bad faith to claim people grumbling about inconveniences are somehow being "harmed".
Also both telemetry and forced updates impose a giant problem for all administrators in every coporation i worked for since widows 10 was deployed, because they want to get rid of it from a security point of view but management refuses. A former colleague of mine quit his job because he couldn't handle the responsibility and had develeoped serious problems with sleep and general wellbeing
I absolutely hate Edge, and feel the same away about being unable to uninstall Safari on iOS/macOS.
> directx12, because why support vulkan when you can just continue locking as many programs to your platform as possible
They've invested decades into DX, I see no reason they should abandon that IP.
> Also both telemetry and forced updates impose a giant problem for all administrators
Windows 10 has group policy controls for IT admins to disable forced upgrades and telemetry. This sounds like complete dramatization.
Only for some SKUs that not all organizations are eligible to license; even for those that are it represents additional expense. And even then, only partially. Telemetry cannot be completely disabled in any Windows 10 SKU. Forced updates can be controlled only via WSUS (by declining them).
> This sounds like complete dramatization.
Please do not inject your value judgements into discussion.
Who is prevented from Windows 10 Enterprise licenses, other than those already under specific sanctions? Windows 10 Server, Education, and IoT also support disabling telemetry.
> Forced updates can be controlled only via WSUS (by declining them).
They can be controlled via Group Policies, not to mention Microsoft abandoned (most) forced upgrades with the May 2019 Update. Explicit action is now required to install these updates (like you mentioned). The only updates that cannot be disabled are end of service updates, which are every 18 months.
Care to explain what's wrong with them? What's anticompetitive about them? Or you just don't like them because "reasons"?
> DX12
What is wrong with it? It's not like it's forced down your throat like Metal. You still have the choice of Vulkan and OpenGL
> deliberately crippling cross-platform Vulkan drivers
Do you mind providing citations, instead of just causing flame wars with no solid argument on your side?
They're all sold in wildly misleading ways.
VSCode is open source according to the website, but the FOSS build is missing a lot of the advertised features (like, say, the extension manager...).
Same with .NET Core. Sure, the compiler and core libraries are open source (if you can find the real thing, rather than the umpteen facades that they pollute their repos with)... But core tools like the debugger are still closed and deliberately locked to only work with (the branded, closed builds of) VSCode. That's not a huge problem if it at least had proper support for printf debugging (like Rust's `Debug`). But of course that's not a thing, because why would you ever use anything else than VSCode?
WSL is sold as a drop-in compatibility layer, but WSL1 was missing a lot of core features (hell, SQLite wouldn't even run with its default settings). WSL2 is somewhat better, but still does a bunch of shady crap (no systemd or equivalent, DirectX-on-WSL, etc).
> What is wrong with it?
For one, there was just no need for it. The only reason for it to exist is to make it harder to port stuff away from Windows.
If they had a problem with Vulkan's API (haven't really used either, so can't comment on the exact differences) then they could have designed it as a layer on top of Vulkan from the start. DXVK already proved that it's a viable approach.
> It's not like it's forced down your throat like Metal. You still have the choice of Vulkan and OpenGL
Xbox?
It's only a few years since they replaced Steve Ballmer. Nadella won't be there forever.
I'd be more cautious too. Big companies have their own interests and it's usually not aligned with ours. I'd be less cautious with a company that has resisted going to the dark side. Microsoft definitely has been there.
As an analogy: A criminal can serve their time and be released. But will you share a house with them?
And? Someone else will take his place, who isn't Gates or Balmer and will have their own way of doing things. There isn't some magical Microsoft central force that pulls things a certain direction, it's people making decisions.
Hate to break it to you, so too does every other company you buy goods and services from. Do you take similar precautions for them?
Where do you get this from? Microsoft has reported annual increases in Windows Server revenues for years. Does no one do research before posting about MS? Azure's market increased by supporting Linux, doesn't mean the Windows side suffered.
> do you think they wouldn't be EEE-ing?
Show me one example of EEE in the last 20 years. Go ahead, I'll wait.
What I loved about Linux was that it was not encumbered by attempts to lock in intellectual property (with the idea of future monetising, of course). Yet all the big players are pushing lots of kernel patches to suit their own cloud-based use and ecosystems. Distributions are trying to lock-in like all the commercial players now. Consider Ubuntu with its snap store that can only be run by Canonical.
I really don't like where Linux is headed now that Microsoft and the other big players love it so much... I'm strongly considering moving back to FreeBSD for a desktop. Yes there's Netflix with some input but that's about it.
That's a Canonical thing, what other Linux vendor does this?
Distros have less lock-in than ever thanks to userland defragmentation.
With this strategy, it doesn't matter what the next platform is because MS will be there.
It's the Microsoft Mantra.
Their turn around with Nadella is incredible and must be a really interesting story.
https://stratechery.com/2018/the-end-of-windows/
For the people that argue that CEOs don’t matter, this is another compelling example of how important top down direction and leadership can be.
Who argues that? Apple already gave us a solid proof of difference between a manager and great CEO decade ago.
Stuff like releasing Vscode and Powershell under the MIT license, porting SQL Server to Linux, developing WSL… Did Ballmer contribute to/enable any of that stuff?
Those specific things didn't happen under Ballmer or kick off under Ballmer, but some their predecessors did (the earliest open source projects happened under Ballmer; the Monaco Code Editor was started under Ballmer even if the trajectory to it becoming VSCode wasn't likely clear back then). You can't necessarily draw a straight line from Ballmer era to those those projects that you like, but you also can't connect all the dots without including some of the dots from the Ballmer era too. Ballmer wasn't the great villain a lot of retrospectives seem to want to make him. He was an infectious cheer-leader and sometimes his cheer-leading led him astray (especially in trying to force people to use Windows/Office, instead of trying to encourage people to want to use Windows/Office; probably the biggest distinction between Ballmer and Nadella, and even then it doesn't seem that wide of a gulf as people make it out to be), but his intentions were in the right place and some of his instincts still lead the company in the right general direction (eventually).
I don't buy the 'things were in motion' while he was there, I think if they were it was in spite of his direction, not because of it.
It helps that Jobs himself hated DRM on music so much that he was willing to give up a legitimate competitive advantage to improve the customer experience.
The difference in attitude is staggering and will not go away.
You blame Microsoft because game companies decided to utilize their DX tech? That's a bit of a stretch, but to each their own.
Nothing stops them from being a good citizen when it comes to GPU APIs too. Except they profit form their lock-in and are hostile to competition (including Linux). So no, MS aren't friends of Linux. They only pretend to be friendly, when it suits them.
Considering they gave the world DX which has been a net win for gamers, yeah definitely not a good citizen.
Has it? What do you know about the alternate universe in which Windows and Xbox did not have a proprietary grpahics API.
I love Linux, but lets stop pretending desktop is a gaming platform when it's not.
There is none. Microsoft won't exactly do a press release to confirm exactly that, but the timing of events is suspicious. Things like this do not just happen randomly.
Android mobile gaming is a multi-billion dollar industry. Guess what isn't? Linux desktop gaming.
> I know plenty of people who maintain a Windows partition purely for games
Selection bias. I'm guessing you know a lot more Linux users than the average person.
> If Microsoft is actually buying studios intending to remove Linux support
My guess is these games need to be compatible with Xbox and Windows, considering that's Microsoft business. That sounds like a completely practical business move.
The name for this is platform politics (and I find it disgusting as a Linux user). Stadia came out having way less users than desktop Linux and game companies run to release for it, while continuing to ignore desktop Linux users. It's not about the size of the market like some like to excuse it with. It's about expected to be paid by "platform owner" or something of the sort.
Linux has no owner, so these companies are at loss from whom to demand extra money, even if they can make profit from actually selling games to Linux gamers. They don't grok open platforms. Better companies which aren't into platform politics release games for Linux fine.
So part of the blame is on those who are hung on platform politics. But it doesn't excuse anti-competitive behavior of MS in the least.
No, the name is business. As it's always been. You seem to have a grudge against companies doing what's best for their business.
> Stadia came out having way less users than desktop Linux
These are not even remotely related. Stadia is a service that is immediately available to a huge market share, and actually has legitimate business opportunity behind it.
> Linux has no owner, so these companies are at loss from whom to demand extra money,
Games is a business, and you go where the money is. Guess where it isn't? Surprise, Linux desktop gaming! Plenty of people are making money (billions) on Linux gaming, just not on the desktop. Because there's an actual market.
And yes, I find platform politics to be disgusting and anti-user in essence. Stadia isn't huge as desktop Linux, not even close yet. And potential is all abstract. Linux has also a lot of potential, if these would release games for it. They don't care.
It's not about potential or size of the market, it's about being paid extra incentives by the platform owner. I.e. dirty platform politics.
> and anti-user in essence
How is it any different that cars having different parts? Is Ford anti-user because I can't snap in Toyota parts?
> Stadia isn't huge as desktop Linux, not even close yet
The potential market size for Stadia is exponentially larger than Linux desktop, and is a safer risk for game developers and publishers to get behind than Linux desktop.
> It's not about potential or size of the market
Of course it is. What else is it about? Google is willing to make the investment to make Stadia successful, to create a thriving customer base which in turn needs content. That's good business. No one is willing to do that for Linux desktop.
Again, focusing on the current state and not what some eventual steady state might be if Linux were better supported.
> Selection bias. I'm guessing you know a lot more Linux users than the average person.
Definitely true (and I had already explicitly called that out), but the fact that those people exist at all lends plausibility to the idea that the set of people who want Linux but want Windows games might be non-negligible and points to the logical flaw in using the current count of Linux gamers to determine what that market might look like if Microsoft weren't interfering (if they are, not sure on that at all -- taking ancestor comments at face value for now for the sake of argument).
Can we stop beating a dead horse? There is no Linux desktop market. The year of the Linux desktop isn't here, unlike mobile. Microsoft isn't interfering in anything, there just isn't good business incentive for Linux desktop gaming. Linux desktop gaming will take off when Linux desktop usages takes off, not the other way around.
In fact, if you want to take aim at anyone it should be NVidia and AMD for Linux driver support.
I did, actually, seeing how I'm the only one of us providing any numbers or research. You're trying to appeal to emotion, and it's not working.
Businesses have decided not to jump on the Linux desktop market, despite you feeling entitled to their services and labor.
That's totally going to convince anyone who can look at actual stores that sell Linux games which supposedly shouldn't exist according to Linux deniers.
It speaks volumes, when to cancel such a product a new owner, who happens to own one of the competing plaforms, has to do that by decree, with no regards to business results.
1. Opt-in to stating your operating system(s), and
2. Put ONLY Linux, not Windows.
Otherwise, you are counted as a Windows gamer. Spelunky 2 developers just released their game without Linux support because of this same misconception.
Do you have a better source on the scale of Steam that can represent a accurate market share for Linux desktop? I'm willing to wager money this number is pretty close to accurate.
DX12 is mostly a clean slate, just like Vulkan.
> Show me examples of Microsoft using coercive practices to make game developers use DX.
Xbox Windows Store
Both prevent you from using open graphics APIs.
Who would have thought why... coming from the company that loves Linux.
So of course they love OSS now that Azure is one of their largest money printers, it's no longer a threat & they've become one of the biggest beneficiaries of it.
And that's not completely wrong.
Also I cannot recommend enough - get a PiHole. That one will help you with Android crap from Google a lot and you can focus on your work instead of policing everything your spouse/kids do in your household.
I was always a windows person on my main computer and I liked 7 a lot but it still didn't quite feel right for devs, but windows 10 really makes us feel at home.
Also you went with a surface pro, which has been a genuinely top of the line hardware serie from microsoft
The dev tooling on Windows for things that aren't .NET has gotten much, much better in the last 10 years. Sometime in the last 4 years I stopped feeling like I had to have a Linux installation available to switch for my programming needs, and I've mostly just been using WSL instead. I still keep the Linux installation around, but now it's months in between when I boot it up.
2.Microsoft loves money.
3.Whatever makes them money gets their love.
4.Making money is the primary purpose of a company.
I'm guessing their current response would be "we have Microsoft 365 for the web" which is becoming less and less adequate.
They'll eventually need to decouple the two and sell them independently. At the end of the day, they're a commercial software company - platform wars aren't in their interest.
You are right that they are a commercial software company but they are trying to keep the Windows as "the platform" for enterprise companies otherwise it will be too hard for them to sell everything as a package.
office 365 for web is unfortunately not the same experience.
My father (70) and his cohort will basically only use outlook while my nieces and nephews (~18) when they get into the business world, they're not only unlikely to have any loyalty to outlook and office but they're unlikely to use email as a first tier communication and may not even use traditional word processing packages as a primary method of long form expression.
Nor will they have any of the branding impressions my father's generation had. A chromebook, macbook, or really anything else will be accepted as "alright, I guess I'm using this".
The desktop metaphor is quite stable as is the software. They will be, and arguably have been for ten years, broadly interchangeable.
Microsoft's current positioning is like 1980s IBM. The mainframe strategy worked so long as the people running the business thought they needed mainframes. As they retired out, IBM had to go elsewhere for profits.
And I'd bet the farm Microsoft knows this. They've diversified into video games, source code management, cloud computing, etc seeing this future ahead.
Also this isn't new for them. They're replaying their strategy from 1975-~1990. They never had the better products, they won because they had the better strategy.
Office was a good cash cow, but it's increasingly becoming a harder fit
When I was young and starting out in a corporate job, I also didn't have much loyalty to anything. This wasn't a function of a new age of tech Aquarius, it was simply because I was young and didn't have experience - no basis to form loyalties.
Youngest workers may not realize the value prop of Outlook and Excel fat clients simply because they haven't been hit with requirements that promote or even compel their use, but in the meantime those young workers are never getting highly proficient with the basics of using them. There's a reason more seasoned knowledge workers (not 70yo, but 30-60 for sure) rely on them beyond inertia - broader requirements, favored features, road experience and ubiquity.
Similar angle on your comment about platform loyalty - a young person won't be invested in a platform until they have the experience to recognize the penalties of swapping. Since I mentioned Excel already, keyboard shortcuts in Excel change across platforms. If Excel is considered a given part of your toolbox, this is one of the most aggravating things about swapping OS's periodically: your muscle memory doesn't apply. That's a real hit to productivity. Young people don't recognize this until they develop a skill level that will be impacted by these kinds of changes and are busy enough to not have time to deal with them.
So I wonder sometimes when I hear about that attitude - are they shooting themselves in the foot by not training early on or even having awareness of the potentially more advanced tooling options?
When I start getting into new hobbies I look at the tools the more advanced people use and learn what may (or may not) make them favored - as well as consider new ways of doing things. I also try to start on the better tools even if the learning curve is higher just so I can start racking up XP with them more quickly - so anticipating that a time will come when I too will leverage and appreciate the features, I'll not be starting from scratch. That's worked well.
Pittman shorthand, for instance, was a pretty useful system that only enthusiasts use these days. The most trivial change that axed this was purely social: men learned how to type.
Similarly, business is being done in different ways that will make other old useful tools less applicable.
The march to the grave starts at the crib, sorry about that.
Look at games. Civ wasn't released on Linux till Civ 5, that too because of Steam OS, four years after the Windows version came out.
Apple released Safari for Windows but never bothered with Linux.
They're building a new Fluent UI for React Native library and it's designed to work across Windows, macOS, iOS, Android. Microsoft Design has blogged [1] about it being the future for the Microsoft 365 initiative (of which Office 365 is a subset).
They've also sunset the "Online" branding for Office Online. It's now simply "Office". [2]
I think they're intentionally blurring the lines here to eventually have everything converge. And save many, many development hours while at it. I wouldn't be surprised if the "Windows Office" will become something hosted in their new WebView2 control (Chromium-based Edge renderer).
[1] https://medium.com/microsoft-design/how-fluent-ui-unlocks-th...
[2] https://techcommunity.microsoft.com/t5/office-apps-blog/why-...
2. MS lost the battle everywhere else.
Software companies want to attract developers and have good relationships with them.
- <-- We are here.
Extend
-
Extinguish
I used various flavors like Ubuntu for the past 8 years and finally got tired of the half-finished UIs, random bugs that were never fixed, and breaking issues with WINE when I upgraded (and had to spend hours of my free time trying to figure it out).
This, in addition to all sorts of driver issues over the years where I had to download test drivers and compile it myself to get it working has made me never look back.
My new setup is a windows 10 machine with the latest version of Ubuntu running (command line only) on Vmware workstation.
I get all of the stability of a windows machine and can still run all of my favorite Linux apps.
The Windows subsystem for Linux also works pretty well for most things.
I also wonder if it has to do with the anti-commercial stance from most people in the Linux community. Besides a couple of niche apps, most open source is not polished in any way and since many OSS developer have day jobs, have no real incentive to fix a specific bug/issue.
I much prefer Linux, not just for development but for every day usage.
Interestingly this matches my experience of Windows 10 rather then Gnome or Plasma although they are all rough around the edges.
Let's not forget the "for my use case, given my technical prowess" is implicit.
I would rather spend the time on my own projects.
On Fedora, I can do this with Ctrl + Tilde. On Windows, there is no shortcut for this. On Mac, this only works if the app is not in full screen mode.
On multi tab you switch between them with CTRL + W.
On MDI you have to disable grouping on Windows settings and then ALT+TAB will let you switch between them just like if they were separate.
I believe this whole 'M$ loves Linux' is a Trojan horse and we should always remain skeptical until the system changes.
The question is: what are M$'s angle(s) - in the long term?
I think the biggest one is ideology.
M$ will say there are some 'limits' as to what can be open sourced... But those limits will be limits imposed by the current Capitalist system, not the organic human limits. Intellectual Property is better termed 'Intellectual Monopoly', or even 'Intellectual Slavery'.
Really, really .. more like shifting it out the door if it compiles sucessfully!
Most businesses follow the Jerry Maguire theory. "Show me the money!".
When it became abundantly clear that Linux wasn't going away, and that cloud computing was growing fast, Microsoft pivoted.
That plus Ballmer leaving along with all his preconceived Windows Everywhere notions.
MS losing mobile killed the idea of Windows everywhere and once they lost that they had to play well with others or risk sales in other places.
Somehow I don't think that's what any "Linux on the Desktop" people had in mind.
As a daily user of Linux desktops for over a decade (and known to be stick-in-the-mud about my desktop experience), even I have awoken to some shifts in my perspective and expectations, but only in an after-the-fact manner.
That's true of every Linux distro as well. Ubuntu pretty much scrapped all the shitty looking window managers of RedHat (FWHM, FWHM95, AfterStep, etc.) and replaced it with another one that looks nice but has shitty usability.
Building a good looking and functional window manager on top of a Linux kernel still is a form of Linux in some sense.
Chromebooks aren't going away, non-Chrome OS Linux desktop usage has seen quite a large uptick in recent years (likely due to ease of installation and compatibility tools like Proton.) Linux, even Ubuntu, has over the years been a headache, however Pop!_OS is the best OS for software development today in my opinion. Regardless of distro, hardware support has greatly increased (as opposed to 2010, where getting a Wi-Fi driver or trackpad to work appropriately on Ubuntu was a chore.)
The primary way people interact with the internet today isn't through a system running Windows, it is through an Android or iOS device. If I have a budget of $200 for a laptop for school, it seems improbable that I'm going to go for a Windows system over Chrome OS unless there is some custom software I can't live without.
I'm sort of outta my lane here, but it seems to me that Microsoft's cloud offerings are going to be far more lucrative and important than revenue earned from desktop OS licensing at some point. Tools like VS Code integrate well with Azure for a reason. Linux dominating the server space isn't something Microsoft is ignoring either.
That's dead now. People don't care. They buy a phone because it has a higher resolution screen and then run it in low-res because that's the default and they don't know the difference. Kids in California fail AP tests because they don't know what a file format is and try to submit HEIC images to a web form that only takes JPEG and PNG.
I agree somewhat but I think the AP test website makers should take some responsibility too. Limiting the upload to only JPEG and PNG is silly IMO. Wouldn’t really cost them anything to install ImageMagick on their server and accept a much wider range of file formats.
ImageMagick can read and convert pretty much any image format you throw at it, check out the list of formats here: https://imagemagick.org/script/formats.php
And yes, ImageMagick supported file formats include even HEIC.
I mean, if we are going to talk about having knowledge about file formats and technical details about computers, perhaps the kids aren’t the only ones lacking – so too do the people that keep insisting on only PNG or JPEG in 2020 IMO. Why force everyone to deal with these things when the server could handle much more of it automatically?
Or... Use proper HTML to limit what file types you site accepts so users can't upload files you don't support. If a site says it only supports JPG/ PNG on the input element, Safari automatically converts it to JPG.
"People" in general were never computer savvy. It just wasn't a thing. We went from completely clueless in the 80s/90s to stumbling and frustrating in early 00s and we've pretty much leveled off at marginally capable for the past 10 years.
> That's dead now. People don't care. They buy a phone because it has a higher resolution screen and then run it in low-res because that's the default and they don't know the difference. Kids in California fail AP tests because they don't know what a file format is and try to submit HEIC images to a web form that only takes JPEG and PNG.
Why on earth should people want or need to deal with nonsense like file formats? This kind of nonsense is what professionals get paid to make clear to end-users. That is a huge part of our job. The case you are talking about is a pretty bad failure on the part of the company running the tests, not a problem with the end users.
> They buy a phone because it has a higher resolution screen and then run it in low-res because that's the default and they don't know the difference.
...neither do I. I have a generic Pixel 4a, $350 off the shelf phone, works great (really - it's nice). I've scoured for this setting, where is it located? I can't find a setting to change screen resolution available on my device, or even if it's possible. It looks great though just the way it is out of the box, I don't know the difference either.
> Kids in California fail AP tests because they don't know what a file format is and try to submit HEIC images to a web form that only takes JPEG and PNG
Apple dealt them this card - you're blaming people for not knowing that a single vendor has introduced a new (image) format by default which really, nobody else is dealing with (I guess?) the way Apple folks are; is it fair to lay this blame on the kid, or does it really belong to Apple trying to push a standard before it was globally (widespread, 80% of the web) adopted? PNG took years to get into tooling, if you recall - it just didn't poof appear, there were many years where JPG and GIF were the only upload formats - in 2020 you're now taking PNG for granted. :) Some day HEIC might be there.
This was Samsung I think who launched a phone, bragged about it's resolution, then shipped it with that resolution disabled so they users would get better battery life out the door. Terribly user-unfriendly and a horrible example since users who buy a high end smartphone would expect the phone ship with it's advertised resolution enabled.
> Apple dealt them this card - you're blaming people for not knowing that a single vendor has introduced a new (image) format by default which really...
IMO if you are passing out blame, look squarely at the developers of the web app who failed to create a standards compliant site. If your service is only capable of receiving specific mime types, you specify it on the input element. `accept="image/png, image/jpeg"` If you fail to do so, you will get arbitrary, random file types. If this were a desktop computer, a student might have sent photoshop images or TIFFs. If they'd specified what file formats their web app accepted, the iPhone would have sent JPGs.
iOS has ~83% marketshare with teens in the US. The fact that AP failed to QA input validation on iOS is an affront to the basic responsibilities of product quality. Submitting an HEIC file would cause the page to time-out with no recourse.
Compounding this is Apple’s decision to abstract file formats to whatever degree possible in iOS. Unless you are an avid tech blog reader or had run into some compatibility issue beforehand, iOS would have never shown you the text “HEIC” anywhere in it’s UI.
What? The CLI was never going mainstream.
In the context of the discussion, Windows does in fact have a significant end-user market share threat posed by alternatives like Linux, specifically in the cases of Android and Chrome OS. They are eating Windows lunch.
Windows desktop is still as massive as ever. Chrome OS has a 0.42% market share, and Android is not a Windows competitor (it's generally a companion).
Ballmer era Microsoft was dogmatic "Microsoft Everywhere". Post Ballmer Microsoft is pragmatic. It's a huge contrast. Desktop Windows is no longer the center of Microsoft's strategy, they are now much more cloud focused. Office and Azure, are their big profit centers. Ballmer would be rolling over in his grave[1] if he knew how far down the pecking order Windows had fallen.
[1] I know
https://gs.statcounter.com/os-market-share/desktop/united-st...
I suspect we'll see this creep up even further in the coming years, if Apple releases cheaper MacBooks (~$700), in response to Apple Silicon lowering their manufacturing costs. They've dramatically lowered the costs of the entry-level iPhones, iPads and watches and I don't see why that same strategy wouldn't come to the Mac.
I'd have to imagine that Microsoft has been thinking of what their Desktop OS play would be at that point, allowing windows to simply die a slow and profitable death would threaten other business lines such as Gaming. Non-Apple hardware vendors will be looking for OSX like systems to compete with a possibly dominant OSX platform.
Only so long as it recognises a potential gain.
Linux just happens to be Microsoft's fly for now.
Ballmers tenure did almost irreperable damage to the thesis that Microsoft was a modern technology company that could operate outside cloth-walled cubicles and quarterly corporate financial planners. With every advance in Linux, Ballmer doubled down on in-house developed analogues and near-Khrushchev rhetoric against not just the OS but the fundamentals of open source. While Google and AWS catered with open arms to all, Ballmers vitriol led redmond down a myopic path of scorched earth ideations in the service of a dead business model.
A cogent argument could be formented that XBox revenue and corporate license renewals were the only thing that kept Redmond from cutting staff and folding divisions during Steves tenure....divisions that in bitter retrospect should have folded far sooner than they did (Phone and the physical store for example.) Steve kept his corner office and the favour of the board thanks to quietly dipping into the coffers of these underappreciated touchstones of the company but in the 21st century...these werent growth.
After saying in 2008 that he intended to remain CEO for another decade, Ballmer announced his retirement in 2013. Microsoft was running on fumes and while Steve managed to coast the hills, Satya was ready at the pump.
The insensitive truth is that the real cancer all along was Steve.
I don't know Satya from the inside, so I will not say anything, other than he does not seem to have initiated a lot. I might be wrong.
You don't turn a business as Microsoft into a super attractive stock overnight. But Ballmer, Kevin Turner, and Guthrie to some extent put in a lot of effort to create what is now being run by Satya.
The Balmer-scream years were very formative to my outlook on the tech world, graduating high school in 2005. Only recently did I bother to see what MS had been up to in the past 10 years ( was down a different industry rabbit hole ). A colleague told me 3-4 years ago what MS had been working on and it seemed to me like there was a light switch flipped somewhere.
Their (once) current bread & butter paradigm (Windows, mostly on desktop) wasn't a growing market. They recognized this and adjusted course.
Call it what you want, but such decisions are the difference between sustained success and fading away.