It is true that Microsoft has been behaving well lately (GWX aside,) but it is also the case that their getting-away-with-it power is at an all time low.
Microsoft would benefit if they could do all of these anti-consumer things, but that recent stampede towards Linux proved that the tennantry can be easily spooked and dangerous when on the move. Developers are leaving windows to work on Linux and Mac, to the extent that I have not seen a start menu on a single laptop in the entire CS department. If, at some point in the future we all ended up locked in to .net, I would be surprised if Microsoft's management continued their support of these charitable projects.
In that light, I think it would be wise for us to avoid getting locked back in - so that Windows might continue moving towards being a nice place to work.
>but it is also the case that their getting-away-with-it power is at an all time low.
Actually, most, if not all of the provisions, of the DOJ settlement have expired. This is why they can ship an AV with their OS and tell OEM's to cut it out with the crapware and offer special installers to remove this stuff.
I think they're just too weak to be the MS of the 90s. They just have to play ball to be competitive like everyone else. This is a good thing. Yet somehow the hysterical anti-MS attitudes are still here. Look at this thread. Its 90% conspiracy theories and axe grinding from things a decade out.
It's not just MS that has changed. Every aspect of every industry that MS competes in has changed.
To argue that the same business practices from the 90s apply today, especially in the technology sector where the entire plane shifts radically every other year... that's just crazy. I don't see why so many people buy into it still.
I suspect a lot of is that people formed that opinion of Microsoft in their formative years. This generation we commonly see on HN has this childhood definition of Microsoft as evil internalized at a pretty deep level. People look for reasons to justify their beliefs more than they look for reasons to challenge them.
There are companies I got a bad impression of many years ago too that I am hesitant to reevaluate. I think it's pretty normal/natural, even though it's kinda unhealthy from an intellectual standpoint.
> Its 90% conspiracy theories and axe grinding from things a decade out.
There are 2/64 comments expressing this view. You're being hyperbolic. The main theme recently has been "Wow, I never thought I would like MS again, but they're impressing me!"
If anything, MS is the new darling brand of the big five here.
If anything, MS is the new darling brand of the big five here.
Eh, they're not as evil as they used to be. But they're not doing as much as I would like to protect user privacy either. So I'm not super happy with them.
Though I really the the Hololens, I hope that goes into mass production soon with a good field of view. That product might get me to buy into more of their ecosystem.
You'll stay with the tech you locked yourself in.
I work in one of several buildings and macs there are just to let others know that you are a manager and you can get away with anything you want.
Beside that everything is a MS pc and will stay that way until the business die.
Getting everyone into the system is infinitively expensive, so is getting everyone out of it.
I work in a large 5000+ software company and were honestly about 30-40% macs. An MBPr is a good investment - I figure my 3yo machine will easily retain about 35-50% of its value if resold.
That's an odd definition of an investment. My MBPr was definitely an expense, and I love it.
I suppose the enjoyment I have received from using it could be considered dividends of sorts, and so I could see calling it an investment from that perspective. But not financially.
> Getting everyone into the system is infinitively expensive, so is getting everyone out of it.
While there are some costs to switching away from an OS/ecosystem, there's a limit to them, and it's often low enough to be economical for a business to switch or to allow some of its employees to switch (I've seen this with development shops that switch from Windows to mostly Linux with a little OS X).
Their 'getting-away-with-it' power hit an all-time low 10 to 15 years ago. They have drastically more room to abuse their position now than they did back when they were universally feared and relentlessly pursued by the US Government. Smart people actually thought Bill Gates was going to seize control of the Internet and put up toll roads, and that Microsoft would own all commerce. The scare-mongering and media insanity was non-stop for years in that regard. If I started listing sensational scare quotes about Microsoft from that time by other industry leaders, journalists, media publications, government leaders and other billionaires, we'd be here all day.
By comparison to what Microsoft went through in the late 1990s, today is a cakewalk, they have drastically more freedom and less of a spotlight on their behavior. Nobody really cares about Windows now, nobody fears Microsoft like they did in 1998, nobody thinks Bill Gates is Satan incarnate, nobody is calling for Microsoft to be broken into pieces.
I think you are misunderstanding... there are multiple dimensions to 'getting-away-with-it' power. While you are correct that Microsoft is under way less legal and public image scrutiny, they also have lost the lock in that gave them the power to do what they want and have everyone else in tech follow.
point blank most families these days have 3 or 4 iOS or Android computing devices for every 1 windows computing device.
the bottomline is that epic wants to make money out of the engine so unless they are willing to lose marketshare to other engines (e.g. Unity) they will eventually have to accept that their engine will be used for UWP apps.
I'm still waiting on seeing some real UWP apps that aren't first-party Microsoft efforts or shovelware. The platform is not exactly new, and I'm just not seeing anybody bothering to adopt it.
I really liked the idea MS had with phones that could scale up to full screen devices if you docked them, but I'm more skeptical of this. Aside from media providers, who else is going to put in the effort to make their UI controller-friendly?
Game engines are just about the only other example I can think of.
I think the goal of scale-up would be to make it so easy people just put it in. Swiping up and down on the phone could be pushing up and down on the joystick on the Xbox, or using the scroll wheel on the PC.
In theory. Obviously if not adjusted some, user experience suffers. But the idea is to make it so easy to adapt to different form factors, it just seems normal to. (People expect these days that their Android phone or iPhone app will work well on an Android tablet or iPad, for instance.)
Well there's also special code that caters for Android and Apple store development, so why not also Windows store. As long as there is choice, there's no problem. If the MS fork were to disallow any other target platform, that would be a problem. Otherwise I expect it to make it to the main branch soon.
I think you could make the case that people should not be locked in to UWP if they wish to develop an UWP app with Unreal Engine, so giving them an easy way to deploy on UWP, Mac, Linux, Android and iOS is beneficial to the engine ecosystem as a whole, even if you think that UWP is not something you should develop for.
If you read the linked post, it explains how you need to connect GitHub to your Epic Games account to see the repo if you see the 404 error. (I didn't know this sort of thing was a GitHub feature, so it was interesting to me. I initially made the same mistake.)
The primary reason to do this is the same as why most logins will just say "incorrect username or password" rather than revealing the username is in fact, correct, and just the password is wrong. You don't want to reveal the identity or location of something that is hidden. If I have an unannounced project at github.com/ocdtrekkie/unannouncedproject, you might discover the name of my unannounced project by fishing for URLs that come up 403 instead of 404.
Confirming something exists narrows what someone looking to find out something secret has to look for. If you know what my username is, you only have to guess passwords for that username. If you know what my project name is, you only need to look for references elsewhere to that project name.
Just knowing the project exists could be telling, if your project was like... github.com/apple/nintendo-igameboyphone it would potentially be a really revealing thing if you hadn't announced it yet. Slack had a big fiasco when entering a fake @whatever.com address would let someone see whatever.com's teams prior to email verification. From just room names alone you could discern some potential acquisitions in negotiation, teams at Microsoft or Apple you might not have otherwise known existed, etc.
A command line timing attack I just tried seems to be able to distinguish between the two. It seems that, as a logged-in user, the real repo takes around 5% longer to respond than a fake one. But this might not be robust - I just used time + loop in bash. Perhaps someone wants to setup a list of nonpublic+fake repos and see if there's consistent difference.
Seems to me if you don't want a project to be public knowledge don't put it on the Internet. Private projects are what internal repositories are for. You can always transition it to Github once it's open to the public.
At the very least, the administrators at Github are going to know about your project, and you don't know who they all are.
That's what it looks like when a repository is made private and you don't have access to it. UE4's code is only available to licensees; I'm sure it was never supposed to be publicly viewable.
I hope it's not an EEE move[1]? MS aren't new to such things. I.e. make Unreal Engine work with certain things only on UWP platforms, putting everything else at a disadvantage. I don't have much trust in good intentions of MS when gaming and 3D graphics are concerned, given how they are dead bent on lock-in with DX12, while the industry is collaborating on Vulkan.
It would be EEE if they, for example, ported DirectX to Linux. But here they're just ported other product to run on their "a-la Android for desktop" system. Possibly evil but not exactly EEE.
They can definitely also EEE with this. Take a game engine, create a fork for your platform, wait until your platform and fork has a significant market share, and then start introducing features to your fork, which are not directly available on the main-branch.
I can't seem to find anything on how UE4 is licensed, but even if Microsoft is forced to release all of their source code, if their fork is different enough from the main-branch, it might be extremely hard to backport all of these features into the main-branch.
> They can definitely also EEE with this. Take a game engine, create a fork for your platform, wait until your platform and fork has a significant market share, and then start introducing features to your fork, which are not directly available on the main-branch.
Exactly. Or if it's not even a fork anymore (let's say it was merged upstream), make those features platform dependent / unportable even in the main codebase. It will still be EEE.
And everybody using the fork still has to pay Epic according to the UE4 license, it doesn't matter for Epic which version they use: they get their 5% profit share either way.
If Microsoft were to discourage Win32 apps Epic would have to adapt UE4, but the MS fork can't really make that harder.
Actually, this may be a valid concern. It does look like Microsoft is trying harder and harder to lock down PC gaming. See Gabe's [0] and Tim Sweeney's [1] complaints.
Newell's comments are from 2012. Sweeney's comments have been pretty much directly invalidated by Microsoft's actions. Note that this year Microsoft made a point to address Sweeney's comments by demonstrating Steamworks in a UWP app, and the Windows 10 update releasing today makes it easier to install UWP apps without the Windows Store. They've repetitively reiterated that UWP apps can be sold through third party stores, the difference is primarily a technical one: That UWP apps are sandboxed (like Android and iOS apps) and therefore less able to act maliciously.
To date, Microsoft has done... nothing I'm aware of... to lock down PC gaming. FUD articles are abound though.
> That UWP apps are sandboxed (like Android and iOS apps) and therefore less able to act maliciously.
Many people don't want Console/phone like restriction on PC. Sweeney feared that MS will try to force everybody to use UWP, and later MS will change UWP policies. But then it will be late for everybody move away from UWP.
How many people you know downloads the apk on their android phone to get the app running? Do you have any say on the Google Play policy changes? MS will certainly repeat how convenient UWP is, but people are reluctant to trust MS because they don't know how MS will behave in the future.
That's what Tim Sweeney fears. Phasing out the open nature of the ecosystem, and replacing it with a locked down one where MS controls everything.
And that's a lock down. Maybe it's not for malicious reason, but people have enough reason to hold on onto the open nature that drove the growth of PC gaming(and malware). Secondly, when they don't know if MS will keep their words.
> How many people you know downloads the apk on their android phone to get the app running?
Most of the PC gamers I know (maybe all) did this on the day Pokemon GO was available in NZ/Aus but not NA. Probably few of them had done it before.
I understand what you're saying, but PC gamers are a different crowd than the average Windows user. We'll do what we have to in order to get control of our own PCs back.
I guess on the one hand the lock down is a rational fear, because it's something MS would attempt. But I'm not actually scared that it would be achievable.
This is only because Pokemon GO made the decision not to roll out in all countries but somehow still work in those countries. i'd like to see the business logic behind that implementation.
i had to help tons of people side load apps on their phones because they were turned off by the malicious warning that most manufacturers place by default now on activating sideloading functionality
There is nothing inherently "open" about the current system. Windows is a closed source platform and Microsoft can write it to block whatever they want to. If they wanted to block Win32 apps, they could. (SmartScreen, by default, DOES, if they're not appropriately signed or commonly downloaded.) The idea that encouraging user level apps to use UWP because it is safer does not inherently mean some draconian lockdown is coming.
If Microsoft is going to lock the platform down, they're going to lock the platform down. There's nothing you can do about it. But Microsoft choosing to use competent security in app design has nothing to do with it.
I didn't mean 'open' or 'close'. I meant 'less restrictive' or 'more restrictive'. Win32 is less restrictive, and UWP is more restrictive.
> The idea that encouraging user level apps to use UWP because it is safer does not inherently mean some draconian lockdown is coming.
All market places (google play, steam, apple app store) has restrictions. I haven't made app for any one of them, but I am certain there are policies people disagree with, and policies that has lead an app to be removed from the store or some app not getting past the the screening process.
On Win32 you don't have that. You can build something, post it anywhere, anybody can download and run it without any hiccup. This fluidity has made PC where it is today.
Also there are trust issues. Sweeney said, "The risk here is that, if Microsoft convinces everybody to use UWP, then they phase out Win32 apps. If they can succeed in doing that then it’s a small leap to forcing all apps and games to be distributed through the Windows Store."
> If Microsoft is going to lock the platform down, they're going to lock the platform down. There's nothing you can do about it.
Yes there is. In fact that's why this thread exists. MS has added UWP feature because Epic games won't do it. Steam has also released SteamOS.
It's more users that MS wants. Security comes later. And as long as the developers are taking a stand against restrictive UWP, MS will leave UWP behind. And hopefully next time they will take a different route.
Least restrictive nature is the hallmark of PC. I can't stress it enough.
So what you're saying is, Windows should not strive to meet any sort of basic security capability? Because that's all UWP is: A more secure design.
Win32 is only "less restrictive" than UWP in one particular way: It allows any application to mess with almost anything in your system without warning or user permission.
Anyone can download and run UWP without any hiccup. In actuality, Win32 does have hiccups: UAC elevation, SmartScreen filtering, etc.
Steam can sell UWP apps if they want. Epic Games can merge Microsoft's fork if they want. (Bear in mind, pull requests start as forks.)
Microsoft has the users, it needs the security so it doesn't lose those users. Tim Sweeney is paranoid in the wrong ways, he's suggesting that you have to leave a poorly secured operating system in order to protect user freedom. That's false.
> So what you're saying is, Windows should not strive to meet any sort of basic security capability? Because that's all UWP is: A more secure design.
Its development is rather weird though. If I understood correctly, part of the build process is out of developers' hands, and is supposed to be performed by distributor which supports UWP (especially for different architectures). Is UWP backend toolchain and implementation even available for other distributors besides Windows Store? If yes, then it's not a distributor lock-in. But if no, then Tim Sweeney was perfectly right criticizing it.
I think you may be misinformed. You can build an app and install it using VS. You don't even need VS, apps are just exe's in zip files. Compile and lay it out right and you're done. The Store isn't required to do any of that, so I don't know what the "UWP backend" is. You might look at Hockey app (pre acquisition) for evidence of an app distributor that wasn't Microsoft.
You compile it to each architecture before you send it to the Store. There's a compilation option to create a .appxbundle, which contains the bits for each arch. It's basically the same as iOS fat binaries.
> So what you're saying is, Windows should not strive to meet any sort of basic security capability?
It should, but at this cost.
> Epic Games can merge Microsoft's fork if they want. (Bear in mind, pull requests start as forks.)
Epic games won't. In fact, our conversation starts following the article interviewing Tim Sweeney. I highly doubt if MS will ever be able to convince Epic games to do that.
Also in the future what's stopping MS from forcing developers to use Microsoft store, followed by restriction on the type of content, followed by...more restrictions. The trust issue is very big here.
It's also true that MS have changed their attitude recently after Satya Nadella became the CEO, but when both Tim Sweeney and Gabe Newell is worried about something - I think the issue deserves more attention.
> Tim Sweeney is paranoid in the wrong ways, he's suggesting that you have to leave a poorly secured operating system in order to protect user freedom. That's false.
He is saying it will be wiser for the community at large to choose user freedom over poorly implemented security policies.
> He is saying it will be wiser for the community at large to choose user freedom over poorly implemented security policies.
To be honest, he ignores the elephant in the room. Despite his criticism of MS, all his talks assume Windows and MS everywhere. To promote user freedom, he should do more for people to ditch MS for good and to switch to Linux for example. While UE is one of the best engines that support Linux today, it's done as an afterthought, more by contribution of UE community, not so much as a big focus by Epic.
Using Windows and complaining that MS can mess things up at any time is somewhat incoherent. Because MS indeed can do it. Make something to reduce relying on MS, that's a fix for it.
> it's done as an afterthought, more by contribution of UE community, not so much as a big focus by Epic.
I think that's wrong. Linux support was on the road map, but since the engine was open source, the community(done by one person if I remember correctly) beat them to it. And today Linux is a fully supported platform.
> Using Windows and complaining that MS can mess things up at any time is somewhat incoherent.
Very true. Our only hope is to talk about it, and hoping that MS would listen.
> Our only hope is to talk about it, and hoping that MS would listen.
I don't really care much whether MS will listen. What even if they won't? The only proper solution is not to depend on them altogether.
> Linux support was on the road map, but since the engine was open source, the community(done by one person if I remember correctly) beat them to it. And today Linux is a fully supported platform.
I red somewhere, that many merge requests were ignored for a long time, and Linux wasn't really much of a priority for Epic to begin with. They should focus on Linux more IMHO, including in such public statements. Otherwise these criticisms by Tim Sweeney seem more like rants, than any practical proposals on how to improve the situation.
I'm sure his attitude has nothing at all to do with owning steam and worrying that Microsoft might actually become a viable competitor. Oh, and unlike EA he can't rely on a customer base that has rabid hatred for MS.
It doesn't contradict what I said above. Supporting multiple APIs is a burden, whether it falls on those who write their own engines, or those who write those engines for others to use. You can view it as a tax which increases costs and reduces efficiency. The point of lock-in is exactly such tax, which will deter some from using anything else, or at the very least make it more expensive for those who will use something else. This tax is passed to the end user either way, in the form of slower progress, higher prices, unavailability on other platforms and so on.
i would argue that having only one universal graphics api would hurt innovation and progress, so it's not perfect either. MS is not even the worst anyway, Apple is even forcing Vulkan do use Metal under the hood.
> i would argue that having only one universal graphics api would hurt innovation and progress, so it's not perfect either.
In general, you could argue that more competition helps progress, yes. But it's not even the case here, since walled gardens (Xbox, PS, iOS, etc.) prevent competing APIs like Vulkan from being used on them. I.e. they don't compete on merit (which could boost progress), they are anti-competitive. So they only slow down progress in this case.
To Windows? Sure. But not remotely in the way Newell and Sweeney are suggesting it'll lock you into the Windows Store. They're primarily concerned with Steam being viable on Windows.
And DirectX 12 is not any more a lock-in than any other version of DirectX since the beginning of DirectX, as far as I'm aware. (And games can and are offered with both DirectX and OpenGL support.)
Responses going forward heavily delayed because I'm rate-limited. If you want to talk elsewhere at any point, I am @ocdtrekkie
> To Windows? Sure. But not remotely in the way Newell and Sweeney are suggesting it'll lock you into the Windows Store.
Yes, they were focused more on distributor lock-in, but lock-in has different forms. Since you were talking about PC gaming in general, saying that MS did nothing to lock it, I pointed out that it's not so. System lock-in is still a lock-in, which MS uses to put competition at a disadvantage.
I singled out DX12 more specifically, because some argued that OpenGL was too much behind / not really moving forward, and so on. Vulkan isn't any of that, and MS not backing the effort instead of making NIH DX12 can't use any of the above excuses. They clearly made DX12 for lock-in purposes.
> And games can and are offered with both DirectX and OpenGL support.
Possibly they don't mean anything evil here, but as I said, MS aren't free of lock-in nastiness yet. Otherwise they would have joined the Vulkan working group already. They simply didn't earn trust in that area.
What could Microsoft gain by helping make Vulkan better? I can see them wanting to support it on Windows but Nvidia and AMD will do that for them so what can MS possibly gain?
> What could Microsoft gain by helping make Vulkan better?
Developers' efficiency. But MS understanding of "gain" is a crooked one. They want to prevent developers from releasing anything for the competition, so they use tools to make it harder (lock-in tactic). Or if they can't prevent it, they want to tax it (i.e. to make it more expensive). It is clearly crooked.
And I don't mean just to make Vulkan better. I mean MS starting supporting it instead of DX12. Same AMD can produce Vulkan support for them for Xbox quite easily.
> MS supporting Vulkan instead of DX would be shooting at their own feet. Makes 0 sense to them.
Same as supporting standard HTML instead of ActiveX and Silverlight? If not using lock-in is shooting in one's own feet, you have some strange notion of what's normal.
I guess you don't remember browser wars. Until MS started supporting standard HTML, they were pushing all kind of junk like ActiveX and Silverlight, in attempt to monopolize the Web with lock-in. Only when browsers competition intensified, they reluctantly started supporting standards (even WebGL). Some still suffer from that lock-in though, such as South Korea who are stuck with very bad decision of requiring ActiveX for bank transactions.
Exactly same thing here. They used DX to lock 3D graphics into their platforms.
Microsoft deservedly lost because they failed to innovate. But that isn't the case with graphics, where they have been at the forefront of innovation for well over a decade now. OpenGL has historically lagged behind DirectX in terms of features, ease of use and performance. Only in the last couple years has anyone (Mantle, Vulkan) actually begun to threaten Microsoft's dominance and if DirectX is inferior it should lose, but to equate that with how the HTML wars went is glossing over a lot of detail.
> Microsoft deservedly lost because they failed to innovate.
Whether they lost deservedly or not, lock-in is always a crooked practice.
See my other comment[1], that with DX12, it's more clear that MS are doing it for the sake of lock-in. Problem with it - it doesn't compete on merits on Xbox for instance. Because it's the only thing supported there.
Sadly it doesn't say anything about motivation. First instinct with Microsoft is that it is probably a dk move. So some explanation would be great. Oh well, let's see how it develops.
How is that a dick move? It's open sourced development that gives you better support if you plan to develop for their platform. They did the same for Unity with their VS/VSCode integrations etc. There was nothing negative about that, as far as I'm concerned.
Microsoft would benefit if they could do all of these anti-consumer things, but that recent stampede towards Linux proved that the tennantry can be easily spooked and dangerous when on the move. Developers are leaving windows to work on Linux and Mac, to the extent that I have not seen a start menu on a single laptop in the entire CS department. If, at some point in the future we all ended up locked in to .net, I would be surprised if Microsoft's management continued their support of these charitable projects.
In that light, I think it would be wise for us to avoid getting locked back in - so that Windows might continue moving towards being a nice place to work.