So, Microsoft just open sourced most of .NET...

[TheBlackCat]

What does Microsoft replace .net with?

Microsoft is already recommending people move away from .net to winrt and html5.

[DeepDayze]

What does Microsoft replace .net with?

Trash

Microsoft is already recommending people move away from .net to winrt and html5.

Winrt is junk while html5 is a good choice (as long as M$ doesn't muck with it)

[LightBit]

Microsoft is already recommending people move away from .net to winrt and html5.

.NET is subplatform in WinRT.

http://en.wikipedia.org/wiki/Windows_Runtime#.NET

[0xBADCODE]

and no it doesn't tie you to windows. i develop commercial apps that need to run on all 3 platforms and so far... i never booted into anything but linux for development

Then, tell us how do you deal with all this widgets idiocy without turning it into clusterfuck. What widget set you're using, to begin with? Somehow dotnet nuts tend to avoid answering "inconvenient" questions all the time. You've got unique chance to tell us your "success story", explaining one of troublesome things around. Come on. Tell us. And no, lie would not work in opensource world. So you better to be honest with us. And IMO Apache license from MS basically means "you're free to do unpaid work and MS is free to close it". Because I doubt you or me would have resources to compete with MS. Not to mention MS haves a long story of backstabbing opensource projects here and there, so only morons would trust to Apache license from MS. Interestingly, unpaid work from morons isn't something valuable and does not helps much. So I fail to see point in fooling and cheating devs. Just not going to work, only causing extra aggravation when devs discover they were tricked to ride dead horse.

.NET is subplatform in WinRT.

The only funny issue there is that this WinRT crap is dead horse. MS learned winblows devs that there is only x86 and win32 for ages. The typical result is some win32-only program which can only run on x86 and can't even run as 64-bit app, not to mention ARM. Okay, now its time to face the result. Interestingly, this backfired on MS itself, putting winblows and win devs into disadvantage. Haha, MS shot their own legs! Nice shot!

[randomizer]

they will take all the changes made by developers on the Open source version and make their next version closed.

So the developers and the end users get to use the improvements. I don't see how this is a problem. If MS chose a copyleft licence like the GPL then they'd simultaneously need to do the same with much of their software. Even if this could realistically happen in a proprietary-happy company like MS, it would be 2025 before the decision was approved by all 36 levels of management.

I don't see anything particularly devious about this move; it's perfectly logical. The only reason MS open sources anything is to increase "market" share amongst developers. Otherwise they risk becoming even more irrelevant.

[TheBlackCat]

.NET is subplatform in WinRT.
http://en.wikipedia.org/wiki/Windows_Runtime#.NET

Yes, you can continue to use .net...for now. However, it is not the approach recommended by Microsoft, and who knows how long Microsoft will continue to personally support it.

[randomizer]

Yes, you can continue to use .net...for now. However, it is not the approach recommended by Microsoft, and who knows how long Microsoft will continue to personally support it.

Probably for a decade longer than they'd like to. There is an enormous investment in .NET (along with Java) for LOB software, and we know how often that is maintained. .NET 3.5 SP1 and later are tied to their respective Windows releases, so .NET 4.5 will be supported until 2023.

[ermo]

Apache 2.0 is a GPLv3-compatible license with a patent grant. What's *wrong* with Apache 2.0?

That depends on your perspective I guess. The only thing I can think of is that Apache 2.0 is not copyleft, even if it is open source. In any case the patent grant is probably very important in terms of placating users of .NET and its implied patents.

Personally, I think the non-/limited-copyleft licenses make a lot of sense in a commercial setting, especially the Apache 2.0 license with its explicit patent grant (in this respect the BSD and MIT licenses show their educational heritage IMHO). The LGPL is workable as well, though a bit trickier to handle in terms of compliance. Apache 2.0 probably works better if you are already a large, well-funded organization and the code you release is not your core (cash cow) business, even if you still intend to leverage it in your products and use it to increase participation in the ecosystem in which you are already an established market leader.

In contrast, I would imagine that you'd want to use the LGPL if you're trying to build an ecosystem around your application's/solution's core libraries, because having them LGPL'ed ensures that no competitor can come along and out-innovate you without contributing back to said core libraries. Conversely, the LGPL levels the playing field for other entrants and ensures that no-one can gain an unfair advantage unless they are willing to maintain their own libraries in-house.

In my view, it would make a great deal of sense to have a Unix-y base OS stack (however that is defined) licensed entirely under the LGPL, where the base system is defined as a FreeBSD-like, self-contained unit. Note that I said 'stack', which is to say that it is only important as a foundation and not as a product in its own right, just like libraries. The idea is of course to use this foundation to build products in the form of appliances, services and solutions.

Having an LGPL-licensed OS stack would ensure a collaborative environment in which to maintain this base stack, while allowing entrepreneurs the legitimate opportunity to explore and grow various niches on top of said stack in the form of both proprietary, open source and copyleft (GPL) solutions. The kicker is that everyone would be free to try to emulate any proprietary solutions using the LGPL base stack and even to make fully GPL copyleft workalikes of the product. This would result in competitive pressure for the proprietary solutions, acting as a balance against monopolistic complacency, which is supposedly what makes competition work in favour of the consumer in a free market. Or maybe I'm just being incredibly na?ve...

Anyway, off topic.

[justmy2cents]

Then, tell us how do you deal with all this widgets idiocy without turning it into clusterfuck. What widget set you're using, to begin with? Somehow dotnet nuts tend to avoid answering "inconvenient" questions all the time. You've got unique chance to tell us your "success story", explaining one of troublesome things around. Come on. Tell us. And no, lie would not work in opensource world. So you better to be honest with us. And IMO Apache license from MS basically means "you're free to do unpaid work and MS is free to close it". Because I doubt you or me would have resources to compete with MS. Not to mention MS haves a long story of backstabbing opensource projects here and there, so only morons would trust to Apache license from MS. Interestingly, unpaid work from morons isn't something valuable and does not helps much. So I fail to see point in fooling and cheating devs. Just not going to work, only causing extra aggravation when devs discover they were tricked to ride dead horse.!

i know this was bad try on trolling. but, i'll indulge you anyway. p.s. don't worry about my honesty, i don't work for MS, i'm not affiliated with any .Net promoting company. from '94 i didn't own one windows computer and so far i'm yet to see win8 in person. my work is more or less maintaining servers and writing cross platform apps, where i test (or better said, i'm sure it works) other platforms on deployment.

and it is not so much about widgets as it is about features. there are 2 kinds of approaches to cross platform. selectively preplanned or panicky porting when product is finished. i'd suggest you watch movies about porting from steamdays. http://www.youtube.com/watch?v=Sd8ie...KFtqR9TeZWMPjm the choices you make before you start your product will make you or break you.

as for tools. not really important if you were thinking about restrictions in your plans. monodevelop, gtk# (now porting to qyoto)... most of your troubles are not of such grotesque nature, but rather small. how do i store data, how do i access file...


now, i'll go further and indulge you even more. if MS would be acting according your words, that would actually be single most dumb business decision of the century.
- .net was was closed,... so why open it, just to close it? sure backfire, since it would piss off more of their customers than anyone else
- while it was still closed, .net had 2 advantages. lead in features supported by compiler and non clear patent issue caused by ECMA. by opening roslyn, they lost both.
i could go on and on how stupid this plan would be, but all basically tell you this "with this move they lost ground and created competition, but they gained larger market"

[chithanh]

They released a lot of libraries, in fact I was surprised. Some of them were released in previous releases but they still count: MVC3, MVC4, DLR, Entity, etc.

"a lot" is not "most" as was claimed by the OP. Sure they have released important parts, but many more are still unreleased (WinForms/WPF being most prominent, also mentioned in this thread already). Until then, I can fully understand those who remain skeptic.

The bit of irony here is that it was Oracle that ended up filing a lawsuit over patent claims on Java. Should everyone stop using Java because Oracle is a litigation-happy company?

It's actually a bit ironic, that by avoiding the GPL'ed OpenJDK and doing a clean-room implementation under the Apache 2.0 license, Google opened themselves to litigation by Oracle. Had Google just used code from the OpenJDK, there would not have been any basis for a lawsuit.