Mono 4 Released, First Version To Adopt Microsoft Code

jones_supa writes: Version 4.0.0 of Mono, the FOSS implementation of the .NET Framework, has been released. This is the first release of Mono that replaces various components of Mono with code that was released by Microsoft under the MIT license. Microsoft itself is working towards .NET Core: a redistributable and re-imagined version of .NET, which has two code drops: CoreFX and CoreCLR. Mono at this point continues to provide an API that tracks the .NET desktop/server version. This means that most of the Mono code that has been integrated from Microsoft comes from the ReferenceSource code drop. Mono's C# compiler now also defaults to C# 6.0.

Comments

Anything unique?

[ThePhilips]

Does Mono provide something unique to grant a look at it?

Re:Anything unique?

[Anonymous Coward]

It absolutely does. You can develop completely cross platform applications with 1 codebase using XWT. The single codebase creates projects for GTK, Cocoa(OSX) and WPF(Windows) that use the native controls for the corresponding platform.

The fact that they are porting most of the libraries over to the first-party Microsoft versions means less bugs, and way more active maintenance. This is very good news for cross-platform developers!

Re:Anything unique?

[Trepidity]

If you mean compared to the official C# release, for the moment it's still Windows-only, so Mono's main advantage is being a cross-platform C# compiler/runtime.

Re:Anything unique?

[smittyoneeach]

If you say that "To Serve Man" is a cookbook, they'll grill you.

Re:Anything unique?

[Njorthbiatr]

Whoa dude C++? FORTRAN is the REAL language. Get your fancy high level crap out of here.

I like my languages like I like my coffee: without synaptic sugar.

Re:Anything unique?

[Anonymous Coward]

GTK is a pile of maggoty old C-shite, the Qt c++ codebase is vastly better designed and infinitely cleaner.

Re:Anything unique?

[Anonymous Coward]

Really? 30 minutes? It'd take me 10 in c#.

Re:Anything unique?

[ThePhilips]

Only after I have pressed "Post" button, I hare realized that I probably given not enough context.

"Anything unique to a Linux developer who looks for RAD on Linux"?

From RAD for Linux, I'm aware only of the FreePascal-based Lazarus (Borland Delphi remake).

Past attempts with the KDevelop (and QtCreator) were pretty abysmal: right after "apt-get install" and few clicks to throw together UI, they were failing simply to compile the "Hello World" msgbox application. (And wild goose chase to install what was missing was counter the whole idea of *rapid* application development.)

Re:Anything unique?

[frank_adrian314159]

Well, it's a better designed language than Java ever was, so there's that. What more do you need?

Re:Anything unique?

[kthreadd]

For RAD I've had some good experiences in the pas with RealBasic (renamed to Xojo some time ago). At the time the GNU/Linux support was quite ok but they haven't really kept up. 64 bit is still missing for example so running anything you build with it will require a couple of 32 bit system libraries to be installed.

Re:Anything unique?

[ThePhilips]

Xojo

Xubuntu 14.04 says that the installation would take 195MB of space. Bit heavy. Worst part: it is BASIC.

On the positive side, Lazarus + FP SDK, requires almost 1GB of space on my Xubuntu.

My ultimate goal is to be able to put together a quick UI, check in the source, and tell in few words to others who going to check it out how to compile it and get it running.

All in all, Xojo gets on the short list of things to try.

Re:Anything unique?

[Foofoobar]

well considering their 'mit license' is invalidated because of the wording saying you can't use without their engine or code... it kind of is a trap. Just a bad one.

Re:Anything unique?

[benjymouse]

well considering their 'mit license' is invalidated because of the wording saying you can't use without their engine or code... it kind of is a trap. Just a bad one.

I suppose you have a source for that restriction?

Can you please help me, because when I view the LICENSE.TXT (https://github.com/dotnet/coreclr/blob/master/LICENSE.TXT) I cannot find anything like that?

Re:Anything unique?

[benjymouse]

I may add:

The PATENTS.TXT (https://github.com/dotnet/coreclr/blob/master/PATENTS.TXT) also does not claim anything like that. It merely says that if *you* bring a patent lawsuit against *anyone* for using the *covered code*, any patent grant is revoked and the promise is revoked.

Re:Anything unique?

[MightyMartian]

There are other ways to develop cross platform code that don't require putting the loaded gun of Microsoft's dubious anti-litigation commitment to your head. .NET simply isn't so good that I'd put my fate in Redmond's hands.

Re:Anything unique?

[Anonymous Coward]

Python.

IronPython

There you go.

Re:Anything unique?

[caseih]

BASIC these days is little different from Pascal so I don't see why this is a downside. BASIC is every bit as modern as any other language and structurally equivalent to any modern static language. It's more verbose than C and similar languages, and maybe as verbose or slightly less verbose than Pascal is. Not sure about RealBASIC, but other BASIC dialects suffer from way too many things being a part of the language with custom syntax, rather than a callable function in the runtime. For example the Pset, Line, Circle, Draw functions. But really, any BASIC hate because its BASIC is about 30 years behind the times.

I don't understand why you couldn't get QtCreator working. Qt is easy to install and use on Ubuntu. And the Qt QUI designer is very easy to work with.

I would say that Python + libglade + glade is also a pretty good combination. It's not quite the RAD experience you seem to want, but it is a fast and powerful way of developing GUI apps, thanks go a nice API and Python.

Re:Anything unique?

[gbjbaanb]

no, it'd take you 10 minutes in Visual Studio - not quite the same time as it takes using anything else, and VS doesn't exactly run on Linux or Mac which is the point of Mono and even then you get a very non-native GUI.

Re:Anything unique?

[ThePhilips]

BASIC is every bit as modern as any other language and structurally equivalent to any modern static language. It's more verbose than C and similar languages [...]

Verbosity is the problem.

If I were fine with the screens and screens of the boilerplate code, I would have simply used the Java.

I don't understand why you couldn't get QtCreator working. Qt is easy to install and use on Ubuntu. And the Qt QUI designer is very easy to work with.

As I heard it was a systemic problem back then: not all package dependencies were declared, meaning that after installation you have to also install bunch of other stuff to make it working. (Many years ago, first time I tried QtCreator, it actually refused to run, because some linked libraries were missing.)

I'm not sure about now, but back then it wasn't even close to anything RAD. It was only a helper to create the GUI in a XML form, which was back then not even properly integrated with the rest: one had to write some code manually to actually tell Qt what resource corresponds to the window. And add resource manually to the resource file.

I would say that Python + libglade + glade is also a pretty good combination. It's not quite the RAD experience you seem to want, but it is a fast and powerful way of developing GUI apps, thanks go a nice API and Python.

Yes, it is not RAD. And for that I already use QT + C++, which provides very powerful, simple and no-fuss API to build GUIs dynamically (without external UI building tools like glade or Qt Designer).

The problem is not me per se - I have no problems with most of the stuff. The problem are the other team-members who are not well versed as me in the scripting languages and building GUIs. On many past projects I have left behind lots of stuff which 95% of coworkers can't support or develop further. And I want to solve the problem by throwing in something that requires as least boilerplate as possible, stays as close to demos/examples as possible. I'm simply trying to find something to help the people start moving in the right direction.

Re:Anything unique?

[RabidReindeer]

java ie write once, compile... once. big failure on cross platform.

For you maybe. I spent 3 years developing Solaris apps in Java on Windows XP machines.

Re:Anything unique?

[MightyMartian]

Indeed. I have written several cross-platform Java apps and utilities that run just fine on Windows, Linux or Mac boxes. One can certainly write Java programs that are locked to one platform, but I've ever seen the need.

Re:Anything unique?

[amorsen]

The patent indemnification only covers you if you are using the code for a .NET runtime or a .NET application. Too bad if you find some of the code useful in an unrelated project and fall for the MIT license.

Re:Anything unique?

[jmac_the_man]

Now they seem to promise cross platform development again, but for how long? It wouldn't be the first time Microsoft changes strategy.

Well, Mono has been around for almost a decade now, and they AREN'T Microsoft. Microsoft submitted .NET for ECMA standardization, and Mono is an alternate, non-Microsoft implementation. Microsoft engineers and Mono engineers have worked closely in the past, but this is the first time that Microsoft developers have contributed code to Mono.

In the same way, Microsoft has contributed some code to the Linux kernel. It's not a majority of the code, so they can't argue that you should call it MS/Linux or something dumb like that, and they didn't change the license on the kernel, so they can't show up and shake you down or anything.

Their contribution to Mono was of a similar size and licensing scope. Microsoft isn't going to show up demanding money for this if you use it.

Re:Anything unique?

[kthreadd]

That means very little in practice. You can be sued for patent infringement by anyone for more or less anything. A license will not help you with that.

Re:Anything unique?

[kthreadd]

Do they still "look just like" or are they "the same thing?" It used to be very easy to distinguish Qt applications on OS X when I used to use it a couple of years ago. The widgets looked very similar but you could instantly recognize that they were not quite right. It would be nice to know if the situation has improved.

Re:Anything unique?

[caseih]

You misunderstand. By verbosity I don't mean boiler plate, java-style. I mean simply that there are more keywords. If/Then/Else/End if, Do Loop, For/Next. Again no worse than Pascal which you don't have any problems with.

QtCreator is much closer to the RAD concept than you state here. Callbacks can be filled out in the IDE. It's not like you state in your post. However, the RAD concept isn't always super flexible, and modern GUI systems like Qt and GTK all let you work on the GUI in a programmatic way that is often much more powerful, but still easy and flexible. If you insist on your definition of RAD you'll likely run into limitations (any RAD system) and be disappointed. It's a good concept but in practice I think you need more than that. Except for certain vertical markets, I don't really see the point of full RAD to be honest. Especially when a half dozen lines of explicit code can do the same thing, but exactly how I want it. The XML gui design is far far better in my opinion. Load the ui file, autoconnect the callbacks to your code, done.

Re:Anything unique?

[HiThere]

IIUC it's worse than that. For a .NET runtime you must implement the complete specification, which they are allowed to change at will. I suspect that the .NET application is safer, but I'd need to study it a bit to find out, and I'm not that interested.

That said, they won't bother to sue you unless you're successful, so most people will be safe.

Re:Anything unique?

[ThePhilips]

If you insist on your definition of RAD you'll likely run into limitations (any RAD system) and be disappointed.

No, I will not be. I have used in the past the Borland Delphi for 5+ years and well aware of the limitations which come with the paradigm (rigid system libraries, "there is only true way to do it", "if there is no button for it, it's impossible", and so on). (And yes, to this day, I deeply hate Borland Delphi.)

I'm interested in RAD for specific purpose, so to say. To show that GUI development can be as easy as writing 10-20 lines of shell, but with the bonus of having a UI which is little bit more than text console. And, well, introduce some GUI into the Linux part of the product.

I don't really see the point of full RAD to be honest.

I do not look at them as a programming language or programming environment.

I see them as a tool to quickly develop and deploy a simple GUI, when the text console doesn't cut it.

Re:Anything unique?

[ljw1004]

well considering their 'mit license' is invalidated because of the wording saying you can't use without their engine or code... it kind of is a trap. Just a bad one.

There is NO wording which says what you claim.

There's an ABSENCE of wording that says the inverse, however.

Re:Anything unique?

[angel'o'sphere]

That is nonsense.

If you compiled a Java 6 App and run it on an Java 1.3 VM, it wont work, for plenty of reasons.

I personally had in 17 years Java development no single cross platform issue.

The people who told me cross platform problems are about a handful. Mostly doing weird stuff, or once an RMI problem between Sun Java 5 and IBM Java 1.4 (which boils down to differences in serialization)

Re:Anything unique?

[Grishnakh]

Microsoft has not contributed any useful code to the Linux kernel. Their "contribution" was drivers so that Linux could work on their hypervisor. If you're not running a MS server with MS Hyper-V, then their contributions are useless to you.

Yeah, it's nice they're playing a little more nicely with Linux now, but don't pretend their "contribution" had any altruistic component to it at all, because it didn't. It was only done because customers were demanding that they be able to run Linux virtualized on MS systems. MS did the bare minimum needed to enable this, and that's it.

Re:Anything unique?

[ahabswhale]

Not unique at all. Java has had this ability for over a decade.

Re:Anything unique?

[shutdown -p now]

I don't think so. They have only just added lambdas in Java 8 (which is something that C# has first got in 2.0 in 2005, and in the form more or less equivalent to what Java has, in 3.5 in 2008). And even then some of the stuff that goes along with it is hilariously bad, like their insistence on type erasure for generics, and the associated requirement to have dozens of interfaces to cover the most basic function types for the most common primitive types. In the meantime, C# is still actively evolving, gaining feature such as "dynamic" (opt-in duck typing) in 4.0, and async/await syntactic sugar for callbacks in 5.0.

Re:Anything unique?

[Beefpatrol]

I seriously doubt MS's customers were demanding that MS make it possible to run Linux virtualized on MS systems. Running Linux virtualized on MS systems has been possible and easy for many years already, and almost nobody that wants to run Linux wants to run it virtualized on MS systems. Lots of people that need to run Windows for something want to run it virtualized on Linux or Mac OS, but that has been possible and easy for years also.

Re:Anything unique?

[Beefpatrol]

The patent indemnification promise is irrelevant as it is basically legally meaningless. Microsoft wants the open source community to write stuff that uses Mono for all the same reasons they court non-open-source developers so aggressively. This post used to be a lot longer, but I am shortening it to simply reference some things that one should look up in a search engine if they want to know what this Mono / patent indemnification / "C# and Visual Studio are the best things ever" bollocks is really about.

Here is what I would look up: Miguel de Icaza, (pay attention to his role in Gnome, Mono, and Novell.) Consider how unlikely it is that the guy who started Gnome suddenly sprouted an interest in an open source version of .NET without money being involved. (.NET was even less interesting to the open source world back then than it is now.) Look up Microsoft's phony error messages on DR-DOS. (I personally witnessed one of these error messages back in the day that was slightly different from messages I see documented now. I also made it go away by editing the DR-DOS command.com with a hex editor to change some english/printable strings in the binary to their MS-DOS counterparts. Yes, Microsoft really was so lame and devious that changing the copyright message string in command.com to read "MS-DOS" instead of "DR-DOS" made the "incompatible" software suddenly work perfectly under DR-DOS.) Then look up a Powerpoint presentation that Microsoft used to explain their "Tech Evangelism" practices to people who would be responsible for carrying out such evangelism. This presentation was made public as part of court proceedings against Microsoft somewhere in Europe related to antitrust violations or whatever they call that sort of thing in whichever country it was. Look up the "Evangelism is War" document.

The point I am making here is that the "it's a trap" posts that always arise when Microsoft says... pretty much anything, and which people often find amusing, are only partially a joke. There is more than a kernel of truth in that assertion. Mono *is* a trap. .NET *is* a trap. If Mono ends up making MS even less relevant than it already is, it is only because the trap backfired. And if the trap backfires too hard, and the open source community embraces the .NET API via Mono and then extends it and accidentally threatens to extinguish the relevance of MS's version of it, you can bet that MS's lawyers will be especially rabid.

Re:Anything unique?

[Beefpatrol]

I might suggest PHP/Apache. Use the browser as the interface. Set Apache to listen only on 127.0.0.1, then you avoid most security concerns. Compilation not necessary. With modern Javascript libraries for complex UI stuff, you might be surprised how much you can get done quickly.

Re:Anything unique?

[jmac_the_man]

Microsoft has not contributed any useful code to the Linux kernel. Their "contribution" was drivers so that Linux could work on their hypervisor.

If you don't find the code useful, that's your business. But if Microsoft's view was that Open Source is a cancer that MS should be trying to kill, they wouldn't have contributed anything to the Linux kernel.

Re:Anything unique?

[benjymouse]

Microsoft has granted patents, to anyone who implements a .NET runtime. The grants were part of the standardization of .NET CLR and core libraries.

It is a misunderstanding that it is bound to Microsofts own implementation. Those grants has always extended to Mono. The anti-Mono and anti-Microsoft fanatics started a FUD campaign based on speculation that MS could just sue anyway, and the mere cost of defending against MS would force Mono underground. It was a response to that FUD campaign that MS also issued the community promise.

The patent grants also are not tied to a full-stack implementation like Java/Oracle. The fact is, the patent protection when using CLR is far more transparent and effective, compared to Oracle/JVM.

Re:Anything unique?

[benjymouse]

Microsoft has not contributed any useful code to the Linux kernel. Their "contribution" was drivers so that Linux could work on their hypervisor.

I find that immensely useful.

When Intel contributes drivers for graphics chips, it is *also* so that Linux can work on their hardware.

Maybe you should take a clue from Linus Torvalds. (hint: It's about scratching your own itch)

Re:Anything unique?

[Grishnakh]

Intel also contributes a lot of money to the foundation which employs Torvalds. Microsoft does no such thing.

Re:Anything unique?

[meta-monkey]

Beware of Redmonders bearing gifts.

Re:Anything unique?

[Beefpatrol]

I imagine that they do, given that there are a lot of people in the world and it is possible to do so. I can't imagine why one would, however. Is there any advantage to doing so that is unique to Azure?

Re:Anything unique?

[mrchaotica]

Is that actually any worse than (for example) whatever indemnification -- or lack thereof -- Oracle gives you for violating patents when using Java code in something else?

Re:Anything unique?

[vilanye]

One in five VM's running on Azure runs Linux.

http://www.computerworld.com/a...

Re:Anything unique?

[vilanye]

IronPython is lagging far behind.

They are stuck on Python 2.6 compatibility which was released in 2008.

Beware Rust, Go, and D.

[Anonymous Coward]

Being able to use C# on Linux and OS X and the BSDs will make languages like Rust, Go and D even less useful than they already are. People who are anti-Microsoft may not like to admit this, but C# is an excellent programming language, and .NET is an excellent runtime. Although they're over a decade old, they were so far ahead of their time when they were first released that they still feel fresh and relevant even today.

Mono was always OK, but not great. Now that Microsoft is releasing code that Mono can also use, everyone is a lot better off. We'll finally be getting a high quality VM runtime for Linux, much better than Java and it's VM, and much, much better than Parrot and the other failed open source VMs. The .NET CLR always feels very transparent, unlike the JVM which is painfully obvious.

It's getting to the point where any sensible software developer will write their software in C#, even if targeting Linux. C# is just the best general purpose language out there. In the rare cases where C# isn't suitable for some reason, modern C++ provides a superb alternative.

Between C# and C++, there's just no need for other languages. Both C# and C++ offer low level functionality, as well as much higher level functionality like lambda functions, closures, generics/templates, OO, memory safety through GC and/or smart pointers, and so on.

There's just no sensible reason to use a language like Rust, Go or D these days. They're inferior to languages like C# and C++ in various ways, but without being any better. So you're inherently worse off when you use them.

If you're ever in a situation where you may be choosing between Rust, Go or D, postpone your decision and look at C# and C++. You will very likely be making a better choice by choosing mainstream, well-supported, portable, mature and efficient languages like C# and C++.

Re:Beware Rust, Go, and D.

[Anonymous Coward]

you better have a very expensive attorney with patent law specialty on 10k per month retainer.....

Re:Beware Rust, Go, and D.

[Anonymous Coward]

It's funny how things have come full-circle.

In the 1990s, Microsoft used to be the one accused of spreading Fear, Uncertainty and Doubt. Microsoft used to be the one pushing unwanted software on the masses. Microsoft used to be the one pushing lousy programming languages like VB.

Now it's the 2010s, and open source supporters like yourself are spreading Fear, Uncertainty and Doubt. Open source supporters are pushing unwanted software like systemd on the masses. Open source supporters are pushing lousy programming languages like Go, Ruby and Rust.

I hate to say it, but I trust Microsoft more these days than I trust Red Hat or the average open source developer.

Re:Beware Rust, Go, and D.

[xonen]

Please don't compare apples to oranges. You are totally right about the trust issue, that's something personal. But it has very little to do with the C# language or .NET.

Pulling JVM into the equation not really helps either, cause the consequent question would be: Do you trust Oracle? Or Google, for that matter if you count Dalvik in.

I do like the C# as a language, having done a few smaller projects with it. The reason why i prefer not to use it is, indeed, because i don't trust or like Microsoft. Having said that, i am totally fine with the Mono project - despite all criticism it's just the language and the VM and has very little to do with MS, and when appropriate (i.e.: someone pays for it) i wouldn't hesitate to develop with Mono or .Net again.

GP totally has a point here: The languages you really need for a certain task already exist, whether it be C, C++, C#, Java and a handful of other niches including but not limited to Perl and PHP. Whatever your choice is, try stick to a steady platform. Code written in any such `proven` language is much more likely to compile in another 10-20 years from now than code written in some obscure actively-developed language which adds little, that couldn't be done otherwise, but headaches.

And AC above also has a point that many OS enthusiasts are guilty of exactly what they accuse their nemesis of. Hence he doesn't deserve the tag 'astroturfer', it may well be his honest opinion. It's totally ok to criticize, but be prepared to accept criticism too, please.

Re:Beware Rust, Go, and D.

[Anonymous Coward]

"We should systematically arrange to depend on the free C# implementations as little as possible. In other words, we should discourage people from writing programs in C#. Therefore, we should not include C# implementations in the default installation of GNU/Linux distributions or in their principal ways of installing GNOME, and we should distribute and recommend non-C# applications rather than comparable C# applications whenever possible."

-- FSF Executive Director John Sullivan

Re:Beware Rust, Go, and D.

[Gadget_Guy]

You sir, are a great astroturfer and deserve a raise from MS.

That's really another type of FUD; that anyone who says something that isn't completely anti-Microsoft must be being paid to say it.

It has been 10 years since Mono was released and 13 years since .NET was released, and for the entire time there have been the predictions that Microsoft will start suing all and sundry for patent infringement. For that entire time it hasn't happened. For that entire time it has been complete FUD, whether you like it or not.

Well, just recently a very interesting article covering Microsoft "open source .NET" license, you should read up on that, especially MS requiring a license to the patents in the code you contribute, but refusing to grant you license for their code, instead, providing a promise not to sue.

So what? None of that means that Microsoft is going to start suing you for using the Mono CLR and Framework. If you don't like their terms then don't add your own patented code to a .NET Foundation-owned project, but feel free to use Mono without any fear of being sued by Microsoft.

If you really trust Microsot more than RedHat or opensource developers, than please, don't let anyone stand in your way, trust is a personal issue, some people trust ISIS, some - the supreme leader, but some prefer to be able to verify the code themselves, and Microsoft throwing their dying platform into opensource stream, hoping for a revival is very far from transparency and verifiability.

Wow, talk about FUD again. Bringing up ISIS is just a modern version of Godwin's law. And "some prefer to be able to verify the code themselves" is FUD because this is all about open source code released by Microsoft. Of course you can verify the code yourself. Or are you mixing up the completely unrelated non-OSS Windows code that you can't see. How is that relevant to this discussion?

Re:Beware Rust, Go, and D.

[guruevi]

I am old enough to have worked with .NET pre-releases. Back then most viewed C#/.NET as a cheap and broken Java clone. It took .NET several years to become mature enough to be useful.

Re:Beware Rust, Go, and D.

[Anonymous Coward]

I tend to agree. I've had to use C# recently, and the .NET stuff is impressively mature and full-featured. I'm an old hardcore Java programmer, and Microsoft copied Java and C++ so closely that I was immediately productive. The whole .NET thing is an excellent rubber-room runtime system with everything you need. You'd be very, very, very hard pressed to make a business case for not using it. And even harder pressed to make a case for some nothing language that has no users anyway.

My pet peeve is that there are too many languages and runtime libraries now and it's hard to mentally switch among so many. With C# and .NET available on Linux, this might change as people consolidate.

Re:Beware Rust, Go, and D.

[gbjbaanb]

Neither are quite perfect though - C++ has plenty of ancient cruft that;s there for C compatibility, but then, C# has plenty of cruft from its old 1.1 days (all those nasty, nasty functions taking char[] parameters for example.

But they're still better than the newcomers simply because the new ones are just as imperfect but without the benefit of a wide user base or tooling. Maybe one day Rust will become mainstream and we'll all start using it, and good luck for that day.

But, there's no reason not to use C++ now for Linux development. Moving people to C# on Linux is just wrong, not because C# is rubbish, but because it's not mainstream on Linux. The reasons not to use Rust or C or Go are the same ones to use when applied to C# on Linux. On Windows its a different matter of course, but Linux - stick with C++ and enjoy the benefits of the same codebases everyone else uses.

Re:Beware Rust, Go, and D.

[MightyMartian]

Have you ever considered the possibility that all those years of misconduct by Microsoft have sowed a considerable amount of distrust in the developer community, and that even where Microsoft has turned over a new leaf, so vile was its conduct "back in the day" (which ain't all that long ago, if you think about the OOXML open standard scam), that it might take years, or maybe never, to convince a lot of people that there isn't some evil plan in the works.

Give me one good fucking reason why I should ever trust Microsoft again?

Re:Beware Rust, Go, and D.

[kthreadd]

That was actually true for Java too. Unfortunately Java development also slowed down significantly once it became mature, around 5.0-6.0. Scala is where most of the interesting stuff in Java is going on these days.

Re:Beware Rust, Go, and D.

[jmac_the_man]

You sir, are a great astroturfer and deserve a raise from MS.

Well, just recently a very interesting article covering Microsoft "open source .NET" license, you should read up on that, especially MS requiring a license to the patents in the code you contribute, but refusing to grant you license for their code, instead, providing a promise not to sue.

Um... Mono is released under an MIT license, which is less restrictive than the Microsoft Public License. But here, take a look what Microsoft's Open Source license says in terms of them licensing you their patents on their code:

2. Grant of Rights (A) Copyright Grant- Subject to the terms of this license, including the license conditions and limitations in section 3, each contributor grants you a non-exclusive, worldwide, royalty-free copyright license to reproduce its contribution, prepare derivative works of its contribution, and distribute its contribution or any derivative works that you create.

(B) Patent Grant- Subject to the terms of this license, including the license conditions and limitations in section 3, each contributor grants you a non-exclusive, worldwide, royalty-free license under its licensed patents to make, have made, use, sell, offer for sale, import, and/or otherwise dispose of its contribution in the software or derivative works of the contribution in the software.

I'm not going to reproduce Section 3, but the restrictions are: You don't get a trademark license or a warranty, you're not allowed to sue other licensees over your patents, you have to retain copyright notices that appear in source code, and you can't re-license MSPL code under viral licenses (e.g. the GPL).

If you want to see the full license, check out the OSI's site on the MS-PL

Re:Beware Rust, Go, and D.

[Loconut1389]

You can't say there is no reason for other languages, ASM and C still have their use in very lowlevel code. VHDL/Verilog are relevant to FPGAs which may or may not hold a very important part in the evolution of computers- dynamic chips that can become anything you want and implement a hardware parallelized version of a software algorithm.

While I agree C# and C++ will start to cover a large percentage of the spectrum, the others are not without purpose.

Re:Beware Rust, Go, and D.

[Anonymous Coward]

Oddly enough, that has turned out to be the case for Java, darling of open source zealots.

Mono has never had any legal problems, and the legal guarantees Microsoft has given have been a lot better than for other platforms.

Re:Beware Rust, Go, and D.

[jmac_the_man]

One's right to life, liberty, property, speech, press, freedom of worship and assembly may not be submitted to vote

So you know, there's a very good reason that the Constitution codifies that the government may not make laws "establishing an official religion, or prohibiting the free exercise thereof." Those two rights are much broader than "freedom to worship."

Re:Beware Rust, Go, and D.

[Anonymous Coward]

If you forget the past you are doomed to repeat it. Microsoft has no reason to sue people over .NET *yet*. They have made sure they have the controls in place to do so if they need though.

Re:Beware Rust, Go, and D.

[Baki]

Better than the Java VM? In what ways, and where are references?

In most large corporations, JEE is 50-80% of new mission critical software nowadays. I work for one of the top-5 banks and we use windows only for desktops, and on the server side mostly for small- to midscale software pacakges. All the companies mission critical systems, either bought or built, are either on the Mainframe (diminshing, but that will take 10 years or more to finish) or various forms of Java (JEE, tomcat, hadoop, various other clusters and compute grids).

I see no evidence at all that C# of .net is gaining traction for mission critical server side software.

Re:Beware Rust, Go, and D.

[bmo]

So much shilling in one post.

>uttering C++ and C# in the same sentence as if they are equivalent

Just... no.

>no need for other languages

Uh huh.

It seriously sounds like you've got only two tools in your toolbox and are looking at the guy with the loaded Gerstner box and telling him all those things are useless, which as a machinist and toolmaker, I have to say that you're delusional.

There is room in the world for C, Lisp, Go, Rust, COBOL, C#, and literal jokes like Brainfuck. Because no one language is perfect for all use cases.

--
BMO

Re:Beware Rust, Go, and D.

[guruevi]

Yet for some reason, the supreme court had to establish this in the above quote in the early 20th century.

Re:Beware Rust, Go, and D.

[Gadget_Guy]

What happened to the Slashdot I used to know? The old crowd is gone, replaced by young 'uns who spent their college years downloading 1000s of music files.

Thanks for calling me a young 'un. Nobody has done that to me in a very long time.

But I have to say that you sound like a leftover hippie from the 60s complaining that everyone who no longer believed in peace and free love had sold out, when in fact they had just grown up. Feel free to complain when Microsoft does something wrong. But after 13 years of predictions of a patent apocalypse, perhaps it is time to face the fact that they are not going to start suing the world for using Mono; especially when there has been cooperation between Mono and Microsoft during its development.

But maybe they still are.

And it all boils down to this. You have no proof in the slightest that they are doing anything wrong or that they intend to. But that doesn't stop the pitchforks coming out because of a feud that dates back decades. Have they done anticompetitive things in the past? Sure. Have they ever turned to litigation after making a public patent promise? No.

Having an open source implementation of .NET and C# legitimises the platform as the standard for Microsoft. They are not going to just turn around and crush it only to suffer a huge PR backlash because they broke their word. And of course, any judge would throw out a claim of patent infringement precisely because they had made public promises about not suing.

Re:Beware Rust, Go, and D.

[Gadget_Guy]

Nope, but a person believing that Microsoft is more trustworthy than global community,

That is their opinion. It doesn't mean that they are a Microsoft shill as you claimed.

that .NET runtime is a silver bullet that will kill Ruby, Go and Rust

The AC didn't say .NET would kill those languages, just make them less useful.

person that keeps insisting that MS won't sue anyone over .NET despite the shady language in the license and a number of restrictions (.NET code can only be used to create a runtime adhering to MS specs and for no other purpose)

There is simply no shady language in the license that is going to affect Mono. If they ever decide to change Mono from an implementation of the .NET platform to something else (eg. JVM) whilst retaining Microsoft's code then they could be in trouble. But do really think for a second that they would change the focus of the project like that? Absolutely not.

And yet that is the main message of your post, that if you don't adhere to Microsoft's spec then they could sue. Well Mono is compliant with the licence, so they are not going to get sued.

Also, for this same time .NET has failed to see adoption the likes of Java did, and right now, Microsoft has even more hooks inside their license allowing them to sue the living hell out of anybody, and (Like with Oracle, Google and Java) they can sue if the code used in .NET will be used for anything other than making a fully fledged .NET runtime (that part is straight in their license, no guessing involved here).

So your response to me pointing out that Microsoft hasn't actually sued anyone for the last 13 years despite all the claims that they would is that Java is still bigger and that if you made something that was unlike Mono that you would get sued. How is that counter my claim that saying that using Mono will not get you sued?

Previous comment was regarding Microsoft and open-source in general - this is an answer in general. Commenter said he trusts Microsoft more than RedHat or opensource developers, I pointed out that trust is a personal issue, ability to verify - is more objective.

Irrelevant. You have the ability to verify code from an open source project.

Re:Beware Rust, Go, and D.

[occasional_dabbler]

Microsoft is really not the company it used to be.

Companies are made of people and they change, grow up/older and move on. It is a huge company and in any such organisation it takes a long time for culture and strategy to change significantly.

Anyone who witnessed the hideousness of the SCO litigation has to look at a new Microsoft where FOSS is actively supported (even if there are strings attached) and where employees can talk about open-sourcing the OS without being fired on the spot has to accept that they are at least heading in a better direction than they were.

Re:Beware Rust, Go, and D.

[Grishnakh]

I thought Rust and D were really aimed more at replacing C than C++.

Re:Beware Rust, Go, and D.

[gtall]

To use C#, I'd have to trust MS. Never.

Re:Beware Rust, Go, and D.

[Chemtox]

Microsoft doesn't needs to sue. _Squeezing_ is much more profitable. And if there are profits to be had, they *will* squeeze. Look at them extorting Android phone makers, squeezing almost as much juice as they are losing with their own mobile efforts, plus Xbox, plus Skype.

And if push comes to shove, unless you're the size of Samson (I mean, Goliath; no, Samsung!), you're going to crumble. How old Mono is is irrelevant; that it's always been a financial nullity, that's what matters. And as long as it remains so, it should be safe to use, sure.

Re:Beware Rust, Go, and D.

[ahabswhale]

I can do the same thing with the JVM and choose from a number of languages (that are better than Java or C#), and I've been able to do this for a very long time. So this really adds nothing but another option but with shittier tools to work with.

Re: Beware Rust, Go, and D.

[senatorpjt]

Rust, Go, and D have completely different use cases than C#. It's like saying that C became irrelevant when C++ came out. Besides, Scala has more to offer than C#, and googling "EPFL lawsuit" gives you "Did you mean USFL lawsuit?"

Re: Beware Rust, Go, and D.

[Shados]

F# is a better functional language than Scala (and isn't plagued by shit like Scalaz and Akka. Also, even though Scala itself is an excellent language, the JVM underneath limits it in weird ways with issues the CLR doesn't have), and C# is better than Java (and in certain, limited cases, better than Scala).

So the combination of both and the CLR, once it is fully cross platform and instrumentation/support ecosystem is available, would blow them out of the water. It probably won't happen because of the microsoft stigma though.

Re:Beware Rust, Go, and D.

[Shados]

There really isn't a neat for that many languages, even though you need quite a few to cover all the relevant use cases.

The biggest issue right now with languages though, is that you have language A, that specializes in a very particular use case. Then you have a group who wants A, but with an ecosystem and focus on another use case, so they create B. For a while, life is good, until zealots of A go and try to recreate A in ecosystem B, and vice versa. Then people need to create ecosystem and language C, and history repeat itself.

Case in point: Ruby/rails plaguing every dynamic language they can.

Re:Beware Rust, Go, and D.

[Darby]

AC blatant marketroid troll at +5.

Truly sad to see this trash around here.

Re:Beware Rust, Go, and D.

[shutdown -p now]

Ms-PL is basically ancient history by now. Most of Microsoft's open source projects are released under either Apache License 2.0 or MIT License. CoreCLR (the current effort to open source .NET), in particular, is under MIT.

Re:Beware Rust, Go, and D.

[shoor]

Java is the darling of open source zealots? I didn't know that. You mean I've been leary of Java all these years for no good reason?

Re:Beware Rust, Go, and D.

[shutdown -p now]

D is aimed more at replacing C++, and definitely ventures onto C#/Java turf as well. It's not a good C replacement because of GC requirement (the base language doesn't need it so long as you avoid some features, but the standard library effectively makes it compulsory). Rust is aimed at both C and C++, with an emphasis on being usable for system programming (kernel etc).

Rust and C# are most certainly not competitors.

Re:Beware Rust, Go, and D.

[shutdown -p now]

Rust and C# target two different niches. Rust is basically a "better C++", and is suitable for kernel development and such. C# is more high-level and is best for userspace UI and web apps.

Re:Beware Rust, Go, and D.

[Beefpatrol]

It would probably be more reliable for Microsoft to just write front ends for their compiler for these languages rather than relying on FUD to prevent people from using them.

Re:Beware Rust, Go, and D.

[Beefpatrol]

One theory about why the big distributions are pushing systemd is that they want to make all the Windows admins currently migrating over to the Linux world more comfortable. They may see Mono, and the resultant large pile of Steam games that now run on Linux as the biggest opportunity open source has ever had to pound nails in MS's coffin. Another theory is that various people in the open source world have been bought in one way or another. Still another theory is that there is now a generation of otherwise great programmers that are basically clueless about the virtues of the UNIX way and think that they have made a significant improvement on those flat config files in specifically named directories and flat log files and cron by making a big monolithic brick that does all the same stuff only less conveniently.

Re:Beware Rust, Go, and D.

[Beefpatrol]

"Pulling JVM into the equation not really helps either, cause the consequent question would be: Do you trust Oracle? Or Google, for that matter if you count Dalvik in."

There is an open JVM. There is not a fully open CLR or a fully open clone of the CLR. So no, I don't trust Oracle, or Google, or IBM, (you forgot IBM's JVM,) but that doesn't really matter. Nobody who cares about their software stack being maximally open should be super-enthused about Mono.

Re:Beware Rust, Go, and D.

[Beefpatrol]

I would love to believe that MS really is a different company. Realistically, if MS isn't a different company, then it is circling the drain, (albeit, from a great distance still.)

But nobody should believe that a company that has for decades made tons of money by selling units of something that has no incremental cost of production, (copies of software or licenses to use them,) and which is publicly traded and whose investors expect it to do more of the same, will suddenly embrace a project that promises no direct revenues and which will compete with their existing product.

One of the ways you can tell that MS doesn't understand open source is that they are pushing Mono. There is a reason that Mono has been around for more than 10 years without gaining any real traction. It is a clone of .NET. Why on earth would the open source community care about cloning .NET? Particularly back when Mono started, and still now to a large degree, this is what the open source community mostly seems to say about Mono: if you want a .NET runtime, run Windows. What is the point of running .NET stuff on, say, Linux?

Re:Beware Rust, Go, and D.

[Beefpatrol]

Java is *not* the darling of open source zealots. Java, (the various JVMs,) is just the least annoying of the cross-platform, business-friendly choices.

Re:Beware Rust, Go, and D.

[Beefpatrol]

Java development also slowed down a lot when Oracle bought Sun and thereby gained ownership of the most commonly used JVM. Something similar happened to MySQL.

Re:Beware Rust, Go, and D.

[Beefpatrol]

The GPL does not prohibit anyone from producing software and then not distributing the source code. The GPL prohibits someone from distributing a binary derived from GPL code without also making the source that compiled into that binary licensed under the GPL and available. An entity can take GPL code, modify it, write stuff that uses it, whatever, and as long as they don't distribute binaries of the derivative work, they are not obligated to release any code at all.

Re:Beware Rust, Go, and D.

[jma05]

> Both C# and C++ offer low level functionality

Not really. Can you write a device driver in C#? How about a plain DLL? CLR is a VM. Its CPU performance is OK (2-8 times slower than C).

http://benchmarksgame.alioth.d...

But programs written on it have memory requirements that are higher than ones written in plain systems languages. The runtime footprint on the disk is also massive. I don't think you can really make a case that C# is a low-level language. It is not that much more CPU efficient than Java. Mono performance is worse than Java.

http://benchmarksgame.alioth.d...

Of course, CLR is better than dynamic language aka scripting language runtimes. But that's about it.

Re:Beware Rust, Go, and D.

[MechaStreisand]

Holy shit, you're right. That's at +5. They need to get rid of the shills around here.

Re:Beware Rust, Go, and D.

[jmac_the_man]

Exactly. But it's the MOST restrictive Open Source license MS has ever used, which is why I picked it.

Re:Beware Rust, Go, and D.

[david_thornley]

The FSF is for Free software, which is also Open Source. They'd rather everybody use some GPL version, or at least something GPL-compatible. The statement about C# reminds me very much of rms's statements about Java a long time ago: that anything controlled by a large corporation is dangerous to use, which seems like a reasonable statement to me.

Re:Beware Rust, Go, and D.

[david_thornley]

Lots of the ancient C-compatible cruft in C++ is no longer necessary, and can be avoided, with the avoidance enforced by code review.

No non-smart pointer should ever be used for ownership. C-type arrays and strings should only be used for communication with non-C++ entities. Pointer arithmetic is almost always avoidable.

Re:Beware Rust, Go, and D.

[Tablizer]

It would be nice if the new CEO made a general and spirited speech about MS mending its ways. I know speeches are just words and not deeds, but the first step in recovering from Evil Addiction is to admit you have/had a problem.

Re:Beware Rust, Go, and D.

[Pi1grim]

>> There is simply no shady language in the license that is going to affect Mono. If they ever decide to change Mono from an implementation of the .NET platform to something else (eg. JVM) whilst retaining Microsoft's code then they could be in trouble. But do really think for a second that they would change the focus of the project like that? Absolutely not.

Imagine a situation where MS introduces changes making .NET incompatible with Linux (or making it run several times slower) on specification level, so any implementation of .NET stack must implement them. Mono can't cut the changes out, they'll get their license to use .NET code revoked, you can't fork .NET for the same reason. Endresult? Embrace, extend, extinguish.

>> So your response to me pointing out that Microsoft hasn't actually sued anyone for the last 13 years despite all the claims that they would is that Java is still bigger and that if you made something that was unlike Mono that you would get sued. How is that counter my claim that saying that using Mono will not get you sued?

Neither Sun nor Oracle sued anyone for long time, untill one time - they did. As for "using Mono" - as long as you are end user and use it to develop your apps - MS is happy, since you're tying into their platform.

>> Irrelevant. You have the ability to verify code from an open source project [github.com].

Actually, it is relevant, as significant portions of whole stack are still closed sourced. Also, license of MS .NET strictly prohibits use of .NET code in projects that don't fullfill MS requirements for .NET stack, which, surprise-surprise, they control. So should they feel the need to introduce specific changes preventing .NET usage (or hindering it) on platforms other than Windows - they change the specs, good luck then. Microsoft vision of open-source - is they let you look at the code and promise not to sue for using it, but if you want to contribute - you mus give them the rights to the code and license for the patents and any changes you make must be made according to MS vision. Effectively, this clause about .NET code being usable only in .NET stack adhering to MS standard means that should MS feel the need to change anything - you won't get to fork it or do anything with it.

Re:Beware Rust, Go, and D.

[jma05]

I am aware. .NET Native just removes some JIT costs and improves load times (60% is the claim). It does not make actual program execution much faster. I don't expect the benchmarks to change much. Although, MS CLR was probably a bit better in performance than the Mono implementation. Let's see how this code merge fares. There are free and commercial native compilers for Java. They don't help all that much. I imagine things to be similar. No idea about how good the disk and memory improvements will play out.

the Qt is vastly superior to .net

[Anonymous Coward]

the Qt is vastly more stable and useful cross platform than this patent lawsuit pending bullshit....

Re: the Qt is vastly superior to .net

[Entrope]

I haven't looked at Qt 5 -- does it still insist on duplicating the STL except with a horrible naming convention and poorly justified design changes, making it painful to use any non-Qt C++ library?

Re: the Qt is vastly superior to .net

[Urkki]

"Insist duplicating"? What do you propose as alternative, when you don't want to destroy source compatibility of old applications? Qt 5 API is almost identical to Qt 4, no big break like after Qt 3, and having it any other way would have been very bad, because underneath Qt 5 has major improvements (starting from being able to connect C++11 lambdas to Qt signals, and being able to construct QStrings at compile time). So most apps still under development should and can easily be upgraded to Qt 5 (while staying Qt 4 compatible if desired, with a few #ifdefs), and this is a great thing.

It's not Qt's fault it took so long for C++ to get its act together. "Horrible naming convention" is also rather subjective. I find using undescore in names ugly, and boost-like nested namespaces hurt my eyes. Yet many seem to like them, and I don't judge them for it.

Note, tone of your post sounded like you're trolling, but I'm giving you the benefit of doubt. If you are trolling, ignore above two paragraphs, and if you aren't, ignore this paragraph :-)

Re:the Qt is vastly superior to .net

[RogueWarrior65]

Yup. Aside from being written in common C/C++, not only does it work cross-platform for desktop environments but also for tablet/phone environments.

Re: the Qt is vastly superior to .net

[Jezral]

Qt's containers expose STL-compatible iterators, making it quite easy to use with any other well designed library. Sure there's some baggage due to history, but you can mix Qt into other code very nicely these days.

Re: the Qt is vastly superior to .net

[Entrope]

As you point out, Qt 4 broke source compatibility in a major way -- so obviously it is not the kind of showstopper you suggest it should be. I would propose using idiomatic C++ design approaches, rather than sticking to architectural decisions that made sense 20 years ago before there was much consistency between C++ compilers. Nowadays, there is no good reason to prefer QString over std::string or std::wstring (and many good reasons to prefer the latter), and the same applies to every Qt container type. The Qt idiom of pass-by-copy-on-write-value makes runtime performance hard to predict, requires care in multithreaded use (do all types implement COW in a thread-safe manner?), and is very much at odds with the standard C++ library. Qt's efforts to make things "just work" end up hiding build-time, storage and execution-time costs, making it hard to figure out how to optimize code.

As you say, it is not Qt's fault that C++ took so long to really be a cohesive, modern language -- but it is Qt's fault that it continues on as if the state of C++ were the same as it was 15 or 20 years ago.

Re: the Qt is vastly superior to .net

[Entrope]

Non-iterator types? In *my* C++? It's more likely than you think!

Re: the Qt is vastly superior to .net

[walshy007]

As you point out, Qt 4 broke source compatibility in a major way -- so obviously it is not the kind of showstopper you suggest it should be.

The transitional period was quite painful for a great number of projects. It was only done because it was a necessary evil at the time.

Compatibility is not something people should jump at the chance to break if it can be avoided unless there is sufficient benefit. It has been judged that there is insufficient benefit to doing so currently.

Re: the Qt is vastly superior to .net

[Entrope]

What made the Qt4 breakage "a necessary evil" but also prevented them from adopting remotely modern C++ principles at the same time? Why was that breakage good, but making it practical for developers to adopt standard C++ practices could, and can, be so lightly set aside?

The sad thing is that Qt people are probably going to remain stuck in a 1990s mindset about C++ as long as people like you are willing to make apologetics for their misdesigns.

Re: the Qt is vastly superior to .net

[Urkki]

The Qt 3 to Qt 4 transition was painful enough for too many projects to make similar breakage very unpalatable for the community. Sort of "been there, done that, didn't enjoy it" situation. And what you are suggesting (replace every container etc with a different one) would be even bigger breakage.

And of course from cost (be it money or time) point of view, touching basically everything in the Qt source code to strip out Qt classes would be gigantic work. Not going to happen, there's neither business case nor developer demand for breaking everything, so there's almost nobody who has any interest in making it actually happen.

Qt COW is thread safe, programmer does not need to worry about that. For performance critical parts, feel free to use something else, but for the usual GUI code, COW is very convenient. Also remember, Qt 5 is already so old that it couldn't require move semantics support from C++ toolchain, which is also a performance issue.

Moving forward, what I hope for future is, C++ enabling reflection and introspection (in ways required by Qt features) without extra code generators (like Qt's moc). That'd be great for Qt 6, if such a thing ever comes. Another thing which would greatly benefit Qt is C++ removing the need for manual implementation of the PIMPL idiom. And supporting C#-like extension methods might make transition to standard C++ containers more likely, with missing functionality added as extension methods (current need to mix methods and free functions is not very nice IMO).

Re: the Qt is vastly superior to .net

[Pseudonym]

It's so cute how the young people of today think that Python is a real programming language because "bindings".

Re: the Qt is vastly superior to .net

[Pseudonym]

And a related question: Is Qt 5 exception-safe yet?

Re: the Qt is vastly superior to .net

[Pseudonym]

The sentence is correct, because professional linguists.

Only competes with other OO languages

[Anonymous Coward]

Go isn't an OO language, and events in the C# world have very little impact on it.

People who have programming preferences specifically in OO or non-OO languages on principle aren't really affected much by what goes on in the opposite camp, as the design and programming mindsets in each one are really quite different.

There isn't even any strong interaction through the jobs market, because the skills and design concepts used in each sector are really quite distinct. An OOP person talking to a non-OOP person about design don't really speak a common tongue.

Re:Only competes with other OO languages

[Daniel Hoffmann]

The CLR like the JVM can host other languages, there is even a lisp for it ( http://clojure.org/clojureclr ). I am not that familiar with the CLR but I think that only "system languages" can't really be implemented in it.

A promise?

[Opportunist]

A "promise" not to sue? I'll take that as soon as the various studios rip out their copy protection in return of their customers' promise not to copy their crap.

Goodbye Mono, your license stinks.

[goruka]

While Microsoft released CoreCLR under MIT license, Mono runtime is LGPL. This makes it unusable on mobile (according to their own words, unless you pay) and more closed platforms, and beats the purpose of a platform independent VM.

CoreCLR has no such restrictions and is gathering an enormous developer community, so I can't wait until we can kiss Mono goodbye.

Mono practically useless

[Eravnrekaree]

After looking at Mono I failed to see the point in the whole thing. Thats because it does not support the WPF. Since a large number of .NET applications are GUI, not having WPF pretty much destroys the value of Mono in allowing Windows .NET programs to run on Linux. Otherwise, there is no point in using Mono. If you have a .NET program written using WPF its not going to run on Mono. If you are writing a new program there is no reason to use Mono instead of another application language such as Ruby. Using a development environment designed by the Evil Empire does not hold special appeal over the FOSS plartforms such as Ruby. If one has to write a program that can run on both Windows and Linux i would probably be better to use Ruby or Python or such.

Why didnt Microsoft Open source the WPF. Instead, they open sourced the parts of .NET that Mono already had implementations of.

Re:Mono practically useless

[Dwedit]

I've never seen WPF used before. Forms are used much more than WPF.

Re:Mono practically useless

[Anonymous Coward]

Mostly because WPF is built on top of DirectX which in turn has deep integration with Windows. They might be able to separate it in the future but I wouldn't hold my breath. Xamarin.Forms will probably be the roadmap for XAML, WPF's layout language, on other OS platforms if it gains enough traction.

Re:Mono practically useless

[gbjbaanb]

I've seen a couple of WPF GUIs, invariably they're horrible - like what you'd write back in the 80s. I once saw a GUI that had orange and blue colouring with red highlight bars - yup, highlight the selected orange text with a red bar, reducing the distinction between the colourblind and those with perfect vision.

Its pretty slow too, a colleague had to drop it for his audio GUI as it simply did not respond quickly enough whereas the old C++ GUI worked perfectly well.

But its the latest cool thing so all the cool kids want to use it. I guess it's keeping them from node.js this week, so maybe its a good thing, but I can see why the Windows team says to write GUIs in html5 rather than WPF - if you need a plain LoB GUI, HTML works, if you need performance then you'll be doing it in C++ with something more akin to a game engine. WPF fails on both counts, and is too complex to boot.

Re:Mono practically useless

[gbjbaanb]

WPF is very over-rated. that has poor hardware rendering that doesn't work as well as old winforms

Maybe Microsoft isn't open sourcing WPF because they know how bad it is. Only the .NET fanbois are still going on about how wonderful it is, even though the majority of UIs I've seen on Windows are using ASP.NET or Winforms.

Re:Mono practically useless

[Billly Gates]

Winforms emulate com underneath. I remember a long time ago there was dcom/com on Linux. Do no not know if it is still active or how well it integrates with Gnome or XFCE?

Problem is we excpect it to be like Java and just work across platforms. C# maybe started out as J++ as a set of apis for java, but it is not java anymore. It is not python either that has gui components and frameworks designed to be cross platforms. It is not PHP either which that has quirks which make running it on Linux far best to this day.

The compiler is free but in 2015 it is about including code. That code is win32 specific. .NET would be great with gnome calls or even MacOSX specific ones. But it is not an all around 100% solution for java like cross platform compatibility. MS is discontinuing WPF anyway from what I read to focus on universal apps apis in Windows 10.

I believe Beagle was one such c# program for gnome that was successful back in the day but was all 100% gnome with no emulating win32 stuff in it.

Re:Mono practically useless

[shutdown -p now]

WPF's perf issues don't eclipse the fact that it's one of the best designed UI frameworks out there in terms of just how powerful and flexible it is. Things that take many pages of code to implement in WinForms are often 2-3 lines of XAML in WPF. The only framework that I'm aware of that's close in terms of expressive power (and beats it on conciseness) is Qt in the most recent incarnations of it.

The reason why WPF is not open sourced is more likely to do with the fact that it's all built on top of Direct3D, which makes it non-portable even if it were open sourced. That said, the high-level layers (widgets etc) aren't really married to any particular renderer, and there's an abstraction layer in between that and D3D, so in theory that is not impossible.

The majority of .NET UIs use WinForms simply because it's been there for much longer (and WPF didn't really mature until about 4.0, which was in 2010 - whereas WinForms was there since 2001), so most LOB apps started using that, and there's no reason to switch an existing codebase that just works. It's also closer to VB and Delphi. and so scores of developers using these two have been retrained to use WinForms with minimal hassle; WPF requires a lot more investment (but the results are also generally far better, because it is designed around strict model/view separation from grounds up).

ASP.NET is completely irrelevant here since it is a web framework.

Re:Mono practically useless

[gbjbaanb]

I'm not convinced its he best designed, nor is the chap in the first link I posted where he complains that in 9 years of using WPF it hasn't gone anywhere and is pretty complex and bloated to use.

Now a well designed UI is Qt - where they enhanced the existing model with their QML language, so you could create a new control in QML and drop it onto an existing form. All the power of QML whilst maintaining the existing investment in UIs. Qt got it right, shame Microsoft didn't have those devs working for them! (perhaps MS should buy Qt out and use it for new developments, probably too late now they've released Windows 10, but it would have been cool, wouldn't it!)

Imagine if Microsoft had done that instead of reinventing the GUI wheel - you'd create a control in WPF and could drop it onto a Winforms dialog. That would have been good.

As it is, WPF is just now very well designed, and not vey well implemented. As the other link showed, the chart controls only work for small datasets, if you want something that works well - you use the old winforms one! I'd hazard a guess that they are also hardware accelerated purely by being constructed by gdi calls too.

So what's the benefit in WPF? It mignt be easier to construct new controls than winforms but is more difficult to use as an application developer, it performa much worse, and requires a lot more investment in training. If the end result isn;t acceptable to users too.. then its no wonder people are sticking with Winforms.

Or HTML GUIs, which is what I meant by ASP.NET.

Re:Mono practically useless

[shutdown -p now]

I'm not convinced its he best designed, nor is the chap in the first link I posted where he complains that in 9 years of using WPF it hasn't gone anywhere and is pretty complex and bloated to use.

The "it hasn't gone anywhere" claim is plain wrong. He gives a very simple "Hello world" sample which hasn't changed (and why would it?), and then contrasts it with the same in ASP.NET and ASP.NET MVC - which are of course different, because they are two vastly different frameworks! There were plenty of major incremental improvements in WPF since then, as the changelogs clearly show.

What hasn't changed is the basics of the framework. And yes, I agree that some of it is not too well designed - e.g. XAML syntax is unnecessary verbose, and QML is much better in that regard. But most of those are surface issues. The core design - the notion of element tree, layout engine, layout and render transforms, data binding, styles, triggers etc - is solid, very powerful, and very flexible.

He's also complaining about things which don't have much to do with WPF - for example, MVVM, which is a generic UI design pattern (which I personally find unnecessary in WPF, because its data binding facilities are flexible enough that it can bind directly to a properly designed model, with converters effectively covering the VM layer).

Some of the point-by-point small things that he brings up are plain wrong, e.g. "Allow binding events directly to methods instead of via commands" - you could do it since WPF 1.0, right in XAML. Some are valid, but can be easily extended by third-party libraries, e.g.: "Allow C# expressions like basic boolean logic instead of requiring converters all the time" - as it happens I have implemented this personally. Some points (again, mostly about XAML syntax and verbosity) are valid but are minor. In any case, the only other comparable framework, even taking into account all these idiosyncrasies, is Qt+QML - and it took Qt a few years longer to even ship a first version of that, and then a couple more years to get it to usable shape. I'm not aware of any other viable competitors in this area.

Imagine if Microsoft had done that instead of reinventing the GUI wheel - you'd create a control in WPF and could drop it onto a Winforms dialog. That would have been good.

You can actually do it (and also the other way around).

The problem is that it's very hard to marry a UI framework that uses immediate, resolution-dependent, non-compositing 2D rendering (WinForms) with a transformation- and composition-based pipeline (WPF). It is possible only if you rewrite the former to be like the latter, but that also breaks backwards compatibility. Qt did that during the 3.x -> 4.x transition, and if you remember that, it wasn't exactly a smooth move.

As it is, WPF is just now very well designed, and not vey well implemented. As the other link showed, the chart controls only work for small datasets, if you want something that works well - you use the old winforms one!

WPF grid is slow largely because it uses the WPF layout engine, which is optimized for at most a few hundred widgets on the form, rather than many thousands of small cells. There are plenty of third-party grid controls for WPF that are blazing fast. By the way, the original WinForms grid control was also painfully slow (for some of the same reasons), and it took until .NET 2.0 to make a new one that was fast.

And remember, WPF development began some time in 2003, and the first alphas went public in 2004 - and the first release was in 2006. This was back whe

Re:Mono practically useless

[benjymouse]

- e.g. XAML syntax is unnecessary verbose, and QML is much better in that regard. But most of those are surface issues. The core design - the notion of element tree, layout engine, layout and render transforms, data binding, styles, triggers etc - is solid, very powerful, and very flexible.

The beauty of WPF and (especially) XAML is that they really are 2 very different technologies. WPF - like many GUI frameworks - describes an UI through an object graph. You can "new" up the controls yourself, wire the event handlers and achieve the exact same result as describing it through XAML. Which also means that you can dynamically change the graph through code.

XAML on the other hand is not in any way specific to WPF. It is simply a way to describe an object graph. Only, XAML can describe very, very complex object graphs with wired event handlers etc. XAML is also used to describe workflows in Workflow Foundation. And anyone can use the format to describe object graphs of any other type. XAML uses a convention for mapping XML namespaces to CLR namespaces. That is a really, really cool.

Re:Mono practically useless

[shutdown -p now]

Kinda true but with a bunch of important gotchas. WPF XAML is not the same as general-purpose XAML (try using some of the new XAML 2009 features in your WPF markup, like say generics or parametrized constructors!). WPF XAML pipeline has BAML in it, and all the associated tools are very specific to WPF. It does not use System.Xaml.

In any case, XAML is not a particularly good way to describe object graphs, mainly because it's just way too verbose (especially the element property syntax) - largely because it is XML. And it's not even like you get any usual benefits of XML - because XAML gives a bunch of ways to describe the same things (e.g. attribute syntax vs element syntax for properties), and sometimes even uses non-XML DSLs requiring special text parsers for some bits (e.g. markup extensions inside attributes), in practice you can't reliably use XSLT and other general-purpose XML tooling to reliably process XAML graphs - you need to use a XAML parser that will construct the actual object graph for processing, or at least deal with object graph events. So basically you're suffering the XML verbosity without getting any ecosystem advantage.

Something closer to JSON would be a much better fit for this purpose. Take a look at QML if you haven't seen it already - this is what XAML should really look like.

Re:Mono practically useless

[rdnetto]

So use a different GUI library. WPF is pretty bad even when compared to WinForms. The Qt libraries are cross-platform, have existing bindings for C#, and are an absolute pleasure to work with. IMO, they blow the pants of WPF.

The value in Microsoft open sourcing .NET isn't in running existing Windows programs under Linux - they probably wouldn't work without WINE anyway, due to P/Invoke calls or lazy programming like concatenating backslashes instead of using Path.Combine(). The value is in being able to use C# to write new, cross-platform programs. And while Mono did have implementations of them, they didn't necessarily work the same way. I gave up on it completely after a particularly annoying bug prevented .NET binaries from running as a less privileged user.

C# is an excellent language, and arguably the best-positioned language in its niche - C++ has too much cruft, Java is struggling to catch up (and seems to lack a decent GUI library), D's userbase and funding is close to non-existent, and Objective-C/Swift aren't used outside of Macs. All the other languages are too immature, or lack the requisite feature set, to be competitors.

Newbie Mono question

[RogueWarrior65]

Can you run a .NET application that currently resides on a Windows-based web server on a Linux-based shared hosting server using Mono?

Re:Newbie Mono question

[benjymouse]

Can you run a .NET application that currently resides on a Windows-based web server on a Linux-based shared hosting server using Mono?

In general: Yes. Need more context about the application to give a definitive answer, e.g. if it uses Windows specific infrastructure such as AD, Workflow Foundation.

Re:Newbie Mono question

[Anonymous Coward]

Probably not. The vast majority of .NET libraries only work on Windows and there has been absolutely no indication that this is going to change any time soon.

Re:Newbie Mono question

[Billly Gates]

Problem is let's say there is a bug that is causing your web app to constantly run out of threads or restart?

Who do you call for support? Let's say you think it is mono causing it? WIth VS.NET on Windows you see the bug is not there. Or if it is you can go to MSDN and find others with the same issue and work arounds and a promise from QA that it will be worked on if it is big enough in a Windows update.

I do not like Mono for this reason.

Like C++ there is a big difference between one written for Unix and one with Win32. .NET is great but not if you make calls that emulate Windows. That is not good. Winforms is an example too which uses dcom/com underneath. It would make more sense to use GTK calls if it is a Linux app.

Re:Newbie Mono question

[nikhilhs]

If you're using ASP MVC, then yes. Check this out. http://gunnarpeipman.com/2014/...

Re:Newbie Mono question

[benjymouse]

Problem is let's say there is a bug that is causing your web app to constantly run out of threads or restart?

Who do you call for support? Let's say you think it is mono causing it? WIth VS.NET on Windows you see the bug is not there.

Isn't that a concern with open source in general? That argument could be used against all open source projects where commercial support is not available. Yet, many open source projects thrives despite of this.

.NET is great but not if you make calls that emulate Windows.

But Mono does not make calls that "emulate Windows". In general the call upon native and/or open source libraries. Certainly you'd be hard pressed to come up with any examples of this behavior in the Mono server stack.

Winforms is an example too which uses dcom/com underneath. It would make more sense to use GTK calls if it is a Linux app.

But you are wrong about that. Winforms is definitively NOT based on COM (much less DCOM).Winforms is a thin wrapper around Win32 APIs. When you create a text box in Winforms, you'll get an actual native Windows textbox.

You may be confused by the fact that Winforms also allow ActiveX controls to be used. When you use that capability you will be using COM (not DCOM), as ActiveX controls are implemented using COM. Interestingly, the part of COM that makes this possible is remarkably similar to the object model of Gnome, almost binary compatible. Basic COM is a binary standard which can be implemented on any platform out there.

Re:Newbie Mono question

[shutdown -p now]

It depends. If it uses, ASP.NET MVC, almost certainly yes. If it uses classic ASP.NET, probably yes. But it all depends on the libraries being used. Also, perf may vary wildly.

The .NET Core / CoreCLR effort by Microsoft is aiming to change that. It's basically an open source reboot of CLR and ASP.NET MVC that starts from scratch, ditching backwards compatibility (in particular, a lot of the older parts of .NET Framework libraries, esp. those that were Windows-specific) and generally embracing modularity over monolithic design, with only the very basic libraries being a part of the framework, and everything else installable packages. It specifically has Linux and OS X compatibility as one of the goals, and is designed to give the same set of features and good perf on all platforms.

Re:Patents?

[benjymouse]

Quite simply, a patent "promise" is not the same thing as a license. You see, even if they're bared by Laches, they can still drag you through the courts and you've got to prove they're barred by making the promise. If you had a license...you could make a single motion at the first hearing or in the pretrial motions to dismiss because of being licensed if they sought to sue you.

Having this crap in there means Mono's toast without a real license to any valid patents, combined with a covenant to license all tech as it becomes apparent, that ends up in this common core of stuff. Otherwise, you're INSANE for using it- because you can and most probably WILL be sued over it.

No - it is actually stronger (look up promissory estoppel). But leave that aside, because the patents have also already been granted.

The *promise* was issued because fanatics cried foul at the patent grant, arguing that Microsoft with it vast army of lawyers could just sue any OS project out of existence, patent grant or not. Hence, Microsoft issued the promise, all but ensuring that such a case would be outright dismissed since you've acted in good faith on a promise. The promise in that case is actually one of the strongest contract forms imaginable, as it is one-sided: you do not have to sign anything to be covered.

Re:Patents?

[Bengie]

Don't worry, Microsoft owns the patents on several IPv6 RFCs, but they promise not to sue anyone. It's the exact same promise that they gave for .Net. Really, the exact one. They have a spot on their website with this promise and the list of all the patents they promise not to sue against, which includes both C#, .Net, and IPv6, among many other things.

Let me guess, now you refuse to use Linux because Linux implements several of these MS patents in order for IPv6 to work.

Re:Patents?

[MightyMartian]

"All but insuring" is not the same as "impossible." I wouldn't touch Mono or any other .NET runtime or compiler with a ten foot pole. There are other cross-platform/cross-architecture development tools out there, so one can completely avoid .NET and still write software that runs on all major platforms. In fact, pushing .NET in the way you do makes me rather suspicious. There's nothing Microsoft would love better than heavy entrenchment in other platforms so it can be turning the screws.

Microsoft has many times expressed its visceral hatred of open source. It is not to be trusted, not ten years ago, not five years not, not today, not ever.

Re:Patents?

[Bengie]

All kinds of patents, including DNS, DHCP, DNSSEC, ISCSI, RADIUS, SMTP, POP3, DSA. MS owns the patents on many RFCs, but are under their generic "Community Promise".

Has 4.0 actually been released yet?

[kthreadd]

From the linked release notes:

THIS IS A DRAFT OF THE 4.0 RELEASE NOTES

I also can't find the 4.0 tarballs on the download page, which still says that 3.12.1 is the latest Mono release.

Re:Has 4.0 actually been released yet?

[kthreadd]

Looks like that's the case

It is a shame that this got submitted before we actually published the code. It is Easter and many of our engineers are taking these days off.
The release notes are also incomplete and not ready for publishing
Miguel

Re:Patents?

[benjymouse]

Microsoft has many times expressed its visceral hatred of open source. It is not to be trusted, not ten years ago, not five years not, not today, not ever.

BS again. Microsoft has NEVER expressed visceral hatred of open source. Ballmer has compared one open source license - GPL - with cancer, because of it's viral nature. The intentionally viral nature.

Ballmer is not at the helm any more. But even he never expressed hatred at open source, as you claim. You could construe his comments about GPL as hatred against that particular license type. And indeed, Microsoft has always opted for other OSI approved licenses when they had the choice.

But if you have any other sources for your made-up claim - say other MS top executives, maybe even present ones - then please feel free to post them.

off topic

[louden obscure]

[rant] Enough of this ms cheer leading. Just stop. /. is being assimilated.and no one seems to notice/care. Jeeze even debian is being assimilated. by systemd. WTF is happening?

"Don't fight it, Miles, it's no use. Sooner or later, you'll have to go to sleep." [/rant]

Re:off topic

[SuiteSisterMary]

So freedom to use whatever software you want, as long as it's the software you approve of?

Re:off topic

[Beefpatrol]

I have been very concerned about this as of late. (Not slashdot, I mostly gave up on it a long time ago, there are other sites that have sort of picked up where slashdot left off, however.) The only thing that makes me feel a little better about systemd is that it is actually GPLd, and the distros I have looked at that are switching/have switched to systemd still have full-fledged init.d packages. The "roll another distro" is probably the right answer. That being said, I think the real reason to be concerned is that there has been a stunning lack of transparency on the part of the distros that are switching. Reasons that have been given are clearly BS. It appears that docker and similar things benefit from systemd, but it isn't obvious how. I wish I knew. Gentoo FTW.

Re:Patents?

[MightyMartian]

It would take a delusional lunatic not to know the long history of attacks against commercial and open source competitors. Microsoft isn't trustworthy, and as there are alternatives to .NET, the easiest way to protect against future bad behavior by Microsoft is to use those alternatives. Why risk future woes when you have no need to?

Re:Patents?

[benjymouse]

It would take a delusional lunatic not to know the long history of attacks against commercial and open source competitors.

Then you should have no problems finding a few examples that illustrates Microsofts visceral hatred of open source (your words).

... long history of attacks against commercial and open source competitors

Fear!

Microsoft isn't trustworthy,

Uncertainty!

Why risk future woes when you have no need to?

Doubt!

As I suspected: Nothing but FUD. But pretty textbook FUD, that much I have to give you credit for.

Title is wrong...

[Anonymous Coward]

Mono 4.0 was not released yet. This is a draft of the release notes. Is it so hard to read the page?

Re:Patents?

[benjymouse]

Cases in point:

1. The ridiculous FAT long-filename patent
2. The subpixel rendering patent (despite prior art being shown)
3. Outright patent-troll behavior: Refusing to disclose a stack of patents its using to extort for-profit Linux distributors behind closed doors.

Thanks.

Which of the above illustrate Microsofts visceral hatred of open source?

(For the record: I believe that software patents should be abolished and I do not condone Microsofts patent litigation)

Re:Patents?

[HiThere]

Not ever is too extreme. I thnik that a decade of good behavior would cause me to be willing to trust them in non-critical matters. Of course the timeing of that is dubious. Should I start counting from the last time they threatened people with patents without saying which patents, or from the last time they extorted a payment?

Re:Patents?

[Burz]

Cases in point:

1. The ridiculous FAT long-filename patent
2. The subpixel rendering patent (despite prior art being shown)
3. Outright patent-troll behavior: Refusing to disclose a stack of patents its using to extort for-profit Linux distributors behind closed doors.

If MS comes out of the closet and enumerates #3 and opens a dialoge with the community about them, THEN I will believe their hype about being open-source friendly. Otherwise, they are in the business of growing their Android-derived revenue using submarine tactics.

Also, explain to us why MS shuts out FOSS AV and document formats (the consumer-oriented ones); Not only from their products but from standards-making processes.

Have a nice day, mods! :)

Re:Patents?

[Beefpatrol]

Actually, I believe Ballmer was the one who first publicly referred to open source as "open sores". I think that qualifies as at least some amount of disgust at open source, if not visceral hatred. And yes, the GPL's intentionally viral nature is the one thing that pisses MS off the most about it. I'm guessing it is because MS was started by the offspring of lawyers and they didn't think of it first. :) Don't forget Bill Gates' "Open Letter to Hobbyists", wherein he mentions that he is opposed to sharing software because it deprives developers of royalties. This is at least, against the "free" in "Free Open Source Software", though not against open source explicitly.

Re:Patents?

[marcello_dl]

> Then you should have no problems finding a few examples that illustrates Microsofts visceral hatred of open source (your words).

LOL alternate universes exist and you come from one. Or you troll very badly. Nice to see how a blatant troll collects mod points.

Microsoft CEO Steve Ballmer (...) calls Linux "a cancer that attaches itself in an intellectual property sense to everything it touches."

But these are words from a guy prone to monkey dance. Let's see.

From: Bill Gates
Sent: Sunday, January 24, 1999 8:41 AM
To: Jeff Westorinen; Ben Fathi
Cc: Carl Stork (Exchange); Nathan Myhrvold; Eric Rudder
Subject: ACPI extensions

One thing I find myself wondering about is whether we shouldnâ(TM)t try and make the âoeACPIâ extensions somehow Windows specific.

It seems unfortunate if we do this work and get our partners to do the work and the results is that Linux works great without having to do the work.

Maybe there is no way to avoid this problem but it does bother me.

But wait, poor Gates just wants it to be windows specific not to boycott FOSS, right?
Except that the opposite happened:

Some of us remember the story of why Linux kernel responds "False" when ACPI BIOS asks if the operating system is Linux. We have found yet another case where mimicking the Windows behavior instead of writing to the spec is the right choice if you just want your machine to work properly

What about the 50million cash to SCO?