Why Mono Is Desirable For Linux

[XorEaxEax]

I think it is the best for games. Just look at games like World of Goo

World of Goo was written in C++, they (2d boy) released a trimmed down version of the framework they used for WOG as open source.

[kazetsukai]

You've made it abundantly clear that you don't know what you're talking about, that post should be the final nail in the coffin of your alleged expertise.

Don't even bother, these people are paid-by-the-hour Microsofties spreading their Koolaid, even though noone's drinking. Linux developers don't need C# or .NET/Mono, there are clearly better alternatives like Java and Python.

They refuse to recognize that the apprehension towards Microsoft isn't the only reason Mono isn't taking off. Hell, they don't even recognize that the apprehension towards Microsoft is justified (which is just insane).

2. C# is intended to coax better code out of unskilled developers(aka: code for dummies)

They also refuse to recognize this. Like it or not, "automagic" memory management doesn't always produce the best (or even desired) results. It has always been up to the programmer to manager their own memory resource and clean up after themselves. Meanwhile, .NET pumps out "developers" that don't even know what memory allocation is, and we can guess how that will turn out by looking at the aftermath of VB6.

There is no C# program in existence that couldn't have also been done in C.

Yeah, they truly believe C# is a magic language that can do things beyond the language that it itself was written in. See, useless.

C#, however, is more limited than C because some of the low-level access to the CPU and memory is abstracted away.

They also refuse to acknowledge this- there are times where you want to or have to do your own memory allocation, or optimize special CPU instructions. They consider this an exotic, fringe case.

It is, beyond the shadow of a doubt, possible to create types and functions in C that can do literally anything that C# and the Mono library functions can do.

Would the M$ trolls care to try and debunk this? I think not, since its a solid brick wall of logic and reason.

There is no reason for a C/C++ developer to switch to using Mono/.NET to do development. None.

[directhex]

Don't even bother, these people are paid-by-the-hour Microsofties

Is that a formal allegation?

[kazetsukai]

Is that a formal allegation?

You tell me. Did you consider yourself a target of that statement?

-You're blaming the community (which is often very vocal) for the lack of adoption of Mono.
Well, knowing the community culture, its history with Microsoft up to now, you honestly believe that scaremongers in the community are the sole reason for the lack of its adoption?

-You seem to assert that Linux needs Mono to solve real world problems.
Developers targeting Linux have decent toolkits to get the job done. They have several high level and low level languages of which development of is directed by the community, not by a few corporate entities. Java, PHP, Python, Perl, Ruby, C/C++. These languages are used successfully for almost everything. Meanwhile, what use case do these languages not fill, that Mono is trying to solve?

Mono is a broken concept in Linux. You don't see a trillion entities trying to sell you a zip archive util or CD burning software under Linux, because free alternatives that are constantly updated exist.

[directhex]

You tell me. Did you consider yourself a target of that statement?

Evasion. How utterly unprecedented.

To flesh it out: the anti-Mono cult cannot begin to conceive of a world where people might disagree with them unless they're being paid to do so. It's a strange, somewhat egotistical view of the world, but it's very common. Since I'm not anti-Mono, I've often seen vague allegations of payment thrown around. Yet every single time - every single one - that I ask for someone to make a formal black and white "I am accusing you, directhex, of receiving X amount in payment from Microsoft for advocating a reimplementation of their technology on forums and here is my evidence" claim, silence. Almost as if it's just an effort to poison the well, to make others think "oh, I can't trust that guy, Microsoft pay him, it must be true because someone on the Internet said it"

-You're blaming the community (which is often very vocal) for the lack of adoption of Mono.
Well, knowing the community culture, its history with Microsoft up to now, you honestly believe that scaremongers in the community are the sole reason for the lack of its adoption?

Pretty much. When the scaremongers start applying their ire equally to other Microsoft-derived technologies like JavaScript, then there might be something to discuss. Otherwise, it's pure (weird) directed nonsense. I know of at least one project which was written in Python purely to avoid the bullshit politics that would have come with being written in C#, despite the original author's major preference (not least for performance reasons) of using C#.

-You seem to assert that Linux needs Mono to solve real world problems.
Developers targeting Linux have decent toolkits to get the job done. They have several high level and low level languages of which development of is directed by the community, not by a few corporate entities. Java, PHP, Python, Perl, Ruby, C/C++. These languages are used successfully for almost everything. Meanwhile, what use case do these languages not fill, that Mono is trying to solve?

Each of the languages you name has upsides and downsides - typically some massive plusses and massive minuses. For example, Python is easy to write, but hella slow. Java is fast (as long as you don't use the dreadful GUI toolkits), but uses all the RAM ever. Mono is an extremely compelling middle ground - not as fast as Oracle Java, but competitive. Not as memory-light as C++, but competitive. Not as "easy" as Python, but competitive. And so on. It's really a very compelling choice for a "pretty much good enough" high level language with no major downsides.

Mono is a broken concept in Linux. You don't see a trillion entities trying to sell you a zip archive util or CD burning software under Linux, because free alternatives that are constantly updated exist.

I... totally don't understand what you're trying to argue here. Mono is broken because there are unzip tools for Linux?

[XorEaxEax]

Java is fast (as long as you don't use the dreadful GUI toolkits), but uses all the RAM ever. Mono is an extremely compelling middle ground - not as fast as Oracle Java, but competitive. Not as memory-light as C++, but competitive. Not as "easy" as Python, but competitive. And so on. It's really a very compelling choice for a "pretty much good enough" high level language with no major downsides.

Looking at the benchmarks over at shootout I see no major difference in memory usage between Java and Mono, overall it's about the same (scroll down):

http://shootout.alioth.debian.org/u64q/benchmark.php?test=all&lang=java&lang2=csharp

Performance-wise though Mono C# is far behind.

[directhex]

Looking at the benchmarks over at shootout I see no major difference in memory usage between Java and Mono, overall it's about the same (scroll down):

http://shootout.alioth.debian.org/u6...a&lang2=csharp

The Shootout figures for Java memory consumption come out weird. One of the things that helps Java's performance a lot is the static heap size, which in the real world often needs to be very large. It's not uncommon for a "basic" webapp to need several gig to operate normally. I've filed bugs against Java apps before where the maintainer has fixed a small heap to try and keep the app memory usage sane, yet the numbers they use are too small for whatever reason resulting in crashes. The single-task nature of the shootout means the heap size issue doesn't come up in recent Javas (where the default is small & the size *can* increase if required)

Oh, and the 32-bit and 64-bit numbers are very different - in 32-bit, Mono does much better (or more accurately, Java does much worse) with a median of about 66% less RAM used by Mono. Basically Java uses about the same amount of RAM whether in 32-bit or 64-bit. Mono uses more in 64-bit (roughly twice)

Performance-wise though Mono C# is far behind.

The median is "Oracle Java is twice as fast". How about Python, which was the one I implied was crazy slow by comparison?

[XorEaxEax]

The median is "Oracle Java is twice as fast". How about Python, which was the one I implied was crazy slow by comparison?

Well the apt comparison is that of Java and C# as they are similar languages technology-wise and generally target the same domain (rapid development in enterprise), no surprise it is faster than python which is an interpreted dynamic language, just as there is no surprise that both Java and Mono is much slower and uses alot more memory than C/C++.

[kazetsukai]

Perhaps I can apologize for the accusations and bring you back to the several arguments you evaded previously. I apologize.

When the scaremongers start applying their ire equally to other Microsoft-derived technologies like JavaScript

That's obviously because of the amount of (zero) control Microsoft has over the standard. I did make a suggestion previously in this thread, and that is to get Microsoft uninvolved as rapidly as possible. That would solve more issues that people have with the standard than anything. A good example of this working is Java. I think at this point at least, the OpenJDK has a brighter future ahead of it than and closed version of Java, which is enough for developers to believe that its OK to actually write Java applications. Oracle wants a bite of Android (and further, a bite of Java), but they're just not going to get it. What about Microsoft in this case? As long as they steer where the spec goes, and the main implementation on Windows is the only functional one, it isn't going to take off. Mono attempts to solve #2, but not #1.

When the scaremongers start applying their ire equally to other Microsoft-derived technologies like JavaScript, then there might be something to discuss. Otherwise, it's pure (weird) directed nonsense. I know of at least one project which was written in Python purely to avoid the bullshit politics that would have come with being written in C#, despite the original author's major preference (not least for performance reasons) of using C#.

I think you're too quick to dismiss the reasons people hate, with passion, Microsoft.

For example, Python is easy to write, but hella slow. Java is fast (as long as you don't use the dreadful GUI toolkits), but uses all the RAM ever. Mono is an extremely compelling middle ground - not as fast as Oracle Java, but competitive. Not as memory-light as C++, but competitive. Not as "easy" as Python, but competitive. And so on. It's really a very compelling choice for a "pretty much good enough" high level language with no major downsides.

The major downsides are platform support, API completeness and ___MICROSOFT___.

I... totally don't understand what you're trying to argue here. Mono is broken because there are unzip tools for Linux?

I'm talking about the abundance of commercially sold software out there that are as rehashed and utterly mundane as archive management in the Windows realm. There just isn't a market for that crapware here.

[directhex]

That's obviously because of the amount of (zero) control Microsoft has over the standard. I did make a suggestion previously in this thread, and that is to get Microsoft uninvolved as rapidly as possible. That would solve more issues that people have with the standard than anything. A good example of this working is Java. I think at this point at least, the OpenJDK has a brighter future ahead of it than and closed version of Java, which is enough for developers to believe that its OK to actually write Java applications. Oracle wants a bite of Android (and further, a bite of Java), but they're just not going to get it. What about Microsoft in this case? As long as they steer where the spec goes, and the main implementation on Windows is the only functional one, it isn't going to take off. Mono attempts to solve #2, but not #1.

Do I need to re-paste the list of companies which steer the .NET specification? The list you clearly ignored when I posted it last time?

• Eiffel Software
• Kahu Research
• Microsoft Corporation
• Novell Corporation
• Twin Roots
• Borland
• Fujitsu Software Corporation
• Hewlett-Packard
• Intel Corporation
• IBM Corporation
• IT University of Copenhagen
• Jagger Software Ltd.
• Monash University
• Netscape
• Phone.Com
• Plum Hall
• Sun Microsystems
• ActiveState
• CSK Corp.
• Jaggersoft (UK)
• Mountain View Compiler
• Pixo
• University of Canterbury (NZ)

I think you're too quick to dismiss the reasons people hate, with passion, Microsoft.

I'm a technologist. Hating corporations blindly just doesn't work for me. Hate on actions, and hate on technology. Hating on a company - especially one with almost a hundred thousand distinct employees - is over-simple and pointless.

The major downsides are platform support, API completeness and ___MICROSOFT___.

So which platform support is the issue, exactly?

And which APIs? Hint: The only people I've *ever* seen moaning about API completeness in Mono are the anti-Mono cult. It's a red herring; an excuse. "Mono is terrible because it doesn't implement this piece of Windows API; also EVERYTHING TO DO WITH WINDOWS IS TERRIBLE". It's just speaking out of both sides.

I'm talking about the abundance of commercially sold software out there that are as rehashed and utterly mundane as archive management in the Windows realm. There just isn't a market for that crapware here.

I still don't see what on earth this has to do with Mono, other than tiresome redirection. "Windows has paid apps, and Microsoft makes .NET, therefore MONO IS TEH DEVILZ!"