Along with the stable maintanance release of Mono 1.0.2, Novell released the first beta of the upcoming Mono 1.2: The Mono JIT has been ported to a new architectures: AMD64, SPARC v9, and S390. In the runtime detection, support for side-by-side execution of applications that require different runtime versions was implemented. Mono now will detect the runtime version that an application requires, and will load the appropriate mscorlib.dll and machine.config.
Mono 1.0.2, 1.1.1 Released
About The Author
Eugenia Loli
Ex-programmer, ex-editor in chief at OSNews.com, now a visual artist/filmmaker.
Follow me on Twitter @EugeniaLoli
yay! amd64 and SPARC
this makes me happy
http://forge.novell.com/modules/xfmod/project/?bsd-sharp
The maintainer of lang/mono and other Mono projects, John Merryweather Cooper is currently unavailable. Tom McLaughlin is taking it over and maintainer it unoffical in novell.com project similar to MarcusCom CVS (GNOME project). It works very well on only FreeBSD 5.3-BETA and above, because of better thread, TLS and etc support.
Some apps that are written in C# might have some problems. If you find any, please report it to help get Mono support better in FreeBSD.
thx except XSP , life with mono on freebsd is far easier compared to java. but XSP performance realy sucks. (no ofence same with linux version). any info when we can expect performance boost?
thx except XSP , life with mono on freebsd is far easier compared to java. but XSP performance realy sucks. (no ofence same with linux version). any info when we can expect performance boost?
Have you read in the release note of 1.1.1? It has something what you are seeking.
http://www.go-mono.com/archive/1.1.1/
maybe people will stop flaming Microsoft and give them recognition that C# and the .NET Framework is a great platform.
Last time I was looking at mono on my gentoo system, I could not get the mono debugger working. Is that problem solved in either 1.0.2 or 1.1.1? I could care less about any new features if I have to debug by writing message to the console. =(
Or perhaps I’m just stupid and am missing something obvious…
The debugger has been broken for long stretches of time. I’ve never got it to work, but haven’t CVS updated in a couple months now.
Do we have Windows.Forms or whatever it’s called yet, otherwise I’ll stick to Python/wxWidgets ta.
I’m guessing there is still no FreeBSD 4.x support. I’m a longtime Java developer and I’ve been curious why Eugenia is so rabidly favourable to C#. (I prefer Perl personally, so I’m looking forward to Parrot.)
However I can’t even get Mono 1.0 to install.
— Begin Jaded Rant Mode —
Also from a brief examination of the documentation I can’t say that C# is inherently any better than Java. Everyone also says native C# guis are heaps better than Java ones, but I don’t see it yet. A thousand toolkits and associated dependencies doesn’t inspire me either. And what exactly is the point of a virtual machine if the apps aren’t cross platform.
I must say that so far I’m decidedly underwhelmed. I’ve been pretty underwhelmed by Java for a decade now (it finally sort of works on FreeBSD though) and .Net looks like being just as underwhelming.
— End Jaded Rant Mode —
writing virtual machines for different processors is not really a big deal. without real cross platform libraries and API’s, things goes hardly further than “Hello world” applications.
I am not going to lie, FreeBSD 4.x is useless for today’s technology but still very useful for basic stuff like webserver and other servers (no desktop). For the new technology like SMP, desktop, new languages and stuff should use FreeBSD 5.x that have have more support in libc, thread and etc.
Windows.Forms is under construction. But you can use GTK# or wx.net.
If I read it right with mono >= 1.1.* I could develop an application that communicate with the rs232, right ?
Also the _same_ code could run under windows too, right ?
Last time I checked the serial drivers for Java were different for Windows and Linux.
The Serial i/o is interesting for me so that I could write a cross platform GUI application that uses the rs232.
Rigth ?
I’m trying hard to understand the enthusiasm some people embrace mono, but I fail to feel the same:
* .NET has working versioning mechanisms. Since Mono relies heavily on native libraries (that are even unstable like Cairo) this major advantage of .NET depends on linux dependency and library versioning management
* Performance of mono is really bad. Run scimark and compare to java, c++ or MS .NET. There are lightyears inbetween!
* There’s no useable IDE for mono. Last time I tried MonoDevelop I couldn’t open aspx files and crashes were common.
* Support for GDI+ via Cairo was depending on a define with a name like MANLY_ENOUGH_TO_INCLUDE
* The garbage collector is in no way comparable with the generational java garbage collector. With 1.5 there’s even a flag -XX:MaxGCPauseMillis that tells the vm that no pauses longer than X milliseconds should happen due to GC.
Maybe Mono will get it right, but that will take time. For the time being it’s not an option for me.
I am excited to see open source community being able to do such a big project, and a product that is based on a MS Framework, although I’m not sure if it would be possible without the help of Novell.
Mono was well over the hump before Ximian was acquired by Novell. Having big corporate money obviously doesn’t hurt though.
I could care less about any new features if I have to debug by writing message to the console.
Slightly off topic, but integrated debuggers are eiter a tool for VERY SPECIFIC investigations into runtime code, or kludges for people who don’t know what they and their code is doing (aimlessly looping around in statements and checking variable values).
Debugging with writing messages to the console is more prodective and efficient. You can search for quotes from several major programmers regarding printf (console) debugging vs debuggers.
Furthermore, in practices like TDD (Test Driven Development) there is even less reason to debug with a debugger.
Indeed it is true, debuggers are not as useful as before. Especially in the agile development disciplines. However, this is not an excuse for not to provide one.
Besides, you need good logging frameworks if you do not wish to use the debugger. i wonder if there is a strong mono (ahem.. C#) logger equivalent of Log4j..
ah.. i forgot, i am sure there is someone lurking behind all java open source projects and frantically trying to convert it to C#.Net ..no sorry mono..
What are you looking for in a logger, aside from “equivalent to log4j”?
The built-in System.Diagnostics namespace provides many facilities to help with logging (via the Trace, Debug, TraceListener, Switch, and related classes). See: http://www.go-mono.com/docs/index.aspx?link=N%3aSystem.Diagnost…
There’s also MSDN: http://msdn.microsoft.com/library/default.asp?url=/library/en-us/cp…
For a more log4j-like library, there’s log4net: http://logging.apache.org/log4net/.
There’s also NLog, which tries to be simpler: http://nlog.sourceforge.net/.
I haven’t used NLog or log4j, but I wrote Mono’s System.Diagnostics trace implementation, and find it quite useful.
It seems like System.Diagnostics is more like a low level mechanism than hig level loggers. the others however, are more like the one i asked. whic are derived from log4j..
Debugging with writing messages to the console is more prodective and efficient. You can search for quotes from several major programmers regarding printf (console) debugging vs debuggers.
If all you do is write simple code, writing console messages is fine. That’s how I get by when writing small python scripts.
However, there’s no way I could have developed the things I have at work on a .NET implementation without a debugger. I don’t think I’m alone either. Most people who rave about VS.NET will mention the debugger, and there’s a reason for that.
Agile development also has yet to be made The One True Way ™. You can’t assume or demand that everyone will work like that. Personally, I’ve yet to be on a project that had enough time to maintain an automated test library in parallel with the application and all of the changing requirements that come with it and still have a chance of meeting the ‘aggressive’ timelines. Workplaces aren’t run on ideals, unfortunately.
Besides … just for the argument’s sake … Just consider me a sub-average programmer that doesn’t give a squat about ‘several major programmer’s’ oppinion of printf debugging. Half the people are below average, you know. A debugger is a requirement of losers such as myself to be able to accomplish project goals.
Claiming that you don’t need a debugger is somewhat like saying that using vi or emacs is more productive than VS.Net. Simply ridiculous.
I do not use a debugger that often, but when I need one, I *really* need one. Besides, a debugger is also a nice way to find out how some sparsely documented class libraries *really* work.
That said, I really like the mono project. It has already saved us some money since it allowed us to stay with our linux-based hosting provider instead of buying an expensive win2k3 dedicated server.
Also mono has been ported to SHARP ZAURUS. I’ve installed it in my zaurus SL-5500 model with ARM processor.
Photo of a zaurus SL-5500 model:
http://www.tuxmobil.org/Mobile-Guide.db/images/zaurus1.png
Download mono for zaurus:
http://www.killefiz.de/zaurus/showdetail.php?app=1082
Debugging with writing messages to the console is more prodective and efficient. You can search for quotes from several major programmers regarding printf (console) debugging vs debuggers.
printf and logging messages to a file are one of your tools in debugging, but the aforementioned statement is brain damaged and its obvious the poster is insecure in himself and considers it “manly” to use printf.
relax people…this is an open source proj… that feature will come up when someone decides to volunteer to do it
“Claiming that you don’t need a debugger is somewhat like saying that using vi or emacs is more productive than VS.Net. Simply ridiculous.”
Agree that you need a debugger..but _that_ is one of the worst examples in history. It comes close to marking the rest of your post as “ignorable”
.
> Also the _same_ code could run under windows too, right ?
…
> The Serial i/o is interesting for me so that I could write a
> cross platform GUI application that uses the rs232.
> Rigth ?
I’m already using Serial I/O with .NET on Linux with Portable.net . It doesn’t work on windows with Portable.net, but of course there’s MS.NET .
My stuff’s pretty ordinary (text AT commands and all that)… I don’t know what you need, so maybe it works , maybe it doesn’t.