DotGNU‘s Portable.Net is a portable suite of tools (including a C# compiler, assembler, and runtime engine). In this release, the C to IL bytecode compiler is now in good shape, and many bugs were fixed in System.Windows.Forms and System.Xml.
DotGNU‘s Portable.Net is a portable suite of tools (including a C# compiler, assembler, and runtime engine). In this release, the C to IL bytecode compiler is now in good shape, and many bugs were fixed in System.Windows.Forms and System.Xml.
Does it works with GTK# yet?
It seems to be somewhat hit or miss depending on what Mono is doing. It has worked in the past, but hasn’t recently. I think it has to do with Mono changing things around. The best thing to do is ask on #dotgnu on irc.freenode.net or the mailing list.
That c to IL compiler seems rather interesting, but from what I know you can only use static functions from the class library with it. I might be wrong there though. It’s good to see SWF coming along and also that System.XML is shaping up. XML has always been a bit of sticking point for DotGNU. For those that have had problems getting Mono running from CVS you might want to give DotGNU a shot. There’s no bootstrapping involved like Mono and none of the infamous “your corlib and runtime are out of sync”, so its just a matter of ./configure, make for treecc(the compiler generator), cscc(the compiler which is fast because it’s written in c), and the libraries.
Thanks, I guess I can install dotNET on FreeBSD. There have the ports of it, which Mono will never work correct.
it should work fine with wx.net (wxnet.sf.net)
How are DotGnu and mono related? Are they different projects with the same goal?
yes. the have the same goal but they work on it with different ways. DotGNU is more GNU (eg. you will never find docs like this in MONO’s site: http://www.dotgnu.org/danger.html )
Mono on the other side is said to be evolving faster and is almost ready (that IMHO has to do with money).
Also read this:
http://www.southern-storm.com.au/pnet_faq.html#q11_1
It’ll be hard to get an objective opinion on that matter and I try to be as objective as I can, but before you read this let me tell you that I am fully on DotGNU’s side and don’t take too much on mono itself, for the following reasons:
1. DotGNU is a GNU project without commcercial background, Mono is developed by a for-profit company (Ximian/Novell) with $$ in mind willing to license it a way it becomes proprietary (DotGNU would never do that with its components)
2. DotGNU Portable .NET has no bootstrapping problems because the compiler, the runtime etc. are written in C and the class libraries in C#. Mono has some low-level stuff written in C# and thus, current versions (as of 0.30.x) still have many problems.
3. Partly because of 2, portable .net is more portable, it is known to work on >10 CPUs and lots of operating systems without problems, while mono is very hard to get working(+bootstrapping) on non-x86 (just ask the people who tried compiling it from source, those using distribution binary packages of course won’t have that problem).
4. Most parts of DotGNU are GPL (+ linking exception) or LGPL licensed, the mono class libraries are X11 licensed (nothing wrong with that so far).
Miguel de Icaza said if Microsoft ever wants to have money for an API or whatever of .NET, Novell is going to pay for it so there *is* a risk that some parts of mono turn proprietary. DotGNU would *never* accept this while mono would do this “for the sake of users and customers” (as miguel de icaza said).
5. Many parts of Mono were started “as an excercise to learn C#” while Portable .NETs compiler was started by someone who has many years of experience (Rhys Weatherley) in compiler construction and was started using the better base (aspect oriented using a tool called ‘treecc’).
Of course, many parts of mono have been re-written over time so I don’t know if that still applies to many, or to what extend.
6. When Miguel de Icaza got interested in the .NET Platform he noticed Rhys had already written some code, much more and better than what he achieved so far (this has nothing to do with mono or DotGNU yet, we are talking about early 2001 now) and miguel got interested. But as soon as he convinced some ximian guys to work at it full time and they advanced pretty quick (don’t know the exact happening, since I am no Ximian internal he started calling rhys’ code useless and he was destroying the committment of Rhys, a guy who was working for free software full time, living on his savings.
That’s one of the main reproaches against Ximian I have here, really!
7. Mono is only developing the framework, DotGNU also is looking into webservices and things related to them (DGEE, goldwater…).
8. Ximian officially spreads myths about DotGNU, just read their FAQ..They list what’s unique about mono and talk about how great their developers are (just as DotGNU didn’t have great developers) or how “Portable .NET remains a pretty untested platform”. Also, Miguel and lupus indirectly sometimes attacked DotGNU within mailing lists etc. (just read the archives).
Some DotGNU folks did so, too, of course, but I think it’s quite dis-placed that Ximian officially does this as a company.
9. Mono more relies on more external dependencies (ICU for internationalization, winelib for System.Windows.Forms etc. [while Portable .NET only needs XFree for it to work])
10. In contrast to Mono (only a few days ago, Miguel said that C is dead to him) DotGNU is also working on a C compiler, which eventual goal is to compile many free software applications and libraries and bring them to the CLR.
Personally I think Ximian/mono partly destroyed some of the good things that could have gone into people’s minds about .NET/Linux and unfortunatly since Ximian/Novell is the bigger marketing machine a lot of people will follow them without looking at the alternatives (and as a result of that, mono evolves a bit faster than DotGNU Portable .NET does, at least at the moment).
Anyway, you perhaps should go and read mailing list archives or do some more research in case you are really interested and draw a picture on your own.
Also, I don’t want to attack any individuals here, it was needed to name them for completion.
As I said, I don’t want to encourage anyone to give up on mono or whatever, this is just *my* point of view from what I got out of the last 1,5 years “.NET and Unix” and if you don’t agree, feel free to disagree with me, I don’t claim that my 10 points are right, I just claim that it is my deepest feeling, experience and impression I got over time.
Still, i’ve gotten much more enjoyment from actually messing with DotGNU so far than MONO.
I hope they will both be successful!
There’s a slight error in gibbon’s point 9: XFree is needed for winforms only on non-windows systems. The rendering part is modular, and there’s already a backend which uses the native windows libraries for drawing, which is being developed at about the same pace as the X-based backend. There’s also a postscript backend for printing support (it still has a lot of TODOs so feel free to volunteer). Other backends (e.g. a gl-based backend) are also possible (just need the volunteers, as always .
Because the winforms controls are being implemented in C#, rather than simply wrapping native widgets (or wine widgets as mono is attempting), theming support was added early on. The control classes implement the behavior of the control and call on the current theme painter to draw them. The only theme painter at the moment has the win95 look, but other themes are planned (e.g. XP themes) once the more important stuff (like getting all the controls implemented) is finished. Feel free to volunteer if you don’t want to wait.