[pacman-dev] Maelstrom(Pacman in C#) project brief

Bozhidar Batsov lordbad at e-card.bg
Mon Jul 23 04:11:50 EDT 2007


"As i know, C# has not even open standards. " - that is not true, the 
specification of the language is open and freely available and it was 
used to build the original mono C# compiler. But you're right - mono is 
not feature complete yet, but is very close to being 100% compatible 
with .NET 2.0. The project's web site claims this will happen around the 
end of October. My biggest difficulty during the development cycle was 
the lack of documentation for the project. Monodoc would claim that 
something is not yet implemented, but when I look through
mono's class hierarchy I'd see what I was looking for...
My greatest concern is actually your remark about the popularity of the 
language in the linux community. Personally I have never in my life used 
C#/.NET with windows(for the simple reason that I don't have windows 
anywhere). But I felt in love with mono from the start. I had done 
development in Pascal, C, C++. Perl, Java, Python, Lisp, Prolog and some 
others. I didn't think that I would enjoy C#, but I was wrong. Its a 
great language, and .NET Framework is very capable environment so I 
believe that it can empower a package management system even better than 
pacman + alpm. But time will tell...

Georg Grabler wrote:
> I can give you another point why NOT using alpm at the current state:
> - API Interfaces still change a lot (all 2-3 weeks)
> - The interface still is "ugly"
> - A lot of functionality is missing for a proper GUI client to be written
>
>
> You have done a great work on the C# implementation. I actually 
> dislike the thought of having mono on the base, not for the reason 
> that c# is anything good or bad, but for the reason that c# / mono 
> isn't complete. I wouldn't even consider it as feature complete, 
> they're always behind the current C# standards, often for a year.
>
> I lately tried to port an application to GTK# (a quite complex 
> client), but it already failed on backend system functionality. The 
> program is completely .NET 2.0, and SHOULD work out with mono. I found 
> some articles / post on their groups that this is a bug of mono - i 
> don't think a framework should come in a stable version with broken 
> functionality, neither it should need more than half a year (as it's 
> for now already) to get fixed.
>
> I can see C# making the deal into one of the main linux application 
> development languages, but the time hasn't come yet. All of the 
> applications you mentioned were either broken when i tried them 
> (muine), had a bad or even worse performance (beagle)
>
> Currently, i also don't see any advantage of C# compared to python. In 
> linux i see disadvantages, since there isn't too much software around, 
> and not too many libs you can use. Also, we would set on a language 
> which is not very popular in the linux community. As i know, C# has 
> not even open standards.
>
> Yours,
> STi
> ------------------------------------------------------------------------
>
> _______________________________________________
> pacman-dev mailing list
> pacman-dev at archlinux.org
> http://archlinux.org/mailman/listinfo/pacman-dev
>   





More information about the pacman-dev mailing list