[Oberon] Zonnon

[Dan Parnete]

Alan Freed ha scritto:

>Doug,
>
>Zonnon started as an experiment, i.e., Oberon for .NET.
>
>Zonnon has been funded by Microsoft.  Their intention was, 
>and still is, to have a compiler for Visual Studio that is in the 
>Pascal family of languages, and that makes maximum use of 
>the Common Language Runtime (CLR), which is Microsoft's 
>commercial implementation of the Common Language
>Infrastructure (CLI) specification. The CLI specification is an 
>international standard for creating development and execution 
>environments in which languages & libraries work together 
>seamlessly. At least that is the design philosophy behind .NET.  
>
>So, the advantage of Zonnon is that you are able to import 
>modules written in any other languages, where there exists 
>compilers that compile down to the CLR.  There are about 
>a dozen or so such languages right now, and will likely be 
>many more in the near future - Zonnon amongst them.  
>
>Why is this important for me?  Because I can work on projects 
>with people who prefer to program in other languages, and I 
>in a language of the Pascal family.  Also, the .NET framework 
>has a huge standardize library that I can make use of, thereby 
>minimizing the code I have to write.  Novell's MONO platform 
>is an open-source implementation of the same international 
>standard that the .NET framework is based upon.  MONO is 
>currently targeted for UNIX, Linux, Mac and Windows platforms, 
>which means that programs written for MONO, which should 
>also run in .NET, 
>
This is partially true. The most core modules are already compatible, 
but they fall on GUI. Mono developers have spent a lot on GTK# 
interface, witch will never be adopted on .Net, and less on 
Windows.Forms. The most today Mono applications are not running on .Net, 
and Zonnon is not running on Mono. Of course, you can use Mono on 
Windows, including GTK#, but today it is 5 times slower then .Net, and 
twice on it's native Linux. And I don't consider .Net a speedy.

>will be able to not only make use of code written 
>in numerous different languages, but it will also be able to 
>run on the major platforms of today.  This is why I'm spending 
>the time to learn Zonnon, and to assist Eugene Zueff and others 
>at ETH in the debugging of their compiler.  There is also the 
>fact that I really like the logic behind Zonnon - I like how it 
>makes me think when I program.
>
>  
>
I like Zonnon ideas too. Unfortunately ETH didn't push it too much. When 
Patrik Reali and Peter Muller have gone, they left an almost complete 
and stable Aos. Probably more stable then today. Thomas Frey was very 
productive and closed the cycle. Eugene Zueff will leave soon, but 
Zennon is not ready for the jungle.

Dan Parnete