The first Mono beta release [LWN.net]

The first beta version of Novell/Ximian's Mono .NET implementation is available; see the release notes for the details. There's a lot of stuff there, including a C# compiler, the runtime virtual machine, support for several architectures, various database adaptors, a "complete cryptography stack," Apache integration, and more. Regardless of whether one agrees with Mono's goals, it looks like an impressive body of work.

to post comments

The first Mono beta release

Posted May 5, 2004 16:30 UTC (Wed) by mmarq (guest, #2332) [Link] (11 responses)

Okay... it could be a better Java than Java... but would everybody start to build their "vital important" and "super featured" applications to run on top of a VM or a JIT,... or would they continue to compile it to bare metal ??...

I belive the second is correct,... although C# and CLI could be a sucess for the fast growing web services... it will be XAML and NGSCB/Paladium that will force the users decisions towards M$ or not !...

Microsoft as always used technologie as a mean to domination and lock-in of users, and not for real innovation,... their company is bull??? propaganda driven (a.k.a Marketing), and .NOT (a.k.a .NET) was a "candy" trowned in with the mean to try to maintain the large community of ISV and freeware developers connected to M$, when Microsoft was ordered to split in two,... and a perfect companion for a major shake out, and or in any kind of technologie shift in the OS part... after all it runs on a Virtual Machine.

But i belive "Real Programmers" would prefer something like D, (http://www.digitalmars.com/d/sdwest/paper.html)
that not only has all the facilities and security features of C#/CLI, but also is sintaxe identical to C, and can be much more optimized for bare metal compiling...

In the mean time, IMO, what Open Source needs is a IDL, some way like XAML,... and for the first time in recent history Open Source has the possibility of "EMBRACE & EXTEND" Microsoft and define a XML based IDL that will build the interfaces of the next wave of "Distributed Applications" fullfiling the promesses of CORBA and OPEN DOC... and who knows!?,... perhaps XUL+ or XUL2 could be a superset of XUL and XAML, and also a CORBA IDL !!...

For the legacy Windows, a "shell replacement" environment ala Litestep or DesktopX, would be IMO a perfect strategy to bring a large protion of those Win9x/2k users, in and outside of entreprises to the Linux/FOSS side... and also the missing link to transform CoLinux into the next Win OS WRAP...

I always thought that "Ximian Destop 2" would evolve to be a Windows Shell replacement,... it could had!?, no!?... and in a contructive criticism, altought giving credit to the importance of something like MONO, i can only hope that Ximian would be more interested in stepping foward than following Microsoft.

The first Mono beta release

Posted May 5, 2004 17:16 UTC (Wed) by penguinista (guest, #308) [Link] (1 responses)

For the legacy Windows, a "shell replacement" environment ala Litestep or DesktopX, would be IMO a perfect strategy to bring a large protion of those Win9x/2k users, in and outside of entreprises to the Linux/FOSS side...
---
A strategy that I am taking is to get users happy with mozilla firefox and thunderbird. Most of what people need these days can be done with these apps along with open office.

Start with Mozilla and then ...

Posted May 5, 2004 18:54 UTC (Wed) by AnswerGuy (guest, #1256) [Link]

...add Evolution, or Thunderbird, and then offer OpenOffice.org.

If you successfully wean them off of Outlook, Office, and IE, then switching the OS out from under many of them will be MUCH easier. Of course you'd start by making it "reversible" (booting from Linux but mounting up their "home" directories off their MS Windows partitions or servers (smbmount).

This process won't happen overnight and we can't push it. Just take it in baby steps and back off (temporarily) whenever we can't show clear advantages to our users.

JimD

The first Mono beta release

Posted May 5, 2004 18:08 UTC (Wed) by rmini (subscriber, #4991) [Link] (6 responses)

In the mean time, IMO, what Open Source needs is a IDL, some way like XAML,... and for the first time in recent history Open Source has the possibility of "EMBRACE & EXTEND" Microsoft and define a XML based IDL that will build the interfaces of the next wave of "Distributed Applications" fullfiling the promesses of CORBA and OPEN DOC... and who knows!?,... perhaps XUL+ or XUL2 could be a superset of XUL and XAML, and also a CORBA IDL !!...

You mean something like WSDL and SOAP? Not the same exact thing, but you can get the same results. WSDL needs a little work (too complex, for starters), IMHO, but the idea and groundwork are there.

Where is the XAML threat?

Posted May 5, 2004 22:45 UTC (Wed) by simon_kitching (guest, #4874) [Link] (5 responses)

I see no resemblance between XAML and WSDL at all.

A fair bit of XAML is very similar in concept to Mozilla XUL - just an updting of the HTML standard to the 21st century. These tags should not be very difficult for Mozilla + potentially other browsers to support. I don't see much threat here (any differing opinions?).

The major threat seems to me to be the ease that .NET code can be referenced from xaml documents via the <bind> tag, resulting in the ability for users to access a website URL, and end up with a real application running locally on their PC (with the .net code downloaded over the network).

The closest thing I see to the above functionality is Java WebStart. You "run" a URL, and the application at that URL is automatically downloaded to your client machine and runs locally, with the ability to communicate with the server it was downloaded from as/when the application desires.

Java WebStart, of course, is cross-platform and cross-os. And because of Java's "sandbox" security model, it is safe(ish) to run applications from remote sources; their priveleges can be limited. I believe that XAML will use the ActiveX-style "signing" approach instead, where you either trust the whole app or don't, leading to large and well-known companies having a major edge over smaller companies. Or does .NET have a sandbox security model too?

If only java tool designers could just get their act together and make use of the new "long-term serialization" feature of java1.4 which now allows GUIs to be designed using drag-and-drop tools and saved into xml format (unlike the existing "code-generation" approach).

As I said previously, XAML's presentation tags should be no big deal to emulate. And Mono is going great on providing a .NET runtime. Of course there is still the issue that any downloaded assemblies will be using the Microsoft .NET libraries. While the .NET runtime and the c# language are official standards, I'm not sure to what extend the libraries are. Certainly Windows.Forms won't be. Will xaml documents commonly contain .net assemblies that reference the Windows.Forms assembly? I don't know...

Where is the XAML threat?

Posted May 6, 2004 9:36 UTC (Thu) by hansl (subscriber, #5086) [Link] (1 responses)

> Certainly Windows.Forms won't be. Will xaml documents
> commonly contain .net assemblies that reference the
> Windows.Forms assembly? I don't know...

Mono has an implementation of Windows.Forms based on Wine.

Where is the XAML threat?

Posted May 6, 2004 9:48 UTC (Thu) by simon_kitching (guest, #4874) [Link]

Yes, but I have my doubts this will ever be reliable.

There are vast chunks of Windows functionality missing from Wine (mostly because Microsoft don't provide documentation for many public APIs). And chunks of Wine missing from the current Windows.Forms wrappers. And Microsoft can exploit this if they feel like, just as they did against DRDOS. If the world moves to XAML, and *if* XAML apps commonly depend on Windows.Forms, then I think non-windows platforms are all in deep trouble. But I don't know if this is indeed the case - can anyone say what dependencies xaml apps encountered on the web in a few years time will have?

Where is the XAML threat?

Posted May 6, 2004 15:35 UTC (Thu) by tjc (guest, #137) [Link] (1 responses)

No, but if you know of a page explaining the similarities and differences between XUL and XAML, that would be appreciated. I've read various postings here and there comparing the two, but it would be nice to read a more comprehensive explanation. I'm sure there's one out there, I just haven't found it yet...

XUL will fail because it's too hard to get into

Posted May 7, 2004 9:29 UTC (Fri) by rwmj (subscriber, #5474) [Link]

I've posted before on the subject of XUL's shortcomings:

http://lwn.net/Articles/80684/

Since that posting, I've bought and read the RAD for Mozilla book, upgraded to Mozilla 1.7RC1, and gone back to the application I was designing. I actually stand by most of the things I said originally. I have found that XUL is:

Still mostly undocumented, beyond the very simple stuff.
Unstable - I can crash Mozilla easily with simple XUL programs.
Hard to set up right: it took me a full 3 days to install a XUL demo program under a 'chrome:' URL, because the 'installed-chrome.txt' and 'chrome.rdf' files are incredibly sensitive to minor changes, and there is absolutely no feedback when things go wrong. The chrome: URL still sometimes 'just breaks' on its own, and I end up endlessly restarting Mozilla and deleting 'chrome.rdf' files until it works again.
Non-trivial XUL programs depend on the internals of Mozilla - try using the <editor> tag for instance.

I like to think I'm a pretty experienced programmer - I've been doing programming in many different languages for 20 years or more - and if I can't write a simple XUL-based editor after two weeks, then there is something deeply wrong with the environment.

Rich.

Where is the XAML threat?

Posted May 6, 2004 16:07 UTC (Thu) by allesfresser (subscriber, #216) [Link]

These tags should not be very difficult for Mozilla + potentially other browsers to support. I don't see much threat here (any differing opinions?).

If the XAML schema is patented, it would be impossible for anybody else to implement it. And it seems a no-brainer that this is exactly Microsoft's plan, for the express purpose of once again using their monopoly power to strongarm their way into forcing people to use their 'standard'.

The first Mono beta release

Posted May 6, 2004 16:48 UTC (Thu) by mmarsh (subscriber, #17029) [Link]

This could actually be a big help to some academics. For example, some of the projects out of MS Research are written in C#, since it has features that keep you from shooting yourself in the foot. Sure, the researchers could have used Java, but if they're going to use one or the other, they might as well use the one produced by their employer. This puts those of us who don't use Windows in a bind, though. If we want to play around with a system like Pastry, we either have to use C# or port the code to another language.

The first Mono beta release

Posted May 19, 2004 0:39 UTC (Wed) by peter_pilgrim (guest, #5448) [Link]

Kewl! I am a professional J2EE contract developer. I now have a training
kit to learn C# without a large investment in my own firm's cash,
or running Windows XP / Visual C# etc etc

I shall endeavour to get into Mono/C# this summer holiday!!!

Wow. Linux is fun