Update: Snippets from Bill Hilf interview @ http://www.networkworld.com/news/2007/051707-hilf-microsoft-wont-sue-over.html

What we heard back after the Novell deal was “Give us more transparency. You say that there is IP involved, give us an understanding of what that is.”
…
… we have no plans to litigate. You can never say we’ll never do anything in the future, but that’s not our strategy. That article spins it on the attack. The only new piece information in that article is that it just put a number on the patents.
…
The people in the open-source community that I know well . . . they contacted me right away. All of the European guys I know called me at 2 a.m. I told them what I told you. They said “Okay, that’s what I needed to hear.”
…
I personally believe that there’s a tremendous amount of work that needs to be done in software patent reform. However, the current rules still apply. It’s still the way we do business today and how all other businesses work. So we still have to find ways to work in the current system even though we do want it to be improved in the future.
…
There’s no other strategy. There’s no other hidden agenda. I’m trying to be as clear as I can to people that this isn’t a threat. We’re not going out and attacking people. We’re trying to solve an IP issue.

Update: I couldn’t help but smile when I saw the pic that John Lam attached to his post congratulating the Mono team on their spectacular achievement,

Which reminds me. If it wasn’t for penguin “pics” such as this,

… reminding us all that there are those who see FLOSS as something more than the freedom to adapt, change, tinker, and create, I wonder if the attitude on MSFT campus towards apparent FLOSS patent infringement might suddenly change?

Folks: There are *MANY* reasons people go to war, but generally speaking it can be boiled down to two,

1) Freedom.

2) Power (AKA Control, Money, etc.)

The FLOSS “War” isn’t an exception to the rule.

I expect to be quite wealthy once the dust from the Linux IPOs has settled

Eric Raymond, Doing It For The Cause, December 12th, 1999

People who invested in the Linux IPO’s of the late 90’s/early 00’s didn’t do so because of a belief in the FLOSS cause. They did so because they believed their was profit to be made.

Please don’t lose sight of that.

Update: It’s now official. Thanks to Zoltan Varga’s recent check-in to SVN, Mono now successfully runs IronPython 2.0A1/DLR,


[mdavid@domU-12-31-37-00-03-10 Debug]$ mono -V
Mono JIT compiler version 1.2.4 (/trunk/ r77478)
Copyright (C) 2002-2007 Novell, Inc and Contributors. www.mono-project.com
        TLS:           __thread
        GC:            Included Boehm (with typed GC)
        SIGSEGV:       normal
        Architecture:  x86
        Disabled:      none
[mdavid@domU-12-31-37-00-03-10 Debug]$ mono ipy.exe
IronPython console: IronPython 2.0A1 (2.0.10427.02) on .NET 2.0.50727.42
Copyright (c) Microsoft Corporation. All rights reserved.
>>>

And to make things even better…

7:54:34 AM sanxiyn: It seems that there’s no more serious Mono runtime bug hiding.

7:55:12 AM sanxiyn: I’ve just run HTTP sanity test (which uses urllib and BeautifulSoup) on IP2/Mono.

7:55:25 AM xmlhacker: and?

7:55:34 AM sanxiyn: Passed.

7:55:39 AM xmlhacker: nice!

7:56:03 AM sanxiyn: DNS test passed. (This uses dnspython, which implements all of DNS protocol marshal/unmarshal in pure Python.)

7:56:37 AM sanxiyn: XML-RPC passed. Well, it looks good.

7:57:03 AM xmlhacker: XML-RPC passed? That is *very* encouraging!

So there you have it folks. 16 days for the Mono-Project hackers to implement support for the DLR.

That *ROCKS*!!! :D

Thanks to Seo, Zoltan, and *ALL* of the rest of the folks who pulled this together! *VERY* nice work!

Update: There’s a vicious rumor floating around the GoogleTalk ether that the necessary fix to the problem showcased below was recently checked into SVN by none other than Zoltan Varga. The problem? Let’s see if you can spot it,

Firstly,


Author: zoltan
Date: 2007-05-16 08:54:35 -0400 (Wed, 16 May 2007)
New Revision: 77476

Modified:
   trunk/mcs/class/corlib/System.Reflection.Emit/ChangeLog
   trunk/mcs/class/corlib/System.Reflection.Emit/FieldBuilder.cs

Okay, now…


mdavid@domU-12-31-37-00-03-10 mono-source]$ svn up mono mcs gtk-sharp
At revision 77475.
At revision 77475.
At revision 77475.

Hmmm…. Drat! Well, anonymous SVN updates every half hour or so. < So… We wait. ;-)

Update: (07:25 MDT) This just in…

mdp:comment: an impatient Seo…

Sanghyeon Seo: (07:10) *go hand edit the file and rebuild.*

Impatient Seo about 13 mins later…

Sanghyeon Seo: (07:23) Patch works as advertised.

Vicious rumor apparently has teeth! SWEET!!!

[Original Post]
So I just got a ping from Seo (Sanghyeon, you know, of IronPython Community Edition fame ;-)) with all of the juicy details…

The Dynamic Language Runtime now runs on Mono!

Want proof? I thought you’d never ask ;)


[mdavid@domU-12-31-37-00-03-10 Debug]$ mono ipy.exe
Traceback (most recent call last):
SystemError: The invoked member is not supported in a dynamic module.
IronPython console: IronPython 2.0A1 (2.0.10427.02) on .NET 2.0.50727.42
Copyright (c) Microsoft Corporation. All rights reserved.
>>> 2 + 2
4
>>>

The above comes courtesy of the patch and makefile attached to Bug# 81638 of the Mono Bugzilla site. Full instructions to reproduce the above results are located on this same entry.

Okay, so there are a few caveats at the moment.

Firstly,

Sanghyeon Seo: SystemError is from Mono’s reflection class library.
Which is simply unimplemented.

Also,

Sanghyeon Seo: Any class instantiation hits unimplemented code in Mono mscorlib.dll.
So you can’t get very far. Function calls work though.

And finally, after realizing it was time to broadcast this info to all of those listening in on Channel O’ReillyNet,

M: is there anything specific you think I should point out?

Sanghyeon Seo: The fact that this thing is not yet fully working.

So, with the above firmly in place, lets disspell some myths,


Myth: Who cares if the DLR is open source. It’s not like it can be run on anything other than the proprietary MSFT .NET runtime, so it’s not really “open”!

The Truth? Yes it can, so yet it is.

Myth: The DLR has some special hook into underlying CLR or something.

The Truth? No it doesn’t ;-)

Now you try…

Myth: (Please fill in your suggested myth)
The Fact? (Please fill in your suggested fact)

(now *this* aught to be interesting ;-))

Now before all you haters start filling in the blanks with all your FUD, let me please be the first to congratulate Seo and *ALL* of the Mono hackers out there that have gotten the DLR capabilities this far in 15 16 (mdp:comment I did the math incorrectly the first time) days since the DLR source was first made available. < That, my friends, is a remarkable feat!

Nice work Mono-Project hackers!

Okay, haters… Have at it!