The One True Object (Part 2)

re: The One True Object (Part 2)

Justin — Sat, 05 May 2007 02:05:52 GMT

I'm enjoying this series so far. I don't have much to add to the discussion, but I like the direction you are going. I'm mostly intersted in IronRuby, so of course I wonder how this architecture will deal with Ruby's pecularities. Especially the concept of "metaclasses" (aka "eigenclass") and modules.

re: The One True Object (Part 2)

Jonah — Sat, 05 May 2007 04:25:20 GMT

Thanks for these posts, Jim. A nit: The S in CLS stands for specification, not subset.

re: The One True Object (Part 2)

G.H. Xiao — Sat, 05 May 2007 05:17:01 GMT

DLR is so cool. I'm expecting for the next release of IronPython.

JVM's invokedynamic

murphee (Werner Schuster) — Sat, 05 May 2007 05:19:06 GMT

Sun is considering adding a new bytecode "invokedynamic" to the JVM and hopes that it'll somehow speed up method dispatch in dynamic languages.

What's your take on this? Could some VM support really give a big performance boost, considering that the actual lookup still has to happen in software (ie. the language implementation)? Maybe with some VM supported (polymorphic) inline cache or other methods from Strongtalk/Self (type feedback, ...)?

re: The One True Object (Part 2)

Casey Marshall — Sat, 05 May 2007 05:31:54 GMT

What is the overhead involved in the message passing? How does it compare with a simple function call? I'm dying to see the implementation!

re: The One True Object (Part 2)

Haacked — Sat, 05 May 2007 09:59:38 GMT

Great series Jim! I very much enjoyed meeting you and Jon Lam at Mix.

p.s.

You're the 300th blog I'm subscribed to. I should send you a prize or something. ;)

Dynamic Language Runtime 微软打出的王牌

自由、创新、研究、探索…… — Sun, 06 May 2007 04:21:28 GMT

Dynamic Language Runtime(DLR)。DLR和IronPython全部开源，如果你微软这样的动作吃惊,请看看Microsoft 的 OpenSource Licence,可以到codeplex下载。新的动态语言运行时（Dynamic Language Runtime，DLR）向CLR中加入了一小部分核心特性，使之得到显著改善。它向平台中加入了一系列明确为动态语言需求所设计的服务，包括同享的动态类型系统、标准托管模型（Standard Hosting Model），以及轻松生成快速动态代码的支持

support for dynamic languages on .NET

Walter Stiers - Academic Relations Team (BeLux) — Sun, 06 May 2007 20:36:54 GMT

Jim Hugunin's Thinking Dynamic has a series of blog entries on a new level of support for dynamic languages

New and Notable 165

Community Blogs — Mon, 07 May 2007 17:36:56 GMT

It's time for a New and Notable Update when my flagged posts in FeedDemon gets past a screenfull

re: The One True Object (Part 2)

Alex Blewitt — Tue, 08 May 2007 13:34:52 GMT

I'm very interested in the work that the DLR is kicking up, and where it might lead.

http://alblue.blogspot.com/2007/05/jvm-or-clr.html

The one question I have is given Microsoft's past record of platform compatibility, how are they going to make the DLR available to the world? I guess the Silverlight technology is going to be cross platform to start with, but is there going to be any kind of guarantee that it will continue? That, more than anything else, is what will keep people away from the technology (though for Microsoft shops, it's a godsend).

re: The One True Object (Part 2)

Attila Szegedi — Thu, 10 May 2007 22:34:43 GMT

I've been thinking about providing a unified runtime support for dynamic languages on the JVM for quite some time - I work on two such projects, the Rhino JavaScript interpreter and on the FreeMarker template engine, and keep in touch with people working on JRuby and Jython as well.

My opinion is that instead of providing interfaces on objects (and thus internalizing the knowledge required for cross-language interop into object implementations), it is a more flexible approach to define interfaces for metaobject protocols, have each language ship its implementation of it (I call these metaobject protocol implementations "object navigators" or just "navigators"), and allow languages to use other languages' metaobject protocol implementations to access and manipulate those languages' native objects, in addition to using its own MOP implementation.

The article is at http://www.szegedi.org/articles/wrappersOrNavigators.html if you're interested.

re: The One True Object (Part 2)

Alex Ward — Tue, 15 May 2007 17:58:49 GMT

Excellent work Jim. Writing .NET 2.0 desktop apps in the one true language has been a very satisfying experience. Clearly for WPF the Python/Xaml interface has been a bit of a road block so I'm delighted you good folks are putting this effort in.

Could the Xaml parser from Silverlight be used on the desktop? If not, do you think that this would be a WPF 3.5 thing or an IronPython 2.0 thing?

IronLisp, MOP and more

Andrew Stopford's Weblog — Mon, 03 Sep 2007 16:09:57 GMT

Scott's last post had a link to a new opensource Lisp project for the DLR, called IronLisp (in homeage

New and Notable 165

Sam Gentile — Wed, 12 Dec 2007 03:08:20 GMT

It's time for a New and Notable Update when my flagged posts in FeedDemon gets past a screenfull