Comments on: Where is “Oslo” Going, IV

By: William Vambenepe’s blog » Blog Archive » With M (Oslo), is Microsoft on the path to reinventing RDF?

Mon, 15 Jun 2009 22:44:04 +0000

[...] JSON replacement! all of the above!). Douglas Purdy makes a valiant 4-part effort (1, 2, 3, 4) but it’s still not crisp enough for my small brain. Even David Chapell, explainer [...]

By: douglasp

douglasp — Sat, 09 May 2009 21:29:25 +0000

If that happen to occur, it would be stored in the catalog (and in the MX file).

I say happen to occur, as one of our goals is to have a “natural SQL database” for everything in our meta-model.

I am loathed to say the meta prefix however.

By: Joe

Joe — Fri, 08 May 2009 02:14:16 +0000

Reading this all with interest. One thing that bothers me still is the repository technology. What happens when SQL Server doesn’t support a construct for a schema defined in M? Is the table schema just there as a ‘cc’, and the master is stored in the .MX file?

By: What Does Oslo Mean By Model-Driven? | Coded Style

What Does Oslo Mean By Model-Driven? | Coded Style — Fri, 17 Apr 2009 19:18:20 +0000

[...] Does Oslo Mean By Model-Driven? There’s a good discussion between Jean-Jacques Dubray, Doug Purdy, and Charles Young about if/how Oslo relates to MDA, MOP, UML, SOA, metamodels and a number of [...]

By: Tune Up Your PC » Post Topic » What Does Oslo Mean By Model-Driven?

Tune Up Your PC » Post Topic » What Does Oslo Mean By Model-Driven? — Fri, 17 Apr 2009 18:47:32 +0000

[...] a good discussion between Jean-Jacques Dubray, Doug Purdy, and Charles Young about if/how Oslo relates to MDA, MOP, UML, SOA, metamodels and a number of [...]

By: Ben Gillis

Ben Gillis — Wed, 15 Apr 2009 06:09:14 +0000

As for all the M0..N talk, just a practical example that will save me writing another autodoc app:
http://bengillis.wordpress.com/2009/04/14/where-is-microsoft-oslo-going/

By: Charles Young

Charles Young — Tue, 14 Apr 2009 01:31:01 +0000

Well, I don’t know. Maybe it may bring in a whole new audience

By: Dody Gunawinata

Dody Gunawinata — Mon, 13 Apr 2009 08:54:07 +0000

M4M is a common abbreviation for “Men 4 Men”. You are going to lose this space on search engines

By: Jean-Jacques Dubray

Jean-Jacques Dubray — Sun, 12 Apr 2009 21:00:11 +0000

Incidentally, you may want to revert to naming this level “M3″, “M4M” comes with lots of unrelated materials as you are trying to search for it.

By: douglasp

douglasp — Sun, 12 Apr 2009 06:11:21 +0000

We spend a bunch of time with the management team to ensure that what we are building will work for them.

Keith can comment on that as he is involved in a lot of that work as well.

By: Jean-Jacques Dubray

Jean-Jacques Dubray — Sun, 12 Apr 2009 06:08:06 +0000

Doug:

thanks for your answer. Yes I have heard of Keith Short and I’ll wait for him to bring some clarity to this question.

I know that Oslo has included a “service implementation” as part of its scenarios and I am very excited about that.

I just want to point out that I have no agenda with respect to the OMG, MOF or MDA. MOF for instance is incomplete and as a result lots of people are using UML as an M3 layer, which it was never designed for.

On a similar note, Microsoft has a similar problem in the “Operations and Management” space. It recently came out with SML (Service ML). The standard before SML was way too complex for an M3 layer and became unusable. That’s the only reason for me pointing out the importance of creating an explicit metametamodel.

By: douglasp

douglasp — Sun, 12 Apr 2009 04:14:48 +0000

I couldn’t agree more.

This is why we are doing things like MUrl, MService too.

We want this foundational technology to apply to architect, developers, and ITPro pros alike.

By: Shawn Wildermuth

Shawn Wildermuth — Sun, 12 Apr 2009 03:23:51 +0000

I had taken the wrong assumption that “Where Oslo is Going” is the only place that Oslo is going. Where I think that textual DSLs can end up in a model driven architecture, that’s not the only use. I am trying to drive the point that DSL’s aren’t just for architects and the like…but typical developers can really be helped by them (and by extension the ease at which you can write Grammars in Oslo makes them viable, instead of dealing with Lex/Yacc which seem to be more driven to GPL’s than DSLs). Thanks for the vote of confidence

By: douglasp

douglasp — Sun, 12 Apr 2009 02:23:32 +0000

I read it. Sells sent it around before it was published. I love it and the follow on piece at http://msdn.microsoft.com/en-us/library/dd548006.aspx.

I am not too sure why you would think I would disagree with your assertions. I think that what you outline is a great way to use the “Oslo” stack.

This thread (”Where is Oslo going?”) has largely been about using Oslo for modeling connected systems and what sort of support that we have in the stack to do that.

Modeling connected systems is just one domain that Microsoft will have tDSL/vDSLs (I am stealing JJD syntax, which I like), models, and runtimes for.

By: Shawn Wildermuth

Shawn Wildermuth — Sat, 11 Apr 2009 21:51:42 +0000

So do you fundamentally disagree with my assertions about Oslo and DSLs or are you just talking about a higher level of abstraction (i.e. that DSLs will drive the models)?

See my first article (in case you haven’t seen it yet) as my assertion:

http://msdn.microsoft.com/en-us/library/dd441702.aspx