the dead parrot

It seems that Microsoft has dropped Oslo and it's focusing on OData and EDM as a more loosely coupled approach to modeling. I was really interested in "M" and creating DSL's in the language, was impressed with Don Box and Doug Perdy at PDC2008 but it seems that we'll have to wait for another fad to occupy our dsl-thirsty minds. Weirdest of all, it seems that Doug also vanished with the Oslo.

Nevertheless, after a few posts on SQL Server Modeling, Oslo, and "M" and a long pause posting, I'm here to write about the next technology I'm trying to learn and apply.
This time it's Python with Google App Engine.This should be a big change after years of using C# and Java in my everyday work. This is somehow weird, because actually I'll still continue to use MS technologies at work, but examining the Google side of development world is like a new hobby.

Right now I'm going through the tutorials and docs on Google code and I really like the language and the approach. Provided services give a lot of power instantly to your applications. You don't have to worry about authentication - authenticate users with Google account, don't worry about the database details and query optimization - just use the easy-to-use Datastore service. Emails, cron jobs, XMPP, Google Apps, all together gives a great platform for development and enable you to focus on your.... Idea!

If you don't have an idea for the next I believe you should try developing an application that you need and you would like to see available. Taken this approach what I miss really is an application to easily organize and track my tasks. I get and assign tasks through several tools, Outlook, issue tracking tools at work, Google tasks, and of course RTM. At the end of day, each application is missing some feature that I think I really need. Thus, I'll try - as an exercise - to start with a task tracking application. Not a great idea, but at least - it could be a real enough project to learn Python and development for Google App Engine.

Journey ended, a new one begins

Few days ago I wrote in my previous post that my impression from the PDC2008, especially while talking with development team, that Microsoft is not quite sure what to do with "Oslo".
In yesterday's post on his blog, Douglas Purdy writes that "Oslo" will be shipped as "SQL Server Modeling", as part of the releases of SQL Server.
Here's a quote:

"The components of the SQL Server Modeling CTP are:

• “M” is a highly productive, developer friendly, textual language for defining schemas, queries, values, functions and DSLs for SQL Server databases
• “Quadrant” is a customizable tool for interacting with large datasets stored in SQL Server databases
• “Repository” is a SQL Server role for the the secure sharing of models between applications and systems

We will announce the official names for these components as we land them, but the key thing is that all of these components are now part of SQL Server and will ship with a future release of that product."

So, it seems that journey to "Oslo" as a "homeless", "general purpose" modeling platform has ended and it has found its "foster home" in MS SQL Server product. Nevertheless, I still like it.

Journey to "Oslo"

First time I heard of Microsoft's "Oslo" was last year while making final arrangements with my colleagues for attending the PDC 2008. My two colleagues and I were strongly recommended the "Oslo": Building textual DSLs session by our company's Chief Software Architect. Before that, I've learned a little bit about DSLs by reading some of the chapters of Martin Fowler's DSL Book WIP (work in progress). I liked the idea of DSLs very much and was eager to see and attend all of the PDC 2008 "Oslo" sessions.

I attended three of the "Oslo" sessions then and also talked with development team at "Ask-the-experts" night (Chris Andersen and Doug Purdy were present, too). My strong impression was that it is definitely a powerful concept and it is worth learning, but I didn't not get a clear idea of how to put all this into a working business application. I also asked some of the "Oslo" team members during "Ask-the-experts" nights about intended use and whether Microsoft will build some DSLs for common domains like finance, insurance, etc., but answer I got was that "developer's community" is expected to build DSLs. All in all, I left L.A. a bit confused about "M" and "Oslo", but still with strong interest in the technology.

A year had passed since and I didn't follow "Oslo" related news really, until yesterday I read this article by Kraig Brockschmidt. After reading the article I've finally decided to try and learn "Oslo", build a simple DSL, build a runtime for it and execute an application.

The aspects I'm really interested with "Oslo" are:

• Ease of model use in .NET applications
• Ease of reuse of models/DSLs
• Flexibility, change management, model extensions, etc.
• How will this increase developer productivity?