MEFz - The Little Blue Pill For That Certain Part Of Your Application...

Glenn Block threw another MEF post up on his Technobabble blog. This one corrals a couple of links to examples beyond those up on the MEF CodePlex site. He also has some links to new posts in the ongoing [community] attempts to differentiate MEF from the usual IoC suspects which I believe is beginning to get some traction. I'm personally very excited over all the potential MEF's metadata interrogation unlocks. This should be a fun ride between now and when .NET 4.0 hits the street.

No Unit Tests For Rocket Scientists...

Bummer. The rocket scientists at SpaceX beefed up the the power and runtime of the their first stage booster before Flight Test 3, but then forgot to add a comensurate delay to the timing of the Stage 1 seperation. The result? The first stage separated clean, but then still had enough juice to come back at the rocket and bump it wildly off course. I would have thought they'd be modeling their systems and simulating launches, but it looks like maybe not so. A remarkable video of the unfortunate flight can be seen here. Flight Test 4 will be at the end of the month and I bet they make orbit.

Programming For The Rest Of Us...

An interesting discussion has twittered by and flowed on to Oren's blog with his post titled: 'Coddling is consider harmful'.

In reality it's a difficult problem and one that highlights the current state of software development. I liken that 'State-of-the-Art' to us having emerged from the Dark Ages into a Renaissance where a relatively small number of great minds feverishly trade ideas amongst themselves, each putting out a great product. But that is still a long way from an 'Industrial Age' where knowledge then becomes something easily systemitized, packaged, and delivered in forms useful for ordinary folks to accomplish heretofore remarkable tasks.

In other words, we're a long way from the day bright sixth graders or entry level devs are writing solid enterprise apps...

Roman Kiss on CodeProject - Contract Model for Manageable Services

The redoubtable Roman Kiss has dropped another big one on CodeProject - 'Contract Model for Manageable Services'. Pretty much a pre-Oslo, manageable services tour de force (akin to Gregory Leake's .NET StockTrader Configuration Service 2.01 reference app). This is just the latest in a spate of great articles he's crafted for CodeProject. Here's some recent gems:

VirtualService for ESB
NullTransport for WCF
Fire WorkflowEvents from AJAX

Great stuff, and you should head over to CodeProject for a complete listing of Roman's contributions. I personally can't wait to see what he does with .NET 4.0 and Oslo (and things like MEF, if he's so inclined).

Oslo, Red Dog, and Virtual Computational Substrates - Oh My!

Of late I've been contemplating the interplay between Oslo and lower level aspects of Microsoft's cloud initiative. This recent Microsoft job listing provides some insight into those basement levels of Microsoft's 'cloud stack':

"The Cloud Infrastructure Services (CIS) team is responsible for creating the Microsoft Utility Computing Platform, also known by its early codename Red Dog (RD). This platform is one of the lowest levels of the services software/hardware stack and includes an efficient, virtualized computational substrate, a fully automated service management system and a comprehensive set of highly scalable storage services. The platform will enable services to scale to millions of machines distributed globally throughout Microsoft data centers. Further, it will provide the lowest operating costs per-node, and will lead the marketplace as the best platform for rapid development, deployment, and maintenance of internet services and applications. CIS is a young and hungry team that is on the path to delivering a V1 product to external customers in the coming year."

What I'm curious about is what role Oslo's modeling and repository plays within this 'v1 product' and in the command and control of 'virtual data centers'. I'm interested because that is the implication of what is represented in the following image - just how far down the 'cloud stack' will Oslo's modeling language and repository reach?

My own guess is - really deep - as in the possibility we'll see some new infrastructure variants of XAML related to the configuration, deployment, provisioning, and administration of not only the 'Red Dog' cloud application layer, but also the lower 'virtualized computational substrate' layer. One can also guess PowerShell will play a leading role managing this 'cloud stack' as well, so I'm equally curious about how PowerShell will be woven into the mix with Oslo's modeling and repository.

All in all, it will be interesting to see exactly how things unfold as it looks like Microsoft is counting on a high level of [virtual] data center automaiton being key to the success, reliability, and differentiation of their cloud initiative.

.NET StockTrader Configuration Service source to be released...

Gregory Leake of the Connected Systems Division, and author of the .NET StockTrader sample app, has announced that the previously unreleased [full] source for the accompanying Configuration Service will indeed be posted up on MSDN in the coming weeks. Should be interesting to get a look at this code.

But with PDC2008 around the corner it does beg the question: what will configuration look like in a .NET 4.0 / Oslo world? Inquiring minds really do want to know! 

Welcome to the shelter for wayward musings...!

With occasional diversions, this blog will be devoted to software development with an emphasis on .NET technologies. I have to date successfully resisted joining the wording fray, content to simply lurk and devote any time available for blogging instead to rock climbing and other futile pursuits. Why the change? I'm not entirely sure, and I also can't say that this blog won't offer more questions than answers over time, but then I suppose that's the nature of the journey we're all on...