July 2007 - Posts
When debugging LINQ to SQL code, did you wish you could easily see the SQL that gets executed? Of course there is the DataContext.Log property that is available for that. You can assign any TextWriter to this property. Console.Out is a good candidate for example, but it doesn't help much for web applications and doesn't integrate with Visual Studio nicely. Fortunately, Kris Vandermotten has a nice solution: DebuggerWriter. It's an implementation of TextWriter that writes to the debugger log.
With this solution, plug an instance of DebuggerWriter into your DataContext's Log property, and you'll be able to see the SQL queries appear in Visual Studio's Output window.
By the way, Kris also has other interesting posts about LINQ. For example, he has a series of posts about the impact the use of LINQ to SQL has over a classical Data-Access Layer (part 1, part 2).
Cross-posted from http://linqinaction.net
You may have seen a lot of messages about the launch date for Visual
Studio 2008, SQL Server 2008 and Windows Server 2008. According to this announcement, Microsoft has planned a grouped launch event in Los Angeles for February 27, 2008.
This does not mean that all the products will be released at that same date. Some may be released after the event, others may be released before.
It
seems that the RTM of Visual Studio 2008 is still planned for the end
of the year. Doug Seven, Senior Product Manager for Visual Studio Team
System, writes on his blog: "While the launch events are scheduled to kick off on February 27, 2008, Visual Studio 2008 will be released before the end of the year."
Scott Guthrie confirms this on a comment to this blog post: "Our
hope is to release the product earlier than that - in which case it
will be available for MSDN subscribers to download as well as for
people to buy prior to the launch events. We haven't announced a
target date for this yet - but I suspect it will be several weeks or a
few months before the launch event."
Regarding VS 2008 Beta 2, Scott writes in another comment to the same blog post that the plan is to make it available for download in the next three weeks with a go-live license. This means that we are only three weeks away from being able to use .NET 3.5 in production!
Cross-posted from http://linqinaction.net
I've just updated my profile on proagora.com.
It is now available in English: 
and in French: 
We have made several improvements to the site over the last weeks.
For example:
- we have added the ability to create search agents that you can use to receive e-mail notifications for your favorite searches
- we improved the home page to make it less static by including carousels with sample profiles and a search box for jobs
- we have started to rework the About page but we need to improve it further and translate it into French
- several small fixes (no major problems until now!)
What are you waiting to create your own .NET and Java profile on proagora.com and promote it with links like the ones I use above?
You can refer to this previous post of mine to learn more about proagora.com.
Patrick Smacchia, C# MVP, author of the best-seller Practical .NET2 and C#2 and creator of NDepend, has started to blog.
In his second post, Patrick proposes a simple trick to rationalize your code environment and build process. Basically, he explains why and how your code should be easy to move between machines.
I would add that setting up a build server for your projects helps you to improve this. In fact, as soon as you have to build your projects on a second machine, you have to ensure that all the dependencies can be resolved and all the required tools are available.
If you use continuous integration, your projects don't build only on your development machine. You have already made a step in the right direction towards code mobility.
Ok, well, this is a catchy title, but this is the way I feel right now. When Acropolis was announced, it was supposed to become the industrialized replacement for the CAB (Composite UI Application Block) and the SCSF (Smart Client Software Factory). Unfortunately, the design focus seems to have shifted a bit. When I see the announcement of the second preview of Acropolis (July 2007 CTP), I'm very disappointed to see that the main new "features" revolve around the "fun" aspects of software applications more than around "enteprise" features. If Acropolis is here to help you create bling-bling applications, well it's not the application framework we need.
In the new preview, the major improvements are related to the support for transitions (fancy animations when switching views - rotations, 3D effects and the like...), and the support for theming (custom themes and styles). Of course there are also improvements on the design-time support, but when I see "for instance, you can now use the Application Designer to select your themes.", this does not look like something critical.
It's very disappointing to see that Microsoft is focusing on the eye-candy features at this stage of the development...
At my client - a major bank - we need to create modular applications because all the software pieces created in-house have to work together and with the current growth we need to foster reuse through the creation of components. We also need to speed up the development of new applications and provide guidance for the developers. This is why we developed an application framework using the CAB and a custom guidance package. We provide the developers with a custom shell and templates for their projects. The modules developers create are hosted in a common shell that provides a standard set of services. Considering that my post in the Acropolis forum didn't receive any answer, it looks like this is not the way Acropolis is heading.
Also, in the user interface we build, we tend to rely on Visual Studio-like docking. This allows us to create integrated business environments, similar to the IDEs (Integrated Development Environments) we all know. Again this does not seem to be part of the Acropolis plan.
I've just noticed that someone has already
raised concern about this in the Acropolis forum. Brad Abrams's reply to the concern aims to be reassuring, but I'm not fully convinced. "Keep in mind, we are still very, very early in the acropolis project...there is time to develop the core ideas more fully." Personally, I'd like to see the the "core ideas" developed first, before the decorative features.
We'll wait to see where Acropolis goes next, because it's to early to throw it away like I appear to do with the title of this post. However, if it doesn't move in the right direction, we may have to create our own tool to build composite and standardized applications. Maybe we'll stick to CAB and improve it, or start from scratch after all.
Some quick links about LINQ:
Cross-posted from http://linqinaction.net
More Posts