Hi, I’m Leslie Koorhan, and I’m doing the .NET blog. This is very exciting for me
as we are transitioning (as I’m sure many of you are) into the latest development products
from Microsoft, Visual Studio 2008 and the .NET Framework 3.5. So I have a
bunch of things to talk about, with all that new technology, but also some great
things that I like about Visual Studio 2005 and the .NET Frameworks 2.0/3.0
But, let me talk a little about myself as an introduction: I have been in the computer
business for too many years to admit, started with PCs back in 1982, using dBase
II on a CPM platform machine. I developed and taught FoxPro and Visual FoxPro for
8 years, spanning at least 6 versions in that time span. I have been working with,
and teaching, Microsoft SQL Server for the past 13 years, spanning now 5 versions
of that product. I learned VB6 and then have been involved with .NET since it was
in beta, back in 2001. I “converse” in both Visual Basic and Visual C# with no particular
preference. (There’s something to like about both.) And I have been a Microsoft
Certified Trainer for over 12 years. In the future, I will be talking about the
new certification paths for Visual Studio 2008, some of the new .NET technologies,
some of the old ones as well.
If you have anything to say, feel free, and let me know any questions you might
have about developer products. For certification information now, try
http://blogs.msdn.com/gerryo. And as always, for learning, use my favorite
site: www.microsoft.com/learning
Programming with Abstractions
Tags:
ADO.NET,
ASP.NET,
Declarative Programming,
Procedural ProgrammingIn the book
"Dreaming in Code" (by
Scott Rosenberg), there is a section on programming today being done as a layer of code on top of another layer of code on top of another layer of code, etc. Most developers realize this because they work with compiled languages that are then turned into machine code by the compiler. In the case of .NET, whether you program in Visual Basic or Visual C#, the compiler turns the code into MSIL, an intermediate language that then becomes machine code at runtime. So there are 3 layers of code with .NET.
In the .NET Framework 2.0, and especially with ASP.NET, Microsoft programmers created another layer, especially in working with data through ADO.NET. Instead of the procedural programming in creating connections, commands, data adapters and data sets, you can use a Sql Data Source object to encapsulate all of that behavior. This object does not show up as procedural code, but as tags with the markup for the web page. Then you can bind controls on the web form to the data source object. This creates another layer, of abstraction, using declarative programming. The huge advantage of this is that I can design a relatively complex web page with NO code whatsoever. The downside is that if the page does not work, there is no code to debug, just properties of controls to adjust. This is the problem with layers of abstraction in programming.