Why we didn’t convert to .Net. And perhaps we never will
wings-of-wind.com
wings-of-wind.com
I don't understand why he's rejecting C and C++ as unsafe yet using Delphi. Doesn't that also have pointer arithmetic and casts and dangling references?
I've never used Delphi, but I wouldn't let layout generators build my interface for me regardless of the technology.
I have tried writing a few applications in .net and I liked it. I think it's really great for the programmer because C# language has lots of features that I really want to see in Delphi's pascal. Still, Delphi (I'm still using the old Delphi 7) is my favourite development tool.
As for the merits of the article - meh - it seems to mostly be based on the writer's experience with early versions of .NET and the framework has come along way. An interesting read, but perhaps not quite relevant anymore.
Some of the arguments are reasonable when applied to desktop development, but they're still pretty light on facts. None of them are particularly reasonable when applied to web development.
He laments that GDI+ isn't hardware-accelerated, but if I recall correctly, WPF (the newer UI toolkit that came out in 2007) is.
"Almost every programmer said you have to write XAML by hand" -- I won't defend Microsoft's obsession with visual tools, but I learned that I have to write HTML by hand years ago, and I'm not complaining.
String processing is slower? Probably because .NET strings are immutable, which actually creates large performance gains in the vast majority of cases.
If your performance concerns are so stringent that you need to worry about whether the type system in a language is rooted, you're either over-optimizing or you should be working with C.
I chose ASP.NET MVC for my startup, and the ONLY downside is the slightly higher server costs.