This blog has moved. Please update your bookmarks.

Delphi 2005 Impressions

I just came back from the Delphi 2005 preview in Stockholm and my first reactions are very positive. Jason Vokes from Borland Europe held a seminar where he (rather quickly) went over the new details of Delphi "Diamondback", and although the IDE still seems to have a few bugs (which led to some amusement among the spectators), it was very professional.

Delphi 2005 combines Win32, .NET and C# in a very efficient platform. It introduces several new behaviors such as unit testing, refactoring and a lot of goodies from the .NET platform and integrates everything in one, clean GUI.

What really strikes me how neatly everything has been folded into a single IDE. Whether you want to develop, debug or test a Delphi Win32 project, or a .NET project (or C#), it doesn't matter - it's all contained in the IDE. You can even combine and interact between them in a single project group.

A few techniques were also demonstrated with .NET data remoting that really impressed me. Although I don't fully believe that this will be of immediate interest to my company, I got the feeling that much is happening right now in the .NET arena that will be commonplace (and necessary) in a few years.

I think Delphi 2005 will be a version which we can't really afford to miss out on. We've been staying with Delphi 5 for a long period of time, but I think this will be the version that will finally begin pushing Delphi 5 to the sidelines. I think from this point, the Delphi community will begin the shift towards .NET in a big time and there won't be any turning back. Luckily, the way Borland has implemented it with VCL.NET, the old source code will still be running fine under .NET (most of it, at least) and they ensure that this will provide a smooth migration process. A lot of businesses still have tons of legacy Win32 code to maintain and I think Borland provides just the right "cushion" against new technologies - while at the same time moving quickly forward - to embrace the new ideas without dropping ten years of legacy applications.

While I don't think the transition will happen immediately, the .NET features and the new goodies in the IDE will be important enough to warrant an upgrade from our part. I don't think we can afford to lose out on this. If we don't stay on top, someone else will; with Delphi 6 and 7 this wasn't so much an issue - with Delphi 2005, it may turn out to be critical.


Post a Comment

<< Home


Blog contents copyright © 2005 Mats Gefvert. All rights reserved.