In response to Michael Pence's message to TDAG (The Delphi Advocacy Group, I responded with the following. It is mostly about the longevity of Delphi.
I think you hit the nail on the head when you questioned the need for all the abstraction in a *single* language environment. You only need the abstractions when you are going across language and platform barriers.
As far as the longevity of Delphi, I think it will be around for a long time. And let me tell you why (I knew you wanted to know). For those of you who read Joel's article on How Microsoft Lost the API War you may have noticed the paragraph that mentions Borland. It not then read it again.
In a nutshell it said that Microsoft needs to keep Borland around. As I am sure you know Microsoft invested a large amount on money in Borland a few years back, right after Dale took over, and shortly before they fixed their name (I think I have that timing right). Borland was not doing to hot, and Microsoft bailed them out (figuratively speaking). Now why would Microsoft support Borland? Call it coopetition (cooperative competition). While they are competitors at one level (both make Development tools for MS Windows), Borland still is one of the largest suppliers of development tools for MS Windows. So not only does Borland directly support MS Windows development, they also provide lots of field research for Microsoft Visual Studio. The other side of the coin is that Borland helps prevent Microsoft from looking like a monopoly of development tools for their environment.
Look at Sam Goodie and Musicland. For those who don't know, they sell music. At our local mall they have both store. Some people don't realize that the same company owns both stores. It creates the illusion of competition. People are more likely to come to the mall to buy music because there are not one, but two stores that sell music.
For MS Windows, the more companies that provide development tools for Windows the more appealing it is for developers to support that platform. This is one of the reasons that you don't see much support for the Mac since there are very few (comparatively speaking) companies that provide development tools for the Mac.
The amount of revenue in sales of development tools that Microsoft looses to Borland is less then the value of these other benefits that Microsoft reaps from Borland's existence. As long as that is the case Borland will be around.
Frequently I talk to project managers who are currently using Delphi and fretting how difficult it is to find experienced Delphi developers. They are considering jumping ship. I think that now is the worst time to do that. With Delphi for .NET the language is becoming both less and more significant.
Look at the VCL. If you are a C++ Builder developer and you are looking at an application written in Delphi with the VCL you can tell exactly what is going on. The FCL (Foundation Class Library) provides the same glue for .NET. The language, for the most part, provides syntactical sugar. What is your favorite flavor? This, and the fact that assemblies written in different languages can interact seamlessly is what makes the language less significant.
What makes the language more significant is the developer’s productivity within that language, and the suitability of the language for specific tasks. For example, Perl.NET would be a great choice for writing a text parser because it includes regular expressions in the syntax without requiring an object reference like C# does. As far as individual developer productivity, Delphi can’t be beat. Beyond the language itself, many developers today took a Pascal class back in school, and it sure does look familiar. I am always surprised how quickly someone can "pick it up".
So don’t drop Delphi now.
You also questioned how long DevExpress and ASTA would be around. My biggest worry about DevExpress is they will go after .NET at the expense of VCL. I see that they have Code Rush for .NET, but not for Delphi 8. As long as they support both .NET and Delphi I would expect them to be around for a long time, it is the big component developers that don’t support .NET that I expect to have a rocky future. ASTA already has support for .NET with ASTA Safari and ASTA IO, but you should talk to Steve Garland for more information about that.
Love it or not, .NET is where a lot of money is to be made.
I could go on for hours, but I’d better not.