A Case for WinForms
When DevExpress released their roadmap for 2011, WinForms barely got a mention. As a ten year old technology that is basically abandoned by its creator this isn’t too surprising. But what it interesting the amount of negative feedback that it generated. A lot of DevExpress’s customers just don’t see WPF or Silverlight as a viable replacement for their major applications.
Seven years ago Avalon was announced. Now known as WPF and Silverlight, it represented a major evolution in the design and development of graphical user interfaces. Combining the declarative style of HTML programming with WinForm’s ability to create custom controls has proven to be very effective. With three major versions of WPF and four of Silverlight now in production, one would expect the technology to be pretty solid.
Unfortunately there continues to be serious performance concerns, both in terms of processor and memory usage. While very real on their own, these concerns are exaggerated by the steep learning curve. Even minor design flaws have a bad tendency to multiply the amount of resources needed for a given screen.
It is sad to see Winforms pushed back so much. WPF is still too slow on most computers for major apps and SL is not mature enough for a complete ERP app.
[…] what's the meaning of much faster graphs if my customers use Citrix and Terminal Services... This is where WPF and SilverLight fail to deliver and why we need further improvements on WinForms too.
Sorry guys, but with this roadmap I will not extend my subscription. I use only WinForms and ASP.NET and I'm not interested in WPF/Silverlight - WPF at this time for me is not suitable for my kind of applications (larger business Apps). Silverlight in my eyes is a dead technology - HTML5 is the future for rich internet applications.
I have to agree to the comments on WinForms.
SL and WPF are not ready for real business applications. WinForms is a robust and well understood platform and has broad support.
WinForms is not going to be replaced by WPF or Silverlight or HTML 5 any time soon. WinForms apps work great with internet services on Amazon, Google and Microsoft Windows Azure. CUstomers love and expect responsive business applications which work as fast as Microsoft Office apps. For now, WinForms is still the best method of developing and delivering this.
Our team came to DX some 2 years ago mainly because of the obviously superb winform tools, and winforms remains and will remain the main focus of our business for the forseeable future.
Experimenting with Silverlight, WPF etc., is good for us to keep an eye on what might become mainstream one day (or maybe not if Ms dev people find some new super toy to play with) but the technology is not yet mature enough to base our business on.
While this may certainly be just a case of the unhappy minority, they do represent real development shops that still don’t see WPF/Silverlight as mature enough for large scale applications.
Awful 'article', really...
A lot of DevExpress’s customers just don’t see WPF or Silverlight as a viable replacement for their major applications.
A lot of developers don't want to learn WPF/Silverlight. I know many of them. Personally, I've been working on a large business application for a multi-national food services company and WPF is the least of our concerns - and our application's look & feel 1000x better than its WinForms predecessor. WPF was the right decision for us; yes there's a learning curve because it's completely different than WinForms. But it's not rocket science. There are many lazy developers out there who don't want to learn new tech, it's nothing new. You could easily replace the title with 'A Case for VB6' and you'll find many shops still unwilling to transition to .NET. That's not to say that there aren't legitimate business reasons to stick with whatever technology is being used (VB6, WinForms, straight Win32/C, etc.) But when it comes to to rewrite said apps, it's just penny-wise and pound foolish not to take the time really to learn SL/WPF.
With three major versions of WPF and four of Silverlight now in production, one would not expect the technology to be pretty solid.
Three/four major versions in production, which correspond with major .NET releases - has what to do with the SL/WPF stack being 'solid' or not? Does that make .NET a shoddy framework because it has 4+ major releases?
Heiko Mueller: WPF at this time for me is not suitable for my kind of applications (larger business Apps).
Does this guy back up his statements with some proof?
Jens Necker: ...SL and WPF are not ready for real business applications. WinForms is a robust and well understood platform and has broad support.
Based on what - his lack of experience with SL/WPF? WPF & Silverlight are robust, the real issue is some developers are scared/unwilling-to-learn of new technology.
As for 'broad support', it's obvious the commenter doesn't have a clue. All the major component vendors (Infragistics, Telerik, DevExpress, ComponentArt, etc.) have rich SL/WPF suites (and have had them for years.) So what 'broad support' is he talking about?
I would consider Microsoft's Expression product line to be as complex as any real-world business client application. It uses WPF through-out, and I haven't had a complaint about its performance (speed-wise anyway.)
Come on InfoQ, I expect more from you guys than this...
I don't see it
Re: Awful 'article', really...
Re: I don't see it
I couldn't have said it better, although I would have been sarcastic. ;-)
Re: Just me?
WPF is doing very well, thanks
It's also about new things for the developers. For example, my team loved binding mechanisms, all style things, MVVM and so on...
Sticking to WinForms is ok for business.
Businesses in general do not care about the interface and want an application as grey as possible. This might change however because of smart phones and tablets being used more and more by the business.
Starting a new business app with WinForms doesn't look like a good infestment and should at least be architectured in such a way that it's easy to change the interface.