Welcome to WindowsClient.net | My Blog | Sign in | Join

Rob Relyea - XAMLified

WPF, Silverlight and XAML

Syndication

Sponsors





  • advertise here
Software should adapt and improve when UI paradigms change – rethink, don’t just port

During WPF’s first version (v3, of course) we considered building a porting tool to enable WinForms developers to easily migrate to WPF. We didn’t build that migration tool because:

  • We thought that the UI paradigm changes didn’t deserve a simple port, and believed people shouldn’t just port their battleship grey apps to WPF, but they should rethink and upgrade their user experience.
  • We were aware of the VB6 –> WinForms converter and had the impression that it cost a bunch, but didn’t provide great customer value.
  • Tooling inside of Visual Studio for WPF was very early in development, compared to the very complete set of design services for Windows Forms.
  • Our platform had many holes that WinForms developers would find lacking, with DataGrid, perhaps the biggest hole.

I was reminded of that first reason when I read “Extreme Makeover, Phone Edition: Comcast’s xfinity”…an example of a what a UI rethink for a new platform can do.

A couple of the other reasons have now changed:

  • Visual Studio 2010’s WPF & Silverlight Designer experience is much improved, approaching WinForms productivity.
  • DataGrid’s were shipped by several 3rd parties + eventually by our team as part of the WPF Toolkit for WPF 3.5sp1 and included in the box with WPF 4.

Interestingly, a MVP is working on a WinForms to WPF/Silverlight/WP7 Converter today. It may be a helpful tool, but it is an interesting question about what role it should play in your apps move to new UI platforms.

What do you think of the “port” vs “rethink” debate? What have you ended up doing with WPF/Silverlight/etc…

Published Tuesday, April 27, 2010 6:48 AM by Rob_Relyea
Filed under: , , ,

Comments

# re: Software should adapt and improve when UI paradigms change – rethink, don’t just port@ Tuesday, April 27, 2010 12:01 PM

The only thing I could see an automated tool solving would be for prototype work - if you were interested in quickly porting a particular dialog to experiment with some WPF features, or to at least get a general sense of the migration path you might take if you're unfamiliar with WPF.

But I completely agree with the rethink approach.  As engineers, we are often tend towards the most 'efficient' path - but this isn't necessarily the most effective long-term solution.  At my component, we have a lot of teams that strive for cross-platform UI solutions; most people don't realize that cross-platform solutions aren't truly cross-platform - they are just Strangers in a Strange Land on every platform.  I think that if you think about your UI in terms of thin clients (even if it's a desktop app) then you can target the specific strengths of any platform.  So I would take the rethink argument further:  How can you further separate your UI from your business logic?  WPF is a huge help in achieving that goal, particularly with strong support for the M-V-VM architecture.

by David Cunningham

# Twitter Trackbacks for Software should adapt and improve when UI paradigms change ??? rethink, don???t just port - Rob Relyea - XAMLified [windowsclient.net] on Topsy.com@ Tuesday, April 27, 2010 9:11 PM

Pingback from  Twitter Trackbacks for                 Software should adapt and improve when UI paradigms change ??? rethink, don???t just port - Rob Relyea - XAMLified         [windowsclient.net]        on Topsy.com

by Twitter Trackbacks for Software should adapt and improve when UI paradigms change ??? rethink, don???t just port - Rob Relyea - XAMLified [windowsclient.net] on Topsy.com

# re: Software should adapt and improve when UI paradigms change – rethink, don’t just port@ Wednesday, April 28, 2010 1:16 AM

I've alway been against migrators/converters in special case when there's a huge technology shift (e.g. VB6->VB.NET or WF->WPF)

The biggest problem in WPF is the paradigm shift, is a total new way to embrace UX programming and migration tools don't help you on this.

Instead of spending energy on migrators/translators I'd rather invest more on tooling to offer a better developer/designer experience.

by Corrado Cavalli

# re: Software should adapt and improve when UI paradigms change – rethink, don’t just port@ Wednesday, April 28, 2010 4:54 AM

I agree with this sentiment whole-heartedly, and I am thusly quite disappointed by the Windows Phone 7 team's rather careless adoption of several Zune UX paradigms.  They seem to be carrying over UX designs without even considering why those designs were chosen for the Zune in the first place, no less whether they're actually appropriate for a smart phone with a multi-touch screen.

by Mike Strobel

# Dew Drop – April 28, 2010 | Alvin Ashcraft's Morning Dew@ Wednesday, April 28, 2010 8:34 AM

Pingback from  Dew Drop – April 28, 2010 | Alvin Ashcraft's Morning Dew

by Dew Drop – April 28, 2010 | Alvin Ashcraft's Morning Dew

# re: Software should adapt and improve when UI paradigms change – rethink, don’t just port@ Thursday, April 29, 2010 6:34 PM

Winforms belongs to the 1990's. It would be a waist of resources and talents not to move to a new user experience.

We don't design and write applications for us, but rather for the users...and those users are not the same as those of the 1990's, they are more demanding and - if i may say so - sophisticated when it comes to hardware and software.  

by Ghayth Hilal

# re: Software should adapt and improve when UI paradigms change – rethink, don’t just port@ Friday, April 30, 2010 4:12 PM

Rethink is important, but in my past life as a consultant, that almost never rang a bell with anyone above the developer paygrade.

Converters are most useful for getting the UI moved over. Once that task is done, you can certainly rethink, but you can do it in phases vs. having to rethink the whole thing at once.

It also gives a leg up to WPF/Silverlight/WP7. If you have to rewrite everything, you sometimes get pressured to write it in an inappropriate technology. If you can show some percent of progress right away, then you're better off.

We went through a similar process with VB6->VB.NET. Most folks used the automated tools just to pull the UI over, then rethought the code behind it.

by Psychlist1972

# re: Software should adapt and improve when UI paradigms change – rethink, don’t just port@ Friday, April 30, 2010 4:13 PM

@Mike

The WP7 team didn't just carry over Zune stuff, they really did think through everything. Have you ever seen that neat storyboard/comp wall in the office there? I believe it shows up in at least one of the WP7 videos.

by Psychlist1972

# Aconteceu no Twitter 14 - 25/04/10 a 01/05/10@ Sunday, May 02, 2010 2:16 PM

Aconteceu no Twitter 14 - 25/04/10 a 01/05/10

by Oneda

Leave a Comment

(required) 
(required) 
(optional)
(required)