Possible paths to convert windows form application - hybrid of controls and html - to web

Question

The existing application mixture of native and third party controls and hosts IE in a webbrowser control for rendering html. Also there is bi-directional update of UI - selecting anything in native controls ( e.g. treeview ) updates html , and vice verse javascript or hyperlinks in a webrowser may cause certain actions which update UI on a form or an application level.
I see 3 possible paths for conversion:

1. Silverlight seems to be the best candidate for conversion, but if try to mimic existing interface we can only do it in SL4 and also only in out of browser mode which makes this conversion a bit pointless. I know about existence of controls which will display HTML not in windowless mode such as C1RichTextBox, but quite clearly it is not going to happen in a near future that such a control will be able to render any arbitrary html properly as IE does ( not all html under my control ) . Plus as I mentioned before in Windows Forms it is possible to call external C# methods from Javascript ( selecting row in a table by mouse ) ) and update UI accordingly ( webrowser in SL4 does it , but as mentioned it out of browser mode).
2. I guess, all can be done purely in ASP.Net ( Web Forms or MVC ) with AJAX and jQuery UI, but, I believe, it is MUCH more work.
3. Possibly ASP.Net facade as a main site/application to show HMTL and self-contained Silverlight modules spawn in a separate browser tabs ( I am not sure whether it makes things simpler than 2).

Perhaps, I am missing something, and there is another better and easier way for such a conversion for all-rounder who has some familiarity with all these technologies, but no real application delivered yet ?

Answer 1

My comments for the 3 points:

1. Keep in mind there'll be a huge change of user experience. Even thought it might look similar visiually, SL gives a more desktop feel than typical app. User who are used to using browser 'back, forward, refresh' buttons will be 'disappointed' of course unless you can address it which can be painful.

2. Definitely do-able. How much work? That's quite subjective. I'll go for this though as it might be 'more-work, but likely to be more straight-forward and easier to manage.

3. This might just blow up the cons of 1 and 2 and not so much or the pros. I might be wrong :P