10 Most Important Points From Dr. Tim Sneath’s and Scott Guthrie’s Silverlight videos on Channel 9
I started writing this post as two separate summaries, however these videos contain so much overlapping information, I decided to just combine them. If you want to watch the videos yourself you can find Dr. Sneath’s video here and Scott Guthire’s video here.
Top 10 Important Points
#1) Silverlight 1.0 is not a second class citizen to 1.1.
Dr. Tim Sneath drove this point home and Scott Guthrie gave us the reasoning behind the smaller scope of the 1.0 feature-set. According to Scott, after some research, they realized that consumption of audio and video in the browser enticed users to install plug-ins like Flash more than anything else. So, they focused on having a solid solution for delivering audio and video on the web, hoping to help drive adoption. And, as much as I want to see a Silverlight 1.1 release RIGHT THIS SECOND!, from a business perspective, I can certainly understand.
#2) Charles asks some stupid questions.
Update: Criticism without the preceding “constructive”. Big no, no. See Dave back-pedal here.
Like…REALLY stupid. I mean, seriously…sometimes I get the slight inkling that this man doesn’t understand any of the technologies he does interviews about. And, judging by the rather frequent, barely veiled to outright slack-jawed looks of flabbergasted surprise on his interviewees, they would agree with me.
#3) Writing browser plug-ins ain’t easy.
As he mentioned in another Channel 9 Silverlight video, Scott re-iterates that writing the Silverlight browser plug-ins presented one of the most challenging aspects of the development process. He also mentioned that Microsoft might put together a knowledge base for writing such browser plug-ins. I’d love for THAT to see the light of day. Since they promised to help the Moonlight guys, perhaps we’ll see that knowledge base in the form of some open source code. Speaking of which…
#4) Silverlight will work on Linux.
As a matter of fact, it already does. Even Scott Guthrie himself has demoed “Moonlight”, Silverlight’s Linux incarnation as dubbed by the Mono team, to a live audience. In true Mono fashion, all code for Moonlight will remain open source, however Microsoft will only offer their video decoders in binary format and for browser-based applications only.
#5) Buy Halo 3
Talk of Silverlight’s support for VC-1, MP3, WMV and WMA always led to mentioning the new Halo 3 trailer. While not an intentional product placement, we get the message: Download Silverlight. Watch the trailer. Buy the game. 
#6) Silverlight is not a panacea.
Although WPF sits at the core of Silverlight, Dr. Sneath makes it quite clear that plenty scenarios exist where you’d want to write a full WPF client application: accessing the full filesystem, integrating with other applications (like Microsoft Office) and hardware accelerated 3-D visualizations, just to name a few. As a man who “moved his family 8000 miles to work on WPF”, I can see why Dr. Sneath wants to let everyone know that WPF on the client remains alive and well.
#7) XBAP is still here.
Besides one or two Channel 9 videos, I haven’t heard much talk about XAML Browser Applications (XBAPs). However, Scott did mention them once. For those who don’t know about it, think of XBAP as the heavy weight brother of Silverlight: it still runs in the browser (IE only, it would appear actually, Adam Kinney left a comment letting me know that as of .NET 3.5, XBAPs WILL work in FireFox.) but requires a full installation of the .NET Framework on Windows. Perfect for intranet applications where you might desire more client-side interaction than Silverlight can provide, but still have restrictions on what you can install on employee’s desktops. I’d like to see this particular solution mentioned more by Microsoft evangelists.
#8) Silverlight is not an ActiveX control.
Update: Dr. Tim Sneath has clarified this in a comment here.
Huh? I think Dr. Sneath meant that the browser doesn’t use it like an ActiveX control embedded in the page, but like a true plug-in. However, under the covers, it reeks of blatant ActiveX-isms. I haven’t tried it myself, but some people have gotten statements like var AgControl = new ActiveXObject(”AgControl.AgControl”); to work and have used regsvr32 on Silverlight DLLs to fix installation problems. Looks ActiveX-ish to me. He also says that because they implemented it as a plug-in you don’t see the “yellow warning bar” appear in your web browser. However, by default, I get the yellow bar in IE 7 all the time.
Even Scott couldn’t provide adequate clarification. When Charles asked Scott about hosting Silverlight in an application as a migratory stopgap, I fully expected him to explain some sort of ActiveX magic, but instead he looked taken aback by the question and fumbled through a convoluted solution involving hosting IE in your application and then loading Silverlight into that. What? Is that really necessary? Why not just host the ActiveX control yourself? Perhaps Microsoft just doesn’t want developers doing that…?
Weird.
Anyway, I know the Mono guys see Silverlight in desktop applications as a viable solution (look under the heading “Gtk# Widget and the Gnome Desktop”), so at least we’ll see some development around that.
#9) JavaScript is and will forever be, a first class Silverlight development language.
Yes, managed code will bring orgasmic programming pleasure, but right now, officially, we have JavaScript and it can handle more scenarios than even Dr. Sneath admitted he originally gave it credit for. While we hardcore Silverlight developers will undoubtedly spend much of our time in the managed world, designers and “UI-focused” web programmers will integrate with our code primarily using JavaScript. So, knowing how JS works with Silverlight can help maximize productivity and make it easier for others to innovate on top of our work. And, there probably exists a number of problems that JS can solve quicker and easier than C#.
#10) Tim Sneath (and by extension, the whole Silverlight team) needs a vacation.
…don’t we all. But, these guys definitely deserve it.
If you didn’t know much about Silverlight these videos will help familiarize you with what Microsoft has to offer. If you did, the few juicy nuggets of information scattered throughout the videos will keep you watching. Interestingly enough, even though these two videos supposedly focus on Silverlight 1.0, the interviews always seemed to gravitate towards 1.1. Microsoft must hear developers chomping at the bit to use managed code with WPF in the browser. Hopefully, we’ll see them make a Silverlight 1.1 announcement soon.
By the way, at the end of his interview, Dr. Sneath mentioned using the Scriptable attribute to call managed code from JavaScript. I just wrote a tutorial on this so feel free to check it out.
Related Posts:
Comments
7 Responses to “10 Most Important Points From Dr. Tim Sneath’s and Scott Guthrie’s Silverlight videos on Channel 9”
-
Featured Links
-
Featured Posts
- 10 Minute JavaFX Tutorial - Develop and deploy JavaFX Applets and Applications while online and offline
- 10 Minute Tutorial - Silverlight: Building a Silverlight application with MSBuild (C#)
- Introduction to game programming in Silverlight
- 10 Minute Tutorial - Silverlight: Hello World
- 10 Minute Tutorial - JavaFX: Hello World
-
Recent Comments
-
Category Feeds
-
Recently Written
- 10 Minute Silverlight Tutorial - Implementing Silverlight 3 out-of-browser support
- Word Invaders Flash Game Post-mortem
- 4 JavaFX tutorials for beginners
- 10 Minute JavaFX Tutorial - Develop and deploy JavaFX Applets and Applications while online and offline
- 10 Minute Silverlight Game Programming Tutorial - Shootorial Conversion #5 (C#)
-
Archives
Nice write up, entertaining and informative
I just thought I would let you know that XBAPs in .NET 3.5 will run in FireFox as well as IE. ( http://blogs.msdn.com/tims/archive/2007/07/27/what-s-new-in-wpf-3-5-here-s-fifteen-cool-features.aspx )
Now I’m off to take a look at a few of your tutorials…
Thanx for the comment. And, thank you for correcting me about XBAPs running in Firefox with .NET 3.5. I’ll update that post.
This is a fun write-up: thanks for putting it together!
Let me just clarify #8 really quickly. Yes, Silverlight is implemented as an ActiveX Control as far as Internet Explorer is concerned (in Firefox, it’s a plug-in). If you go to IE / Manage Add-ons, you’ll see it listed there as an ActiveX Control (listed as “AgControl Class”). You’re also right that we don’t support its use as an ActiveX Control outside of the browser.
In the interview, the point I was making is that the Silverlight runtime is installed as an executable, with the location triggered from a JavaScript file that you include. That’s why you don’t see the yellow “information bar” in IE when you go to a page containing Silverlight content but rather a little “install now” web button; we picked this approach to offer the simplest, most seamless way of getting Silverlight onto a machine. It’s physically installed as an EXE, rather than through a CAB file, if that helps explain.
Take care,
Tim
Care to elaborate on my specifically stupid questions? Thanks. Since they are so common, you can just target these two interviews.
C
Dr. Sneath,
Thanx for the comment. The post now points to your comment for further clarification.
Charles,
I committed the “cardinal sin” of providing criticism sans the “constructive”. My bad. Happens to the best of us.
I’ve got a few posts in the queue, however, once I find the time, I will review the videos that generated the feelings which led to that statement, so I can provide you with accurate feedback. I will say that the questions…in question…don’t occur as frequently as your comment implies. The ActiveX question in Scott Guthrie’s video targeted exactly the type of stuff I want to hear about. Also, I thought the MSR Cambridge video went extremely well, and I liked several of the “going deep” shows, especially the one about security and the NT kernel. But then comes that one comment or question that belies an assumed understanding, built from previous interviews and statements, and completely derails me. And, it also appears to, at least momentarily, do the same to the interviewee. But maybe that’s just me projecting my own understanding onto someone else’s (another cardinal sin
).
Anyway, I’ll let my original statement go unaltered because I don’t believe in rewriting history for the sake of back-pedaling (I’m not a politician
). But for my own personal reckoning, I will link to this clarification in the original post. And, I’ll make sure to review those videos and give you some more detailed feedback…and provide healthier criticism in the future.
Thanx for the comment.
Miles, thanks for the feedback. Keep in mind that my interviewing philosophy and implementation involves carving questions from the conversation as opposed to achieving a premeditated agenda. I remain behind the camera because it is the interviewee matters most. Always.
I go into my interviews without reading up on the technologies at hand because I. like the viewer, want to learn in real time. My questions must be created on the fly and as a direct result of listening to the interviewee. Much of the art of conversation involves listening. Sometimes, my style invokes stupid questions, based on lack of understanding, but this fine. This limits the possibilit of these questions being created in the minds of the viewer. So, it my hope that my questions accurately reflect those of the viewers/listeners of the conversation.
At some point, I may blog about this conversational, “real-time” questions(see above for definition) based interviewing style.
C