Is that a Silverlight in the Web Interface future?

I have been working with both of these technologies (WPF and Silverlight).  Beyond the "glitz", they are compelling for me to use as a developer.

I used WPF in developing my "PimpMy For SoftGrid" product, released earlier this year.  This requires installing .NET 3.5 on the target system (but fortunately the installer takes care of that) which is the runtime implementation (The developer can also target .NET 3.0 to use WPF, but I needed some of the newer features).  Going with WPF in this way allowed me to have the same interface as a non WPF app (plus a nice but unnecessary zoom feature thrown in for bling) in about one half of the code.

Developers will be moving to this.  They can spend more time working the business logic and less on the plumbing.  You on the other hand, will be buying more memory.

[Had to break this up because of the comment size limit]

I like Silverlight (2.0) as well.  It allows me to update a portion of the web page without redrawing the entire page.  Not that their are not other ways to accomplish this.  But since I didn't jump on the earlier methods Silverlight just makes it easy enough to make that jump.  Client side graphics management just makes sense in the web world.  I had a web app that wanted to display some "realtime" graphs once every 5 seconds - it was awfull.  With silverlight I could send the info to the client every second and have the client browser implement the logic to build the graph from the data.

Better buy more CPU for while you are shopping for that memory.

Yup.  We're making good use of .NET 3.x too, for instance using WCF as a foundation for new communication paths in XenDesktop.  But thus far we haven't planned a switch to WPF for UIs, since our admin consoles can happily use WinForms for a while yet and end-user UIs are a much tougher case to consider (given the breadth of client devices we want to support, and the CPU & memory costs on the server/session side).

That's partly why I'm very keen for us to build up a strong story around web services that are useable by many front-ends.  We can't make the standard WI depend on Silverlight probably for years to come, but if we can dramatically lower the cost of building alternative front-ends then people who can stomach the pre-reqs and want it can have it.

When someone can write a Yahoo/Vista/Leopard/... widget in 20 lines of code, what does it matter that each of them only appeals to a few % of people!

Cheers,
AndrewI

Am I reading Thomas Koetzing's blog/article right?

"After Web Interface 4.0, 4.1, 4.2, 4.5, 4.6 and 4.7 Citrix will release version 5.0 together with the Delaware release of Presentation Server. The new Web Interface version has to run on Server 2008, which means Internet Information Server 7.0, but what else is new with WI 5.0? This article describes the details of the next release."

So the next version of PS/XenApp will only run on Server 2008, and not 2003!?  Is Citrix really looking at the news and facts that the most businesses currently have no interests in moving to Vista/2008?  Their really shooting themselves in the foot with this one. Where I work at, I don't see us moving to Vista/2008 for at least another year or two (maybe even three), and in my personal opinion I would like us to sit out till Windows 7.

Silverlight is not about the server side of the equation pe se, although your point is as much about toolsets and that is very relevant.  It is competing against Flash, and more particularly AIR - and since AIR is new too, that is shaping up to be a pretty intense two-horse race.  I expect this to play out rather like it has between Java and .NET on the server side; Microsoft takes a good hard look at what Java did well, and what could be done better a different way now, and they do that.  It won't wipe out Java, partly because the Java community is strongly motivated to innovate and improve too, but .NET caught up really quite quickly.

From my perspective, I'm not trying to place bets on either Silverlight or Flash/AIR - both are interesting, and I want it to be possible to use either to build front-ends.  I don't even want to predict too much what those front-ends will want to do - the point of services is that the service doesn't care any more than it absolutely has to what the caller is doing.   (BTW, that is part of what is not quite right about the PNAgent Services support in Web Interface today; PNA Services are real web services, but the interfaces were not conciously designed to have minimal coupling in the interface contracts.  Ie. PNA Services assumes it is speaking to PNAgent, and expecting PNAgent to behave in certain ways based on the config.xml settings on the server.)

Just to clarify since the name "XenApp" has been thrown around a lot:

Yes, Delaware is Windows 2008 only. In fact, that's kind of the whole point for Delaware, since there are so many under the hood changes from MS, Citrix really had to do a major update to make Presentation Server work on 2008.

PS 4.5 will still exist for Windows 2003. Delaware / XenApp is like PS 4.5 for Windows 2008. Citrix will also continue to release Feature Packs for PS 4.5 for Windows 2003 to add as many Delaware features as possible. (Although some, of course, will be features that require Windows 2008, so they won't make it into new FPs for PS 4.5... Like Windows special folder redirection for example.)

I wouldn't really classify Delaware/XenApp like PS 4.5 for Windows 2008... because Sciota/XenApp is the next version of PS for Windows 2003. A lot of the new features of Delaware and Sciota parallel each other, it's just that Delaware runs on 2008 and Sciota runs on 2003. Likely Delaware will be released as XenApp 5.0 for Windows 2008, and Sciota will be released as XenApp 5.0 for Windows 2003... but don't quote me.

Great article and info on the upcoming WI release. The article Thomas did was good too. It's great to see a new look and feel for WI...it was beginning to feel a bit dated. Does anyone know if the Secure Gateway will still be available/supported with WI5? Any ideas whether if/when SG be phased out, thus forcing some of us to move to Access Gateway or other solutions?

 Thanks!

For those that are interested in an example of these technologies together, check out this site from Narenda Wicaksono:

http://narenda.mvps.org/coolts/

 and the related blog posts:
http://wss-id.org/blogs/narenda/archive/2008/02/05/ts-web-access-extensibility.asp

Thanks James.  There are no plans that I know of to drop Secure Gateway; as ThomasK just noted , a slightly updated version is included in the Delaware preview that runs on Windows Server 2008.  Of course, Access Gateway continues to be the focus point for most of our ongoing development around gateway functionality, and part of its value is to take away the overheads the customer would otherwise bear for hardening and maintaining a server in the DMZ.

Cheers,
AndrewI

For reference, the second link is missing an "x" at the end: the working link is http://wss-id.org/blogs/narenda/archive/2008/02/05/ts-web-access-extensibility.aspx

Cheers,
AndrewI