Andrew,

I think this raises some rather interesting discussion points.

    Silverlight provides us with a whole new palette to design with but, as with anything worthwhile,  it comes at a cost. While Silverlight will be platform independent (should work on Mac, Linux, etc ....) we can't assume that performance would be acceptable from any client machine. A Silverlight XenApp UI would require a fairly modern workstation with a respectable amount of memory and a decent video card. I wonder how many customers would balk at the minimum requirements.

    The other point I would like to make is about the actual design. If we embrace Silverlight  the end result could be much different than what we showed in the demo. We could craft something more subdued or we could create something even more modern and cool. I would like to refine the question to one about our strategy and direction. Should Citrix continue to push the limits of Web design in our upcoming releases of XenApp ? 

Al Grandville

Product Manager - Citrix XenApp

This is one of the coolest things I have seen from citrix since the first time I ever saw Project Charlotte (how's that for a dated reference).

You have to at least make this available as a download, like the oild nfuse add ons jay tomlin used to write.

Silverlight provides us with a whole new palette to design with but, as with anything worthwhile,  it comes at a cost. While Silverlight will be platform independent (should work on Mac, Linux, etc ....) we can't assume that performance would be acceptable from any client machine. A Silverlight XenApp UI would require a fairly modern workstation with a respectable amount of memory and a decent video card. I wonder how many customers would balk at the minimum requirements.

I realize citrix is a business and needs to make money, but this post and this comment from a business guy at citrix sums up the company for me. I always thought citrix has so much potential to do such incredibly cool things and be a thought leader, and you get right on the edge of doing it, and some suit rains on the parade. As long as I have been a consultant, it has always seemed citrix was just on the verge of absolutely launching to the stars, only to fizzle out on the launch pad.

As long as this attitude pervades, citrix will never be as good as it could be. You have to take some risks and do a few crazy things to be a thought leader. Just my two cents...

Unfortunately I'm not sure we can release this, as it is based on Silverlight 1.1 Alpha which doesn't have a "Go-live" licence. If there is enough interest then we can investigate the legal issues, or reform the demo around Silverlight 2.0 Beta.

Ouch ! That suit comment hurt.

Thanks, Billy Walsh from Entourage

    For the record, I am a supporter of pushing the limits of End User Experience and design. I would think it's obvious that  companies that don't embrace these trends are doomed to become irrelevant. Just look at how much time and energy Microsoft, Apple, Google, etc .. are spending on their user experience. This movement isn't any different than when we transitioned to graphic UI's back in the 80s. Software companies ignore this space at their own peril.

    However, we are running a business, it does take time and resource to turn something like this into shipping product and we have to test our market to see if it's ready. So, is it ? If we build this will they come ?

I'm all for posting these types of things out on CDN and letting the download count speak for itself. Like John says this demo is based on Silverlight Alpha code which we can't redistribute. MS has Suits too My understanding is that the Beta 1 of Silverlight is due out this quarter. Mix is next week, you can do the math. I doubt there would be much in the way of us releasing a version based on the beta code. So, to quote my father, "We'll see."

Al- 

Andrew's right on target.  If we drop that cost, anyone, including Citrix, can come up with the next cool thing without getting too worried about the business impact. 
However, from the Usability Design and Evaluation Team's perspective, we have a UI that is cost and resource accountable as well. We need to make the right choices for our design dollars and focus on the features and functions that our users want and need and are not supplied in our products (or our competitors) today.

A  UI that makes users say, "I want to use THIS!" might very well turn out to be what the users want and need. We know our customers can and do customize but if our look and feel is up to date, easy to use, intuitive  and yes, "cool" they might not throw it out and just make minor changes allowing us to keep our brand identity. "We employ the User-Centered Design process (UCD). The purpose is to give our users what they want, not what we think they need. To do this we need to design and validate the UI with actual users throughout all stages of the process.
This process was used recently to design the latest WI (Caxton) demos you've seen and  validate the design with our users at the Usability lab in Edinburgh.
We use a multi-disciplinary team (interaction designers, developers, product managers, tech writers, architects, visual designers, etc.) for most or all of the processes.  What this entails (and has entailed for the latest WI) is:
-          Figuring out who the "80% use case" users are and what makes them tick (interviews, customer visits, observations )
-          Developing Personas,  A "snapshot" of who the users are and what they are trying to accomplish
-          Determining the users' mental models and expected tasks
-          Task Analyses and documenting how tasks flow and interact

-          An additional process was undertaken during WI Caxton design was - 'Role Model Evaluation'.  A usability comparison of best of breed products to ensure we learned from their strengths and weaknesses and avoided re-inventing the wheel

-          Iterative design and evaluation sessions including whiteboarding sessions (to flesh out design concepts), storyboards (screen to screen depiction of the user tasks) and low to high fidelity mock-ups or prototypes using tools like Visio, PowerPoint, Axure, and, occasionally, actual WinForms/WPF prototypes or existing code.  We've done evaluations remotely (over the phone and with GoToMeeting), and we've done them live.  Those of you who have been to iForum (US or UK) or App Delivery Expo may have seen us conduct evaluations in the "Usability Lab" with session participants

-          We recently completed the construction of our own Usability lab in Fort Lauderdale to bring in representative users and run them through some typical scenarios while we record their feedback on the look and feel

The UX team (and even the "suits" ) are welcome to observe, hear the unfiltered comments from the end users and they will not bias the results as they observe through one-way glass  from another  room.

This lab will allow us to engage actual users and solicit feedback on our designs in a cost effective, streamlined and iterative UCD process.
So, if you're in South Florida and you would like to participate in a study in our Usability Lab, feel free to contact me.
Doug Bloch
Manager Usability Design and Evaluation
Application Virtualization Group
Citrix Systems, Inc.
(954) 229-6391 office
(512) 789-5791 mobile
doug.bloch@citrix.com
"It's hard to make things easy" - Alphonse Chapanis