The fact that the Blend tool was used to design the non-trivial UI of Blend itself, is pretty convincing evidence that the first release version is probably already suitable for use on many real-world projects, but there are a few practical concerns that always seem to be glossed over in Blend-related discussions.

How exactly is this new designer-developer coordination supposed to work?  An interesting Channel9 interview with members of Microsoft's User Experience Team in the UK details some valuable lessons learned through trying to apply this developer-designer collaboration on actual projects (jump to minute 16:00 to skip intro material).  The team discovered that designers needed some level of understanding of Windows development and WPF, in order to avoid problems such as designers creating image elements where they should have restyled control templates, or designers breaking links to underlying business logic by "replacing" controls with restyled versions.  The developers also found the auto-generated Xaml provided by the designers to be difficult to read and edit in those inevitable situations where it became necessary or beneficial to do so.  The UK team finally settled on the very interesting approach of having the designers generate a "wire frame" DRAWING of the overall UI, which the developers use as a guide in beginning the actual project, while the designers go on to create separate resource library projects, featuring the production quality UI elements that the developers plug into their own carefully-crafted production code.  Other development teams have reported success using a single set of shared project files, but its helpful to be aware of a full range of real-world proven alternatives before rushing to apply Blend on your company's next mission critical project.

Realistically, what are the chances that your company will hire professional desigers, or approve funds to hire one of the few consulting firms with experience in this area, like Vertigo?  I recently worked at a chemical equipment manufacturer, which contracted with a professional industrial design firm for their complete line of hardware enclosures, while relying on the artistic talents of internal developers using MS Paint for their accompanying software, and a marketing manager at a different company once asked why I couldn't just use clip art throughout the UI for the "commercial quality" digital video capture and editing application they'd hired me to write (using DirectShow, to accompany one of the first generation 1394 Host Bus Adapters).  For many organizations, this attitude is likely to continue -- at least until that first trade show where a competitor demonstrates a professionally-designed product that takes full advantage of new WPF graphics capabilities.  Even if a company is willing to allocate the funds, there is still the issue of trying to FIND designers who have the combination of skills described in the UK group video.  This is not a situation where you can simply retask your company's web developer or marcom person, which still leaves it up to the DEVELOPERS to acquire expertise with Blend, at least in the near term.

On the architecture front, Blend promises to enable designers to "reskin" an application at will, but this separation of UI from business logic occurs at a much higher level than what is currently supported through code-level frameworks such as CAB and SCSF. Whereas these code-level frameworks often attempt to create extra abstraction layers (presenters, controllers, etc.) to decouple UI "views" from the underlying business logic, they still assume that developers will be required to "code" the UI.  Microsoft has already announced that CAB/SCSF will be replaced by Acropolis, which will offer some degree of support for Blend (currently, limited to the ability to open "PartViews"), but is not expected to ship "for quite some time". 

Finally, Blend is specific to building WPF applications, and WPF is a brand new technology, with a steep learning curve, which will not even enjoy a significant degree of Visual Studio designer support until the release of Visual Studio 2008 at the end of this year. Some corporate managers may feel it is safer to hold off on any commitment to Blend, or designers, or possibly even WPF, until things sort themselves out.  I disagree, in cases where a change in technology clearly represents a major advance that will need to be adopted eventually anyway, and I'll have more about THAT attitude in a future post.  A beta of Blend 2.0 is already available and, as the tool continues to mature, and the practical issues I've mentioned are resolved, we can expect Blend-based development to become the standard for a significant percentage of Windows dev. projects in the not-too-distant future.  I recently decided to check out an online course in Blend at the local junior college, and the text we are using, "Microsoft Expression Blend Bible", seems to be a pretty good introduction to its features