The topic of whether to execute a project by developing in Dexterity came up in a requirements discussion I had this week after The Dynamics GP Blogster and then Mark Polino opined on the subject. I have spent some time thinking about this myself as we're also working on plans for 2011. I don't think Dexterity is going anywhere as long as GP is around although it will continue to get depreciated over time.
When Microsoft purchased GP and consolidated ERP solutions under the Dynamics brand all the buzz was about both consolidating those solutions into 1 best of breed solution and/or replacing Dexterity with (insert your favorite .Net language here). I don't even remember how long ago that was. We're not really any closer to realizing either vision even though it's clear that .Net is the future for GP.
With the introduction of the Visual Studio Toolkit we're able to finally do some GP Development work in Visual Studio. That's great but we still can't create Alternate Dynamics Windows or really modify existing GP windows in Visual Studio. Dexterity still provides the tightest integration. For some development projects, you just can't get around that. I can't really speculate on how reasonable it is to expect that to change anytime soon.
Mariano makes great points in his IMHO post that it is basically (my opinion, not his) just not reasonable to expect Dexterity to be replaced by .Net unless GP is abandoned as a viable ERP solution. The last I saw the Dynamics GP roadmap extended beyond v14 (2016). When it comes to selecting a development tool for GP it's inevitable that Dexterity will rise to the surface as the best tool for some projects. It's not always my first choice but sometime it is the best choice. I don't foresee that changing anytime soon.
There's a good white paper on choosing the right development tool for your Microsoft Dynamics GP 10 Development Project. http://www.microsoft.com/downloads/en/details.aspx?FamilyID=a5b6c523-0add-48fd-9deb-2c0ef39b5673