I'm new to dnn and am learning more about how it works each day.  I am excited about some of the changes that have been mentioned.  Being able to change permissions on admin modules and change their locations will allow me to give users access to peices of the website I would not have let them see before.  I see a lot of changes that seem small when you first look at them but when you realize how much they change in the long run you realize how important some of it is.  Now this is from what I have read and I haven't actually used it yet.  Widgets sound great and would love to hear more about them specifically.  It seems like some of the changes will open DNN up to more social networking possibilities. 

Bruce Chapman wrote

I just wish it was done in C#   )

Each time I see this sentiment I have to disagree.  As someone who prefers C# over VB for a host of reasons, I'm baffled when people want a C# DNN version.  Read the quite valid reasons given in the sticky 'where's the c# version' thread as to why supporting two versions would be a nightmare with little benefit.

If you like C# there is nothing stopping you from using it in your DNN install.  You shouldn't be modifying core code - and any enhancements you make to DNN should be done through either Modules, Providers or HttpModules. Each of these can be written in C# and used without issue.  Personally I write all my DNN code in C# and have never encountered a problem.    It all gets compiled down to the same IL code anyway, so there's no advantage at all to having the core in C#.   In addition, VB is very readable for learning programmers, at the expense of annyoing more seasoned developers with it's verboseness and occasional syntax ambiguity.  If one of the aims of DNN is to introduce an advanced web platform to a less-than-experienced users, then you could successfully argue that VB is the correct language choice.

The only time I ever have to look at VB is if I need to understand what is going on in the Core code to solve a problem or just 'get' what is going on.  That would represent about 1% of my total code-facing time with DNN.

And onto the topic of the OP - for me the direction the core is taking with 5.0 is the right one.  There aren't enough people and resources on the core team to cater for every whim in end-user demand.  However, if they setup the platform to allow for greater freedom of skinning and development, the rest of the community can get to work in developing the kinds of things that people want - whether released for free or as commercial modules.  All successful platforms ensure that developers for that platform have all the tools necessary to do and create what they want.  I would say that is more important than glitzy UI features - because you can always build UI bits and pieces, but it's much harder to re-engineer a platform to position it for future capabilities.

I don't want to get into C# vs VB debate. I know one can create modules in C#. I am a C# developer and I like to look at C# code instead of VB code. This releases me from doing JIT translations in my mind! Yes I am one of those people. I don't want to change core code but understand how it works. 

There are already tons of modules and skins out there so the current DNN structure seems to be a big enabler unless developers are complaining about its inflexibility and complexity.

As for the roadmap someone quoted (the Gemeni page) , it's extremely verbose and I can't get my head around it. We just need a bulleted concise, up to date list of major features (1-2 pages). I don't want to see a huge list of bug fixes.

you're both correct, framework capabilities and UI features & functions are both essential. My earlier point was that a large part of the 5.0 changes are framework features to enable future functions e.g we needed to overhaul the installer so that you can install providers -i'll be using the ability to upload providers in a future release to allow swappable security providers and many other core (and 3rd party vendors) will be able to use this feature as well. That's why despite the long timeline there isn't a huge amount of new features (though there are 1000+ checkins so theres some new toys to play with and many bugfixes) - it's vital to have strong framework capabilities that the new features can build upon.

Cathal