First of all I must say I like DNN, I don't love it but I like it. And I use it.
I think the core is just great, now it is fast, stable and covers almost all my needs. The problem is in the modules. I think a project like this, which is aimed at people who don't want to stay for hours to create a single page, post a blog entry or uploading their photos should have "killer modules"... Blog, Picture Gallery, HMTL edit, Forums and one or two more. It is fine that there is more modules, but some of them should have an extra effort to make them perfect.
People look little to overall performance (well if performance is crap they don't use it, but this is not clearly the case), they look if when they post a blog entry they see it, if they can upload theis files easily, if they can type fast and create the page...
Today production DNN let them do that things? In an overall sense... yes, but not in a proper way:
If you post a blog entry and you're not in the same timezone, your entry will (or not) appear the next day, because there happen extrange things with that. If you click on the forums breadcrumb you can ramdomly be redirected to another forum, If you upload a portrait photo, the frame of the photo is likely to be badly painted. If you type your ads html code in the html editor the tags get re-encoded the second time you edit them...
What I want to say? DNN is a great framework, and it seems it has reached maturity, but the modules clearly haven't. When I tell people to install dnn for their sites, they say: "Well, the blog is not as good as XXX's" "The forums have problems"...
I think is time to stall the core development for a couple of months (or whatever) and help the modules teams to pulish the things, then we'll really have a killer framework with killer apps.
Just my 2cents.
Enrique Blanco
www.eblanco.com