I realize that those two things can be seen as incompatible in some circles ;)
Although it is all in Japanese
Web site Design, at least to me, can be very subjective . . . and its appeal may differ from person to person not to mention from browser to browser...
Check out these ACS/OpenACS sites and see if they fit your definition of "visually stunning" ...
Deds and me did majority of semizone. With the amount of custom
code that is put on it... I am not too sure I would like to migrate
that to dotLRN. But then again I will evaluate it and if the client
pays us for then we must consider dotLRN.
Initially Semizone did not have that focused stuff in e-learning and
it then evolved to this "Education on Demand" thing. Then
everything focused on it. This is by the way the first ACS site and
Tcl site that me and Deds did. Not too bad for a couple of newbies
Doesn't www.infiniteinfo.com count? It looks ok although I must
admit it is the least functional site that I ever done.
If aolserver only counts without ACS code I think www.foxsports.com
has got to be the best looking and functional aolserver site.
Personally I don't see what the visual experience has to do with OpenACS. You could use wget to 'download' any site and have the static content served by an OpenACS install. OpenACS probably doesn't help or hinder the visual experience at all.
I've learned the hard way that non-technical clients somehow think that there's a correlation between the capabilities of the software they're using and its ability to wow them visually. You can say, "But we can make it look like whatever you want; the important thing is the underlying software engine" until you're blue in the face, but it won't make any difference.
The problem is that many people associate a nicer interface with better-quality software. Here in Israel, I've had a number of clients ask how Linux can possibly store Hebrew in the database if it can't open Hebrew Word documents. Luckily, Mozilla has made it possible for me to view Hebrew sites, which has made it much easier to convince them that I'm not talking nonsense.
At the same time, I've found that giving a potential client a list of OpenACS sites demonstrates the common functionality that is hidden behind diffeent designs. Once they've seen 3-4 sites that look quite different but do the same thing, they begin to understand just what's going on, and that the interface can appear in just about any way they please.
Joel Spolsky has an interesting take on this at his site.
wgbh.org is not bad, it's ACS4.