That's an interesting discussion going on, or I don't know anything.
1/
Recognizability of a product. I think it requires some thought. I for one did recognize SMF when I looked at Penguin. It had the same color scheme, and basically the entire layout is 95% the same. Now, when you look at Wedge, the layout is also similar, but it offers a homepage by default, so it's not exactly SMF either, but it shows signs of being a fork. The color scheme is different precisely because it's not SMF, and it needed to have a scheme of its own. Those who like SMF's better can stick to something closer in spirit, like the Wuthering skin. (You can see it in action by selecting it in the sidebar's drop down menu. It should work for guests, too.)
All in all: SMF 2.1 should be a recognizable SMF iteration because it's SMF, and because people are going to update one day or the other. When someone goes from SMF 2.0 to 2.1, they want to benefit from the latest additions and tweaks and security fixes. When they go from SMF 2.0 (or whatever) to Wedge, that's because they want to try something entirely different. It's not the same state of mind.
For that reason, I think it's best that SMF keeps in line with its earlier themes when it introduces a NDT, and waits until version 3.0 (with the smCore rewrite) if they want to attempt something different -- which I'm sure they'll do, somehow.
Good recognizability examples: phpBB 3 (default theme has the userbox on the right, which is something no other forum had tried before, and it's still usable, even though I prefer it on the left), BBPress, IPB.
Bad recognizability: XenForo, Vanilla, myBB. For the first two, the default themes are just too unconventional and full of odd ideas. So, yes, finding a generic forum you'll immediately recognize XF or Vanilla, but your second reaction will be 'too bad it's that default theme...'
For myBB, the problem is that it's not recognizable enough... When I see a myBB forum, I'll either think it's phpBB or some vBulletin theme.
2/
Content versus containers. Oh, that... It's only one of my many obsessions. When I'm on a forum, as a power user I want to have easy access to any possible actions or information. Especially when I'm administering that forum. When I'm just a regular user, it's not so important. But if you start thinking like a noob user, or just someone that gets to your forum through a google reference and only cares to read a post and then go on with their life (which is absolutely fine with me!), they don't want all the clutter.
Think of the last time you followed a news link and found a news website that had tons of 'Follow us on twitter/facebook/iwetmypants/google+' buttons, multiple ads everywhere (header, sidebar, content side, footer, popups...), things like that. Websites so crowded that you have to search carefully for the comments area. They don't encourage discussions, they just treat you like you're stupid and all you want to do is share their URL to all your friends.
So, that's one of the things that bothers me: there should be enough side information to allow for both noobs and power users to be happy, but still make the content king.
I fixed this in my own way, with several tweaks:
- having a mandatory sidebar. It can actually be removed by a skin, but these skins should provide for a fallback or expect the sidebar content to be hidden entirely. Having a sidebar by default means that modders will have a place to target when they want to insert content, instead of having to add it where they know it can be pushed -- i.e. where the main content is. I know it can be disputed (emanuele doesn't share my views on this IIRC), but it's still a decision I made that I'm happy about, two years later. Most notably the fact that if your screen isn't wide enough, Wedge won't attempt to cram the sidebar into its usual place, and will instead move it to the bottom, where it's unlikely to hinder your reading habits.
- user mini-menus. I like to think that my implementation is what inspired the same thing in Penguin
After all, it's not a very original concept, what's original in mine is that the menus are generated by JavaScript so as not to take any space in the HTML, and thus make it faster to load. Anyway, user menus definitely help with the userbox clutter. I still haven't gotten to removing the visible icons, which kind of defeats the menu's original goal, but that's only because I want to be able to see if a user has a website icon or not. I'll probably move that one next to the username because it's important to me, more important than a user's PM or mail links. You may also notice that these icons, even if they're still there, are now dimmed (reduced opacity) so as not to take attention away from the content. It's a good compromise and I'd recommend that if icons are going to stay, they're dimmed the same way in SMF. It's only 3 lines of CSS, really... (including the hover code.)
- more padding! It's not up to Bloc's tests in this area, but it's still nice to have some breathing space I think... If you have to take away padding from an element for it to it, it's time to start thinking about the reason why it needs to be so cranked.
- obviously, no facebook/twitter/etc clutter in the default theme Weaving... Not only does it save space, it allows make pages more responsive.
- and generally moving all JS to the bottom (at least when a user has no session opened yet, meaning they don't have any JS in their browser cache), things like that make the page more responsive and thus you can start reading the content earlier.
These are things that I'd encourage SMF to try out as well...
Good clutter-free forum: FluxBB (although it takes it a bit too far... Given that even avatars are a plugin, that's a bit far-fetched in my opinion...)
Bad forum (i.e. not clutter-free): vBulletin. It's nice to offer plenty of features... But it's a very bad idea to enable them all by default. I know, what they're after is your money, but to me it's always been a no-no because the focus seems to be on forum features rather than the post content...
So, all in all, I'd take FluxBB as a good example of both recognizability and clutter-free content, with BBPress a close second (it does have interesting things going for it, but overall it's too WordPress-centric for my taste). Had FluxBB had a nicer codebase (i.e. clean and flexible code), I might have worked on it rather than on an SMF fork... I really did evaluate it. It just wasn't enough for me. But OTOH, SMF was so slow in comparison, which is why I focused so much of my work on making Wedge faster than anything... I'm not sure it'll be noticeable for most users, probably won't, but it's still going to be something I'm very happy with. Plus, the work I invested in Wess (the CSS preprocessor) is probably going to be distributed with an OS license in the future, so that it can be reused in other projects.
3/
Going for originality in the default theme. Ah, this one is tricky... One could argue that Weaving is 'original' in many aspects, but I really did my best to make it look 'dull' to the untrained eye. Not exactly dull, but at least 'normal'. The idea, which is valid for SMF and every other web software product really, is that recognizability should only be a plus, but it's more important that when someone installs a forum and starts using it without applying any tweaks, it shouldn't look like everyone else's forum. It's something I never really fixed in Wedge, so I'm definitely not going to boast about it... Half of the work should be done by admins, i.e. installing a logo/header is the first thing they should do. The other half should be made by developers, e.g. by ensuring that it's actually very easy to change the default look. I think that even just changing the background or windowbg colors would be enough to guarantee a 'unique' effect, as long as the default theme isn't too unique itself in its design. In Wedge, you can easily select one of the other available skins, which are quite different from the root one, but I'd like to make it easier to just change the color, so I'll probably also ship a 'test' skin that just does that -- changing some colors -- and will have the very bare minimum needed to do that task. If people look into that skin and see it only has a few lines, they'll probably want to play with it and change the colors, and then make it the default skin for their forum.
I think that SMF's 'variants' aren't used much on SMF forums because (1) they aren't advertised properly, (2) SMF never really encouraged to use them (for instance by shipping a sample variant), (3) well, they're probably a bit complicated to grasp for the untrained eye... Even though it's not exactly true.
I think I'll stop here for now... I've given you enough food for thought
