Nevertheless, these would not even be in my personal wishlist to deal with the first-hour experience 🙂

From my perspective, I have to take a look at what people actually do in online social networking environments, Web-based or otherwise. And I think I can narrow it down to two major uses: connect to (existing) friends and make new friends; find out things that interest us (note that this is quite different from telling people what is interesting! More on that later) and spend time doing what we enjoy (again, this is different from telling people what is fun to do in SL).

Now of course this isn’t what every single user in SL does. A computer geek will want to know how to develop the latest MMOG on top of SL; a design nerd will want to learn how to glue prims together; a business guru will want to enter the real estate business immediately. But let’s be honest: the vast majority of SL residents are not in any of those groups. Should SL be designed only for the “niche of the niche”? I think not. It should think first and foremost about what most people actually do on online social networking environments, and not on what a niche does. Prokofy Neva, on an excellent article about the difficulties of designing a virtual world properly, claims that this was what completely ruined Metaplace: it was the perfect developer’s environment, Raph Koster had all the techie stuff ready in place, and it was insanely well developed, but… Metaplace is not only about developers. Regular users — the majority of us — want to have fun there, too. Blue Mars, for instance, seems to be way too focused on pretty buildings — to the point that veteran SL residents go there and find nothing to click on (because, well, interactivity comes second place after nice buildings) and get frustrated about the lack of choices in avatars and the exaggerated animations (because, well, avatars and gestures come in the third place…).

So, what do the vast majority of the successful Web-based social networking engines do? The first step is to give you a minimalist set of personalisation tools — this usually means just slapping in your nickname or real name and a picture of yourself. The next step, however, is finding out if any of your friends are already online. These days, since we start having APIs for doing that on all major websites, this is easy. You click a button, all your contacts from AOL Mail, Yahoo Mail, Google Mail, or Hotmail are ported over. You can immediately connect with the ones that have signed in, and spam the remaining ones with invites to connect. Almost all social sites work that way. These days, they also get your friends from MySpace, Facebook, or even Twitter. Sometimes it’s as easy as just having a few fields in your profile to show what other tools you use, and the site will present you all your friends that way.

In Second Life, however, you have absolutely no clue which of your friends are already registered. And there is no “automated” system to send them invites. Sure, you can send them a snapshot-by-email (how many people do actually know that this provides them with a link to join SL?); or you can join the Affiliate Program (there was a different one which was far easier to use and understand; I wonder if it’s gone…). I noticed this only after finding a RL friend in SL who has been in-world for three years, has been quite active for ages, and even has a good reputation as a solutions provider. I had no clue, and neither had he — although we both have been “friends” on several social tools out there 🙂

Now you might claim that this shortcoming is actually SL’s big advantage, because it allows anonymity. Facebook, the typical counter-example, loathes anonymity because it makes a profit from profiling “real people”. Well, it’s all a matter of checks and balances. What about giving people the option to place their contacts on their profiles and be searchable that way? What would happen is that if you’re willing to share your contacts in-world, all your friends would be notified with an IM “Jane Doe, your friend from MySpace, has just joined Second Life. Would you like to add her as a friend (Y/N)?”

Most of the big social sites actually have several layers of visibility and privacy. You could have your contacts totally hidden. You could have them open for your friends on other social networks. You could have links to those sites (e.g. not only a single web link for your profile, but links to each of every other social networking site). And you could switch them on and off at will. And, of course, there should be an easy way to spam all your other friends with invites to SL 🙂 — possibly with a small L$ incentive if they join (a more cleverer mind than mine would have to think how to prevent it from being abused).

So, this would possibly be the first thing a new user should do — not enter a “welcome area”, but deal with interconnecting with friends.

The next thing is personalisation. I would also suggest that people are given more choices of starting avatars. The least that SL should offer — since SL has the most amazing abilities in avatar personalisation, but that is hardly visible on the registration page! — is what Blue Mars does: a very simple interface where you can at least pick a skin colour (BM has far too few choices), eye colour, hair style, and a few different styles of clothes. It could be a very simple tool, but not one that conveys the impression that there are only 12 types of avatars to pick from. An opportunity to write a few lines from your profile (possibly copying them over from your profile from any of the social networking tools you have listed) as well as an image (again, it could simply pull a picture from any of your profiles on other tools) would at least get rid of that “newbie with empty profile” look.

But of course this is not all. Now you know who your friends in SL are. The next step is to find out what they enjoy — the theory is that your friends might share something in common with you: so what they like is very likely what you like as well. Sometimes it’s too simplistic to assume that everybody will be interested in seeing nice buildings, attend live music events, or go to a party — which is pretty much the choices that many helpers give to newbies, in the attempt to encourage them to find something worthwhile. It’s far better to assume they share things in common with their friends.

So the next step would be to gather all the groups that your friends are in, and give you the choice to see each group description (and which of your friends are in that group) and join them if you wish. Again, this could have a privacy setting: right now, you can flag a group as being invisible on your profile, and of course this would mean that group would never appear to a friend as a choice.

But groups are just part of the social environment. The rest are locations. Thus, the last step would be to gather all picks from all your friends and place it in a very easily searchable interface. Allegedly, the new “SL 2.0” browser will allow landmarks and SLURLs to be used far more efficiently, “just as a web browser”, it is claimed. So, well, I can assume that there is some space on the UI for a link showing “your friends’ picks”. And, of course, since this is the age of push media, as soon as a friend joins a group or adds a group, you’d be notified — and could add it to your list as well (again, assuming that the group join/pick add has been made visible).

All this should be obvious to a company like Big Spaceship, who allegedly are consultants to Linden Lab in terms of interface design. I’m sure they’ve registered with Facebook, LinkedIn, or even YouTube. All of these have “timelines” showing what groups/causes/organisations your friends have joined, befriended, or became fans of. SL should work just like that. Perhaps the difference is that on sites like Facebook, spamming all users with as much information is the goal, since that’s what Facebook wants: more profiling data. While on SL, the goal would be to make privacy the priority, and you’d have to explicitly check what you’re allowed others to see.

Let’s see how this fits in with SL Groups. It has an insane amount of different uses — chatting is one of them. One thing I never figured out is why LL doesn’t show a tiny profile picture next to the avatar’s name on the IM chat box. This is something that even Gtalk, very reluctantly, did after a few years. But the whole concept of notices — which get received even if you’re not online, but it’s impossible to place URLs in it — and “group spam” to keep people informed — which you get only if you’re in-world, and you can’t send attachments, but you can send URLs — should be totally redesigned. For instance, why are Events — often hosted by groups! — not part of this system at all? Why do you need to subscribe to specific events to get a notice? Why can’t you automatically subscribe to all events by one performer? (Or on one location?) If you’re a group member, and there are several performers in the same group, as well as several locations where they do their events, why can’t you simply click on a checkbox and get notifications for all of them?

A “timeline” is what you need for the groups — granted, you can click on the Group tab and look at the past notices (if you remember that they are never sorted and you have to click first on the date tab to get them sorted…), but that’s cumbersome. We should have a “timeline” window where you get information from groups — either notices, attachments, or events. This doesn’t need to be the intrusive aqua box popping up on the top right corner… but just another window somewhere. Imagine a timeline that says: “[Group notice]: Jack Sparrow added a new event: Live Jazz tonight at the Hot Salsa Club. Subscribe? [x]”. Add a few search tools, a few filters, and hey presto, you have SLbook or Facelife, or whatever you wish to call it: a perfect social networking tool just created on top of Second Life.

And, of course, a timeline should be… an RSS feed, so you can keep up to date with it. Or it might send subscription notices by email or send them to your Twitter/Facebook account, so you don’t ever miss anything. Make it an open API, and soon people will have lovely mashups showing your personal agenda on your blog, or listing the concerts on a venue, or having a catalogue of the latest goodies given away from your favourite designer, or sending messages from Facebook to appear as group notices, or something even more crazy than that.

In fact, if the “timeline” has the ability to add all kinds of content as attachments, this might look more like Facebook or Google Wave than anything else. That’s all right. These concepts are familiar to whomever logs in to Second Life. Familiarity, specially when it comes to connecting people together, to get them easily jump-started, and keep in touch with what actually really interest them, is what will make people come back to SL and stay here.

While we’re talking about the interface, there is naturally another thing that really requires a major overhaul — the Inventory. I know it’ll be very hard to change this aspect of SL, since it’s very closely tied to the underlying technology, both at the asset servers and on the communication layer. Let’s see what improvements we had in 6 years: you can click and drag on more than one item (although not to drop them anywhere; sometimes this doesn’t work); you can have two windows side-by-side; you can order items by name or date (with folders sorted separately); you can search for part of a word; filters; and the most amazing changes in 2009, the “Recent Items” tab and the “Worn” tab (the latter is only available on Snowglobe so far, as well as on all third-party viewers).

However, the whole concept is still the same. You will have a trillion items with yellow cubes, and have no clue which is which. Hair, shoes, skirts, jewelry, and more and more accessories and clothes become yellow cubes too, since they all become sculpties over time; with meshes, everything will be a yellow cube except perhaps for your shape and skin.

Landmarks and SLURLs will soon merge, so I wonder how they will be kept apart (if at all); on the other hand, separating between Pictures and Textures makes little sense since… you have no idea which is which from the name anyway (a picture someone else took and gave to you often becomes a texture on your own inventory).

So the whole concept of categories requires some deep rethinking. “Clothing” ought to be just any wearable item, and instead of the silly icons we have now, why don’t we get — a tiny icon representing the actual piece of clothing (or attachment?). You might claim that this would put a superfluous burden on the server side, as everybody now has to download pictures for their 20k-sized inventories — and that for at least a million users. Sure, but that happens just once, and then gets locally cached. The impact on the first day will be tremendous, but, over time, even if you get a handful of new items every day, this will be a small download — after all, each item will just have a 16×16 icon or so (although I’d certainly prefer something slightly larger, so that I can actually identify the items!). Well, I can imagine that doing that for 3 trillion items will be a massive investment in bandwidth on Day 1!

But consider that almost all other virtual worlds (which, granted, have way less items…) work that way. IMVU does it on an external web browser, which is clever — you can have far larger images, and they get easily cached since it’s all HTML.

And of course there should be a far wider range of categories to classify items in. The existing classifications come from the underlying communications protocol. But we humans think differently. A vehicle is not a piece of furniture which in turnis not a prim skirt nor a HUD… so we ought to allow the content creators to label their items on a broad category, and, when these are bought/sold, they’d remain under a folder (or a filter?) with that category name. And what categories should be there? Well… why don’t we start with the ones on XstreetSL? 🙂 They might not be perfect, but they’re a good start. Also, an item without a specific icon, could get a “default” icon for each of that category. At least it would prevent newbies from attaching cars to themselves or dropping skirts on the ground in the attempt to drive them…

Then you ought to have a way to easily place your favourite items (or outfits) somewhere. Imagine having the equivalent of a sidebar/toolbar where you simply can drag your favourite styles, like a link. They’ll be there at a glance and be easily accessible — not unlike Apple’s Finder sidebar on Mac OS X.

And, of course, I’ve already suggested what the Search engine should look like. Instead of a “general purpose” workhouse from Google, it ought to be a “marketplace engine” just like eBay’s (or, to a degree, Amazon.com’s). I didn’t even mention the notion of giving the ability to rate on everything — I still think it’s a good idea, if you can prevent it from being abused. Somehow, eBay and Amazon.com are able to get away with it, even though the blogosphere is full of complaints. So this might require some thought. But certainly residents ought to have the ability to flag something they like, and have that appear on their friends’ timeline — just like on Facebook and any other social website which has copied this cool idea of theirs. It’s an important way to keep in touch with what your friends like. And sometimes that is even better than a good search engine…

All this is not us “eye candy” for the veteran user. It’s a way to get new users starting quickly in Second Life, by immediately knowing what to do and with whom — SL’s current nightmare. Once they get engaged with their friends, go to their locations, enjoy what they enjoy, they have been given a huge jump-start into Second Life, and a very good taste of the reasons that make us stay here. And then, with the “timeline” model of keeping you updated with your whole network, you’ll be constantly in touch with what is going on in the virtual world — and that hopefully will encourage people to remain around for far longer than just 15 minutes.

My only concern is that people might then spend more time scrolling through their timelines and organising their cute new inventories than actually enjoying the 3D virtual world… 🙂