In our case we compared the different environments including Java and Shockwave 3D, programming environment and multi-platform (Mac, iPhone, PC) support, also weighed up ease of install and stability (v important), and finally decided on a browser plugin.

Even though there is a disadvantage of being a barrier to entry for many users, the advantages for us at least outweigh the disadvantages, and actually more people than expected are ok with browser plugin which is very lightweight and simple to install.

Looking from the outside – it is on almost 60% of PC’s, mac or windows, provides genuine 3D, a very well recongised consumer brand and is easy to code. When I see how easy it is to use MaidMarion virtual worlds then it seems surprising to me that other companies do not use this platform.

I develop in the corporate and education world where client installs, exe’s and large downloads are a complete no go. Shockwave 3D is probably the best option for this. I guess there must be reasons why it is not used for mass consumer virtual worlds.

Avatar portability is one of the next big things that won’t happen along with the other types of portabilities, e.g. socialgraph for social networks. You could see it working in products from NHN (some casual multiplayer games at ijji.com) and few other Korean portals, but it’s a rare exclusion, that doesn’t make any sense as you can’t have you avatar in both 2D and 3D games, especially when he’s both a warrior and a belly dancer. Being a generic Mii jerk would suck, let’s see what 360 will come out with and what level of in-game adoption we’ll see besides Xbox Live Arcade.

With Meez API contest, we’ll see broad adaption of unified 3D (ok, prerendered, but the term 3D in browser is blurred) avatars in standalone Flash games.

I don’t think we could hope for avatar portability between competitors, with more than 120+ titles f2p MMO titles out there in USA from 30+ developers and publishers, there’s no chance anyone would do that during this iteration… We’ll have to wait for the next major leap to come so everyone will be prepared. What will it be, hmm…

I think some portal clients from the large multi-game operators can still be viable in the future – especially as you mention, the brand is trusted and they don’t “take advantage” of the client by spamming unnecessary ads.

One benefit I can see in this is account (including identity, in-game currency) and user 3D avatar portability between different games, something that for example in Asia, 9You has been talking about for 2 years now..

Large single product downloads will be a lot tougher going forward, especially given how much you can do in browser (instant action is just the beginning) The big games / Worlds like WoW / AoC where the product is solid and people already know what to expect will be ok, but the rest?

On the large client downloads – I’ve recently wanted to trial european World of Warcraft and was really surprised by the streaming technology behind it – just in 10 minutes from getting to the site and creating account, I was already in the game!

Everyone should also look at how Asian portals require installing ActiveX download clients to fix those traditional IE download f*ck-ups. ActiveX is a regional sensation, but that works in any case – when the player sees that his game is downloadaded by some native installer with publishers logo, it adds some calmness and trust I guess. And if that installer pushes the right message (not an ad), it adds a value to the brand.

As you are going to force the player to download few gigs anyway and if you have several games out there, is it worth to have a unified experience?
Take a look at NCSoft client. Some players might be grumpy and get annoyed by the extra installs, if they are overbloated, but don’t forget that opinions exposed on forums (I’m tired of ‘em!) always belong to vocal minority and most of people wouldn’t care to keep their list of programs virgin with Solitaire and Minesweeper.

All the casual portals do it – some do better job, some worse. BFG, iWin, RealArcade, WildTangent with their own clients – big or small, with long annoying autoupdates after install or minimalist functionality. The other way is gamefools.com – minimalist downloader application is great, but including ads at this stage seems like an overkill.

One of the the other aspects of having “portal clients” for companies who are self-publishing their products – is it worth to cross-promote own products? Should a company provide a player an easy access to SO many games, or let a player focus on a single product and get the most out of it? I haven’t seen such kind of researches, but I guess there’s no universal recipe and each case is unique.