« Mixing It Up For a Good Cause | Main | Michigan in October »

Jul 28, 2005

Comments

1.

Nate>What does this say of the current technical vision of virtual worlds?

You say "the" current vision as if there were only one.

I'm sure that with the WWW, there were indeed people who predicted that participation and sharing were going to be the key factors. I'm sure there were other people who predicted other things which have yet to happen beyond that.

Some of the people predicting the future of virtual worlds are undoubtedly right. The only thing is, we don't yet know who those people are.

Richard

2.

I'm very confused about what you're asking here. For starters, I'm not sure what the *technical* vision is, and why it needs to be asked about as discrete from any other kind of related vision.

And in terms of the rest of the post, it seems like you're asking an unanswerable question, due to the fact that we are not yet in the future. Generally speaking, if the question is "do we fully understand the medium and its potential trajectories in the future?" the answer is always going to be "no," pretty much regardless of the medium in question, because we cannot possibly.

Visions of the future are inevitably flawed, usually deeply so, and if for no other reason than that any future event will be affected by any number of other future events that arise from other trajectories we cannot be aware of, whether those events be catastrophic or euphoric in nature.

All that said, personally I think that if the technology (by which I mean the software specifically and the approach to developing that software) being used to create MMOGs does not change dramatically and soon, it's very stupid. We're in a rather silly place right now as far as I'm concerned, in the technical departments, because we are forced to focus on absurd battles like how to create and subsequently fill - manually - giant landscapes with automaton mobs. The cost of developing the games (I am speaking specifically about MMOGs, though it may be relevant to other games as well) is rising, and it's the software's fault. Or, rather, it's our fault for not looking to develop software differently, which would require developing the games differently. And as far as I can tell there's all of one person really making ANY strides in this area - Will Wright, with Spore.

In general, though, I do admit that I believe there is a startlingly (and to me deeply frustrating) limited set of viewpoints from the people creating current games (or at least the people talking about creating current games, some of whom are also creating current games.) I don't see people pushing many boundaries, and I have furthermore found that when posing questions or proposing discussions that I find lead me to a very different set of answers (not to mention lots more questions) and propositions, they keep being answered by others within the same highly limited contexts of what is being done now. I don't feel like many people are going anywhere interesting.

Uncertain whether what I'm saying is the kind of comment you're looking for. I think you may be making a false distinction about the kinds of pertinent visions. In this field I'm not sure the philosophical and the technical can be separated, and I'm likewise not sure that it is meaningful to ask something of one vision without taking the others into consideration as well.

3.

I would *love* to see an "Imagining the Metaverse" forecasting/prediction project in the vein of "Imagining the Internet". (BTW a search for "MUD" on the Imagining database brings up a 1991 "prediction" by Richard Bartle. I use quotes because he raises a cultural issue to be dealt with in the future (which is now now), and not a claim about any future state. For those making the pilgrimage to Accelerating Change this year, the project leader of Imagining the Internet will have an interview room where you can drop some future knowledge onto video for the ages. Next best thing to a drunken tattoo! ;)

I've been brainstorming a Metaverse scenario done similarly to this Flash Googlezon media scenario (great format, great scenario, great narrative, great soundtrack--not to be missed). The historical milestones would include things like (--and you'll see how these would be presented after viewing the piece--):

"In 1978 Richard Bartle and Roy Trubshaw create MUD1, the first multi-user virtual world ... In 1998 Ultima Online attracts 100,000 subscribers, proving that virtual worlds can gain a mass audience ... In 2001 Sony bans the sale of in-game items for real world currency. Under threat of lawsuit, eBay and Yahoo remove EQ listings from their auctions. A group of EQ players threaten countersuit, claiming their rights have been violated ... In 2003 Second Life allows users to retain legal ownership of their creations, paving the way for legally profitable markets and entrepreneurship in virtual worlds ... In 2004 the global secondary market for virtual goods is estimated at $880 million ... In 2005 Google and Microsoft begin a fierce competition to virtualize the entire planet, paving the way for the Geospatial Web ... In 2007... In 2009..." and then you're off to the races.

I think this would be a useful exercise to get people thinking seriously about the future of virtual worlds, from technical and cultural and business sides (groping the elephant, in Nate's words ;). And fun too! What say you?

4.

MUD1, the first multi-user virtual world

Is it just me, or does anyone else wish the mid-1970s D&D inspired multiplayer online games on Plato (Moria, Oubliette, DND, etc.) got more recognition as firsts of their kind? They also had graphics, even 3D wireframe in some cases, whilst most all of their successors were text-only until the 1980s.

Later pioneers don't often seem to get their due either - Shadow of Yserbius, the original Neverwinter Nights (the AOL one, based on the SSI Gold Box D&D software, not the recent game of the same name), even Meridian 59 doesn't get mentioned too often.

As for the future of the medium - I have to say I agree with babylona that there's an extremely limited set of viewpoints amongst those making MMOGs professionally. It seems that most have the tacit assumption that every game must involve killing monsters (or some form of bad guy), getting loot, and going up levels. Personally I suspect that in 50 years, the vast majority of time spent in virtual worlds/games will be in ones where none of those things are even possible - or are restricted to small, speciallized gaming areas within those worlds.

The number one thing most human beings want to do for entertainment is talk to each other. After that, gambling is big - far bigger than movies, videogames, and television put together. Sex is also popular, but society constrains that market by telling people it's not supposed to be for your entertainment - or if it is, you're supposed to hide that from absolutely everybody, even your friends and family!

Killing a bunch of orcs and dragons and becoming wealthy enough to buy a virtual castle is definitely a niche market kind of thing, compared to the basic human forms of entertainment. Or even compared to the widespread popularity of Windows Solitaire. We've just scratched the surface of virtual worlds so far.

-- Dr. Cat

5.

Dr. Cat> Is it just me, or does anyone else wish the mid-1970s D&D inspired multiplayer online games on Plato (Moria, Oubliette, DND, etc.) got more recognition as firsts of their kind? They also had graphics, even 3D wireframe in some cases, whilst most all of their successors were text-only until the 1980s.

Yeah, that's true. There's a kind of canon of MMORPG history with standard cites, and a kind of canon of things that are standardly ignored. Sometimes the reasons for the line between those isn't entirely clear. And I'm complicit -- I always put Plato and M59 in footnotes.

6.

Not only was -oubliette- a pioneering multiplayer online game, but it also pioneered the first virtual property marketplace. I distinctly recall "The Mithril Crew" selling oubliette items for real life currency (the UICC and Wright College people were some of the buyers). The prices were advertised in a notesfile called =mithril, the Crew would advertise "monthly specials," and so forth.

I believe the Crew made ~$2,000 or so in an 18 month period just selling oub items.

7.


>You say "the" current vision as if there were only one.

Yes, this is simplistic. However, I think it fair to say that not all visions are created equal. Whether the gatemaster be the marketplace, academe, futurists, blogosphere, literati - consensus themes arise.


I'm not sure what the *technical* vision is, and why it needs to be asked about as discrete from any other kind of related vision.

It doesn't have to be. And beyond the technical one, there are many other types of visions I feel even less comfortable with. Having said that, I feel that this bias for all its merits permeats this and internet spaces in general. In part I think this has to do with the nature of the business and its participants. In part it is also the IP rocket-fuel of business plans which fuels these fires. But it is also, I speculate, a comforting and misapplied metaphor: we're acruing, building our stairway to heaven... I wonder though if lessons and inventions are much more fickle, less tidy in their progress... because of cultural forces.


I think this would be a useful exercise to get people thinking seriously about the future of virtual worlds, from technical and cultural and business sides

I agree

8.

Prognostication beyond the current technological horizon (currently at 20-30 years) is always perilous. We're good at making straight-line projections -- we assume that In The Future we'll be doing exactly the same kinds of things we do now, except that they'll be easier/cheaper/faster/more fun. And over the near term, those projections can be reasonably accurate because human nature itself doesn't change. We want to keep doing the things we know because learning to do truly new things is hard.

But there are far too many counterexamples visible in hindsight to think it's safe to make straight-line projections beyond, say, 25 years. And yet we still do it, because it's just so damned hard to accurately guess what revolutions (technological or otherwise) will happen that suddenly allow many people to do things that were never before possible.

So it's understandable to think that virtual worlds over the near term (call it the next 10 years or so) will look pretty much like today's virtual worlds, only with better graphics and sound and maybe a few evolutionary enhancements in the capabilities of avatars. Considering just the MMOG branch of virtual worlds, why assume that gamers will suddenly decide en masse that they're bored with mindless NPC-killing and token-collecting and rise up as one to demand more cooperative games, more social games, more exploratory games? Wouldn't it be safer to assume that we just need better versions of what gamers want today, and make those games?

But consider just some of the ways that question could become irrelevant:

* Will any society still be affluent enough to have time to design and play games?
* Could some other channel for gameplaying become more popular than networked games?
* Could a revolution in AI make single-player games sufficient?
* Could security concerns make networked games unacceptable?
* Will a simple/powerful/cheap "build-your-own" world creation and distribution system emerge?

Any of these events could render a straight-line projection meaningless. So could many others -- in fact, there are so many revolutionary things that could happen (especially in such a technology-driven field as virtual worlds) that it's hard to see how anyone who needs to pay back a big development investment could justify planning beyond five years or so. The likelihood that what you're trying to create will be overtaken by unexpected changes is just too great.

Hasn't "ahead of its time" become synonymous with "money-loser"? (If it makes money, well, it wasn't ahead of its time, was it?)

Isn't a lack of innovation really just understandable financial caution?

Or is there some realistic reason to believe that real innovation doesn't have to ride solely on Will Wright's shoulders?

--Bart

9.

The rest of your post notwithstanding, this is an absurd statement:

"Or is there some realistic reason to believe that real innovation doesn't have to ride solely on Will Wright's shoulders?"

Realistic reason? Innovation has driven profit since we chipped sticks into spears. Technological innovation has made more money for more people in the past twenty years than Bill Gates to an order of magnitude. We're not talking just case studies here, were talking entire economics textbooks and full course loads.

Understandable financial caution has rarely made anyone a LOT of money (some money, sure. But the big bucks go to the big risks.) If it looked that way, it was probably creative accounting disguised as understandable financial caution.

The single most basic economic equation is high risk -> high reward.

Not only that, but it is consistently and repeatedly initially costly technological innovation that drives costs DOWN. Edison? Henry Ford, perhaps? Bill Gates? Du Pont? The list of tycoons is nigh-equivalent to the list of innovators.

10.

babylona> "And as far as I can tell there's all of one person really making ANY strides in [developing software/games differently] - Will Wright, with Spore."

Really? What is so different about Spore? (That's not an attempt at Socratic Irony; I really am curious.) I'm always a little hesitant to believe in the revolutionary nature of a game that hasn't shipped. I've bought and played too many "revolutionary" games that ended up being more of the same.

As for MMOG's, my inevitably flawed vision for their future is that focus will increasingly shift from strictly content-driven entertainment toward context-driven entertainment. At some point, the interaction between NPCs and environment and the possibilities for building and shaping the world will provide a rich enough context so that canned hand-to-mouth content from developer to player will no longer be as necessary as it currently is. No doubt I'm quoting from the Simulationist's Manifesto, but that's just where I see things heading.

--Paul "Phinehas" Schwanz

11.

Regarding Spore:

The implication of Spore is the impact of the gameplay model on the MMOG space. If implemented successfully you'll have an integrated platform of Sim Everything Online from Sim Life to Sim City to Cilivization (Sim Earth) to 4X (Sim Galaxy?).

The Spore software/client then becomes the New New web browser for online entertainment where at the Spore "lobby" you can focus on a particular gameplay youlike and successfully share with other players your lifeform.

The procedural method of creating and storing user-created lifeforms is just an elegant way of marketing decades old AI Life computer experiments to the masses. It's going to be a supercharged Neopets.

Thus, Will Wright, IMO, has the right concept, the right resources, and the right support to make a major impact on how online games will be played.

But of course the market may decide to stick with the grind in ways similar to how we stick to our RL rat race. It's familiar :)

Regarding Culture vs Tech:
Tech is tech. It's people that applied tech. Therefore it's people's culture that drives the adoption and usage of a particular tech.

The quoted statement by Kevin Kelly, IMO, just reduces everything to be about communication... like saying Architecture is about communication or like saying Biology is about communication or all quests are fedex.

So does the MMO Grind implies based on my theory of tech application that culturally we like the MMO Grind? Maybe. But, the MMO Grind has so far been profitable :)

On the other hand, will the growing discontent with current models of virtual worlds (broader sense), the current great interest in the addressible world (Geomaps, IPv6, etc) and social software (Teamspeak, Friendster, IM, etc) mean that people will live in RL more connected by technology, but also living on different planes of existence?

Let me elaborate a little on what I mean by living a more connected life yet on different planes of existence. As we are more technologically connected to everyone, we'll have mechanisms to /ignore and filter out 'noise'. There is a possible future where we will live in both highly connected and highly filtered world.

People could be walking on the street permanently attached to their mobile devices talking to someone elsewhere and completely ignore the people right next to them. Perhaps their Wi-Fi earphone will filter out anything except just audio from people on their buddy list.

Anyway enough babbling for me.

Frank

12.

Dr. Cat>Is it just me, or does anyone else wish the mid-1970s D&D inspired multiplayer online games on Plato (Moria, Oubliette, DND, etc.) got more recognition as firsts of their kind?

I do describe them in my book, as one of half a dozen instances where virtual worlds were invented from first principles.

The reason they don't get a lot of publicity is mainly because they didn't have much of an impact on what followed. It's historically interesting that they were seeds that were planted around the same time as MUD1 and Sceptre of Goth, but nothing much grew from them.

Richard

13.

A few comments on Will Wright and Spore since they've been brought up a couple of times. First, IMO Will is a bona fide genius (and also a heck of a nice guy). He also has leverage that almost no one else in the industry has: in short, EA needs him a lot more than he needs EA. It's no surprise that when the rest of the EA universe contracted to Redwood Shores, Will and his small team stayed in offices in Emoryville. What is Larry Probst going to do, fire Will for not wanting to make the cross-bay commute?

And this is how The Sims survived multiple ninja-like attacks to finally be released as a product. Even immediately prior to the game's release, internally few people believed in it -- the buzz was that it would net maybe 300,000 units if it did well. And then shortly after it shipped EA began to realize the pre-orders weren't a fluke and that they had seriously underestimated the market (so much so that comp'ed copies given to employees were taken back so they could be shipped out).

This is also how Will has been able to champion Spore over the past four or so years and get it to where it is today. He has been able to bring in major hits consistently, so EA management gives him time and room to innovate (that others would kill for! :) ) even if they don't know quite what to make of him.

Will Spore's innovations survive the homogenizing pressure-cooker of EA's development process? I don't know, but I can't think of anyone better than Will to oversee the creative side of this (and the rest of his all-star team to execute on it).

To me, what makes Spore unique and potentially a rules-changer is the combination of its procedurally generated worlds and how they have managed to turn "massively single player" into a virtue -- if it all comes to fruition.

14.

But enough of the future for a moment: Spore has a ways to go before its out, and who knows what will happen then?

So, here's my prediction: in the future when we look back and search for dramatic technological changes that no one seemed to see (e.g., the rise of countries like Brazil and Costa Rica as major players in the online space -- something else that's just beginning to happen while most are distracted by China and India), there are two complementary trends that I think we'll point to, and that we can see today.

The first is procedural content generation. Yes, yes, everyone breathlessly talks about Spore. But that's the future. If you want to see this in action today, check out SpeedTree. This is truly innovative software that enables procedural generation of, well, trees. From individual leaves waving translucent in the sun to hills and valleys filled with thousands upon thousands of trees. Not quite as glitzy as procedurally creating space aliens, but SpeedTree points the way: broaden this out to parametric creation of buildings, vehicles, animals, people, dramatic contexts, etc., and you can see where this goes. Say goodbye to painstakingly handcrafted and yet unsatisfyingly homogenous content (well, not just yet, but soon).

The second trend is middleware. I know, people have talked about this one forever too. But consider how 3D rendering engines have changed in the past 5-10 years. This area has gone from black art to near-commodity in an amazingly short amount of time. We're about to see this happen across the technological spectrum, from 3D renderers to billing and CRM systems. Companies like Micro Forte with front-to-back products like BigWorld have begun to make serious inroads into MMOG development. And this trend is just getting started; I think few now realize how much need there is for full solutions (of which renderers are a significant but small part) that offload the risk and expense of infrastructural development from MMOG developers.

The combination of these two trends are to me like gunpowder and spark. Yes, they're both entirely technological. But if by using these more developers are able to try more innovative approaches to virtual world construction, if we're able to reduce our cost, time, and risk, then we will finally be able to focus on new forms of gameplay, community, and attendant cultural changes.

15.

Mike Sellers wrote - The first is procedural content generation. Yes, yes, everyone breathlessly talks about Spore. But that's the future. If you want to see this in action today, check out SpeedTree.

I disagree, somewhat. In my own project I'm using a huge amount of procedurally generated stuff, including procedural buildings and some-what procedural character models. Procedural stuff only goes so far...

The human brain pretty quickly determines what is procedural and what is hand-done. The procedural part of it is then dismissed as "uninmportant", and not really seen any more. A piece of content is only as interesting as the thought/effort/skill that went into it. A huge procedural forest that takes a developer 1 day to produce is roughly as interesting (content-wise) as a single detailed room that takes 1 day of work. (Assuming that time is used efficiently, etc.)

However, using procedural code to create background scenery works, if it's just there to round out the content. A procedural texture for a stock dungeon wall works okay (not great) because the user assumes/knows the exact placement of stones in the wall isn't important. However, a blank wall (with no texture) would stand out.

The second trend is middleware. I know, people have talked about this one forever too. But consider how 3D rendering engines have changed in the past 5-10 years. This area has gone from black art to near-commodity in an amazingly short amount of time. We're about to see this happen across the technological spectrum, from 3D renderers to billing and CRM systems.

Yes, and no. Technologies only go completely middleware when they stop advancing. Graphics engines have a LONG way to go before they stop improving in quality... NPR, ray tracing, soft shadows, radiosity, realistic lighting, fabric simulation, physical modelling of animation for characters instead of cut-and-pasted segments, collision detection good enough that it determines how a character sits on a horse instead of the current hacked model, etc.

Plus, once something becomes middleware, like printer drivers, some new must-have feature pops up. For many years to come, a bleeding edge MMORPG will be technically superior (in the user's eyes) to one created with stock middleware.

16.

It's really way too late for me to be typing, so I'll probably make lots of typos :) - but had to reply briefly at least.

What makes Spore special is pretty much what Richard Bartle just said. Consider:

- Proved that a small group of "ninja" programmers can create revolutionary software.

- Proved that player/environment interaction is both possible and fun.

- Proved that dynamic content that is generated via (and perhaps with) AI is feasible.

IMO, each of those three things (once I get some sleep I'll probably come up with lots more) have the potential to dramatically change the MMOG market specifically and the game market generally.

17.

Keep in mind that Spore hasn't proved anything -- it's a year or two from release by my guesstimate; we'll know what it shows us then. It's easy to get caught up in the thrill of an early demo like this, and I too have high hopes for it... but we don't know how it'll play yet. The experience of seeing an amazing demo and then confronting the product that doesn't live up to the expectations set is an all-too common one.

I don't disagree with your points; I just don't think we can claim that Spore has proved them. Nor do I consider Spore an MMOG in any sense. The experience (if it stays as demoed) will be single-player only; the only interactions with other players will be anonymous via content.

18.

Mike Rozak wrote: Procedural stuff only goes so far...

The human brain pretty quickly determines what is procedural and what is hand-done. The procedural part of it is then dismissed as "uninmportant", and not really seen any more.

Often but not always the case. This is true of many current procedural systems, but I don't believe this will remain true. Moreover, the real current utility of procedural content in current MMOGs (e.g. DAoC) is that it reduces risk and leaves additional resources available to focus on the more hand-crafted parts. I don't have any hard numbers other than their purported budgets, but the difference between what Mythic spent and what Blizzard spent was about an order of magnitude -- how much of this was because WoW has little to no procedural content and thus required an army of artists and level designers? As we go forward, I expect to see those hand-crafted parts shrink and shift in emphasis.

Technologies only go completely middleware when they stop advancing. Graphics engines have a LONG way to go before they stop improving in quality...

I disagree. Look at the Unreal engine, Gamebryo, and other graphics enginse. Or Havok's physics engine. Look at Big World's offering. None of these areas have stopped advancing by any means, and yet they are highly successful as middleware.

The requirement for successful middleware is that an area becomes known, common to many developers, sufficiently difficult (as technology advances), and outside of the developers' focus or expertise. This allows other vendors who do focus on these areas (e.g. graphics or physics) to create better products than could the developers. It's a classic market niche: the developer spends less and reduces their risk by buying the middleware; the vendor is able to focus on a known area and provide a valuable product/service.

In addition, while graphics may have a long way to go, everything is relative. The verisimilitude of dedicated (note: mostly middleware) graphics engines so far outstrips the gameplay, AI, UI, and other infrastructural systems that people are beginning to say, "so what?" EQII has gorgeous graphics -- but so what? That doesn't make it a better game.

For many years to come, a bleeding edge MMORPG will be technically superior (in the user's eyes) to one created with stock middleware.

Again, I disagree. I think in fact that the days of major MMOGs not being made with significant middleware are about over. What benefit is there to the developer to create their own graphics, sound, physics, AI, QA, security, metrics, server balancing, billing, CRM, and many other software systems? Again using DAoC as an example, I suspect the fact that they used middleware for their graphics (NDL) and for some of their zones (SpeedTree) has helped them compete in significant ways. I fully expect this trend to continue and increase, not go away.

19.

> Mike Sellers wrote:
>
> The second trend is middleware. I know, people
> have talked about this one forever too. But consider
> how 3D rendering engines have changed in the past
> 5-10 years. This area has gone from black art to
> near-commodity in an amazingly short amount of time.
> We're about to see this happen across the
> technological spectrum, from 3D renderers to billing
> and CRM systems.
> ...
> I think few now realize how much need there is for full
> solutions (of which renderers are a significant but small
> part) that offload the risk and expense of infrastructural
> development from MMOG developers.

I am pleasantly surprised to see you raise this point. When I raised the same point a few weeks ago, you seemed to disagree.

These are the types of technological advancements that will make it a lot easier for smaller companies with smaller budgets to make excellent MMOs. When every single company does not have to re-invent every aspect of not just their MMO, but the management of the MMO, it will be a lot easier for people to make them.

One of the most interesting results of this, at least to me, will be more niche targeted MMOs with more varied gameplay.


> But if by using these more developers are able to try
> more innovative approaches to virtual world construction,
> if we're able to reduce our cost, time, and risk, then
> we will finally be able to focus on new forms of gameplay,
> community, and attendant cultural changes.

(Emphasis mine)

Amazing. That is exactly my take on the issue, and it is precisely the point I made a few weeks ago in a different thread about future MMO trends and consolidation.

It is nice to see you have changed your stance from:

https://terranova.blogs.com/terra_nova/2005/06/developments_in.html
>> Mike Sellers wrote:
>> Several companies are working on comprehensive toolsets
>> for creating MMOs. Whether any of these are sufficient
>> to create low-cost commercial-grade MMOs remains to be
>> seen. I'd say it's extremely doubtful given the
>> development risks and costs on the one hand and the
>> difficulty of monetization of low-cost MMOGs on the other hand..

I am glad to see you've seen the light. :)

This development will be, in my view, the second "boom" of the MMO genre. People are always asking themselves what is it about MMOs that turn so many people off or simply do not interest them. There are zillions of possible explanations. Sadly, we never see if any of them are right, because big companies just make the exact same MMO over and over. They slightly refine it, add a few new features, hype the heck out of it through their absolute control of PC Gamer, Gamespot.com, and other gaming information outlets, then stamp it done and go gold.

The big companies seem afraid to make anything other than Gahhh, Another Grind Game (GAGG). I actually enjoy grinding on occasion, but as a player I really can't take another grind game. As a developer, I refuse to make one.

I am really looking forward to the time (which I believe will happen in 5-10 years) when we will have hundreds of new MMOs covering tons of genres and tons of styles of gameplay. Sure, none of them individually will compare to the user base of World of Warcraft. Taken as a whole, they will contribute to the long term health and growth of the market.

Finally, this is all good news for the PC, since the PC does MMOs much better than the console (I won't go into why- that is a matter for a different thread).

20.

Earlier, as Michael Hartman quoted, I said that increasing use of middleware would help reduce our cost, time, and risk enabling us to focus on new forms of gameplay, community, and attendant cultural changes.

Earlier I had also said (as Hartman quoted): Several companies are working on comprehensive toolsets for creating MMOs. Whether any of these are sufficient to create low-cost commercial-grade MMOs remains to be seen. I'd say it's extremely doubtful given the development risks and costs on the one hand and the difficulty of monetization of low-cost MMOGs on the other hand...

In these two comments, I'm discussing different things. I believe strongly (and have for several years) that the appropriate application of middleware will reduce cost and risk -- but I'm talking about reducing costs down from tens of millions of dollars to something more reasonable. Maybe as low as a million or so for a viable boutique MMOG. But I'm not talking about the advent of any kind of "U-Create-It" MMOG construction sets.

Middleware technology and content generation have a long way to go before truly low-cost garage projects become viable in the marketplace. Ten years? Maybe. My speculoscope doesn't typically reach out that far. And certainly such tools won't make turn-key creation of MMOGs possible in the near term, as I believe was your earlier contention. But I continue to believe that such technology can help prevent project bloat and shave significant time and dollars off of a serious commercial project.

21.

Mike Sellers wrote: Moreover, the real current utility of procedural content in current MMOGs (e.g. DAoC) is that it reduces risk and leaves additional resources available to focus on the more hand-crafted parts.

True, but the procedural content ends up being filler between the hand-created stuff. I suppose you could start a world out with procedural and over the years add more hand-generated content. Perhaps the question begs: Is the frosting on the cake the hand-generated content surrounding a heap of procedural content, or vice versa?

I do expect smaller developers to emphasize procedural content. Larger ones have the bucks to pay for hand-generated content.

This allows other vendors who do focus on these areas (e.g. graphics or physics) to create better products than could the developers. It's a classic market niche: the developer spends less and reduces their risk by buying the middleware; the vendor is able to focus on a known area and provide a valuable product/service.

Yes, but by the time a technology is exclusively middleware, it's not longer a product differentiator. Something new steps in. In 20 years (or whatever) when completely realistic renderings are commonplace, users will say, "Sure, you have good looking goblins, but so does everyone. How well does their AI work?" Smaller companies still end up being behind the times and unable to compete against the big companies.

Actually, seeing as goblin AI hasn't really changed since Ultima I and Wizardy I (20+ years ago), I'm not holding my breath for anything better 20 years from now.


Again using DAoC as an example, I suspect the fact that they used middleware for their graphics (NDL) and for some of their zones (SpeedTree) has helped them compete in significant ways. I fully expect this trend to continue and increase, not go away.

Middleware is useful for medium and low-end developers, but it is also a limiting factor. Because DAOC uses speedtree, its trees will never look any better than any other MMORPGs trees (because they also use speedtree). I expect the trend to continue, but bleeding edge MMORPGs will only use middleware where they don't think they can make signficant improvements, and/or areas that are not as strategic. (By the way, what happens if EA buys speedtree? Is there a speedtree competitor that non-EA games could use as backup in case EA decides to use its hypothetical tree monopoly as a competitive advantage?)


... I think I'm (being) too cynical... As I write this E-mail I'm building different versions of my text-to-speech voice (procedural voice talent). With a 100 MB voice file, the voice actually sounds like a recording of me. (At 10MB it sounds like a compressed/distorted recording.) The prosody is still lousy though... For all my negative proclamations about procedural content and middleware, I'm using it.

22.

--- start
I tend to agree with Bart’s comments above, that the problem with future casting is, to be trite, the unknown. I can guess about the trajectory of current socio-technical trends but these are so easily set off course in the medium term.

But in the short term I guess I see a bunch of things that I want of VWs and see no good reason why they should not emerge eventually. So I see a slow growth of pervasive virtual spaces, call this the 3D web, the meta-verse whatever. Though I feel that these will be heterogeneous for some long time just like the web is kind of a single space but then again not as my bank site recognises me in a very different way than, say the TerraNova community does.

What puzzles me is the degree of persistence in the network that will occur when we are not active. This is part of what I was getting to in virtual after life thread. Some years ago there was a lot of talk about ‘agents’ that would go off and do our shopping for us etc., they would be virtual nuggets of preference with the capacity to choose and buy. This does not happen much, we don’t seem to like a machine being us even in this small way – a fear of the lack of control I guess. But some people macro. The stakes are lower if your bot is just grinding. Maybe this will start a wider trend, we will see a group of decisions that bot’s can handle – after all I let my Out Off Office notification tell people I’m not around (when sometimes it might be lying – but don’t tell it that), I’d be happy for it to let people know that I love cheese and hate fish.

No this really is not very MMORPGish, but that’s not where I see virtual world going. Well I do, but that seems to be more of the same to be, its building the better dragon, and sure I want better dragons but they aint gonna be changing my life style.

- click --
-- end transmission from the Ren bot ----- Ren is not around right now, these may well be his opinions unless a software patch is due ---- If you’d like to leave a message please do so -- and he really does like cheese ---

23.

> Mike Sellers wrote:
>
> But I'm not talking about the advent of any
> kind of "U-Create-It" MMOG construction sets.

Nor was I. Anything that extreme could actually have a very negative effect on MMOs. As usual, an applicable analogy from MUDs can be drawn. When it got *too easy* to start a mud, get a server, etc. the quality suffered outrageously. This had a severely negative effect on the medium as a whole. If someone was looking for a MUD to play, their odds of landing in a piece of junk "StockMUD" was extremely high. After seeing a few of those (or heck, even one), I imagine many people ran screaming from MUDs forever.


> Mike Sellers wrote:
>
> And certainly such tools won't make turn-key
> creation of MMOGs possible in the near term,
> as I believe was your earlier contention.

That absolutely was not my contention. I specifically stated that I believed smaller _companies_ would be able to make more specialized, niche targeted MMOs when there are tools/middleware that make it more affordable.

In fact, I know of a few small companies in particular that are well poised to seize such an opportunity if they continue to build operating capital and reputation in the meantime.

Basically, I said exactly what you've said here. But when I said it originally, you vehemently disagreed. Perhaps it was just confusion or a misunderstanding. Regardless, I'm quite happy to see that you do indeed agree, since it lends credence to my earlier predicition.

Developers (more design opportunities) and players (more variety in gameplay) will both reap the benefits of this development that we both believe will happen.

I also hope Mike Rozak's pessimism turns out to be incorrect. It is my feeling that middleware of the sort described in this thread tends to narrow the gaps between huge company and medium/small company, without making the smaller game look amateurish. Also, we should not ignore the beneficial impact of middleware that helps in the business side of things. Doing things wrong there can be as ruinous as a poor graphics engine or weak mob AI.

24.

Upon reading my post I wanted to clarify one thing:

Mr. Rozak, I am not implying you are generally pessimistic. I was just referring to the fact that you seem to be pessimistic on this issue. Specifically, that middleware of the sort discussed here will help smaller companies produce successful, "boutique" MMOs.

25.

Mike Rozak wrote: I do expect smaller developers to emphasize procedural content. Larger ones have the bucks to pay for hand-generated content.

Given that we're talking about budgets that have swollen to the $10M-$30M range, the amount of hand-generated content needed for MMOGs is an issue for even the largest developers.

But our main difference seems to be that you see procedural content being useful only for "filler." I think that's an unwarranted and outdated assumption. And even when the content is background or filler (e.g., SWG's terrain, all procedurally generated) this can still represent a substantial savings that allows for less risk and/or greater innovation elsewhere.

by the time a technology is exclusively middleware, it's not longer a product differentiator.

True to some degree -- but then do you really want your billing system, security, server infrastructure, etc., to be your USP? Not having to focus on graphics or physics or AI allows the developer to focus on their game.

OTOH, it's not the case that middleware = moribund. Epic, Havok, and other companies have stayed at the forefront of their respective areas precisely because they've been able to focus on their technology. In effect, using the Unreal engine is a differentiator for their clients, and will be so long as it stays convincingly ahead of others marketing their product based on graphic superiority.

In 20 years (or whatever) when completely realistic renderings are commonplace, users will say, "Sure, you have good looking goblins, but so does everyone. How well does their AI work?"

Heh. I think that time is a lot closer than you do. Here for example is a quote from a recent Popular Science article on the the Cell processor being used in the PS3:

Videogames have never looked more hauntingly realistic, yet many don’t seem to have the artificial-intelligence oomph to act realistic. (You can practically smell the city burning in Half-Life 2, but shouldn’t the guards flinch when you blow the head off one of their squad mates?) Now a powerful new chip will add brains to the games’ beauty...

Mike Rozak continued:Middleware is useful for medium and low-end developers, but it is also a limiting factor.

You'd call Mythic, Firaxis, Electronic Arts, Microsoft, Sony, and similar companies medium or low-end? I agree that middleware tends to be used in areas that individual companies don't see as differentiators, but that's sort of the point: why spend your time and money on something that isn't a major selling point and that someone else does better anyway?

(Good luck with the prosody, btw. I've worked on the recognition side of artificial speech, and became convinced that appropriate prosodic forms wouldn't emerge without a lot of semantics behind them -- sort of a frying pan and fire problem. :) )

26.

I think the 80/20 rule applies in the application of procedural content and middleware. Both get you to 80% where you want to be cheaply and quickly. It's the 20% that probably going to cost 80% of your budget.

But this is where the vision for Spores break the rule with attempting 100% procedural content (the creatures at least) yet providing a large enough possibility space that minimize duplication (two monkeys typing up the same play) and pattern recognition ("I saw that coming").

Just as there are now very cheap homepage creation tools, one day there will be cheap MMOGs creation tools. There are going to be lots of bad homebrew MMOG, but someone will apply the technology and create something profitable and sustainable with a budget of $1m and making 80% gross margins.

Just do some research on Tencent Holdings on Yahoo Finance or other financial websites and look at their history and strategy for the future.

Or just go browse Bill Bishop's blog: https://bbb.typepad.com/billsdue/ for the China perspective.

Frank

27.

Mike Sellers wrote - Given that we're talking about budgets that have swollen to the $10M-$30M range, the amount of hand-generated content needed for MMOGs is an issue for even the largest developers.

You're thinking like a small developer when making this statement. Large companies often use the following thought process... "We want to beat our competitors." "What advantages do we have over our competitors?" "Technology, blah blah blah yackity yack... Lots of money and marketing." "How do we use our piles of cash and marketing to our advantage?" "Find features that users like that are also expensive, and emphasize them as product differentiators. For example: Eye candy."

Hollywood makes special effects movies because they're expensive, and have huge sales. Smaller movie studios (like in Europe) can't front up the cash and avoid special effects.

Which is why I claim that as soon as one technology is middleware ready, it's no longer important to the consumer. The large companies' marketing budgets convince the consumer of the importance of something else. (BTW - I completely agree that billing systems and non-differentiators will be middleware for even large companies.)


But our main difference seems to be that you see procedural content being useful only for "filler." I think that's an unwarranted and outdated assumption.

I would love to use procedural content for more than just filler. After all, I'm much more programmer than content creator. Every time I do a thought experiment of a world created automatically, it ends up being pretty boring. Do you have any examples? (Other than the mythical/hypothetical Spore.) Diablo is one.


Good luck with the prosody, btw. I've worked on the recognition side of artificial speech, and became convinced that appropriate prosodic forms wouldn't emerge without a lot of semantics behind them -- sort of a frying pan and fire problem.

That's exactly the problem. Unforuntately, if I spend a lot of time and incorporate semantics for English, then I limit my TTS to english-only. I don't want to do that. The way I've designed it is so that a semi-intelligent person with a few hundred hours can create TTS for their language. Adding semantics knowledge requires a skilled person, which then breaks my localization scheme.

Which brings up one future aspect of MMORPGs that no one has mentioned... All MMORPGs are in major languages, like English, Japanese, Chinese, Russian, and Korean (not so major). What about a Romanian MMORPG? A Finnish MMORPG?

28.

> Mike Rozak wrote:
>
> Which is why I claim that as soon as one
> technology is middleware ready, it's no longer
> important to the consumer.

How can you say this when things like Havok, Unreal, Quake, etc. are all middleware that are major selling points for consumers.

Also, as Mr. Sellers stated:

> but then do you really want your billing system,
> security, server infrastructure, etc., to be your USP?

These are all things that are invisible to the player. They just need to work well and efficiently. If you can save money and time in these areas, that gives you more time and money to spend on the parts of your MMO that players will more directly experience.

29.

Michael Hartman wrote - How can you say this when things like Havok, Unreal, Quake, etc. are all middleware that are major selling points for consumers.

3D and physics engines are still in their infancy, and still improving in leaps and bounds. They're selling points to customers. Printer drivers used to be selling point to word processor users (20 years ago) but no one mentions them anymore; you don't see word processors advertise "Compatible with over 2000 printers!" anymore. Nor is "Wysiwyg" a big thing; Microsoft Windows wysiwyg code has become middleware.

Plus, the Havok, Unreal, and Quake engines, are mostly used by 2nd tier developers (with the exception of the original developer that created the technology for their own games). In MMORPG-land, EQII, WoW, (and lineage 2?) are all rolling their own 3D engines. EQII even has a very limited physics engine for cloth and hair. 2nd tier MMORPGs like DAOC, Hero's Journey, etc. are using more middleware.


These are all things that are invisible to the player. They just need to work well and efficiently. If you can save money and time in these areas, that gives you more time and money to spend on the parts of your MMO that players will more directly experience.

As I said previously, I completely agree that invisible or low-visiblility things like billing systems are going to be middleware for everyone.

30.

> Mike Rozak wrote:
>
> Plus, the Havok, Unreal, and Quake engines,
> are mostly used by 2nd tier developers (with
> the exception of the original developer that
> created the technology for their own games).

Um.... I must politely say: no. :)

Half Life 2 and Age of Empires III use Havok. Are those 2nd tier games or 2nd tier developers? It doesn't get much bigger than those games.

Other games that use Havok: Max Payne 2, Ghost Recon 2, Medal of Honor: Pacific Assault, Halo 2, the list goes on of major, 1st tier developers and games that use Havok.

Then there's Unreal. Pariah, Star Wars: Republic Commando, Tribes: Vengeance, Splinter Cell and Splinter Cell: Pandora Tommorrow, Lineage II, America's Army, Rainbow Six: Raven Shield, and more.

The truth is actually the direct opposite of what you put forth. Some of the most enormous games by the biggest developers use licensed graphics engines, physics engines, etc.

Furthermore, what about all the games that use middleware like MAYA?

> Mike Rozak wrote:
>
> In MMORPG-land, EQII, WoW, (and lineage 2?)

No. Lineage 2 uses the Unreal engine. You're making this easy on me, Mr. Rozak. :)

> are all rolling their own 3D engines. EQII even
> has a very limited physics engine for cloth and
> hair. 2nd tier MMORPGs like DAOC, Hero's Journey,
> etc. are using more middleware.

DAoC was definitely not a second tier MMORPG. It was close to 400,000 users at one point. In the pre-WoW days, it was huge. Also, the graphics engine used by DAoC (NDL) was also used by Morrowind and Sid Meier's Pirates- two huge games big major developers.

31.

Mike Rozak wrote: 3D and physics engines are still in their infancy, and still improving in leaps and bounds. They're selling points to customers.

That's sort of true. 3D engines are no longer in their infancy -- they were 5 years ago maybe; now they're a much more established technology. And yet virtually all of the big game developers are using middleware to get the job done. Which refutes your earlier contention that middleware was useful only in areas that users don't care about.

Plus, the Havok, Unreal, and Quake engines, are mostly used by 2nd tier developers

As Michael Hartman says, that's simply not the case. Unreal, Havok, NDL, and other middleware engines have been used by EA, Microsoft, Sony, Mythic (I can't imagine on what basis you call them second-tier!), Firaxis, Activision, etc. Middleware engines may also be used by smaller/second-tier developers, but the days when big game companies rolled all their own technology are long gone.

Blizzard is the primary exception to this, but then Blizzard is an exception in a lot of ways. They're definitely not in the mainstream for technology uptake.


32.

Uncle! I admit defeat on the middleware issue.

33.

Soon our victory will be complete. ;-)

34.

Resistance is futile! :D

The comments to this entry are closed.