So, about those predictions . . .

What's the point in making predictions unless they are graded? Clearly it's time to shine the harsh light of history on my fearless 2004 predictions that were posted on January 2nd of last year. Read on for the predictions and the grades:

1) At least one MMORPG will be anointed "the MMORPG for the mainstream" but fail to deliver

So this is an interesting one. Certainly The Matrix Online was the product I was thinking about, but it hasn't launched yet, so that doesn't work. Certainly CoH, WoW or FF XI might fit the bill, but none of these have really crossed over. I'll take this close one.

Score: 1-0

2) A handheld device will allow portable connection to a major MMORPG

Ragnarok Online Mobile launched this year and allows Asian players limited access. Correct!

Score: 2-0

3) There will be a successful online-world game on the Macintosh

While a bunch of online games are available for the Mac, including SL, ATiTD, and Planeshift, it hadn't been a good year for large MMORPGs. Sony announced that they were doing a code freeze on the Mac EQ and things looked glum. Then, lo and behold, WoW on Mac! Give me another one.

Score: 3-0

4) An online-world will come under unflattering media attention on par with the current GTA3/Vice City coverage

MMORPG addiction has been the topic of the year, building on the original "EverCrack" stories to reach a pretty high level. I has expected that the attention would come from other directions, but I'll take it.

Score: 4-0

5) Richard Bartle will flame somebody for bringing up a topic discussed on MUD-Dev in 1997

While Richard actually was preemptively flamed several times, he showed admirable patience throughout 2004 didn't draw from MUD-Dev very often. Either TN's discussions are moving forward or he just got tired of providing links. Wrong.

Score: 4-1

6) A successful MMORPG will be released by a small, independent developer

Of course, "successful" is a poorly defined term here. However, 2004 saw no significant MMORPG launches by small or independent developers -- or at least none that show up Sir Bruce's chart. Wrong again.

Score: 4-2

7) A game designer will reference psychological, behavioral, economic, and game theory inaccurately in the same talk

This was, of course, tongue-in-cheek. In fact, I think that 2004 saw some of the best game related talks that I've heard, especially Ted's keynote at AGC. Happily wrong.

Score: 4-3

8) Real-time voice masking technology will become good enough to mask the gender of the speaker

This one really bugs me as the tech seemed to be moving along quite well last fall but hasn't gotten over the hump. Damn. Wrong and back to .500.

Score: 4-4

9) The 2nd Austin Game Conference and State of Play 2 will be both more contentious and more interesting than their first outings

OK, this was a pretty easy one, although it would have been possible for either or both conferences to jump the shark. Fortunately, neither did. A return to correct answers!

Score: 5-4

10) At least one of these predictions will be laughably incorrect!

Phew, my CYA prediction keeps me over .500! Clearly my predictions need to be more aggressive for next year.

Score: 6-4

Comments

[GSH]

1.

6) A successful MMORPG will be released by a small, independent developer

Wouldn't Cryptic Studios (City of Heroes) qualify for this? They seem reasonably small. Or does their association with NCSoft disqualify them?

[hunter]

2.

Hmm, i might be tougher on #1 and #4.

>> #1 At least one MMORPG will be anointed "the MMORPG for the mainstream" but fail to deliver

I don't believe that any MMORPG in 2004 was anointed for mainstream success, so perhaps a "n/a" score would be more appropriate. TSO and SW:G were the last two to be reasonably debated as heirs to this false throne and neither succeeded in doing so. As a result, mainstream media is rightfully gun-shy about the space and have moved on to the "mobile gaming for the masses" meme.

For all their success within the gaming community, CoH/WoW/FFXI don't pass the sniff test on this one.

>> 4) An online-world will come under unflattering media attention on par with the current GTA3/Vice City coverage

No way. The EverCrack Widow buzz percolated from time to time but that story has been in the ebb and flow for the last few years. You set a very high bar by seeking GTA level controversy and in my opinion, 2004 passed without anything close.

[Darniaq]

3.

> Of course, "successful" is a poorly defined term here. However, 2004 saw no significant MMORPG launches by small or independent developers -- or at least none that show up Sir Bruce's chart.

This is tricky, in my opinion. I would put CoH in this category. While NC Soft is certainly not small, the developer Cryptic Studios fits the bill of a small and independent developer very well.

In effect, I "reward" the independent developer for the work that matters most to the game player. At the same time, I acknowledge it wasn't likely to launch without the ability for NC Soft to recognize a good thing when they see it. But I don't think this is a bad thing. 2005 promises even more relationships between separate publishers and developers.

While CoH is not an innovative title in terms of game play, it does what WoW does: works. I still craved something that worked at *all* in the beginning of 2004. Entering 2005, I don't at all lament the rise of games that are better tested, and more stable at launch, even if they are just offering more of the same wrapped in more frenetic lining.

We're a long way from mainstream acceptance, but part of that is only achieved when the games don't crash on launch day. :)

[Michael Hartman]

4.

Like previous posters, I don't think you got #1 or #4 right. Sorry Cory. :)

> 1) At least one MMORPG will be anointed "the MMORPG
> for the mainstream" but fail to deliver

This didn't happen. For 2004, this would mean either FFXI, WoW, or EQ2 would have needed to fail. None of them did.

> 4) An online-world will come under unflattering
> media attention on par with the current GTA3/Vice
> City coverage

This didn't happen.

> MMORPG addiction has been the topic of the year,
> building on the original "EverCrack" stories to
> reach a pretty high level.

Sorry, but you can't use recycled stories that never fully fade to fulfill this prediction. :)

As hunter said: "You set a very high bar by seeking GTA level controversy and in my opinion, 2004 passed without anything close."

[Cory Ondrejka]

5.

I had originally docked myself on #4 but a quick review of google showed a surprising number of stories. However, on further review I'll agree that it hasn't reached GTA levels. Down to .500.

On the mainstream MMORPG, I wasn't arguing that any of those MMORPGs failed, just that their popularity didn't tip into the mainstream in the way that single players games do (5 million copies of Halo 2, for example). As to whether FFXI, CoH or WoW were annoited as the game to do it? Not sure, although I've certainly heard a lot of discussion along those lines. Not to quibble, but I didn't predict that the mainstream media would do the annoiting ;-).

Yes, since CoH received a bunch of NCSoft funding I didn't count them, which is too bad. However, I completely agree that the reasons CoH is as original as it is directly stems from the fact that they began as a small, independent company.

[Ola Fosheim Grøstad]

6.

Cory> However, I completely agree that the reasons CoH is as original as it is directly stems from the fact that they began as a small, independent company.

So Ultima Online wasn't original?

[hunter]

7.

So, when do we get to pick apart 2005 recommendations? :-)

[Cory Ondrejka]

8.

Ola> So Ultima Online wasn't original?

That's a trick question, right?

[Jim Purbrick]

9.

Phew, my CYA prediction keeps me over .500!

Don't be so hard on yourself, this is still a relatively wild frontier and I'd be happy with a score of 0.1 (possibly because of my academic background!)

Real-time voice masking technology will become good enough to mask the gender of the speaker

Surely voice synthesis is the way to go here if we want to mask identity or create a new one?

OK, everyone sounds a bit robotic, but you get gruff robot or squeaky hobbit robot and you save tons of bandwidth and allow translation, filtering, logging and interoperation with text clients.

I stumbled across an interesting 10 year old book by Nicholas Negroponte while looking for something to read at my parents over Christmas, which has interesting things to say about voice recognition (amongst other things, it even talks about VR and MUDs).

It struck me that digital worlds are particularly good applications for providing the kind of context that makes voice recognition much easier. We have a complete model of the world that the users are likely to be talking about and should be able to use that to provide clues. Even when users aren't talking about the virtual world there are often themed spaces for discussion which could help. A virtual jazz bar is more likely to feature discussions about Miles Davis than some other places for example.

I know this has been discussed before, but I've started thinking about it again. It would definitely make an interesting PhD if it hasn't already...

[Mike Rozak]

10.

Jim Purbrick wrote:
It struck me that digital worlds are particularly good applications for providing the kind of context that makes voice recognition much easier. We have a complete model of the world that the users are likely to be talking about and should be able to use that to provide clues. Even when users aren't talking about the virtual world there are often themed spaces for discussion which could help. A virtual jazz bar is more likely to feature discussions about Miles Davis than some other places for example.

I know this has been discussed before, but I've started thinking about it again. It would definitely make an interesting PhD if it hasn't already...

Some speech recognition comments:
1) Get a command & control system working before attempting arbitrary speech. A command & control system would allow simple commands like "equip sword of shining" or "run away". This should be able to achieve 98%-99% accuracy. (IE: Replace/augment keyboard accelerators.)

2) Arbitrary speech (for voice chat) requires a lot of data collection. To do this well, collect all chat logs for several years. (If you have voice-chat, collect voice-chat recordings too.) I don't know if any of the SR companies make their language-model tools publicly avaialable, but you feed the chat logs through their tools to get a more SR system.

3) Aribtrary speech can be made more accurate by having your client monitor context and predict what words/phrases are likely based on the context. (Such as the jazz bar, or the names of characters in the room.)

Text-to-speech:
1) Raw text without prosody tags sounds robotic, or at least unexciting. (AT&T has some decent research demos.)

2) If the user types an ALL-CAPS word, you could have the synthesized prosody emphasize it... this will improve prosody somewhat.

3) Add transplanted prosody from the speaker, and TTS will sound reasonably decent. However, to do transplanted prosody without forcing the speaker to type in a transcript of what he/she just said, you need arbitrary SR, which won't work so well.

4) You can use transplanted prosody for NPCs though, which saves you from recording professionals and shipping gigabytes worth of audio recordings.

Voice masking technology:
1) If you wait for it to be perfect (or even good), you'll be waiting a long time.

2) At least one of the MMORPGs coming out at the end of 2005 (Tabula Rasa?) has voice-chat listed as a feature. I suspect they'll include voice masking.