Recommendation System

I'd like to try a little experiment with an 'interest based recommendation system'.

If you have visited this blog more then once, please continue reading. If not, this experiment doesn't apply so well yet (but thanks for visiting!). Before participating, please read through the blog a little more and see if you find the subject matter here interesting. If yes, continue, if no, sorry. ;)

If you have read more then one article I've written and found it interesting and appealing to your own personal interests, please continue reading. If you come back for my sharp wit and good looks -- again, thank you -- but you don't apply.

Final step: if you are already a fan of Battlestar Galactica, you don't apply. You have exquisite taste in television shows, but you can't help with this experiment.

So if you are still reading then my experiment is based on this statement: You will like Battlestar Galactica. If you give it a chance, if you watch even one or two episodes, there will be something about it that appeals to you. Maybe you'll like the characters, the setting, the sci-fi backdrop or the morale quandaries and dilemmas that they explore on a weekly basis, but you will enjoy it. To help convince you to give it a chance, I suggest checking out The Story So Far
or the free online webisodes. If you have the time, this 44 minute primer will provide an excellent summary.

Okay -- so now the experiment: given that you and I share some interests (which we obviously do, or you wouldn't have found at least 2 of my past blog posts interesting) and given that I really like Battlestar Galactica, does my statement that you too will like the show have any weight or not? If you are willing to participate in this little experiment I'd greatly appreciate it if you would post your answer ("yes, I gave it a shot and do in fact like BG" or "nope, sorry, even though we share some interests, this is obviously not one of them").

The point I'm obviously trying to explore is (1) whether the fact that we share some common interests means we'll share others (seems obvious to me) and (2) how many (and which) interests do we have to have in common for me to be able to safely recommend you something new.

Obviously the question of interests and taste is not black and white and is vastly more complex then my 2-point analysis above, but I'd love to get some first hand information about this so would appreciate if you'd humor me. ;)

Finally, if you find this experiment interesting I'd greatly appreciate it if you'd consider forwarding on this blog post to a friend who you think would want to participate. Thanks in advance.

Demolition Tycoon

How is this for an idea for a casual game? Demolition Tycoon.

The 'tycoon' genre seems to be on something of a downturn lately (at least in terms of quantity of titles) but I think there will always be room for a new entry provided it meets the player's core needs: is easy to play, rewards them quickly with accomplishments (makes them feel like a superstar) and has some depth for replayability.

Demolition Tycoon would see the player starting out as a teenager looking to make a few bucks by assisting with small-scale renovation projects. Demo your grandmother's aging shed in the backyard using nothing but rudimentary tools and brawn. Earna few bucks and you can buy some tool upgrades (sledgehammer anyone?) and move on to more challenging projects -- a neighbors' kitchen wall, etc.

The joy of wanton destruction is paired with the entrepreneurial spirit that makes the tycoon genre so popular and could create an interesting dichotomy not often seen in competing products.

As the player advances in the game the choices could start to include personal preference -- when demolishing a skyscraper, do you choose a wrecking ball, or explosives placed on the building's supporting columns? The opportunity for fantasy destruction, too, is endless.

A lot of gamers probably list 'seeing things blow up' as an important source of entertainment, so why not let them feel a little virtuous while doing so?

Peekaboom -- google image labeller v2.0

So a friend just introduced me to peekaboom another 'wisdom of crowds' image labelling game similar to the 'ESP Game' or Google's own image labeller.

The goal of Peekaboom is to help train computers how to autonomously identify items in images -- teach them how to 'see'. Google image labeller perhaps has less noble goals and, as I've said before, could be used to build up relevant google addword tags to apply to all web images, monetizing an element of the web that is massive and yet untapped. In general meta-data on most web objects other then text (movies, images, music, etc) is difficult to do automatically.

Cool app -- worth exploring for a few minutes.

Longtail meets Bricks-and-Mortar Pt. II

In one of my first posts I blogged about an idea to use the digital camera on cellphones to take pictures of a product's barcode and access related information online.

A little while later I came across a service called MobSaverthat is not unlike my idea.

Just today I came across yet another -- Frucall. This time, rather then using any of the phone's data services you simply call a number and read out the car code to a CSM on the other line.

Someone needs to take this the next logical step. There are phones out there (smartphones, for example) that open up the camera API to applications. A very lightweight client could make this process even easier (and cooler).

Sony to MS: "Here, have some more money"

Anyone want to take a stab at trying to calculate how much money Sony gave away to Microsoft today?

At stores across North America people are finding no PS3s left but the best deal in the history of bundles waiting for them on an Xbox360.

The 1% Rule

The 80/20 rule is well known and applies to many areas. I've only just now learned of the 1% rule, though, in an article on Guardian Unlimited.

The point is simple: in the world of user-generated content only 1% create. The other 99% benefit from the creations of the lead users passively.

In a post yesterday I mentioned the recent interview with Phil Harrison discussing Sony's participatory culture ambitions. I get the feeling that there are some very smart people around the world with ambitions to take consumer-generated content into the mass market and turn the 1% rule into the 80/20 -- 20% of the userbase for a product creating 80% of the content.

Toblo. Free, Fun. Digipen's Finest

A friend of mine (Steve) recently graduated from Digipen. While a student there his final group project was a game called Toblo which is an absolute blast.

Massive amounts of destruction, capture the flag gameplay, impressive physics and simply endearing graphics make this a huge hit in my eyes. I only wish the distribution rights stayed in the hands of the students who created it so they could find a way to profit from the excellent game they created. I'd happily pay for a copy.

Steve recently saw his baby featured in the new 'Games For Windows' magazine. A crowning point, I'm sure.

If you like games, you'll enjoy Toblo. Free to download and play, so you have no excuse. Give it a try and spread the word if you like it.

The Hype Machine -- Music Zeitgeist

So today someone introduced me to a cool new tool called The Hype Machine.

The premise is simple -- by constantly pinging all of 'the best' music blogs The Hype Machine creates a list of the music that is being discussed at any given time and makes it all available to listen to. The idea is that you'll see some new track/artist listed on the main page that you like, sample the music (streaming, no downloads) and then choose to buy the song (iTunes, Amazon, etc) thus giving The Hype Machine an affiliate's cut of the sale. Of course they also link back to the blog that listed the track in the first place, thus hopefully generating greater traffic for the originating blog, too.

In the top right corner of the page is a Search box. Typing in an artist brings up all the hits in the database, all of which can be streamed in a nice simple flash player.

If you like pandora I'm willing to bet you'll find something to like about the Hype Machine, too. For people who want to stay abreast of the latest and greated in the music world, I see this as being an invaluable tool.

Sony goes Longtail with PS3

I just read thisarticle on Gamasutra and found it interesting. An interview with Phil Harrison on Sony's PS3 'blog', ThreeSpeech, talks about their ambitions for user-generated content on the PS3.

This point in particular got me thinking that they are likely looking to games like Spore for inspiration as to how to take user-generated content to the mass market.

On the most basic level you’re talking about map makers, but how far could user-created content go with PS3?

“Well, I have to be really careful not to give the game away because we’re keeping this secret, but don’t think about it in terms of maps, think of it in terms of behaviours, environments, physics, rules… all the tools that you could want, but in a very consumer friendly way.”

Frets On Fire -- Update

So, you could read my lengthy post from last night, or you could simply download Frets On Fire.

In terms of indie bands using this to promote their song, I can't say for sure, but I did find this resource that certainly contains some bands I'm not familiar with. Perhaps some of them are indeed by bands with myspace pages trying to push their craft in new ways.

A major difference with this game and the idea I explored yesterday is the quality-vs-quantity trade-off. Songs are distributed with the actual audio data included (so 5+ mB downloads and questionable legality) rather then just the gameplay 'track'. The experience seems to more closely mirror that of GH then what I proposed, but I came across at least one person on the forums bemoaning the fact that there weren't more songs available.



Update: Is this a sign of things to come? -- In line with the points brought up in the comment thread of my previous post, I'm willing to bet this is not the last time we will see Guitar Hero (or similar gameplay) used as a promotional vehicle for important pop-culture music brands.

Guitar Hero and Myspace

If anyone reading this knows me personally, you know my wife is not a fan of videogames. That was, of course, until she played the original Guitar Hero. For several months after I bought it every guest we had over was an opportunity for my wife to convert them to Heroes. She batted 1000 and sold the wonders of Harmonix's masterpiece better then I ever could have. Partially due to her passion for the game, but mostly due to the fact that Guitar Hero is a masterpiece, I happily place this title amongst my all-time favorites. It was a no-brainer then that I would buy Guitar Hero II as soon as it came out and spend all weekend playing it (which I just did).

Most people reading this blog probably know at least a bit about Guitar Hero by nature of our common interests, but if you haven't played you probably assume that the only songs available are the popular licensed tracks that were so instrumental in making it the success that it is. GH2, in fact, ships with some 20+ additional 'bonus tracks' that are indeed still licensed, but from much less well known bands. The facts that these bands are 'indie', though, makes the songs no less fun to play. You aren't really a geek unless you have rocked out to Trogdor The Burninator.

All weekend while playing I couldn't get out of my head something I read on gamespot two weeks ago -- an article on using games to launch the careers of new bands and artists -- Games as the new MTV (some would probably argue that the current trend of featuring new artists in serial-dramas on TV has partially replaced MTV, but I digress).

GH2 is effectively doing this in some small fashion by giving smaller relevant (read: guitar rock) bands an opportunity to showcase their talent with a level of stickyness I don't think exists in any other medium. I _promise_ you that a song will never stick in your head as aggressively as if you've struggled through it several times on Guitar Hero. I fell asleep last night desperately trying to get FTK by Vagiant out of my head (you can hear the song at their myspace page here. warning NSFW lyrics.). If it weren't for GH2 I would never have come across Vagiant. Now I might even join their myspace...

So, of course, this got me to thinking -- wasn't myspace originally created as a way for 'small' bands to attract a web audience? Not everyone is as lucky as FTK and can take their song and turn it into a truly interactive sticky ear-worm 'phenomenon', but why not? Why not allow anyone and everyone (in true long tail fashion) to attach Guitar Hero gameplay on to any song they wish and then share it with their friends (and on their myspace)?

The question is obviously rhetorical. First of all there is no system in place to distribute new songs to guitar hero owners other then forcing them to purchase a new DVD full of tracks (at least in the current form on the PS2, but that will all certainly change if/when GH hits the 360). Secondly, there are certainly very significant technical recording challenges that the fine folks at Harmonix had to face in order to make the songs they feature work as well as they do with the GH experience. Third, and certainly not least of all, the legal issue -- a service like this would invite people to distribute non-licensed music with GH gameplay possibly competing with the fully licensed tracks RedOctane is working on for future iterations.

To address the first point -- lack of proper distribution channels. I would love to see RedOctane simply move to the PC for this service. It would facilitate distribution and probably be a requirement to run whatever tool was used by consumers to create their own GH tracks (quick tangent on the tool element, though: just as Wii developers can soon create the control systems for their games by playing with the wiimote, so too should wannabe guitar heroes be able to lay down their own GH tracks using the guitar hero controller directly. Turn the creation of a custom GH track into a game in itself). Perhaps less of a 'party game' sensation standing around your desktop with a guitar in hand then crowded around your TV, but I don't see that as a showstopper. In the best case the tool would be PC based but distribution would be through a networked console.

The second point is important to maintain the full quality of the experience -- something RedOctane and Harmonix obviously take very seriously. Songs can play at several speeds and are certainly recorded in tracks to separate the vocals, bass, lead guitar, rhythm guitar and the rest of the band. This gives a lot of power to the game to alter the audio experience according to the player's actions. Miss a note and the guitar line is audibly impacted. In practice mode (new to GH2) you can slow the song down to a crawl and filter out all of the distracting band elements. I'm sure none of these features would be possible if someone were not to record the song according to very specific technical constraints. However I don't think that would matter, at least not for the masses. Any MP3 could have a GH 'track' created by a consumer and layered over it -- simply a synchronised playback of the GH notes lined up to the beat of the song. If the player misses a note, the engine wouldn't necessarily be able to silence the guitar track, but it could easily play a 'squelch' note to reinforce the failure.

Thirdly -- license issues. True some commercial bands might take issue and RedOctane (or Harmonix or Activition or MTV or any of the other companies with vested interest in the financial success of GH53) might balk for fear it would reduce future sales. Given that these songs would be much less feature rich then the officially licensed songs, though, they would likely be free to download (buyer beware -- no guarantees on the quality). To avoid the distribution of MP3 issues the actual content being distributed could only be the GH track, too. It would be the consumer's responsibility to get their hands on the appropriate song to synch up with said track.

But despite all of these issues (PC based tool, perhaps even player; weaker total experience when compared to the console songs; responsibility on the part of the consumer to own the song in question) the bands who would most benefit from this would go out of their way to make sure the experience would be the best they could possibly making it (spending hours tweaking the difficulty curve of the GH tracks, I'm sure) because they have the greatest vested interest in the content they are providing being high quality. A truly great GH track could quickly become viral and generate incredible press and buzz.

The bands needing attention would love it. Fans would an incredible long tail database of song content to extend the value of their purchase. The suits would get incredible press for being so forward thinking and embracing participatory culture. In general, I think it would 'rock'.

...And really, isn't that what being a guitar hero is all about?

Strategy vs Implementation

So I just had a catch-up call with David Edery (of Game Tycoon fame). Of the many points we discussed one stuck with me after the call -- the question of whether it is better to sit 'high up' and control the strategy (for a game, product, brand, company, etc) or sit 'down low' and control (or influence) the implementation. That is -- BizDev or Production?

As some of you may be aware, a while back David accepted a position as the Worldwide Games Portfolio Manager for Xbox Live Arcade. In this position he has the responsibility to set the strategy for the entire XBL portfolio. Too many casual games? Not enough retro? Need more sports? Less parlor games? etc.

I, on the other hand, am burried deep in production. My bosses tell me the game I'm going to make, and I do everything I can to make it the best game possible. Generally speaking the strategy of the 'what', 'when' and 'why' is pushed upon me. I create the 'how'.

I am envious of David's position at Microsoft because I like to think I have something to offer in the strategy department. David, if I may be so bold, is probably a little envious of me because he would like to get his hands dirty and create something. The grass is always greener...

It reminds me of something Seamus Blackley said at this year's GDC which, to my ears, was the most profound and poignant thing anyone has ever said at a GDC talk: "You guys are the future, and it’s a beautiful future if you open your mind and actually think about business a bit more." (keep in mind he was talking to a room filled with designers and programmers and artists who, if I understand his point properly, could help out industry tremendously if they thought not just about the art of making games, but also the business of it).

The parallel between the two points, to me, is that the reverse point to that raised by Seamus is also true. Producers, Marketers, BizDev (etc) can all do their part to improve our industry if they try to think a little more creatively sometimes and look at the art of what we do as well as the numbers.

Whether you are in Production dreaming of strategy, or a Portfolio Manager thinking of making cool games I think we can all stand to benefit by thinking like 'the other guy' a little more in our daily jobs.