Lately it seems as though the hardcore gamers of the world are lamenting the difficulty levels of current A-list titles, calling them too easy, or too focused on the mass-market. There’s a very good reason why this is happening, and there really isn’t a reason why it should ruin anyone’s fun.

See, I don’t know about you, but I apparently suck at Shadow of the Colossus. If you’re unfamiliar with the game, Shadow of the Colossus (SotC) involves your avatar adventuring through a desolate realm where larger-than-life “monsters” known as Colossi roam. It’s one of the key titles used in the games-as-art debate, and whether looking at it or playing it, it’s hard to deny; the game is simply gorgeous in its design and elegant in its execution.

Shadow of the Plotline

The “game” portion of the title is often described as being a series of boss-battles. In fact, each Colossi is a puzzle that must be solved, as the key to taking down a Colossi involves mounting and climbing the beast in order to reach it’s weak spot, displayed as an ancient sigil, and then striking it with your sword until the creature falls. How to use tools or the environment to get it to appear, how to avoid its attacks, how to get close to it, how to climb and stay on it as it tries to shake you off — these are all part of that puzzle.

In the game’s opening sequence, our hero brings a woman who is presumed dead to an altar inside a temple within mysterious lands. Through dialog with an unseen force, we find that he seeks an entity called Dormin, which legend says can control the souls of the living and could bring her back to life. The price, Dormin tells him, may be more than what our hero is willing to pay.

And that’s where the narrative ends. Each “level” starts with Dormin giving you a hint on where to find the next Colossi, and ends with the creature falling, and a mysterious black force attacking our hero. He falls as well, but then wakes up in the temple, where Dormin tells him of his next challenge. That’s it. No explicit story is told during the body of the game (i.e., once control is finally handed off to you the player), underscoring the mystery surrounding the game’s circumstances.

120’ between victory and failure

I’ve been stuck at the final, 16th Colossus since I first picked up the game back in 2005. In case you haven’t seen what this Colossus looks like, he is a gigantic tower at land’s end. Approaching him, he’ll instantly fire bolts at you, so there is a long process of learning where cover is and how to sneak up to him through a series of tunnels without being blasted to death. Once at his base, you need to scale up levels of platforms and moving gaps of armor. Past the lowest of the armor, you start a series of attacks, leaps and grabs to move from the small of his back, to a hand, to an arm, all the way up to his head. Any slip, and you plummet back to the ground where you are most likely in his sights again, and the bolts can hit you, killing you. And once you are dead, you start from the beginning, from that initial approach. Yes, the beginning. There are no checkpoints.

I really can’t tell you how many times I fell off his arm. As you can imagine, it was enough to convince me that I had seen enough of the game, and that I really didn’t need to make myself even more frustrated by forcing myself to finish it.

But here’s the problem: when you finish the game, which effectively ends at the defeat of the final Colossus, the game gives you a thirty-minute ending sequence. During this sequence, the entire point of the game — all of the story which is alluded to by your missions and actions — is revealed.

I know, because I caved in and watched a video of it online. It moved me, and then, in a way, it annoyed me.

It annoyed me because it turns out the game did have a point to make. A really good point. A salient, well-planned, well-executed story arc that requires your complicity in the actions of the character. Throwing him against larger, more dangerous challenges time after time (level after level) is in fact development around an agreement between Dormin and himself that we don’t get to understand until the game wraps itself up.

In other words, if the story of SotC was like drinking a glass of lemonade, the intro would be getting the pitcher and the glass, the entirety of game play would be sloooooooowly pouring that lemonade into the glass, and the ending would be actually drinking the damn thing.

Cost of entry = your skills

Because of the game mechanics, the skill level, the patience and control needed, I never got to see any of this until now. In fact, once I did see it, it took me a long time to connect the plot points together because it had been 3 years since I had seen the first half of the story, back when I started the game.

So right there and then, the difficultly of the game had become a barrier to the narrative structure that the pacing of the game was meticulously designed to reinforce. In fact, I could say of all elements of the game, this was the most risky and daring move on the part of the developers. The game was designed to give you an ending that was an experience, one that was a culmination of all the actions that you are forced to go through. E.g., his connection to the woman is never explicitly explained because it is not necessary. What he does for her, by way of the game play you have put him through, is more than enough to tell you was his connection to her is. It’s character development by doing, not telling.

But it’s an experience that must be earned, though the challenge of the Colossi and the literal struggle of toppling them. If it wasn’t challenging for the player, you wouldn’t make the connections you need to make for the story to make an impact. The developers knowingly made the game hard,^[1] even though it came with a risk that some people wouldn’t ever see the culmination of their work.

This is a serious game design challenge for modern developers: When faced with a story that matters, with a point to make for players to hear, what is more important: a challenge of skill that makes it a game, or the assurance of completion that will let it succeed as narrative?

We’re all wearing yesterday’s “Large”

What we’re seeing at ground level, is the redefinition of “Normal,” and the reaction the gaming community is having to what “normal” means. “Normal” was challenging. “Normal” was a solid, palm-sweaty, suspense-filled setting for everyone except the super-skilled video gaming masters who demanded everything kill you with only one touch. Today, “Normal” is turning out to be “just hard enough to keep you on your toes and entertained, but not too hard that you’ll stop playing.”

The community has mostly reacted to this in the same way that many Americans react to the continuing enlargement of “Medium” sized clothing. It’s insulting, they may think, because it implies that we are making extra consideration for the gaming equivalent of an overeating majority that can’t handle the emotional pain of needing to wear larger clothing now.

It’s rare to find a gamer that doesn’t want a good story with their game. The hardcore, however, do not want the presence of story to compromise its design. Most games of skill don’t need story anyway — with them, the story is a reward, an acknowledgment of the player’s skill, and solid closure for his travails. It’s more important that the game was beaten.

Sharing is caring

We’ll continue to see some developers like those behind SotC not compromise their narrative designs to make a game more accessible, to make the act of beating a game a truly valuable achievement. But if games need to keep evolving and developing great story lines with their game play, and if developers must spend tens of millions of dollars making them to “next generation” standards, and if gaming wants to be accepted as a narrative medium with other arts, then they need to be accessible to all. That means letting folks who accept that they aren’t the greatest gamers in the world in on the fun, too.

If game developers and designers keep an eye on difficulty levels, by starting tough and recognizing when players are struggling, offering to adjust with them (God of War implemented a very straight-forward version of this), or simply by providing a wide range of difficulty settings from the start without needing to unlock them, we should make this painful process of redefining what “Normal” means that much easier to get through.

[1] Note that I didn’t tell you the length of time it took me to beat the 14th, or 15th Colossi, nor how long I played the game in total before deciding to stop. I have to save some face here.

Edited 2006-08-06, While adding some sub-heads to help pace the writing, I noticed something wrong. Originally I wrote about gaming wanting to be accepted as a “true medium,” when I meant to say “narrative medium.” That was the point, after all.

Companies are only just understanding what a smartphone means in today’s world. I’m reminded of this as I watch Nokia Search take thirty-eight seconds to bring me to a Yahoo! search field.

Thirty-eight seconds is more than long enough to forget what you were searching for in the first place. About 63% of that wait is looking at a screen that says “finding search providers,” a search whose choice is never stored by the app, and is performed every time “search the internet” is selected.

But this is where the problem becomes apparent: Nokia Search isn’t really a web search application, it’s a mobile device search. The app launches quickly and happily, and starts returning video, emails, bookmarks and contacts (and more, even) results immediately as you start typing in your terms.

It says a lot that Nokia feels searching your emails/contacts/documents on your own phone is more important than searching the web. Like many smartphone makers, there’s been a strong emphasis on real computing on the mobile up until now. It’s a big marketing point for Nokia (e.g., the “it’s what computers have become” campaign they ran for the original N95), and it’s been a cultural force in the company that’s been evident in much of their concept and beta work. Windows Mobile is entrenched in mobile computing thanks to business & corporate IT interests, and Palm has always walked the line between users and productivity, balancing their original PDA goals with the needs of the mobile workforce.

Me, I just wanted a web search field.

That’s what was missing until now: finding the things that ordinary users would want to do on a “smart” phone, and focusing only on that. This is usually a matter of returns: it’s hard to retrofit a business platform like WinMo or the Blackberry into a “lifestyle” product. Few companies have the resources and freedom to start from scratch. And no one was comfortable with releasing a phone that did less. Why strip out features, or for that matter, sell to those people who “clearly” didn’t want those features?

The iPhone has been instrumental in putting the writing on the wall: it’s not about “doing things on your phone” anymore… it’s about “doing things with your phone.” And it’s a difference that is reflected in the physical shape of the device and how the software goes straight to making something happen for a user. Who tried to reach me? What is near me? What do I need to do next? Just another task this mutable slab helps you do, just more information it can pull for you. Editing a .doc file? That’s the past. Or at least, it’s not the focus anymore.

It’s a bit frustrating to see the Sidekick get such limited acknowledgment in this evolution of the smartphone. T-Mobile turned Danger’s device into a teenager’s phone in the States, from the webisodes that sold them, all the way down to the illustration of a purple-haired woman that graced the Phone application icon… alienating most adult users and preventing them from discovering its breakthrough features, like push email and OTA synchronization to a web portal where you could access & edit your info on the desktop. Even an (even more) affordable unlimited data plan. All “breakthrough” features or services of today.

Yet, T-Mobile’s research was probably right, that only the youth market was hungry to use those sorts of features at the time the phone was introduced. Could the Sidekick have been a breakthrough success like the iPhone? It was a solid enough product to, sure. But the iPhone did what was hard for the newcomers to do — giving the older market motivation to care about smartphones, by riding in on the perfect storm of iPod frenzy. As Jobs put it, it was an iPod, a Phone, and an internet device. The internet device part, in many ways, was the trojan horse of the set.

As much as it seems like Apple has taken over the market, it says much more about how much potential in this market has been opened up now that consumers care about these kinds of functions on their phone — or more accurately, these functions away from their computer. What is wonderful about the iPhone’s success is that we’re all starting to benefit from it. Not only by proving the viability of these devices, but by setting these baseline experiences.

We don’t always need to edit ID3 tags on our devices, we don’t need to actually do video editing on the device, and we certainly don’t need to run a web server on it — and not ironically, these are all features found in Nokia’s Nseries phones. The core uses for a general consumer are clear now, and simplifying software doesn’t need the hard sell anymore since it’s now evident in the world. Providing distinct solutions can take the place of having to make all solutions possible.

And you can bet that Google and Nokia are more than happy to hear that, too.

When we talk about “usable” or “intuitive” interfaces, Apple devotees and the web app crowd (myself included) tend to bias toward the first-time user….
— 37signals on the TripLog/1040 iPhone app

So true. Web sites are heavily-skewed towards winning people over on first-impression, which puts a lot of importance on perceived simplicity and attractive graphic/UI design. But it also undervalues real usability vis-a-vis accessibility and task-completion, which the app’s author (a veteran Palm developer) spent a lot of time trying to explain in the comments regarding the look-and-feel of his first iPhone application. In the end, all of his principles were valid, they just weren’t styled well, or assembled Apple-y enough for the masses.

It will definitely be interesting to see how developers moving to the iPhone from other mobile platforms weather the transition going forward.

[Edited 11-07-2007] Not sure if this reviewer is being meta, but one of the app’s first reviews on the App Store itself is “This lacks all of the UI elegance that the iPhone has come to be known for. This belongs on a Palm Pilot.”

It’s funny, but I just figured out why FriendFeed just isn’t clicking for me. A long time ago, I wished for RSS-style email, and what FF is giving me is email-style RSS. Every update is delivered to you as being from a user, but what you’re consuming changes context every other line. For me, it’s very disruptive, and whenever I look at FF I feel overwhelmed. Occasionally you can see connections: first a bookmark, then some photos, then a blog post that embeds those photos and refers to that bookmark… but then I didn’t really need to see those photos/bookmarks did I? No, it’s most valuable when there are no connections, when it is random, and that’s when it is at it’s most cluttered. [Edit: Mashable calls it the “FriendFeed firehose”]

I like that it’s giving me an intimate look at what my friends are doing online, but I just haven’t been able to fit it into my workflow in any productive way, even with the new filters, for the same reasons as the above — at some point, someone mentions it elsewhere. It’s Winer’s “river of news” philosophy, if it’s really important it will be echoed by the same person or someone else later. So the micro-view isn’t really that necessary.

It doesn’t seem like it, but I’m not the target user for FriendFeed because I am a “joiner.” FF is at it’s best when you’re not on those other services, where you can just friend/follow people there — it’s obviously more valuable to be connected in the iLike system to see your musical tastes compared than it is to watch it roll by in RSS-fashion on the FF website. FF is for that type of person who can’t stand joining site after site after site following everyone else — the irony being we now have to convince those types to join FF.

I mostly use FriendFeed as a tool, to promote off-site activity on Facebook, but even Facebook is now copying this feature in their new Mini-Feed additions, so I don’t really need it there. I could also embed FF’s output on my blog, but I already have a page using the Drupal’s RSS aggregator.module doing that exact same thing. And let’s not forget that FF is just another source of comments to look out for — because it’s “you” on the system, it leads into a false sense of a commenter thinking they’re being heard. (There’s been progress in the Wordpress world with a plugin that pulls FF comments and turns them into blog comments, however.)

Sadly, this is becoming a bigger problem going forward since were have so many more status services that can copy those updates to other sites, and so many more pinging systems like Ping.fm, which allow you to mass-update multiple sites without ever actually visiting any one of them and seeing the activity therein.

This leads to a weird universe where I can create content on Twitter from a remote service, and someone else can reply from another remote service, and no one actually connects or sees the actual post served from the Twitter servers. I don’t like it, but I can already see it happening, as the level of actual conversation on Twitter feels much lower than it used to be. Although that may be just a side effect of the site being so unreliable for so long, these services have helped accelerated this, because instead of actually moving to a new service, we have simply started stacking services on top of each other.

P.S. Ironically enough, I just received a Twitter reply about Blip.fm and how they had reinstated RSS feeds after their domain name move. I replied back to the user on Twitter, but seconds later realized that all of this user’s Twitter posts were being generated by Blip’s Twitter integration, and because I had the same username in both services (and they both use the “@” convention), that was why I saw it. We need to fix this kind of redundancy, and fast. Will OpenSocial do it? I sure hope something does.

From the Last.fm forums:

I took a look in the iTunes log and it says exactly why it doesn’t scrobble: I had MobileScrobbler installed. That’s true, when my iPod was jailbroken. However, I restored my iPod so that it’s not jailbroken anymore, but MobileScrobbler doesn’t do his job anymore and now Last.FM won’t either.

I also had a jailbroken iPod which I restored two weeks ago so I could be ready for the new v2.0 software. But since I’m running stock v1.1.4 now, this shouldn’t be happening. But there it is: “Mobile Scrobbler detected on iPod Touch - client scrobbling not needed.”

This highlights a fundamental problem I’ve seen with Audioscrobbler from the very beginning: in their quest to prevent spam, they prevent data from getting into their system in the first place, instead of storing the data and applying spam filtering on what was received.

Imagine if instead of Gmail putting spam in a spam folder it simply rejected the message in the first place — sure, it works great in general, but it sucks for all those false positives, right? As a result this site has never worked for anything but the core use-case: someone who listens to their music only when sitting in front of a computer, where they have a live helper application watching their every play.

Because Last.fm outright rejects data that looks “spammy”, there’s no opportunity to correct it when, in fact, it was system-, computer-, or even user-error. That data is lost for good. I’ve had days of listening lost like this because just one play on a desktop had been “newer” than my iPod sync, and the program thought I was trying to sneak extra listens in.

But this is the wrong way to run the service because not only does it severely punish the user for common mistakes, but it forces them to change their behavior on how they consume music to meet Last.fm’s strict patterns — just so they can’t spam the service and manipulate the overall charts. Not to mention, there’s an extremely easy way to handle that: it’s called sandboxing. Does it look like spam? Don’t count it. But don’t throw it out either. Put it in a space that can be counted for the user, but doesn’t affect someone else’s stats.

So we’re back at the quote at the top of the page. In a rush to keep duplicate entries from getting into the system, it falsely assumes the tracks were already submitted just by looking at the iPod itself and nothing else. Of course, Mobile Scrobbler keeps its own log and preferences on the device. There’s probably some magic it can do to see if Mobile Scrobbler was on, or if the tracks it was about to scrobble were logged as scrobbled by the app, but hell, it’s not even doing a good enough job to see if the app itself is still installed on the device. If Last.fm could sandbox submissions, it could submit those tracks and then see that, hey, those tracks were never actually submitted, so let’s keep them around after all. Instead, I just lost three days of tracks.

I should be used to this. It’s happened before. This is what I wrote in 2006 on this very same issue:

I literally think “Oh my god, did I update the iPod?!” before playing files in iTunes. Three or four days of straight iPod listening would be erased in a heartbeat because of a careless what’s-that-song-again double-click on the computer. This is a horrible thing to feel and I’m so glad to have finally stopped caring enough about my profile to finally get away from it.

And to the use-case point, I immediately got a nasty comment from some freetard saying that I should stop whining that the site makes me do things their way, and that I should just write my own.^[1]

Having been a user of the service for over five years total now, I’m finally ready to toss the entire thing to the curb. This doesn’t need to be the end, however, but it requires Last.fm to stop treating all users like spammers, and actually build some real intelligence into track submissions and chart creation.

[1] Oh yes, as a working IA professional, a former employee of a Webby-winning music portal, and a passionate user of social media, I am more than ready to write an alternative. VCs or startup CEOs, give me a call. I’ll hook you up for sure.

[Edited 2008-08-15] Russ from Last.fm responded to this post, clarifying: “The (possibly misnamed) spam protection wasn’t created to stop people falsifying scrobbles — it was created to prevent buggy plugins from mistakenly submitting tracks twice, which was an annoyingly common occurrence for a lot of users.”