Cold Constructs Blog

From the perspective of a game designer, people are stupid. Okay, that can be a perspective anyone can have, but the reason game designers have is that people aren’t buying the games they’re asking for.

Doesn’t make sense at all, does it? As a hardcore gamer (also casual when my Free Time Meter^tm is low), I’m deep into the gaming scene - forums, news feeds, modding, and so on. While in these trenches (actually they’re pretty comfy chairs) it’s obvious that originality is lacking and every gamer knows it. We want more original titles! But hey, don’t pay attention to the sales figures on that last Madden game or MMO expansion - we really don’t like endless franchises (pay no attention to that Final Fantasy XIII).

Yet every time an excellent original title comes out, it falls short of even the most conservative financial expectations. Okami, Rez and Psychonauts are just a few of the many original titles that hit huge critical acclaim but fell flat on store shelves. Why?

It seems there’s a pattern to all this. I’m probably wrong, but I wouldn’t know that until you said something (there’s a comment section for a reason eh). Anyway, it appears that each original yet unsuccessful title had a vague objective (or many diverse objectives) like attack monster in a variety of unfamiliar ways or explore this illogical realm and so on. On the other hand, all the successful titles were very clear - win football games or kill anything that moves or conquer bases by building your own and so on.

So could it be that the reason original games are failing in the market is because we don’t understand what they’re about in the 5 seconds we take to consider it? This isn’t too far fetched, considering the fact that the average American has the attention span of the average goldfish (goldfish are shiny). If we don’t understand something at a glance, well, forget about it. Literally.

But is this really surprising? We’re overloaded with massive amounts of information every hour, so in order to continue functioning well enough to do our jobs (studying or writing memos, whatever) we have to be damn good at filtering out the noise in the signal. There’s already a ton of stuff about the attention economy and whatnot, so this certainly isn’t new. What is new is how we’re coping with it all - we still haven’t quite figured that out yet. We’re still developing the necessary tools to get us back to pre-internet info loads.

So how do we design around this? In games, the way I would do it is to present a very clear objective, but while you’re playing the game, slowly introduce a new and (hopefully) interesting gameplay mechanic, without ever confusing the player about what the goal is. If you try to slap the player with a new concept from moment they look at the cover, forget about it. Alternatively, you could take the Katamari approach and have the only objective be so incredibly simple that it could be translated to the potential player in so few words or pictures.

This could also be applied to anything else really. You just have to make sure your design is a natural extension of what they already know in order to keep people’s attention - grabbing it is easy. Even better though, you could stick to simple designs. (But oh wait you have to have real talent to do that, so it’s an unlikely course of action for most designers, unfortunately.) There’s a lot of hubbub regarding simplicity in everyday designs (software, internet and other types of interfaces, other products or even architecture and so on) and how we should probably scale back the features to the point where the purpose of the object is pronounced so clearly that the user wouldn’t have to think twice about it. Because they won’t even if we want them to.

This is an exploration of effective digital writing methods - i.e. how to get words onto the screen. There are a number of readily-available solutions for a number of digital device designs, but which ones are best? What is the ideal method that we should work towards?

The first one that comes to mind is to simply write directly on the screen. The problem with this is handwriting recognition. Research has a come a long way though and accuracy has generally reached 80-90%. However, while simply writing on the screen is the most natural way, it is noticeably slower than typing, plus the option of writing isn’t always available due to screen real estate.

Typing, on the other hand, isn’t always available either. Small devices such as phone and PDAs don’t have room for keyboards, and when they do, they’re a pain in the ass to use because they’re so damn small and they limit you to just using your thumbs. Despite our love for our opposable phalanges, using them to input thoughts into a computer is, ironically enough, archaic. However, when keyboards are available, it is one of the fastest and most error-free method short of voice input.

While those small devices don’t have the room for keyboards or sometimes even writing, they do have microphones. Speech input has the advantage of being our only real “natural” way to communicate. But as is the case with handwriting, recognition accuracy isn’t yet perfect. (Then again, this can be a really good thing, for example when you’re drunk, you probably won’t be able to make a call if speech is the sole input method, or the same goes if cars were voice-activated. Designing our creations to work with human frailties like this should be Interface Design 101.) Let’s not forget that computers are non-linear systems thus often requiring non-linear input. Since speech is linear, it’s not very effective for primary input. For example, it’s much faster to navigate a visual data tree than it is a phone menu.

Speaking of recognition, there is yet another method for input: drawing shapes. I’ve mentioned this tech before (wasn’t exactly a positive review), but since then the app has improved a bit and it has become commercial. I think the general idea is worth pursuing for a couple reasons: it works on any sized touchscreen (or stylus-enabled screen) and it’s just as fast as typing (with a similar if not easier learning curve). The letter layout was generated by genetic algorithms - using it is much faster since the distance between each key is drastically shortened. Some tests show that it’s faster than a real keyboard if used by an experienced user.

The ideal method is to have computers translate our thoughts directly to the screen, but without that option (yet), it seems the most effective and versatile method is shape-writing or some derivative of it. Despite its obvious downfall of being a terrible game controller it should work well on mobiles, laptops and desktops. Having it work on most platforms allows for consistency of input, making the whole experience easier on our minds (since it’s just one method we need to memorize). It also decreases physical stress that typing usually causes since you’re either moving your finger around the layout or a stylus. However, it’s still slower than voice, which I think has its place in drafting emails and other linear applications. But until recognition gets to 100% accuracy while still being able to distinguish when the user is slurring their words (eg drunk), we should keep voice recognition to a minimum - working in non-critical (or harmless) situations such as email.

Google made cloud computing famous, but it is not a new concept. There were Beowulf clusters before then (and still are), and of course the supercomputers and servers have almost always been comprised of clusters of processors and massive memory banks.

Hardware power is cheap and so is the software needed to string them together into a whole is greater than the sum of its parts system with redundancy and decentralized control mechanisms.

I need a home system that stores all of my work (a personal encyclopedia, images and photographs, movies and music and so on) that never fails me and is accessible from either a terminal across the world or my own handheld device. Being a 3D artist, I would also use it as my personal render farm. It needs to be fast, reliable, scalable, secure and above all it needs to have a natural interface (or better yet, transparent).

This can all be accomplished by buying cheap computers and connecting them together in a LAN that is only accessible by me, myself and I (with the option to share all or some of it with whomever I wish). A distributed filesystem like Google’s cloud FS is readily available (an open version of it, called Hadoop). The problem with it, it seems, is that an idiot can’t use it. (Personal note: add ‘build an interface to Hadoop’ to project list. *Sigh* Anyone wanna help?)

Of course this will be a niche product, but if things are done right (a good system is built around it that integrates seamlessly into people’s lives with a new type of interface)… this is the thing that could dissolve the desktop as we know it and blur the boundary between internet and that damn desktop metaphor.

The ‘killer app’ here would be allowing multiple users to access the home cloud simultaneously. This would eliminate the need for multiple systems all over the house since all you need to access it is any portable device. It also eliminates unnecessary effort and waste (since you don’t have to “upgrade” per se, just keep adding computers to the cloud for more power) as well as the trouble of backing everything up all the time and keeping it all synchronized.

And who wouldn’t like their own supercomputer that they can command with their smartphone? You could hook it up to your entertainment system and use your smartphone as the media browser and remote (which simply commands your computer to run content on the target media system). There are already commercial solutions for this kind of thing, so it’s just a matter of compiling it into a seamless user experience. I plan on building such a system when I get the money–I’ll definitely share my experience with you here when it happens.

Check out more entries in the tech concepts category.

Synthetic paper covered with a thin film of a waterproof polymer that can be dissolved with heat and re-applied (solidified and pasted back on) to another (or the same) sheet of synthetic paper. Ink of some sort is printed on the synthetic paper but it’s kept in place by slapping the polymer film on it. To reuse it, simply slide it into the printer as if loading paper. It will burn it off (not too intense heat however) and wipe off the writing automatically… it’s then ready to be printed on once more.

This eliminates the whole “recycle” process (which is actually cheap degradation into worse products), the shredding of paper, and will save our oxygen supply (trees) which will in turn help with global warming by sucking up more CO2. Not that they’d stop cutting down trees or anything, but hey, we can hope. And threaten their fucking lives (or provide them with better-paying jobs).

This is what will appear before the Auggy.

The concept is largely the same, the only difference is that it utilizes a limited form of AR - information isn’t project directly over your eyesight but rather displayed on a handheld device. In order for this to work, the unit still requires a GPS unit. It doesn’t need to recognize hand gestures for navigation and issuing commands.

What it also needs is wireless connectivity and have mesh network capability. Special software is needed to allow people to tag places with metadata, so that people can look up what others have said, either via websites or personal group of trusted individuals, or ideally both. The software must recognize a GPS coordinate and search the approximate area for similarities. This way one can search for local pizzerias as well as identify a building correctly (the GPS coordinate could be closer to a neighboring building instead of the one of interest).

The best news is this: it’s here. As this post has been brewing, certain developments have occurred. Latest generation smartphones are getting WiFi built-in as well as GPS, and here’s the required software to make the simplified augrecon truly useful. I don’t have that much money, but I’ll be buying one because I already know what kind of technology it will be - disruptive. My favorite. (I’ll be waiting for the one with enough horsepower to run Skype though.)