This is still the header! Main site

Recognizing the Possibility


Back in the early 2000s, my dad had a digital camera he got to take home from work. It existed for the purposes of doing surveys of buildings surrounding construction sites (to tell apart pre-existing issues from ones caused by the construction itself).

It barely had a screen (I think it was designed with usage via a viewfinder on top like traditional cameras); it took compact flash cards and had a somewhat questionable quality, too. But it existed. Sometimes, he would take a couple photos of us, the family.

As it happens, sometime during the summer, we were having a LAN party in the house. He once dropped by and took some pictures. About 14 of them, total.

Which remain the only pictures anyone has ever taken of the numerous LAN parties we had. (Me and my friends took exactly zero of them.)

Fast forward... a number of years. We once had a LAN party, just a single afternoon, gather at one of our places with laptops, play for a while, then go home, not even close to the multi-day endeavors of the glory days. And yet, I counted 35 pictures of just this one event.

Well, obviously, now everyone has phones. They have gotten better, they have much better quality, it's easier to share stuff. So everyone just keeps taking pictures because why not? They are free, they are fun, and they serve as good memories.

But it wasn't only the technology. We had to figure out what to do with this capability first.

If I were to go back in time, despite the lack of refined equipment, there would be a lot more pictures and videos of us playing games during LAN parties. Yes, it's not as convenient, but we would make it work, because we know it is possible.

This seems to be a common pattern. There's a new technology which becomes so popular that it's now obvious for everyone how to use it. But it's not just about how good the technology itself is. If we were to be forcefully downgraded into early 2000s level of internet connectivity, we would still have more online stores than the actual early 2000s.

Which, of course, implies that there are things you could do right now that will be obvious a couple years later, but are already possible today.

Talking to computers comes to mind. It's now entirely possible technologically; in fact, I'm doing it right now, writing this blog post. Being able to do this is, in fact, new. Before Whisper came out, I at some point experimented with some voice recognition models that were publicly available. They were remarkably bad, compared to an average human listening to you.

And yet, most interactions with computers are still based on either typing or clicking on things on screens. Even devices specifically designed for voice interaction mostly serve the purpose of switching lights on and off.

I do expect this to change pretty soon... but not because we needed to wait a bit more for some more advanced level of technology. We just needed some extra time to figure out what to do with it.