Just what is this real-time thing anyway?

Yesterday someone asked me a surprising question: How do you define real-time?

Perhaps more surprising is that I had no idea how to answer it! Uhm, ah, eeeh, hmmm … fuck, what IS real-time?

And it got me thinking. Well, ok, in real life we define real-time as the instant reaction from physical interactions between objects. Throw a ball at the wall and it will bounce back in real-time. Smash a photon into a metal and electrons fly out in real time. Go very very fast and … no wait, time gets very complicated in relativistic physics, let’s not go there.

But in communication real-time is defined a bit differently. When you shout at someone to stop being so loud with the jack hammer, if they respond within a few seconds. That’s real-time. When you tell someone to do the dishes, if they grunt within a minute, that’s real-time. When you call someone in the phone you don’t even notice them up to a few seconds delay (tested this myself, it’s weird). When transmitting video you’d rather drop a frame than have the video seem to lock-up for even a fraction of a second. This is real-time.

And that’s before we even get into the time it takes our brains to process audio signals, transform them into language, parse the language and access any memories needed to form understanding.

In electronics real-time are responses that happen within a few nanoseconds, for the really hardcore amongst hardware people even that is a bit too slow. When booting a desktop computer everything under 10 seconds is bloody real-time.

So what gives? How can we talk about the real-time web, already even about the steps beyond real-time, when we obviously have no clue what real-time actually means?

The answer is that we. Do. Not. Care.

Are you with me?

There is no need to define what real-time means because on the web there is no real-time. There isn’t even any expectation of real-time. The real-time web isn’t about how long updates take to reach us. It’s not about how long we have to stare at the screen to find a response. It’s a very technological concept.

The real-time web is simply how we choose to label interfaces that do not require refreshing a website to sync its data with the server. It means that whenever the server finds out something new, we are notified at the same moment or a fraction of a second away from that moment.

Now granted as far as most of the web goes this is a very big revolution. But IM has had this ability for well over a decade. We just needed a way to push it onto this archane technology called a Web Browser and that other archane technology called a HTTP protocol, not to mention the whole Client-Server set up we’ve got going.

There is nothing wrong with this, don’t get me wrong, I love real-time updated websites and I seriously do not care how real-time the information is. Anything with a resolution smaller than a minute is fine for me, I’m swamped as it is.

But don’t you think we’re beating a dead horse with a stick here? Peer to peer is a much better backbone for doing these sorts of things, HTTP and the web were never meant to do this and as long as we continue tacking on gizmos and doodads we’re just making everything a little bit worse. The web is already a duct tape job, let’s give her a chance and, you know, introduce some proper behind the scenes remarkable innovation for this real-time web thing. I would love to see that.

What do you think the real-time web is?

This entry was written by swizec, posted on January 13, 2010 at 9:33 am, filed under Real-Time web and tagged Business, Data, Hypertext Transfer Protocol, Peer-to-peer, Real-time computing, Technology, Web Browser, Website. Bookmark the permalink.
Follow any comments here with the RSS feed for this post.