The Wimbledon Championships has just happened or may be happening if you are reading this early. Having worked at the event from 1998-2008, whilst at IBM, it is interesting to reflect on both IoT and Virtual Reality being part of that experience at a time when the web was just starting to gain social acceptance. Everything old is new again and exploring these patterns is something I now get to do as an industry analyst at 451 Research.

This story is relevant for all those inventors out there, as well as for those who may think IoT is something brand new.

Scoring a hit with MQTT:

Anyone in the IoT industry will most likely have come across the MQTT publish/subscribe protocol. This, now open source, standard approach to the delivery of data has been around for quite a long while since 1999. It may seem to have sprung up in the last couple of years though as it is on every IoT platform menu of protocols.

Andy Stanford-Clark from IBM Hursley near Winchester (Andy is now the CTO of IBM UK) co-invented an approach for a small lightweight piece of code to be able to send and receive information selectively across low bandwidth network called Message Queuing Telemetry Transport (MQTT). This was when IoT was called originally called pervasive computing. Andy instrumented his house, mouse traps, Alpacas, Cows, Dinosaurs (models or course) and the Isle of Wight Ferries, to capture the imagination of anyone interested in the potential. Here I should also add that MQTT also makes an appearance in my science fiction novels in the Reconfigure series as a very useful way to engage with an unusual quantum computing service.

But what has this to do with Tennis? Back in the late 90’s web traffic was pretty much dealt with by server architectures, or the cloud as we might refer to it today. Any client request would hit the server, each request being referred to as a hit.

In Feb 1998 IBM’s Nagano Winter Olympics site reached nearly 110 thousand hits a minute. Later in July 1998 IBM’s Wimbledon tennis website saw significant traffic surpassing the Nagano number with 145 thousand hits per minute. A considerable increase which showed the societal growth in web usage.

The total number of hits from 1997-1998 grew by nearly four times. Massive traffic lets IT companies prove they can handle scale, but that rapid growth brought challenges to provide enough of the then costly infrastructure to handle the peak traffic, when most of the time it was much lower. Much of the traffic came from the online scoreboard, both web pages and an application would update in a then traditional web way by asking the server every 20-30 seconds, polling. It was suggested that Andy’s fledgling MQTT project might be able to help the following year as he was close to the innovations at Wimbledon and this was before it was exposed to the world in general.

The publish subscribe model means that the server can send a small score packet out when it is ready, and the clients can listen and choose to update at will. It wasn’t something that was shared widely at the time, as this was not a product, not a standard, and no one knew they needed this sort of thing. Plus, in the IT world of web hits it was not about being clever, it was about absorbing the biggest amount possible. The experiment was a huge success, and it is now great to see MQTT all over the place in the world of IoT, which after all is what it was for in the first place 20 years ago.

But there’s more:

As you will see from above, Wimbledon tended to be (and may well still be) a great place to try new things out as a developer and for me, that meant getting people to engage with virtual worlds and explore the possibilities they offered in human communication.

In 2006 I was focussing on the use of Second Life, a user-generated virtual environment, as a metaverse evangelist. Using video game style graphics and interactions received an emotive reaction from many in the corporate world, usually one of shock or suspicion.

In a coming article I will share that journey and its relevance to IoT today. For now consider that it was not about retreating into VR but instead engaging with others and pulling in real world data for visualization and cooperation. This offers some food for thought for today’s VR startups.