? quizzically

Sometimes I have ideas that are brilliant. However, most times that I have these ideas I soon find out that I have just invented something that already exists, but not today. And to be fair as of yet I have still not seen this idea really delivered well. I remember a trip back to my home town and my brother taking me out to a pub that had these speak and spell looking machines that you would get from the bar and then you’d use your machine to answer quiz questions that came up on the TV. It was a fun night even though we were too drunk to win anything.

Coming away from the night I thought hey that’s some I could do, and I could do it much better than the system I had played.

Alternatively I have a friend name Brian who has managed pubs his entire career and has since told me that it was entirely his idea. So whether I stole it from friend Brian or from a random pub back home quizzically was born.

quizzically in its simplest form was essentially a system to supplement traditional pub quizzes. There was a portal/dashboard that was run by the quiz master who would be able to see activity within the room and would give them questions to read out for the participants. Once read the question could be released to the room through mobile apps. The participants would then have the ability to answer the questions via their app with a points system based on the speed of answering the question.

While this was the core of the system we saw a lot of potential to go further with the idea. We saw the ability to easily pull together and score teams based on average scores. So essentially no more having your neighbour score your answers. We saw ways of interacting across teams (so for example if you wanted to “dunk” on your rivals you could through fun animations and nerdy trash talking. Ultimately we saw a time when the quiz could actually respond to the ability in the room offering harder or easier questions as you go through the quiz based on speed of answering and correctness of the room’s responses.

The system also allowed for the game to be played across multiple locations, so allowing say rival pubs to play against each other and encouraging rivalries and competitions. I still think it’d be a pretty awesome fun game to play. So if anybody wants to build this system and get some awesome pub quizzes going I’m game!

But we also went further, and envisioned systems that would automagically generate questions. Instead of relying on the quiz master to write their own questions all of the time we spent a short disastrous time partnered with Napier University in Edinburgh developing a system for automatically generating quiz questions from internet sources. In the end the best academic idea that came through was a system of generating question blanks and then cycling through a range of forms of that question.

So you could imagine a question blank such as, “What is the capital of ______?”

Then the blank is filled in with countries both current and historical and matched with the correct capital using a basic information website such as Wikipedia.

While I’m not going into minute detail, the ultimate take away is I love the idea of small businesses and teams partnering with educational institutions to solve problems, there is a substantial power difference between the two types of organisations. But perhaps more notably there is a huge difference in the perception of time between say a university and almost every other thing on the planet with the exception of particularly lethargic sloths.

I still think there are legs to this idea. I really do like pub quizzes, but even when we were actively looking at this system between 2012 and 2014 there was a sharp decline in pubs as well as pub quizzes.  I seemed to recall that at the time we had roughly estimated that 20,000 pub quizzes took place each year in Scotland. As pubs were shutting so too were the number of quizzes declining, but in the mistake made by every young entrepreneur we told ourselves that if we could only persuade a quarter of those to use our system charging something like £30 a shot then we’d be millionaires in a handful of years.

Public Service Announcement – Do not use a top down market share sales forecast methodology.

While the sentiment that getting a percentage of a market share is technically true is acts almost like a logical fallacy. There are so many factors involved in shifting a product that picking market share doesn’t describe the key issues to delivery of actual products. What ends up happening is that you are massively over predicting the amount of potential sales to be made without considering the barriers to those sales. For example, while we didn’t feel we had any directly comparable competitors we at least understood that the £30 we were attempting to extract was coming from the pub entertainment budget which they were already spending on quiz masters, jukeboxes, bands, and other entertainment.

I built the system in several phases and went through a gamut of technologies. Advice I can offer to anyone going forward with any project is to not use Cold Fusion. As much as the people who use it tell you that it is the prototyping language of NASA and all, nobody on the planet uses it. It is clunky, slow, difficult to deploy and there is virtually no help online except from a guy named Steve living in his parent’s basement in Copenhagen.  Although if you fancy a quick introduction into the politics behind an open source version of ColdFusion then this is a fun read: http://wwvv.codersrevolution.com/blog/railo-and-lucee-hunka-hunka-burning-questions

That said, my first iterations weren’t trying to be too ambitious. I wanted to develop a central web app that could be accessed by a quiz master and he could then design a pub quiz using her own questions or using those offered up by the system. These allowed the quiz master then to read the question out and then release it to the room for answering. It would then be sent out to devices that were polling the server to check if a question was ready for them. Knowing what I know now, this is almost the central use case for web sockets. Nevertheless, polling is how I solved the challenge and it worked.

The system was born and it was successful. We had a few very successful live sessions with friends seeing if we could break the number of users on a pub wifi. At the time wifi wasn’t as prevalent or as robust as we are a bit more used to today and there were pubs that had wifi that would topple over with more than 25 devices. They were essentially just getting home use wifi routers and hoping nobody would notice. I once even (in the name of science) did a pub crawl throughout Dundee testing the strength and ability of wifi hubs in pubs. It would have been incredibly rude not to order a drink in each pub. I was fairly drunk after those tests.

With a working system and a list of the pubs that were using robust enough wifi in hand we were ready to take on the world. This is where we even started a real company with equity shares and all that. I was on the precipice of greatness and wealth. And also I was just about to learn that business is 99% hawking your wares to customers and I being pretty much autistic can not sell an umbrella to a man in the rain.

I still think this system is better than what is on offer. Even the few systems that do attempt to supplement pub quizzes seem to have substantial short comings that are detrimental to an actual live pub quiz.

Sensing I was crap at sales I attempted to get others involved in the company to sell resigning myself to more development. Like I mentioned we had grand ideas of the quiz responding to the ability of the room, or perhaps even responding to individual ability. We still had automated question generation to build and we wanted to build a system for nerdy trash talking.

Unfortunately no one who got involved in quizzically had the ability to sell this to anyone. And we became the statistic you most associate with start ups. Closing down although at least we didn’t grumble about outside factors we knew it was our inability to sell that stood in our way.

How many sheep did Noah take onto his boat (according to the bible)?

The bible actually said "Take with you seven of every kind of clean animal, a male and its mate, and two of every kind of unclean animal, a male and its mate…"

There was a grand moment where one of the crew had a connection with someone who provided gaming and entertainment machines to military bases up and down the country. And this fella had asked us to come and present our idea in a military base outside of Newcastle. We gave all the require specs such as requiring internet and all of that and got excited. We took some time off form our respective jobs and headed down to surely present our little hearts out. However, as you can most likely guess we got there and the venue, despite being told repeatedly that they needed to have internet, did not have internet and our presentation kind of depended on that. We had an alternative presentation, but without being able to demonstrate the system we pretty much watched our idea die right before our eyes. Best part of the trip I believe was a nice cheese toastie at a garden centre on the way home.

You can tell by our marketing spiel that we were bad at this. Maybe one day I’ll get it, but I certainly didn’t then:

In traditional pub quiz terms, you know the same team wins most weeks. You know some teams 'cheat' by using the internet. You suspect you're quicker and better and have a more rounded knowledge, and you just can't prove it. Perhaps you're tired of sitting in one place and not interacting with the other teams. Or you fancy your trivia with a bit of luck involved. Hoping that the team with the fella that remembers obscure facts like a banana is technically a fruit and a herb doesn't win yet again this week. Perhaps you're tired of lazy quiz masters downloading rote trivia questions off the internet and want something with a little more variety.

In quizzically we have the answer to reinvent, reinvigorate, and rejuvenate the traditional pub quiz!

From the days of pen and paper, technology has moved on. The need to get other teams to 'dubiously' mark other team's answers is past. Some quiz teams use technology to 'cheat'. In essence, what the honest quiz goer needs is an even playing field. They want more interactivity. They want a bit of useful analysis - what topics they're good at, how quick they are, where they rank, have they improved, where have they improved, how intelligent they are and where they and their pub ranks locally and nationally.

quizzically is building the next generation of quiz platforms to deliver on all of these player needs and increase the fun to be had in a trip down to the local for some banter and competition.

And I’ve got all of this code sitting here, I’m sure it’s no longer useful, but if you, dear reader, feel more compelled to give it a go and have fun with it, you can have a blast with it. There is no rhyme or reason to it and definitely not instructions or ongoing support, but maybe a start of an idea…

If you fancy attempting to pick this up and do something with it, please do, I’ve linked to a GitHub repository that has what I could still find on my computer relating to quizzically. If nothing else it’ll be a good laugh: https://github.com/matthew-davis/quizically