One of the most amazing and wonderful things in the last couple of years here at Victors & Spoils has been dynamic tension that’s been created between being an advertising agency and a tech start-up. That tension that’s been created by mashing up the two cultures has been profound for me.
When we started we noticed that even the most tech-based agencies were analogue at their core. We asked ourselves a couple of questions. First, what would happen if we took the functions of a typical ad agency and digitized them? The way we communicate. How we work in a massively distributed way across the globe. Hence, we created Agency Machine. Taking an analogue business process and making it digital is a messy affair, full of trials and errors. All said, it’s working pretty darn well and gives us the ability to scale globally and rapidly.
Next we thought about what we could build for our clients. It’s something that has intrigued me for a while, and was an inspiration for Alex Bogusky and me to write Baked-In. Lot’s of agencies talk about building their own IP but their own structure gets in the way. Forty percent of the Victors & Spoils team is here to build proprietary technology. We feel it’s the only way to go beyond just building campaigns for clients to building platforms that scale across media and can be massively scalable. Yesterday we launched our second tech product, Fan Machine, working with Harley-Davidson as our launch client. Mark-Hans Richer, Dino Bernnachi and their team at Harley-Davidson have been great partners in exploring ways to shift the paradigm of how advertising is created. We want to make the experience better for creatives, customers and brands by cutting out fat and facilitating faster, deeper conversations between brands, consumers and creatives.
Agency Machine and Fan Machine are just the start of what we’re building. It’s not just about creating great work, it’s also about building the digital tools to change the way campaigns are born and executed.
First I want to say I appreciate the thsneughroos of your comment and the way in which you delivered it. I find discussions of electoral reform can get pretty heated and sometimes get away from the actual topic at hand.Your majoritarian criterion is important to many but not the only criterion that is important. On this criterion, Approval does not fail but certainly does not guarantee a candidate supported by the majority. Though, other than in runoff voting systems, it is the system that is most likely to produce a candidate that is supported by the majority. For me personally, majoritarian concerns bother me very little since I believe it just makes logical sense that whoever is supported most, wins. We often make the argument that the Conservatives were voted against by a majority. But every other party was voted against by an even greater majority. While majorities are nice, they are better when they are legitimate majorities and not ones cobbled together by runoffs.Approval voting does not mean necessarily that a candidate who fails to win a majority is elected. There's no reason why a threshold requirement can't be implemented and when not met, perhaps no one is elected. This seems fairer than pretending a majority of the population supports a candidate when they don't. Those specifics are to be argued in a different discussion, however.As for the reverse issue of a candidate being able to receive 50%+1 and still not being elected because another candidate hits that threshold and exceeds it further than the first candidate, I again have no personal issue with this. In that case the most supported candidate is still elected and to me that's what representative democracy is all about.Approval is in no way a preferential system as preferential systems require rankings of some sort. Approval is a ratings system, as is its more complicated variant, range-voting. Preferential systems by definition require the voter to differentiate between all candidates. To me, your suggestion that Alternative Vote doesn't require you to rank all candidates suggest that in a model of that sort, Alternative voting is actually more of a hybrid of a preferential system and an approval system because it requires rankings where a mark is written but allows for unfilled boxes unlike all other rank-based systems. In this case, Alternative Vote is actually taking some aspects of Approval into its process. The only difference between this and range voting is that in range voting, it would be okay to give two different candidates the same rating (say 5 out of a system between 1 and 10).I don't think your argument about Approval allowing the least supported candidate to win holds any water. by definition of the system, voters can select any candidates they wish and as many as they wish. Support is strangled under preferential systems that require a choice between two options instead of a binary choice between supporting or not supporting a specific option, irregardless of support or non-support of any other candidate. It has been mathematically worked out through data by American mathematicians that Approval does in fact do better than preferential systems in this regard.As for Arrow's Impossibility Theorem, that theorem states that when there are more than two candidates in an election, vote splits will occur that will distort the overall intention of voters. Vote splits are caused by preferential balloting and any balloting containing a rank will split votes. Approval voting is not a preferential system, does not force the splitting of votes, and therefore very easily addresses Arrow's Theorem.Yes, Approval by design does not take into consideration preference beyond support or not. I am personally okay with that and so are many voters who do not have party memberships nor care about parties the same way we do. Approval is not bad, wrong, unfair or any other descriptor better applied to people than voting systems because of this. It is a matter of personal taste whether that is a concern to you or not. For you it is, for me it isn't. Electoral system design is precisely about this issue. The system, once the math and theory are covered, must ultimately be suitable to its environment and the public who will use it. I don't think we can throw out approval because you personally have preferences that run counter to how it functions.It simply isn't correct to say Alternative Voting guarantees the intent of a majority of voters is represented in the final result. When you whittle down the race to two candidates in a runoff you create a false majority instead of a split-vote plurality. The majoritarian candidate doesn't have the real support of a majority of the voters, they've just been forced to go down the list until they hit rock bottom and the results are derived from that. Hardly ideal and hardly the original aim of those in favour of the system. If you aren't required to rank every candidate that changes it a bit, but also adds in more confusion for the voter and as system design theory has worked out, the most important characteristic of a voting system is that the electorate using it can actually understand the process it employs from start to finish. Runoff systems work a lot like that cup game where the ball is hidden under the cup and the cups are moved around until finally you have to guess which cup your ball is under. Voting should and needs to be more straightforward. The reversal of the democratic deficit by way of low turnout won't be happen if the way we vote isn't explainable in just mere seconds. And by the way – to see someone mention Arrow’s Impossibility Theorem here warms might heart. It’s nice to see such intellectually stimulating and thoughtful debate. I agree about how nice it is to have thoughtful debate on the subject. I've been researching electoral system design theory and mechanics for the last four years in my spare time and am pretty passionate about it. I'm glad to have gotten to engage you and others on options that exist being PR and preferential voting since it is usually assumed those are the only alternatives.Ultimately the things you find uncomfortable in Approval Voting can be answered through its more specific offspring range-voting (with the exception of majoritarian results). In the end, I think the best option would be to do as New Zealand is doing in a few days and provide several options to the electorate.Thanks for the discussion!
Posted by: Boris | May 03, 2012 at 06:41 AM