Team Partners Telecommunications

Understanding Consensus

Posted on Jul 27, 2021

Researching last month’s blog and the story of Xerox’s voyage of discovery into the use of their carefully written how-to-sort-problems manuals by their engineers brought a fascinating subject to the surface. It’s one that could potentially have quite a bearing on how larger companies decide policy, especially in a world where employees have increasing rights to decide their own, and therefore their company’s, future, so we thought it may be of interest.

Decision making, or more to the point, the way decisions are made, has been a topic of discussion ever since drawing the short straw became a problem. One only has to look at the UK’s first past the post system of governmental voting to see how unrepresentative what appears on the face of it to be “fair” can be. This is especially so if the outcome is of the “winner takes all” variety (and we’re referring to any first-past-the-post vote here), where all those who didn’t vote become the losers, as opposed to being simply those who didn’t chose the winners. The underlying question here is, how does one reach the greatest level of consensus so that all involved feel they have been honoured by the eventual outcome.

Amongst any group of people there will always be both agreement and disagreement. It’s seldom possible to reach full agreement on a TV programme in your own sitting room, let alone anything more serious, and so here’s the first thing to think about; not agreeing is not the same as disagreeing. You may not agree with watching programme X, but you may also not disagree with it to the point of refusal. It’s an important distinction, because a rough consensus enables a process to proceed, whilst disagreement does not. Lack of disagreement, then, can be more important than agreement. Consensus, on the other hand, and forgive us if you’ve got the point already, doesn’t require that everyone is happy, it just means they’re sufficiently satisfied to allow continuation.

Having said that, in any attempt to arrive at a rough consensus it important to recognise that it requires the active engagement of all those concerned. A rueful acceptance and then withdrawal from the process because one isn’t getting one’s own way, is not consensus. Neither is the outcome of horse trading so that something is given to achieve an aim. There’s also an important distinction to be made when it comes to compromise. An engineering compromise to achieve an overall better result by giving up on one element of performance to achieve another is acceptable. A personal compromise arrived at as an appeasement to others is not. All these things can lead to an appearance of consensus whilst not being genuine.

So, what defines a rough consensus?

Many would say that the answer to this is a simple majority but, as we have seen above, this makes no allowance for a minority view that still holds merit and deserves consideration. A rough consensus is one where those holding the minority view are happy that their point has been considered and they have been heard, even though rejected. If, on this basis, they are prepared to proceed a rough consensus had been reached.

Reading this, you may feel that the whole subject is essentially trivial, but believe us when we tell you that this very topic is at the heart of one of our most valuable commodities, the World Wide Web. What happens to the www. is determined by a body called the Internet Engineering Task Force (IETF), a relatively loose agglomeration of techies from all over the world. Not being members of the IETF, because there are no members as such, they do not vote, they are all “participants”, and as such they work to reach a general consensus, as talked about above. And, when they meet, instead of a show of hands, they often “hum”, using this to gauge consensus, and potentially, a starting point for discussion.  

We’ll tell you more about how humming affects the way the www. works next month!