[r-t] Extension

[King, Peter R]

> But I'm not with him, Peter and Alan in thinking this means we shouldn't try and find good algorithms for generating extensions.

The real problem I have is that as soon as we say "these are the ways that method X should be extended" then someone has to police it. So we have a committee that spends its time defining and refining the rules on what is a valid extension (because there are always exceptions and oh isn't this a good idea we should include) and another, or the same committee that enforces this. Is this what we really want? Suppose I choose to name a method after a previous stage, do I have to stand before the committee and justify my position, show that it fits the rules or persuade them the rules aren't quite right in this case so they need to add another tweak here and there? What if I refuse to conform, do I get my licence to ring a peal revoked?

Yes there are methods that appear to sit in families. But with a different set of rules they could sit together in different families and they would be just as valid. We have already had arguments that Yorkshire could be an extension of Cambridge the reason it isn't is historical. Fine, that was the past do we want to put shackles on the future?