Joshua O'Madadhain

@alan-isaac I think that's a reasonable point; I think that it's also worth noting that including all self-loops may tend to conflate "transitive" and "reflexive". (Although I have one quibble:...

The quibble is this: if the edges in the graph have weights, and the length of a `u-v` walk is defined as being the sum of the edge weights in...

@alan-isaac I think that there's a confusion here between two related concepts. No resource that I've found has suggested that the transitive closure of a relation (graph) should _not_ be...

I expect that most people that want the transitive closure of a graph have never heard of the reflexive property. :) In any event, I agree that the idea of...

I still agree that we should address this. It's on my list of items to work out with the rest of the Guava team. :)

Here are some issues that I'd expect to come up in the API review. (@jbduncan has already pointed out some of the key ones; thanks for that!) First of all,...

@ogregoire You could think of this as being a special case of #2041, but since selection using a uniform random distribution is such a common thing to want to do...

Yes, I do have plans to support tree topologies in common.graph. Right now it's an open question whether it would be represented as its own type, or as a specialized...

@jbduncan, a couple of things about your last post. First, I'm not actually sure that using a graph (or tree) would actually help you much here. It doesn't sound like...

@liach It depends on which definition you use for "tree"; it's not a settled question in graph theory. Yes, some sources refer to trees more or less as you do...