Gavin Simpson

@EDiLD the issue with current `summary` is most easily discerned if you have a data set of a thousand samples or, as I do now, 13000+ OTUs. No one wants...

@EDiLD That's a side effect; it's not the default so you get bitten by huge output until you've been bitten enough times to remember, assuming you work it out, to...

By adding to the `print.cca` output I mean rather than just showing eigenvalues `summary` should show the eigenvalues, proportion explained, cumulative explained data, split up by constrained and unconstrained axes,...

@jarioksa If people have used `summary.cca` to extract results, that's really their own fault and I will gladly incur their wrath for breaking those things if the result is we...

@jarioksa I don't disagree with what you write. You should be able to extract the data as the underlying R objects that get printed via the `print.summary.cca` method. I'm fully...

OK, let me implement something in a branch and do a PR. We can discuss the detail and implementation there referencing this thread/issue.

@joshgsmith > I'm not sure I fully understand how the distances are calculated without calling `dist()` in that function The key is this bit of code: ``` d

@jarioksa This would be nice to add to vegan; I've had people ask about this on courses before, and just pointed them at the code for `betadisper()` to see how...

I can take a look at this and model the parallel permutations on how we do it in `permutest.cca` and `anova.cca()`-related functions.

What would you want to parallelize over here? We could do the permutations in parallel with some changes to `envfit.default`, but that would still do all 3000 (in your example)...