Suggestion: play_gnm and play_gnp

[szhorvat]

This is just a small suggestion to separate play_erdos_renyi into gnm and gnp versions. This is what the modern interface does as well in igraph, see sample_gnm and sample_gnp (erdos.renyi.game is deprecated). Mixing up the two in a single function is not particularly productive. Also, while today both the G(n,p) and G(n,m) are sometimes referred to as the Erdős-Rényi model, the original paper of Erdős and Rényi actually considered G(n,m), not G(n,p).

[szhorvat]

Thanks! It's good that this was done, as in a future version of igraph, the parameters of sample_gnm() and sample_gnp() will diverge (the former will get an extra parameter, but not the latter).

[szhorvat]

Also, just in case you are preparing a new release, be aware that a new version of igraph is coming very soon, with some deprecations and small breaking changes (no breakage revealed in tidygraph yet due to this but you never know...) You might want to wait for that release.

[thomasp85]

Thanks for the heads up - do you know the timeframe of the igraph release?

[szhorvat]

@krlmlr would know.