Should we make k a generic parameter?

[spaceships]

An aggregate signature requires individual signatures over k different indices.

Should we make k a generic parameter? We seemed to remember (but couldn't find) that we wanted to support multiple settings of k. What is the best way to achieve that?

[abakst]

I think @pyrros mentioned, regarding k, f, and so on:

they may be potentially changed, and/or there may be concurrently running at different params.

[pyrros]

Just to reiterate: Indeed, it would be great if we could indeed support different parameters concurrently (i.e nodes A,B,C produce signatures for [k_0,m_0,f], but node D attempts to aggregate for [k_1,m_1,f]).

It might be possible for f to vary as well, but it's less of a priority.

[abailly-iohk]

Wouldn't this be defined at the node level, hence a parameter set in the library. Seems much simpler than allowing the library itself to support multiple k or other parameter values?

[pyrros]

Indeed, we don't need to make the library explicitly aware of multiple values.

But we would like the library to avoid hardcoding/embedding the k/m values to sigs and certs. That is, we want to be able to handle sigs/certs being created under one set of k/m parameters and evaluated (i.e verified/aggregated) under another.