Benoît Legat

Why would it be more convenient ? This isn't consistent with `HermitianPositiveSemidefiniteConeTriangle`

For the hermitian PSD cone, we can't change before MOI v2 since that would be breaking and I don't remember getting any comment on the choice of ordering (see https://github.com/jump-dev/MathOptInterface.jl/pull/1962)....

> I don't really understand the problem of supporting `VectorAffineFunction{ComplexF64}-in-NormOneCone`. We do it for `ScalarAffineFunction{ComplexF64}-in-EqualTo{ComplexF64}`. What if you add `VectorAffineFunction{ComplexF64}`-in-`NormOneCone` and it gets bridge by the `SlackBridge` which adds `VectorOfVariables`-in-`NormOneCone`...

I don't really understand the `is_maybe_real` function added in https://github.com/jump-dev/MathOptInterface.jl/pull/2468. Don't we instead want `may_be_complex` ? The bridges need to have a guarantee that the function is always real right...

> So I didn't want to break existing code (like Polynomials) for which we don't know. So we had to default to false positives. MOI 2.0 is still far in...

> At the moment, if you use an unsupported function you can hit a MethodError because of a bridge that isn't relevant. That isn't good. I think it's sufficiently rare...

If the default of `is_maybe_complex` is `false` then it's non-breaking (we say it's mandatory to implement it for new functions). We can add a note in the docstring that in...

It seems to be called from https://github.com/cddlib/cddlib/blob/475890c3a760300f5b088c0c308d2b3b95b2acbb/lib-src/cddlib.c#L265 and https://github.com/cddlib/cddlib/blob/475890c3a760300f5b088c0c308d2b3b95b2acbb/lib-src/cddlib.c#L303 It would be interesting to know what exactly causes that.

You can compile it from source with a different constant and then copy the library in `~/.julia/packages/CDDLib/.../deps/usr/lib`

> @blegat, is the following statement still true? You have upstreamed your changes, haven't you? Indeed, this should be removed. **EDIT** updated in 3d86b4ce39db94c8b1e246c9ee9bb748fa8223e4 > Also, if you are using...