Proposal to merge the JuliaLinearAlgebra and JuliaMatrices organizations

It seems like the JuliaLinearAlgebra and JuliaMatrices are very similar in scope. We (Viral, Alex, and myself) would like to propose merging the two organizations.

Any objections?

24 Likes

The purpose to merge is to reduce administrative burden of having linear algebra packages in too many different orgs.

8 Likes

I can only say that it sounds good to me :slight_smile:

1 Like

Pinging some contributors in the org: @dlfivefifty @jishnub @dkarrasch. Want to make sure they have seen this and chime in on whether they think it is a good idea or not.

Anyone else we should ping?

-viral

2 Likes

I don’t have an issue with merging.

Broadly speaking JuliaMatrices is for packages representing special matrix types while JuliaLinearAlgebra is for linear algebra, but I don’t see any real benefit from having this separation.

One could arguably go one step further and also merge with JuliaArrays, though that org tends to have more “fundamental” packages so may wish to have a stricter membership.

2 Likes

I agree with Sheehan. Merging the orgs might lead to more active members participating in packages from the other org, which sounds like a good idea

I also agree with Sheehan. All JuliaMatrices.jl packages are closely enough related to linear algebra.

I also like the idea of merging JuliaArrays, but it has a lot of packages in it. My thought was the same - easier organization management, but given the number of packages in JuliaArrays, I am thinking it is best to keep the two orgs separate.

I have moved all the repos from JuliaMatrices to JuliaLinearAlgebra, and invited all the members to JuliaLinearAlgebra. The next step is to update the URLs in General.

If anyone has lost access to packages, please let me know here and I will fix it.

-viral

2 Likes

I think keeping JuliaArrays separate is a good idea as it has a much broader scope.

Otherwise, we risk taking the merge argument to extreme and end up putting everything in a single JuliaPackages organisation… then we’d argue to merge that with JuliaLang!

I agree again. I think most packages in the JuliaArrays org view arrays as general data containers, not necessarily vectors, matrices, or linear operators/maps.

Sounds good to me.