Ability to modify individual elements of Matrix and Vector objects.

[nihonjinrxs]

I've implemented the ability to modify individual elements of Matrix and Vector objects. Implementation followed TDD methodologies, so appropriate tests are included and all tests pass.

If there are any questions, please don't hesitate to ask.

[nihonjinrxs]

Any issues with this pull request?

[nihonjinrxs]

@jcoglan Have you had a chance to look at this? Any reason not to accept it?

[nihonjinrxs]

Still waiting for some response on this. Is there a reason not to accept this pull request? I'm happy to refactor if needed.

[jcoglan]

The main reason is that I've not had time to maintain this for a long time. I did try to get the repo in shape for someone else to take over but it never worked out. I've still not been able to get the person who claimed the sylvester name on npm to sync our codebases and give me push access, so I can't actually push releases even if I had time to work on them.

[ghost]

+1

[theksmith]

+2

[nihonjinrxs]

@jcoglan this has been out there for a while now. If you don't have time to manage it, can you hand it off so we can get some of these pull requests merged? I'm happy to take ownership if needed.

[jcoglan]

I've not had time to work on Sylvester for a long time. My plan was to get the project into a transferrable state, release what I had done for 0.2, merge things people have asked for, ensure backward compatibility with 0.1 and with the npm release (which I didn't make), then hand over. I started this convo a couple of years ago but it didn't go very far and there's still an awful lot to do.

[jcoglan]

All that aside, I don't actually agree that vectors ought to be mutable, although users can currently do this already since they have access to the elements property. What's the use case?

[nihonjinrxs]

The use case is for matrix math operations on large matrices where you want to be able to do in-place manipulation. My particular use case was an implementation of the dynamic time warping method, which provides a distance measure between two jittered time-series vectors. When things get big, it's not practical (and sometimes not possible) to create a new copy of the matrix or vector on every operation.

This is your project, so you can do what you like with it. However, I think by not transitioning it soon, it loses utility for others, as they see it as a dead project. I think it has great potential, and would like to see it continue to improve. You say your plan was to get it into a transferrable state -- I'm not sure what exactly you mean by that. Anyone who has worked on the code should have enough knowledge of it to set a roadmap for it and push it forward. Either you maintain ownership of it and actively shepherd it, or you give it over to someone else to take that on. I'm happy to shepherd it, or to have someone else do it. I just don't want to see it sit dormant.