Hacker News new | past | comments | ask | show | jobs | submit login

There is a vast amount of talent that has created all the powerful open source numerical platforms we have today but if it could somehow coordinate we'd significantly accelerate these projects to maturity.

Octave is playing in a crowded orchestra, with the python scipy stack, the R and julia ecosystems, not to mention libraries in c++ like armadillo and eigen or even new projects in rust.

The funny thing is that in individual projects we like to apply DRY but we thinking more broadly, when solving numerical / computational problems there seem to be (too) many open source alternatives with significant overlap and duplication

No doubt all them have their pros and cons and optimal niches or historically strong points.

Ultimately though it might be an idea to somehow consolidate and coalesce around fewer and more mature stacks that have large use and development communities and which dont compete against each other for attention




They're tools for different tasks, consolidating them would make a less helpful mishmash


Octave is open source Matlab. If you need Matlab, nothing else helps.


This simply means that there is a need for an open source api that emulates the matlab api. But other components, e.g. IDE, compute engines etc could (theoretically) be shared between various stacks. To some degree this happens already, eg when they link underlying linear algebra libraries written in c++ or even fortran.




Join us for AI Startup School this June 16-17 in San Francisco!

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: