3 ms·
I'm not sure if it's applicable, however; I'd be interested in any association that can be made against previous versions of such libraries, and the newly rele
by gU9x3u8XmQNG 8y ago
I'm not sure if it's applicable, however;
I'd be interested in any association that can be made against previous versions of such libraries, and the newly released open source libraries. Are the 'releases' between the old closed source and that of the newly open source significantly different?
If so, and either way; I'd be interested in knowing what kind of changes were applied in direct action of the source being released. Was it just cleaning/linting of the code? Were there features that were otherwise inappropriate (security? quality?) that needed attention before release.
Things like alignment with child dependency license agreements, or overall quality (in code, testing, and deployment) are things I find extremely interesting. Was there some interesting 'inappropriate' language in the source from developers long gone. Could this have effectively been the opposite spectrum of what happened with event-stream https://github.com/dominictarr/event-stream/issues/116 https://github.com/dominictarr/event-stream/issues/116
I'm not explicitly saying it's any of my business. I can imagine a multitude of reasons that this history would never be provided. But... as above.. I can consider reasons, perhaps; it should..?
I don't mean to discourage release. I really appreciate it. Just some thoughts..
Edit: Minor readability