Ask HN: What if all code was suddenly open source? | Hacker News Reader

Ask HN: What if all code was suddenly open source?

1 pointwarent3 comments

There's this sort of thought experiment that's been in the back of my mind for the past few weeks.

Imagine every technology company, government, programmer, etc. is notified that all code that has ever been written is now going to be open source, licensed under the MIT license. Furthermore, new code will instantly be open sourced and MIT licensed as well.

Even if all of your code is served in an intranet and has no connection to the WWW, it doesn't matter, your source-code is now publicly available and easily accessible (read-only to the public). Don't worry about the cost or logistics; some perfect public entity with enough resources will be hosting it.

This, for some reason, is just accepted as the new reality. Private corporations like Amazon, for example, are now forced to play in this new competitive landscape of transparent, free software.

How do you think this would affect technology and the way we develop it? Do you think this would harm or improve society? What would the short term and long term changes look like?

Edit: Clarifying some of the language

I can't make sense of what this means.

Are my HTML and LaTeX documents code? What if I practice literate programming and interweave my manuscript and the code? Are both released under the MIT license?

The appendix of Cryptonomicon includes a perl script. Is the original manuscript therefore source code? Or only the executable code part?

What about a video game where the engine reads external data files, and it's the data files which specify the level. (Eg, wad files and DOOM.) Are those data files also "code"?

A Game of Life engine can take an array of 0s and 1s as inputs. Including from a 1-bit image file like one in PBM format. Are those image files under the MIT license? Are all PBM files under the MIT license because they could be used as input? Remember, the Game of Life is Turing complete.

It's surprising just how many things are accidentally Turing complete - http://beza1e1.tuxen.de/articles/accidentally_turing_complet... .

Tools like fuzz checkers can generate billions of source code tests. Are those all archived?

If I have my computer generate millions of variations of my source code, will this "perfect public entity" store all of them? How is it indexed? Because it seems one way workaround is to make it hard to find the source code needle in the haystack of false leads.

Wow this is amazing, thank you for the link. I never knew about this.

The patent lawyers would have a field day. I remember reading somewhere this may be one of the reasons that linux video drivers weren't open source by nvidea and/or AMD some years ago.

Reply on news.ycombinator.com