Hacker News Reader

Comment by niggler | Hacker News Reader

I wish licensing was emphasized more in this discussion.

Including a license is the #1 reason myself and many other people stray away from code on the internet. I won't touch a project if it's not stated properly.

And to be clear, just writing that your project is MIT licensed or sticking it in a package.json or making a small remark in your README doesn't cut it. You need to follow the terms of the license you wish to use (my comment on the matter: http://blog.nig.gl/post/48848761220/just-saying-something-is...)

(always a good read: http://www.codinghorror.com/blog/2007/04/pick-a-license-any-...)

View on HN

I'm curious - have you been burned by issues with licenses in the past? I've never had licensing affect me in any tangible way and so they are really more of a vague theoretical, and thus low priority, concern for me. But I am vaguely aware that this is theoretically the wrong position to have while yours is the right one.

It's better to be safe than sorry, as the saying goes. I've never been burned personally. I'm also pretty sure that anyone working for a large software company is warned about using third party code or snippets found on expertsexchange or stackoverflow or random blogs.

Normally people are well intentioned, but sometimes you stumble upon repos with very strange licenses (for example, a javascript repo that has operating system restrictions https://github.com/stephen-hardy/xlsx.js/issues/8 ).

It's usually not of a concern for most people working on small projects (after all, another party has to notice and then decide to take action), but if you are trying to enter an industry with a highly litigious incumbent then you should make sure your ducks are in a row first.

That all makes perfect sense. I think I misread your original comment somewhat - I read it as "licenses are the #1 thing I consider when evaluating projects", rather than "of the projects I decide not to use after evaluating, licenses are the #1 reason for that decision", which isn't the same thing at all.

> I'm curious - have you been burned by issues with licenses in the past?

If you count "not being allowed to use $LIBRARY because your company's lawyers won't let you use improperly code", then this is incredibly common.

The alternative scenario - using code released under an ambiguous license and then later getting sued for it - is much less common with large companies simply because good legal teams won't let that happen (see the above scenario).

I'm sure there have been examples of it, though - I know I've heard of those stories myself, even; I just can't think of them at the moment.

Unfortunately, it seems that most FOSS code on Github is not actually properly licensed: http://www.theregister.co.uk/2013/04/18/github_licensing_stu...

When my company was getting aquired we sent them the license details of some 50 projects as part of the DD, the lawyers emailed back and said to only include licenses to things we that were either not open source or that were substantial to the tech stack. SciPy, NumPy, Python, and GNU/Linux was all that was sent.

That codinghorror.com article is great and is a must read for any developers on the fence about attaching a license to their open source projects. I have to admit I was feeling both awkward and douchey debating whether I should attach a license to a few miscellaneous projects that no one may ever even see.

"That's ironic, considering the whole reason I posted the code in the first place was so other developers could benefit from that code. I could have easily avoided this unfortunate situation if I had done the right thing and included a software license with my code."

Also, a nice summary of various licenses. Thanks

You're completely off on this subject. The copyright holder can do whatever he wants with it - including releasing it under multiple, different licenses. The onus is on the licensee - a person who obtains the licensed code, to honor the license given to him. It's the responsibility of the licensee to comply with the terms of the license and republish whatever needs republishing when (if) he redistributes the code.

None of your statements contradicts points from my comment, my post, or the coding horror post. If you are releasing code and not putting in a license or doing so improperly, don't expect others to use it. CH focused on one facet (missing a license) while I focused on a different facet (how to add a license properly)

> And to be clear, just writing that your project is MIT licensed or sticking it in a package.json or making a small remark in your README doesn't cut it.

My point is that it does cut it. It's up to the licensee to include the license, not the copyright holder. There is nothing improper about specifying the license and not including it in its entirety. That's just your personal interpretation of it, that has nothing to do with copyright law.

Reply on news.ycombinator.com