Hacker News Reader

Comment by smackfu | Hacker News Reader

This is 6 months old, so it doesn't seem like it went anywhere. Probably because the greater internet has no motivation to oppose moving content to a non-commercial license.

View on HN

Original author here: You're right. It hasn't gone anywhere in 6 months, because there's no real angle on getting involved in a legal dispute. There's no money to be gained in suing TVT. And forcing them to take content offline hurts the commons perhaps more than it helps it.

The motivation behind the original blog was at least partially catharsis -- there really are no good options for copyright enforcement of open content. Sure, if it was a few lousy pages, TV Tropes could stand to lose them, but I'm talking about core pages like "Anime" and several tropes of legend initially written by my colleague.

Sure I could send a DCMA request, but they'd have every reason to fight it, and what then? Hire lawyers? Try to delete content that people thought was legal and amended, but actually turned into infringing content? Get them to delete pages on works of fiction of which I am a fan?

All of the options are awful except one: getting people to leave that illegal and immoral site. True morality isn't about protecting children's delicate eyes from textual metapornography, it's about giving back to the people who give to you -- and not just taking more from them.

I recommend either All The Tropes or the Tropes Mirror Wiki on Wikia.

I think it speaks to a different mindset from what we as programmers are used to.

Licensing is a pretty big deal for us. Virtually every programmer has some sort of opinion on GPL vs BSD vs public domain vs whatever. Open source projects can live or die because of their license.

Wikis are basically open source projects with natural language instead of code. But I get the impression that few wiki contributors care that much about licenses.

If I had to guess, I'd say that it's because code is more modular and flexible. If you write a useful open source library, the license you put on it can have huge implications for what products end up using it. There's no wiki equivalent to e.g. large quantities of BSD code being used to build a smartphone that sells hundreds of millions of units.

In programming, when you reuse others' work product it's considered good engineering practice. We call it "not reinventing the wheel."

In most professions where the work product is in a natural language, reusing significant portions of others' work is called "plagiarism", and is considered a serious ethical violation. Reusing small portions for the sake of making a point, on the other hand, is covered by fair use rules.

So yeah, licensing concerns might be a mostly irrelevant detail. The big questions about when you can and cannot reuse others' work product are already covered by a rigorously socially-enforced ethical standard, and the law. Both of those are more powerful than mere contracts in their own way.

Think about free textbooks, which may be commercially printed (of course you have to pay for a physical object) but are also available online under CC-BY-SA. If something is well explained in, say, Wikipedia, I can lift it and put it directly into the textbook, with a credit, making only the small tweaks necessary to make it flow in context. This can be hugely valuable, but tacking on NC means no one can ever sell those physical copies (which some people prefer).

I kind of hope this situation doesn't arise with TV Tropes, but you never know...

I don't see the distinction you're trying to make here.

It's plagiarism to copy someone's code without giving attribution. It's not plagiarism to copy someone's code if you give attribution.

It's plagiarism to copy someone's natural language text without giving attribution. It's not plagiarism to copy someone's natural language text if you give attribution.

In either case, copying someone else's code or natural text is potentially copyright infringement, but that's a different matter.

/ * @author BigDave1974 over at SlashDot */

I wonder if it would make sense to sell reusable chunks of text, like libraries of code.

I'm a programmer and at the risk of sounding like an idiot - I have no opinion on licensing and could not tell you a single thing about any one license. Are there really that many people knowledgable about different licenses?

Nah, you're not an idiot. There's a tendency on HN of people saying "all programmers" to mean "programmers who live in Silicon Valley, use open source tools to write web apps funded by ad eyeballs, and put all their code on Github". Witness the countless posts about "why should I interview when I can point to a Github account", or that article yesterday about "articles every programmer should read" that had something about SEO in it (and then the replies from people unable to see why that was silly), etc.

That's kind of funny, since I'm nowhere near Silicon Valley and don't even identify with that crowd, I rarely do any web work, and write plenty of proprietary code. (Ironically, web apps are an area where you can generally get away with not knowing anything about open source licenses, since you're not redistributing the programs you make.)

I merely observe that programmers generally use third party libraries, and that doing so requires at least a basic understanding of the license terms.

Maybe there are a lot of people out there writing code who never touch external code?

(Ironically, web apps are an area where you can generally get away with not knowing anything about open source licenses, since you're not redistributing the programs you make.)

Except if you happen to use something under an Affero GPL license, in which case you really should know about it.

But yes, those project are rare.

I hadn't heard of that one. Is it actually enforceable? I was under the impression that the GPL didn't require opening server code, not because it was missing something, but because server code doesn't get distributed and thus doesn't fall under copyright limitations in the first place.

The Affero GPL is a modificated version of the common GPL that explicitly handles the case of software used as a service through internet.

It's fairly uncommon, but those projects licensed with this version have specific function built in that allows any user to get a copy of the source code (and you couldn't remove it without breaching the license); and some known names use it (MediaGoblin, MongoDB, Diaspora, Ghostscript, Launchpad, POV-Ray are some examples).

You find a library on the Internet somewhere. What entitles you to copy it onto your machines? Copyright does not allow to do this, so you are reliant on a license to give you permission to do this. GPL gives you permission to do this, as long as when you distribute any modifications of it, you also pass along the source code. AGPL gives you permission to do this, as long as you give the source code to all of your users should they request it. If you can't meet the conditions, than you don't get the rights acquired by the license, and you fall back to default copyright, which says it's illegal to have copied the library to your own machines in the first place.

AGPL enforceability (under copyright) rests on the concept of modification rights. The basic idea is that one must get copyright permission in order to modify a protected work, which then AGPL adds conditions for.

This has been semi-tested in the past. In a court case regarding sold paintings, the court sided with the painter when the buyer wanted to chop the art into pieces and rearrange them. The court decided that the act of modifying the art created an derivative work which the painter had not given the buyer permission for.

I highly recommend learning about copyright and licensing. If you use or contribute to open source software, it's a must, IMO. Here's a post of mine which would be a good start: http://stackoverflow.com/a/11455485/247696

Based on the number of GitHub projects which don't include a LICENSE or COPYING file, I would say you are not alone.

If you intend to use your work commercially, you should be aware of licenses when you select what tools you are building on top of, or you may get bitten later.

You may also wish to license your own work for others to use. If you have an opinion about when others should be allowed to use your work, licensing can be important for you to achieve that.

You may not need to be aware if you have a legal department who do this for you, for example (or if you are willing to take the risk of legal action later).

The reason I care about licenses is because the auditors care about licenses and we can get hit with ginormously hefty fines if we're not in compliance.

So, I wouldn't say you're an idiot, but you're a bit dangerous to work with. Professional programmers should know about licensing.

If you use other people's code, you might want to take a glance at the agreement that you're entering into with the person who wrote it for you, because you're going to be held to it. Within FOSS we've pretty much standardized on a few general forms, with small differences. When you mix them, though, those small differences can become very important.

If you work with Microsoft, though, you may want to hire someone to figure licensing out for you. Figuring out Microsoft licensing seems to be a lot harder than programming.

Reply on news.ycombinator.com