Avoid copyright paranoia

A trend I am observing on Wikipedia is excessive paranoia about copyright infringements. While a certain amount of preventive action is certainly necessary, I strongly oppose simply deleting an entry for "presumed" copyright violations.

Examples:

An entry which contains a few copied phrases from individual websites or other sources. Quotation, even without attribution, is specifically allowed in international copyright law, and single sentences are generally not protectable.
A digitized picture that may be copied from elsewhere, but has in fact been created hundreds of years ago. Sometimes, the companies who have digitized these pictures claim copyright on them, but I find such claims highly dubious.
Screenshots of free software applications, and small illustrative screenshots. It should generally be no problem if a screenshot is copied from the official product page.

These are cases where I would generally not delete the entry unless the alleged copyright holder complains. Given Wikipedia's potential liability, after a complaint, immediate action may be necessary, although later correction is possible. But it is not Wikipedians' job to excessively "police" content for copyright infringements, especially when such may not even exist.

In general, when in doubt, do not delete. When "fairly certain", ask the author first in /Talk. The notion of "intellectual property" is dubious at best, and Wikipedia should not support it beyond the limits given by law. Personally, I will restore any entries which I do not see as copyright infringements, and I encourage you to do the same.

-- w:Eloquence

There's no "trend": we've always been "paranoid" about copyright infringement, and I like it that way, personally! :-)

Copyright infringements put Wikipedia's owners and the project itself at some risk. That's why we care. What if, in a few years, a dozen different websites and paper encyclopedia publishers had editions of Wikipedia--and all of them had copies of some copyrighted article. What if the copyright holder wanted to make a bunch of money on the case? Basically, we want to avoid even having to deal with that, even if it could be dealt with amicably.

You might not realize it, but copyright infringment was a regular, if not significant, problem in the months before you arrived. On a fairly regular basis--a few times a week, perhaps--someone would simply copy large amounts of copyrighted text into Wikipedia. We'd catch it and remove it. I think this doesn't happen quite as often anymore, for whatever reason (maybe we aren't catching it enough :-( ).

Now, if your main point is that we should always ask someone first before removing an article, I would counter that this is both a matter of context and a matter of taste, not something on which we need a strict policy. For instance, if I see that so-and-so has uploaded an article from a webpage that has a clear copyright notice on it, and it's clear that so-and-so is not the author of that text, I feel no regrets about simply deleting the text from Wikipedia. Others might follow other policies. Generally, I feel that there is enough abundance of content for Wikipedia that it is probably better to err on the side of deleting. If it is pretty clear that we are violating someone's copyright by having their content on Wikipedia, delete the text--but explain on the /Talk page, always, in any case! If it's not so clear, but it does appear to be a copyright violation, then ask on the /Talk page; if no reply, delete. Those are my rules of thumb.

Finally, Wikipedia does not have an official policy about the propriety of the institution of intellectual property. We should not attempt to advocate a policy whereby we try to impose on other Wikipedians the view that intellectual property is wrong. So, if the decision to delete or not to delete comes down to whether one respects intellectual property (I can't imagine, but just suppose), then kindly allow those of us who might happen to believe that there should be such a thing as intellectual property to act according to our own principles. --Larry_Sanger

Personally, I highly doubt the value of intellectual property law. However, I remove all clear copyright violations with extreme prejudice. If Wikipedia content were to stay solely on the wiki, I would happily support a "wait for the challenge" policy, since it is trivial to add and remove content in this medium. However, Wikipedia will probably be distributed in other forms (multimedia CD, el-cheapo paperback edition, etc) and we could get in a whole heap of trouble if copyrighted material slips into those. We need to catch it as soon as we can, so that those good guys at w:Bomis aren't bled dry in legal battles. --STG

Wholesale copying should be discouraged, but there won't be any harmful consequences for Wikipedia if action is only taken in response to complaint. In other words, if individuals want to be vigilant, that's fine, but there's no need to encourage people to be so. --The Cunctator

1) Trend: Yes, maybe you're right, Larry -- I noticed that the removal of "violations" seemed to happen in "bursts" at least, perhaps because certain individuals are especially motivated to go after presumed violators.

2) You always have to live with the risk of a lawsuit that has no merit. For example, what if an entry about Shell Oil contains information about their murderous practices in Nigeria that Shell Oil doesn't like? Should we remove it preventively to avoid a lawsuit from Shell Oil, which could get quite expensive? What if a company interprets a Wikipedia title as a trademark violation because our entry ranks higher in the search engine than theirs (such cases already exist)? Should we preventively move all company names to subpages? If you want to be safe from the risk of a lawsuit, running an encyclopedia which anyone can edit is probably not such a good idea ;-)

More generally speaking, while we can find evidence that an entry is copied from somewhere else, we can never say with certainty that it is not. If this general uncertainty is just as big as the "certainty" that a certain entry violates copyright, or other laws, then the entry should not be deleted, because consequently this would mean that everything except what you have written yourself needs to be deleted.

I agree with you that people should be able to make decisions on the basis of their beliefs about intellectual property -- what I wrote is only a police recommendation that nobody has to heed. Unfortunately, that is not true for the edit notice that I am currently seeing below my textarea, which in bold upper case letters states:

DO NOT USE COPYRIGHTED WORK WITHOUT PERMISSION!

As is, this statement and the preceding one disallows even quotes from copyrighted material, i.e. it is more restrictive than actual law. Also, it may be not clear what "USE" means in this context. What if I take an entry from an encyclopedia and extract some meaningful information from it for an entry, perhaps only slightly rewording it? Usually, this is completely legal, as copyright only protects form, not content.

Unlike other texts, this is not a text that I can edit, and it is clearly a recommendation from the makers of Wikipedia, so it is Wikipedia/Bomis that is not allowing people who do not believe in intellectual property to act according to their beliefs. Not that I am criticizing that, but it's the exact opposite of your last sentence. I run a few websites myself and of course do not allow people to post copyrighted content either. However, in the context of Wikipedia, it may make more sense to link to an article that explains several standards of inappropriate content, i.e. a "lenient" and more "restrictive" standard, and lets users choose which one to use when posting or deleting content. This article could serve as a stub for the "lenient" policy.

Which would not allow posting copyrighted works verbatim, either, of course, but recommend different behavior for different rules of "certainty". -- w:Eloquence

I agree that the statement is a bit harsh, but I think it's pretty clear (from earlier discussions of the topic and more detailed statements on policy pages) that actual policy of Wikipedia is simply to comply with the law. If the statement is more restrictive than that, consider it an error on the side of caution. --Lee Daniel Crocker

I think another concern you should have is theft of Wikipedia copyright. Due to the nature of the articles on Wikipedia--very general and generic prose written to avoid idiosyncracies specific to Wikipedia--and due to MeatBall:KeptPages' forgive and forget policy, it's possible for someone to take a copy of a page today, wait some months, run the system maintenance (or wait for someone at Bomis to do this even though this rarely happens), claim original copyright, and then sue Wikipedia for "stealing" the material.

The alternative is to keep every version, every flamewar, every mistake, and every liability. That is, if someone libels today, and you version everything, that libel will be recorded forever. Then you get into deleting individual versions, and that's another major problem. Who has the power to do this? Do you trust the administrators? I don't trust myself which is why MeatBall:PageDeletion is not an administrator action, but a communal one. I don't have answers to this problem, unfortunately, but maybe someone else can think of one. Whatever you do, avoid Cliff's old MeatBall:KeptVersions idea. The key to scalability is MeatBall:ForgiveAndForget. -- SunirShah

I think real deletions will be rare, and affect only empty pages with stupid titles or the like. Otherwise, the very script that is generating this page keeps all article versions. Of course, being SQL-based, a database dump can be easily made before removing very old versions. --Magnus Manske

Urg, that "wait till the old version is gone and claim copyright theft" bit is nasty... and entirely possible. That's something we need to deal with. --w:Stephen Gilbert

I'm not exactly sure what the "KeptVersions" idea is; it seems to be frozen pages or somesuch. I don't think it's a good idea either.

However, the ForgiveAndForget idea is a point I largely disagree with Sunir on, while I agree with most of his/MeatBall's concepts. At least for me, the usefulness of a full revision history of Wikipedia entries drastically outweighs the negative consequences. I believe in forgiveness, but I'm not sure I see the benefit in forgetting. Already it's been a source of annoyance that on the main site the revision history only goes back about a month--I can't remember what changes/additions I made. I like to be able to pick up on entries right where I left off, be it an hour or two months, and a complete revision history helps. Also, sometimes earlier versions of entries take an alternate (but not inferior) approach to the subject, the coexistence of all the early versions create a more comprehensive entry than the fixed page itself. I'm a ForgiveAndRemember fan. --The Cunctator

Would you want all your disputes with Larry remembered and reiterated to you monthly for the next several years? As wikis always exist in the Wiki:WikiNow, flame wars are often continuously reignited unless their context is allowed to be removed. Full version history prevents this.

By the way, it shouldn't surprise you that ahimsa teaches forgive and forget. If you believe in WikipediAhimsa, you believe in building forgive and forget into the architecture of Wikipedia. -- SunirShah

<< Copyright infringements put Wikipedia's owners and the project itself at some risk. That's why we care. What if, in a few years, a dozen different websites and paper encyclopedia publishers had editions of Wikipedia--and all of them had copies of some copyrighted article. What if the copyright holder wanted to make a bunch of money on the case? Basically, we want to avoid even having to deal with that, even if it could be dealt with amicably.>>

This "floodgates of litigation" argument is extremely fallacious. To wit, only the most litigious of individuals would prosecute an unwitting infringement of copyright, as an affirmative defense of "innocent infringement" would defeat any claim of damages. Moreover, it is incumbent upon anyone who uses Wikipedia content to make a reasonable review of that content, and (by necessity) that review would include a competent investigation into potential infringement of copyrights. This is the same standard which Wikipedia should impose upon itself.

As I see it, the primary justification for paranoia about copyright infringement should not be concerns of legal liability; it should be quality control. The only person who *might* be held accountable for copyright infringement is someone who intentionally steals the work of others and/or refuses to remedy clear cases of infringement by others.

<< I think another concern you should have is theft of Wikipedia copyright. >>

Assuming that Wikipedia spends $30 to register its copyrighted material every three months (as most publishers of periodic materials do), this is not a valid concern.--NetEsq

"Wikipedia" is not yet an entity capable of claiming or owning copyrights at the moment, but we're working on it. When the paperwork is done, we will then assert collection copyrights (though individual contributions will remain with authors as they are now, licensed to us under the GFDL). But that's not my concern at all. I am an IP skeptic as well, but as such I tend to scrupulously follow existing law for the sake of personal credibility--i.e., so I can't be accused of opposing IP law just so I can satisfy my own desires to use copyrighted material). But even so, I agree that some people here are overly paranoid. Obvious violations can certainly be deleted; especially since they can be recovered if we make a big mistake. But otherwise, I agree that asking the question first is OK--we have plenty of means to do that now. --Lee Daniel Crocker

An alternative viewpoint: Trying to keep Wikipedia relatively clear of copyright violations isn't something I try to do out of fear of legal action - it's something I do for quality control. I try to replace fair use photos with GFDL photos for the same reasons I try to replace misspellings with correct spellings. MyRedDice

Cheese.