The Dangers of Open Content

gihan_ripper writes "Recently released open movie Elephants Dream found itself in hot water with Catalonians after accidentally using an offensive word instead of 'Català' in the subtitle menu. The cause? Designer Matt Ebb had used Wikipedia to look up the Catalan word for Catalan on a day when the entry had been vandalized. He writes about this experience on the Elephant Dream blog. We may have scoffed at John Seigenthaler over his criticisms of Wikipedia, but it gives us pause for thought when we to heavily on Wikipedia."

Re:Dangers of international content?

[DiamondGeezer]

I understand the dangers from using wikipedia (and like so many slashdotters have said, for serious work, use it as a starting point, not a source.)

Why would I trust it as a starting point if I can't trust it as a source?

Re:Dangers of international content?

[Whiney Mac Fanboy]

Why would I trust it as a starting point if I can't trust it as a source?

You shouldn't trust any single source.

Wikipedia is a useful starting point as it will contain pointers (or at least useful search terms) to begin looking for other items to reference. It's no different to any other encyclopedia in that respect.

Surely you don't use a single soruce for information for an important project?

is open content the real problem?

[a10_es]

I'm catalan. And I can say that lately there's been a lot of hatred against our nation pushed by some spanish political parties (Yet I don't to turn this into a political discussion). This problem appeared because of a vandalized entry in wikipedia, but could also have appeared if a person had modified the film or written it wrong from the start, so the problem here is not the reliance on open content, but the reliance on people's goodness, which in the open [content, source, ...] is mostly there, but can be displaced by some feelings, most of them learnt and fueled since childhood. But the same thing's been happening throughout the history. Surely if you looked on recognized encyclopedias some time ago, a lot of entries about slaves would be unaccpetable by today standards. The same happens over conquered soil after a war, when the losers become the vermin that had to be erradicated and the winners the saviors of the people (and usually end up being as bad as those they overthrew). And many other examples could be given. So the problem here is the open content or close-minded people?

Re:Dangers of international content?

[DerekLyons]

Surely you don't use a single soruce for information for an important project?

I routinely do. But then the source in question is unimpeachable and has stood the test of time and criticism. In fact, in the real world it's very common to rely on single sources, handbooks, references, etc...
 
When writing a program, you don't look up the meaning of a command in three sources do you? When wiring a house, you don't check three different copies of the electrical code. When working on your car, all you need is your Chilton's. Examples abound of routine daily use of single sources.

Proofreading?

[Wonko the Sane]

Isn't this more of a case study of not proofreading the final product rather than relying on an unreliable source? The list of names could have been emailed to all the translators first before finalizing the DVD.

Joeri and thousands of screaming fans here were rightfully pestering me to get it done as fast as possible,

I think I found the real problem.

Been there, done that...

[(H)elix1]

I do a fair bit of international coding. Problem is, I am not fluent in many of the languages I am building software. When putting together my language bundles, I always have someone do a quick walkthrough of the application who knows the language and context. You cannot count on software to give you a proper translation. Last year I was building some portlets for a French company. I added navigation and hit the fish to translate some of the finishing touches. I added a 'back' button - only to find the word I used was a person's back (not return to the previous step) in my i18n resource bundle.

How do they say - nothing is as permanent as that which was deemed temporary? Not uncommon for stuff like this to not get checked by QA.

danger of careless people

[fermion]

This is no different than reading something anywhere and then quoting it as fact. The only difference is that wikipedia is not static, and so the errors can change from minute to minute. Therefore this is not a problem with open content, but a problem with dynamic open content.

All of this can be easily solved by fact checking before the distribution of a static content.

I do understand the problem. I can be careless. But when I am I do not blame my carelessness on someone else.

Authors and Authority

[Doc Ruby]

The problem is not "open content", Wikipedia, or vandals. The problem is people who rely on a single unaccountable source for any knowledge. That is a recipe for failure.

This has also been the problem with "authoritative" sources, like the Encyclopedia Britannica, NY Times or White House Spokesman. Those sources are highly managed, consciously or unconsciously, so they don't usually go as obviously haywire. Instead they mislead to usually workable misconceptions. In the service of the writer/speaker or the organization that produces/publishes them.

Now that the world is finally filling with lots of smalltime publishers, as publishing has become so cheap, easy and scaleable, we're all seeing the limits of sources. So we all must learn what the past publishers learn: power of the press belongs to people with presses, and power corrupts, absolute power corrupts absolutely. The only way to handle the corruption is to match power against power, cross-reference information from independent (of each other) sources.

Wikipedia will be even better when it includes an independent "fact checking" feature, like automated Google/Yahoo/MSN searching of citations. Until then, its superior power to managed press is just raw power that requires users to do that for ourselves.

Quickie Mart

[Joebert]

Why aren't changes highlighted, or otherwise made glowing neon fucking {INSERT COLO[U]R HERE} for the first X days after the change is made ?

apologize, repair, move on

[Simonetta]

Seriously, you got caught in some asshole's juvenile prank. Defacing a public resource (wikipedia) to reflect an immature joke at the expense of the next person to use that resource.

    So apologise, repair the mistake, and move on. Just because some jerk doesn't understand the usefullness of an open source public resource doesn't change the utility of that resource. And anyone who is 'offended' by the prank needs to understand this. This is like sueing the streetcar company for racism because some pissant spray-painted a racist remark on a streetcar. The correct response is to find the person responsible if possible, and if not, then to teach your own children why civilized people don't do such things.

Re:Dangers of international content?

[gameforge]

When writing a program, you don't look up the meaning of a command in three sources do you?

Regularly. And only then do I get a complete description, if not find an error in one.

When wiring a house, you don't check three different copies of the electrical code.

If one, even. Really, if there were multiple versions (not copies) released at the same time, of course I would look at all of them.

When working on your car, all you need is your Chilton's.

And that's exactly why my interior door panel on my old 1993 Grand Am held on for dear life by three screws. Sure it was my fault for not being gentle; but the factory shop manual, I discovered, had a full blown illustration and much more detailed procedure. Chiltons and Haynes both throw five models over ten years into one book, making it useless for anything but drivetrain work. They may as well cut the interior and body work out of their manuals entirely, along with much of the electrical and vacuum system stuff.

Again, if Pontiac made several publications with varying but similar information, I'd want all of them, and I did own both the Haynes and Chiltons manuals, occasionally referring to all of them.

The point is, you really can't trust any source of information unless you've personally witnessed the accuracy of the information (i.e. it's your research, etc.) Information comes from imperfect humans, and you simply can't trust that 100% (if 10% in some cases). That's fundamental, not practical; if it turns out most of the info you research is accurate enough for your needs, which happens most of the time, you'll be okay for the most part.

Wikipedia is ultimately more helpful than it is harmful, but if you choose to use it for a single source of information where it's critical that the information be accurate, you HAVE to double check the info at least, if not simply use it to acquire other sources. Reason: There's no blaming Wikipedia and holding them responsible for your embarassing and possibly consequential mistake in your work.

Re:Dangers of international content?

[Whiney Mac Fanboy]

When writing a program, you don't look up the meaning of a command in three sources do you? When wiring a house, you don't check three different copies of the electrical code. When working on your car, all you need is your Chilton's. Examples abound of routine daily use of single sources.

No - I don't use an encyclopedia for any of them, I would use a specialised source, perhaps using wikipedia/other encyclopedia to find out what that specialised source was. That was the mistake the guy we're discussing made - he should have used wikipedia to point him to a localization authority's page on language localisation. (not that I'm criticising him, heavy workload and all that).

Professionalism

[Brandybuck]

Professionals use professional translation services. 'Nuff said.

Re:Fact-checking and Wikipedia

[Anonymous Coward]

Of course, the Daily Texan "may" be the most prestigious college newspaper in the U.S.

But then, there may really be underpants gnomes as well. Almost anything "may" possibly be true.

But having read the Daily Texan, if it's the most prestigious college paper, then I would worry about the state of college newspapers.

Re:Scoff at Seigenthaler?

[Wordsmith]

Both points are true - my failable memory is to blame. But there's some irony in that you point to Wikipedia to illustrate one of them. :)

But the overall message is right - that Seigenthaler had a very reasonable concern, and addressed it reasonably. And unlike many who've been wronged, he didn't push for the heavy-handed solution of government regulation; on the contrary, he worried that similar abuses might eventually lead to it, and he saw that as detrimental to the idea of free speech.