[Commons-l] Making damn sure image attribution is very clear

[James Duncan Davidson]

Bryan Tong Minh wrote:

> There has been some recent discussion on what exif data we should
> include in thumbnails. It is for us technically possible and generally
> people think it is a good idea. Unfortunately, some cameras add 25 KB
> of exif data to a picture; which is about the size of a thumbnail
> itself. The exif data has to be stripped before attached to the
> thumbnail.

Right. There is some of that metadata that is unimportant in many  
contexts--sensor dust data and the like come to mind. But, the IPTC  
style stuff: title, caption, created by, copyright. That's the stuff  
that one really cares about. And it's the baby that goes out with the  
bathwater.

geni wrote:

> Nyet. Plenty of books put the credit at the end. How many books have
> the credit for the cover art on the cover?

It's pretty hard to snag a photo from a book. I've tracked down many  
violations of my CC-licensed photographs to people who "borrowed"  
them from Wikipedia. WIthout any indication that they are subject to  
any kind of license, well, people don't know. And that's what they've  
told me.

As well, Wikipedia isn't a book. It's not a pamphlet either. It's a  
website. And websites can be used in a variety of ways. The content  
can be repurposed. Print a Wikipedia page. The credit is gone.  
Archive it as a PDF. The credit is gone. Right click and save an  
image, the credit was never seen.

> The medium is wikis the means is mediawiki. Click through is the
> reasonable manner in this case.

The medium is the web. You guys know what the heck a mediawiki is.  
The world that uses it sees web content and _may_ know that they can  
edit it. I certainly don't care that the medium is a particular kind  
of software. It is what shows up on my screen.

> > By crediting in a manner that is
> > accepted and practiced in the photographic industry,
>
> We are not part of the photographic industry. More relevant examples
> would be Encarta and Britannica online. Or just general websites.

Encarta and Britannica make sure that every piece of content,  
including photographs, are licensed in a way that is appropriate.  
They also give credit. You bring up Encarta, here's Bill Clinton's  
page on Encarta:

http://encarta.msn.com/encyclopedia_761564341/Clinton_Bill.html

The image provided there is small and obviously leads to more  
content. Click on it, and you land on a page with a usabel size  
image. And there's the credit.

http://encarta.msn.com/media_461526028_761564341_-1_1/Bill_Clinton.html

If you're going to put up reasonably sized images on the article  
pages, you should give credit there.

> We have no way to know what the photographer's wishes are.

That's what I'm doing here and trying to communicate, as a  
photographer. And I'm not your average photographer. I'm one that  
spent 5 years in the Open Source community and dealing with legal  
issues. Most photographers just want to bury their heads in the sand  
because all this new stuff is scary. I'm trying to communicate to you  
guys how to do things in a way that will increase the participation  
of photographers in the commons. Obviously, it's from my viewpoint,  
and you have to take that into account, but this is my intent.

When it comes down to it, I have two options right now. I can saw  
that my use of the CC licenses over the last few years was a blazing  
mistake and try to find a different way to live in the brave world  
where copyright is changing. Or I can try to communicate how you guys  
can meet us half way so that we can get MORE photographers playing ball.

I believe in the commons. I want it to grow. But if you're going to  
put credit in a place where it's invisible and negates the whole  
intent of the attribution request, then the CC is not a valid tool to  
use. And since Wikipedia is a shining flagship in the commons, how it  
behaves is, in some ways, a standard bearer.

I may indeed be forced into that position. I hope not.

> > Placing that data one click away is not
> > obvious to users and doesn't feel "right" from the perspective of a
> > copyright holder.
>
> Allowing blatent violations of :
>
> http://en.wikipedia.org/wiki/Wikipedia:Ownership_of_articles
>
> does not feel right to a wikipedian.

Understand. But I'm not arguing about ownership of articles. I'm  
talking about my ownership of content that was placed onto your site  
by a third person. This comes up again and again in the blog comments  
I've received. My placing of content under a CC license in no way  
means that I am acting as a contributor to Wikipedia. There's no  
reason you can/should establish that wikipedian ideas pertain to  
content that is owned by third parties who are not a party to  
Wikipedia and bury their credit to the point of invisibility.

> As well as the spaming issue is becomes problematical in cases like
> this where there are three seperate authors to consider:
>
> http://commons.wikimedia.org/wiki/ 
> Image:Caisson_lockenglish.svg#Description

Um, I don't get what you're getting at. But in the end, it's not my  
domain. I can establish ownership of my images. I provide them in a  
way that keeps information along.

> > 3) EXIF metadata should be preserved, even on resized images.  
> Thumbnails can
> > be recreated, so junking those isn't an issue. But stripping  
> unrecoverable
> > information, especially that which may contain author and license
> > information, is a problem when the images are borrowed and used  
> downstream.
> > I wish I had a good way to strip just thumbnails, but I don't  
> currently know
> > of one. Flickr has the same practice as well, and it's annoying....
>
> This would require someone to rewrite the code. In understand that
> mediawiki uses http://en.wikipedia.org/wiki/ImageMagick to resize
> images.

 From my perspective, your exposing internal implementation details  
as a reason to not do something. Understand that I don't care what  
tool is used. I'm simply stating that it is problematic when EXIF  
data is stripped and making a request that important metdata be  
preserved.


James Duncan Davidson
james at duncandavidson.com
+1 503 784 8747


-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.wikimedia.org/pipermail/commons-l/attachments/20070825/9a8878dc/attachment.htm