Picture copyright check (was: Re: [Wikipedia-l] What would Richard Stallman say?)

Andre Engels engelsAG at t-online.de
Sun Feb 22 09:46:38 UTC 2004


"Caroline Ford" <caroline at secretlondon.me.uk> schrieb:
> Sean Barrett wrote:
> 
> >>Well, let me make that suggestion then. I would like to remove all
> >>pictures for which:
> >>1. no author or source is known, and
> >>2. there is no good reason to believe they are public domain
> >>
> >>From my own sample, which unfortunately did not get posted on the list
> >>yet (I sent it from another email address, and it is waiting moderator
> >>approval because it is from a non-subscribed address), I think that
> >>this might include as many as 20% of our current images.
> >>    
> >
> >If so, then I'm wrong and will immediately change my position.  If 20%
> >of our images are from questionable sources, we are in a bad position
> >and need to correct it quickly.  Given the large number of images that
> >have no provenance information, perhaps we should rearrange the
> >image-upload process to require uploaders to click through a "where
> >did you get this image" page before they enter the file name.  Of
> >course that won't force anyone to enter anything, but it will make it
> >more obvious that they should.
> >
> I would say it was greater than 20% from the tagging work we've been doing.
> 
> However: the figures have a bias as we have been tagging our own images. 
> I have uploaded 193 images, Morwen has uploaded 1224 (a thousand of 
> which are her GFDL maps). When I was tagging yesterday's image upload 
> log over 50% had no info.
> 
> I hope this helps.

Here are my results (the original email is still waited to be accepted).
My sample were the first 100 images (alphabetically) of which the name
starts with E, removing images that were different format copies of one
already in the set, and removing all but one images from two series.

Of these 100:
GNU/FDL: 18 images (6 photographs by uploader, 4 drawings by uploader, 8 
    drawings obviously by uploader though not so tagged)
Public domain: 37 images (19 US government, 3 clearly US government though
    not so tagged, 8 public domain because of age, 4 public domain because of
    age though not so tagged, 1 mentioned as 'free of copyright' with source,
    2 no copyright mention, but source site mentions, which says its material
    is PD unless explicitly stated differently)

==making 55 images clearly without problem==

Fair use: 4 images (only images counted which were explicitly tagged as
    'fair use')
Non-commercial: 2 images
Specific permission for Wikipedia: 2 images
Possibly free of copyright: 9 images (1 logo, 2 coats of arms, 4 flags, 1
    Go position, 1 very basic outline-style map)
Uncategorizable: 1 sound file (work played is out of copyright, but nothing
    was stated about the probable copyright on the recording itself)

==making 18 images with possible problems==

Untagged, but possibly ok: 6 images (1 might be an unmarked own photo, 2
    an unmarked own drawing, 2 likely to be from NASA, 1 possible not
    copyrightable)
Untagged, no reason to believe ok for any reason other than fair use: 21
    images

==making 27 untagged images, obvious own drawings not included==

Andre Engels





More information about the Wikipedia-l mailing list