Come across another few thousand edits of copy and paste violations again today. These have occurred over more than a year. It is wearing me out. Really what is the point on collaborating on Wikipedia if it is simply a copyright violation. We need a solution and one has been proposed here a couple of years ago https://en.wikipedia.org/wiki/Wikipedia:Turnitin
We now need programmers to carry it out. The Wiki Education Foundation has expressed interest. We will need support from the foundation as this software will likely need to mesh closely with edits as they come in. I am willing to offer $5,000 dollars Canadian (almost the same as American) for a working solution that tags potential copyright issues in near real time with a greater than 90% accuracy. It is to function on at least all medical and pharmacology articles but I would not complain if it worked on all of Wikipedia. The WMF is free to apply.
This is one of the best ideas Ive read on here!
Date: Sun, 20 Jul 2014 20:00:28 -0600 From: jmh649@gmail.com To: wikimedia-l@lists.wikimedia.org; eloquence@gmail.com; fschulenburg@wikimedia.org; ladsgroup@gmail.com; jorlowitz@gmail.com; madman.enwiki@gmail.com; west.andrew.g@gmail.com Subject: [Wikimedia-l] Catching copy and pasting early
Come across another few thousand edits of copy and paste violations again today. These have occurred over more than a year. It is wearing me out. Really what is the point on collaborating on Wikipedia if it is simply a copyright violation. We need a solution and one has been proposed here a couple of years ago https://en.wikipedia.org/wiki/Wikipedia:Turnitin
We now need programmers to carry it out. The Wiki Education Foundation has expressed interest. We will need support from the foundation as this software will likely need to mesh closely with edits as they come in. I am willing to offer $5,000 dollars Canadian (almost the same as American) for a working solution that tags potential copyright issues in near real time with a greater than 90% accuracy. It is to function on at least all medical and pharmacology articles but I would not complain if it worked on all of Wikipedia. The WMF is free to apply.
-- James Heilman MD, CCFP-EM, Wikipedian
The Wikipedia Open Textbook of Medicine www.opentextbookofmedicine.com _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
It should be relatively easy to catch a significant percentage of those copyright violations with the assistance of automated search tools. The trick is to do it at a large scale in near-realtime, which might require some computationally intensive and bandwidth intensive work. James, can I suggest that you take this discussion to Wiki-Research-l? There are a number of ways that the copyright violation problem could be addressed and I think this would be a good subject for discussion on that list, or at Wikimania. Depending on how the discussion on Research goes, it might be good to invite some dev or tech ops people to participate in the discussion as well.
Pine
On Sun, Jul 20, 2014 at 7:05 PM, Leigh Thelmadatter osamadre@hotmail.com wrote:
This is one of the best ideas Ive read on here!
Date: Sun, 20 Jul 2014 20:00:28 -0600 From: jmh649@gmail.com To: wikimedia-l@lists.wikimedia.org; eloquence@gmail.com;
fschulenburg@wikimedia.org; ladsgroup@gmail.com; jorlowitz@gmail.com; madman.enwiki@gmail.com; west.andrew.g@gmail.com
Subject: [Wikimedia-l] Catching copy and pasting early
Come across another few thousand edits of copy and paste violations again today. These have occurred over more than a year. It is wearing me out. Really what is the point on collaborating on Wikipedia if it is simply a copyright violation. We need a solution and one has been proposed here a couple of years ago https://en.wikipedia.org/wiki/Wikipedia:Turnitin
We now need programmers to carry it out. The Wiki Education Foundation
has
expressed interest. We will need support from the foundation as this software will likely need to mesh closely with edits as they come in. I
am
willing to offer $5,000 dollars Canadian (almost the same as American)
for
a working solution that tags potential copyright issues in near real time with a greater than 90% accuracy. It is to function on at least all
medical
and pharmacology articles but I would not complain if it worked on all of Wikipedia. The WMF is free to apply.
-- James Heilman MD, CCFP-EM, Wikipedian
The Wikipedia Open Textbook of Medicine www.opentextbookofmedicine.com _______________________________________________ Wikimedia-l mailing list, guidelines at:
https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines
Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l,
mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Isn't that what Corenbot does/did? I always found it very confusing though whenever I ran into it, and the false positives are huge (so many sites copy Wikimedia content these days)
On Mon, Jul 21, 2014 at 9:11 AM, Pine W wiki.pine@gmail.com wrote:
It should be relatively easy to catch a significant percentage of those copyright violations with the assistance of automated search tools. The trick is to do it at a large scale in near-realtime, which might require some computationally intensive and bandwidth intensive work. James, can I suggest that you take this discussion to Wiki-Research-l? There are a number of ways that the copyright violation problem could be addressed and I think this would be a good subject for discussion on that list, or at Wikimania. Depending on how the discussion on Research goes, it might be good to invite some dev or tech ops people to participate in the discussion as well.
Pine
On Sun, Jul 20, 2014 at 7:05 PM, Leigh Thelmadatter osamadre@hotmail.com wrote:
This is one of the best ideas Ive read on here!
Date: Sun, 20 Jul 2014 20:00:28 -0600 From: jmh649@gmail.com To: wikimedia-l@lists.wikimedia.org; eloquence@gmail.com;
fschulenburg@wikimedia.org; ladsgroup@gmail.com; jorlowitz@gmail.com; madman.enwiki@gmail.com; west.andrew.g@gmail.com
Subject: [Wikimedia-l] Catching copy and pasting early
Come across another few thousand edits of copy and paste violations
again
today. These have occurred over more than a year. It is wearing me out. Really what is the point on collaborating on Wikipedia if it is simply
a
copyright violation. We need a solution and one has been proposed here
a
couple of years ago https://en.wikipedia.org/wiki/Wikipedia:Turnitin
We now need programmers to carry it out. The Wiki Education Foundation
has
expressed interest. We will need support from the foundation as this software will likely need to mesh closely with edits as they come in. I
am
willing to offer $5,000 dollars Canadian (almost the same as American)
for
a working solution that tags potential copyright issues in near real
time
with a greater than 90% accuracy. It is to function on at least all
medical
and pharmacology articles but I would not complain if it worked on all
of
Wikipedia. The WMF is free to apply.
-- James Heilman MD, CCFP-EM, Wikipedian
The Wikipedia Open Textbook of Medicine www.opentextbookofmedicine.com _______________________________________________ Wikimedia-l mailing list, guidelines at:
https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines
Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l,
mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
This has nothing to do with wiki-research-l, it's mostly about who can pay for the API keys. https://wikitech.wikimedia.org/wiki/Web_search
Nemo
hi james, this sounds interesting. the only obstacle i see is that neither the service nor the software nor the database are open. by using turnitin we would give up on the principle of opening contents, at least in this area. this might be perceived bad for opening content, editor retention, and gaining new editors.
from a turnitin standpoint this might be marketing until they are well known. as soon their goals are reached there is danger this will be discontinued or the terms changed to be not acceptable any more.
technical and license obstacles do exist. to properly check wikipedia editors would need access to restricted content. either content is copied somewhere in the open, permission granted for all, or in the worst case, a restricted number of editors.
not to forget is potential link spam. turnitin would put links to a site not in control of the movement all over the place. they could without problems put other content behind the links at any time.
but, i do think one could find a less risky and more open way to do check for plagiarism. one might find better ways to access wikipedia dumps, better access to recent changes, permission to automatically report back issues, nitification of contributors and admins. all in a kind of standardized way (i.e. well defined interface), and most important, usable by everybody, for free.
rupert Am 21.07.2014 04:05 schrieb "Leigh Thelmadatter" osamadre@hotmail.com:
This is one of the best ideas Ive read on here!
Date: Sun, 20 Jul 2014 20:00:28 -0600 From: jmh649@gmail.com To: wikimedia-l@lists.wikimedia.org; eloquence@gmail.com;
fschulenburg@wikimedia.org; ladsgroup@gmail.com; jorlowitz@gmail.com; madman.enwiki@gmail.com; west.andrew.g@gmail.com
Subject: [Wikimedia-l] Catching copy and pasting early
Come across another few thousand edits of copy and paste violations again today. These have occurred over more than a year. It is wearing me out. Really what is the point on collaborating on Wikipedia if it is simply a copyright violation. We need a solution and one has been proposed here a couple of years ago https://en.wikipedia.org/wiki/Wikipedia:Turnitin
We now need programmers to carry it out. The Wiki Education Foundation
has
expressed interest. We will need support from the foundation as this software will likely need to mesh closely with edits as they come in. I
am
willing to offer $5,000 dollars Canadian (almost the same as American)
for
a working solution that tags potential copyright issues in near real time with a greater than 90% accuracy. It is to function on at least all
medical
and pharmacology articles but I would not complain if it worked on all of Wikipedia. The WMF is free to apply.
-- James Heilman MD, CCFP-EM, Wikipedian
The Wikipedia Open Textbook of Medicine www.opentextbookofmedicine.com _______________________________________________ Wikimedia-l mailing list, guidelines at:
https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines
Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l,
mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
I dont know if we have to work with Turn-it-in per se. Given that most copy-paste is from digital sources, we could start with a program/tool to search the web, which could lessen the burden on editors significantly by itself. (Speaking as a writing teacher, I can tell you this is how I find proof of the majority of plagiarism instances... people are often not-too-smart when they are cheating).
If this works, I bet we could then work with some of the databases to allow the access needed to check, as it would require only excepts from the text, not the entire text... enough to justify a revert.
Granted this would not catch everything (no system does) but it would be a very good step in the right direction.
Date: Mon, 21 Jul 2014 10:15:44 +0200 From: rupert.thurner@gmail.com To: wikimedia-l@lists.wikimedia.org Subject: Re: [Wikimedia-l] Catching copy and pasting early
hi james, this sounds interesting. the only obstacle i see is that neither the service nor the software nor the database are open. by using turnitin we would give up on the principle of opening contents, at least in this area. this might be perceived bad for opening content, editor retention, and gaining new editors.
from a turnitin standpoint this might be marketing until they are well known. as soon their goals are reached there is danger this will be discontinued or the terms changed to be not acceptable any more.
technical and license obstacles do exist. to properly check wikipedia editors would need access to restricted content. either content is copied somewhere in the open, permission granted for all, or in the worst case, a restricted number of editors.
not to forget is potential link spam. turnitin would put links to a site not in control of the movement all over the place. they could without problems put other content behind the links at any time.
but, i do think one could find a less risky and more open way to do check for plagiarism. one might find better ways to access wikipedia dumps, better access to recent changes, permission to automatically report back issues, nitification of contributors and admins. all in a kind of standardized way (i.e. well defined interface), and most important, usable by everybody, for free.
rupert Am 21.07.2014 04:05 schrieb "Leigh Thelmadatter" osamadre@hotmail.com:
This is one of the best ideas Ive read on here!
Date: Sun, 20 Jul 2014 20:00:28 -0600 From: jmh649@gmail.com To: wikimedia-l@lists.wikimedia.org; eloquence@gmail.com;
fschulenburg@wikimedia.org; ladsgroup@gmail.com; jorlowitz@gmail.com; madman.enwiki@gmail.com; west.andrew.g@gmail.com
Subject: [Wikimedia-l] Catching copy and pasting early
Come across another few thousand edits of copy and paste violations again today. These have occurred over more than a year. It is wearing me out. Really what is the point on collaborating on Wikipedia if it is simply a copyright violation. We need a solution and one has been proposed here a couple of years ago https://en.wikipedia.org/wiki/Wikipedia:Turnitin
We now need programmers to carry it out. The Wiki Education Foundation
has
expressed interest. We will need support from the foundation as this software will likely need to mesh closely with edits as they come in. I
am
willing to offer $5,000 dollars Canadian (almost the same as American)
for
a working solution that tags potential copyright issues in near real time with a greater than 90% accuracy. It is to function on at least all
medical
and pharmacology articles but I would not complain if it worked on all of Wikipedia. The WMF is free to apply.
-- James Heilman MD, CCFP-EM, Wikipedian
The Wikipedia Open Textbook of Medicine www.opentextbookofmedicine.com _______________________________________________ Wikimedia-l mailing list, guidelines at:
https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines
Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l,
mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
wikimedia-l@lists.wikimedia.org