Mailing List Archive: Attribution made cleaner?

Attribution made cleaner?

shimgray at gmail

Feb 2, 2009, 12:24 PM

Post #1 of 22 (2238 views)

So, whatever way we decide to go with licenses or attribution
requirements when this debate has settled, at some point our
prospective reader will find themselves confronted with a long list of
names, whether printed on the page or at the end of a URL or
steganographically encoded into the site logo. :-)

On this list, a minority will be real names ("John Smith"); the rest,
if we discount the thousand variants on "anonymous" via our IP
editors, are pseudonyms ("WikiUser") or modified names
("JohnSmith78").

In some cases, users adopt pseudonyms out of a desire for privacy, but
in many cases, it doesn't signify much more than a simple decision
that a username is a lot easier to work with internally, or a general
habit of using some kind of nickname online... or the fact that "John
Smith" was taken. And many of *those* people would, no doubt, prefer
to be credited by a real name (or at least a real-sounding nom de
plume...). Similarly, some of those using pseudonyms who don't want to
use real names, may prefer a different pseudonym... etc, etc, etc.

It would be helpful to figure out some way of (automatically) being
able to have a given username "translate" into a different name when a
list of credits is generated - we would have a list which better
reflects the attribution wishes of our users, and one which looks a
little "neater" for the reuser to put in their Respectable Scholarly
Publication. Win-win situation.

So how could we do it? At a rough sketch, I'm envisaging:

* each user has a "credit" field which they can (optionally!) set
through preferences

* when we generate the list of contributors to an article, in whatever
way we end up deciding to do that, the system can be set to read off
this "credit name" rather than simply using the normal internal
username, if one is available.

I note that MediaWiki already has a user_real_name field - could we
use it for this sort of purpose? Would this be technically practical?

--
- Andrew Gray
andrew.gray@dunelm.org.uk

_______________________________________________
foundation-l mailing list
foundation-l@lists.wikimedia.org
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l

Re: Attribution made cleaner? [ In reply to ]

dgerard at gmail

Feb 2, 2009, 1:20 PM

Post #2 of 22 (2170 views)

2009/2/2 Andrew Gray <shimgray@gmail.com>:

> It would be helpful to figure out some way of (automatically) being
> able to have a given username "translate" into a different name when a
> list of credits is generated - we would have a list which better
> reflects the attribution wishes of our users, and one which looks a
> little "neater" for the reuser to put in their Respectable Scholarly
> Publication. Win-win situation.
> So how could we do it? At a rough sketch, I'm envisaging:
> * each user has a "credit" field which they can (optionally!) set
> through preferences
> I note that MediaWiki already has a user_real_name field - could we
> use it for this sort of purpose? Would this be technically practical?

Default MediaWiki has this as an option anyone who cares can set. On
our work intranet wiki, it specifically says it's for giving credit
under. I couldn't find it in my en:wp preferences, though ... what
happened to it?

- d.

_______________________________________________
foundation-l mailing list
foundation-l@lists.wikimedia.org
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l

Re: Attribution made cleaner? [ In reply to ]

innocentkiller at gmail

Feb 2, 2009, 1:22 PM

Post #3 of 22 (2173 views)

On Mon, Feb 2, 2009 at 4:20 PM, David Gerard <dgerard@gmail.com> wrote:

> 2009/2/2 Andrew Gray <shimgray@gmail.com>:
>
> > It would be helpful to figure out some way of (automatically) being
> > able to have a given username "translate" into a different name when a
> > list of credits is generated - we would have a list which better
> > reflects the attribution wishes of our users, and one which looks a
> > little "neater" for the reuser to put in their Respectable Scholarly
> > Publication. Win-win situation.
> > So how could we do it? At a rough sketch, I'm envisaging:
> > * each user has a "credit" field which they can (optionally!) set
> > through preferences
> > I note that MediaWiki already has a user_real_name field - could we
> > use it for this sort of purpose? Would this be technically practical?
>
>
> Default MediaWiki has this as an option anyone who cares can set. On
> our work intranet wiki, it specifically says it's for giving credit
> under. I couldn't find it in my en:wp preferences, though ... what
> happened to it?
>
>
> - d.
>

It's disabled on WMF wikis afaik.

-Chad
_______________________________________________
foundation-l mailing list
foundation-l@lists.wikimedia.org
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l

Re: Attribution made cleaner? [ In reply to ]

neuro.wikipedia at googlemail

Feb 2, 2009, 1:39 PM

Post #4 of 22 (2180 views)

Any reason why? I can't seem to find anything on it.

- Chris

On Mon, Feb 2, 2009 at 9:22 PM, Chad <innocentkiller@gmail.com> wrote:

> On Mon, Feb 2, 2009 at 4:20 PM, David Gerard <dgerard@gmail.com> wrote:
>
> > 2009/2/2 Andrew Gray <shimgray@gmail.com>:
> >
> > > It would be helpful to figure out some way of (automatically) being
> > > able to have a given username "translate" into a different name when a
> > > list of credits is generated - we would have a list which better
> > > reflects the attribution wishes of our users, and one which looks a
> > > little "neater" for the reuser to put in their Respectable Scholarly
> > > Publication. Win-win situation.
> > > So how could we do it? At a rough sketch, I'm envisaging:
> > > * each user has a "credit" field which they can (optionally!) set
> > > through preferences
> > > I note that MediaWiki already has a user_real_name field - could we
> > > use it for this sort of purpose? Would this be technically practical?
> >
> >
> > Default MediaWiki has this as an option anyone who cares can set. On
> > our work intranet wiki, it specifically says it's for giving credit
> > under. I couldn't find it in my en:wp preferences, though ... what
> > happened to it?
> >
> >
> > - d.
> >
>
> It's disabled on WMF wikis afaik.
>
> -Chad
> _______________________________________________
> foundation-l mailing list
> foundation-l@lists.wikimedia.org
> Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
>
_______________________________________________
foundation-l mailing list
foundation-l@lists.wikimedia.org
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l

Re: Attribution made cleaner? [ In reply to ]

removed at example

Feb 2, 2009, 1:41 PM

Post #5 of 22 (2180 views)

Following this line of reasoning in both directions, many users who
contribute to an encyclopedia that "anyone can edit" may not want their name
reprinted on every conceivable medium that their contributions could be
replicated on. In other words, many users probably don't care even a little
bit about the attribution requirements of the CC-BY-SA. They contribute
under the implicit assumption that their work is in the public domain. An
argument can be made that printing their username all over the place is an
invasion of their privacy, since with a bit of Googling its often possible
to relate that to their real identity. I've got a collection of references
to algorithms that show its possible to link users across social networking
sites. Some of these methods would apply to a user's edits as well.

My honest intrepretation of the 5 authors or less rule else a hyperlink is
that it's silly.

On Mon, Feb 2, 2009 at 1:24 PM, Andrew Gray <shimgray@gmail.com> wrote:

> So, whatever way we decide to go with licenses or attribution
> requirements when this debate has settled, at some point our
> prospective reader will find themselves confronted with a long list of
> names, whether printed on the page or at the end of a URL or
> steganographically encoded into the site logo. :-)
>
> On this list, a minority will be real names ("John Smith"); the rest,
> if we discount the thousand variants on "anonymous" via our IP
> editors, are pseudonyms ("WikiUser") or modified names
> ("JohnSmith78").
>
> In some cases, users adopt pseudonyms out of a desire for privacy, but
> in many cases, it doesn't signify much more than a simple decision
> that a username is a lot easier to work with internally, or a general
> habit of using some kind of nickname online... or the fact that "John
> Smith" was taken. And many of *those* people would, no doubt, prefer
> to be credited by a real name (or at least a real-sounding nom de
> plume...). Similarly, some of those using pseudonyms who don't want to
> use real names, may prefer a different pseudonym... etc, etc, etc.
>
> It would be helpful to figure out some way of (automatically) being
> able to have a given username "translate" into a different name when a
> list of credits is generated - we would have a list which better
> reflects the attribution wishes of our users, and one which looks a
> little "neater" for the reuser to put in their Respectable Scholarly
> Publication. Win-win situation.
>
> So how could we do it? At a rough sketch, I'm envisaging:
>
> * each user has a "credit" field which they can (optionally!) set
> through preferences
>
> * when we generate the list of contributors to an article, in whatever
> way we end up deciding to do that, the system can be set to read off
> this "credit name" rather than simply using the normal internal
> username, if one is available.
>
> I note that MediaWiki already has a user_real_name field - could we
> use it for this sort of purpose? Would this be technically practical?
>
> --
> - Andrew Gray
> andrew.gray@dunelm.org.uk
>
> _______________________________________________
> foundation-l mailing list
> foundation-l@lists.wikimedia.org
> Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
>
_______________________________________________
foundation-l mailing list
foundation-l@lists.wikimedia.org
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l

Re: Attribution made cleaner? [ In reply to ]

removed at example

Feb 2, 2009, 1:45 PM

Post #6 of 22 (2173 views)

Also, has it been discussed that the minimum number of authors rule
effectually only applies to stubs and some starts? Even these have often
been edited by many more than a handful of bots.

It would be useful to have an SQL query that output the number of articles
on en.wp with more than a handful of articles. It's probably fairly small.

So the effectual rule is that attribution is done by a hyperlink.

On Mon, Feb 2, 2009 at 2:41 PM, Brian <Brian.Mingus@colorado.edu> wrote:

> Following this line of reasoning in both directions, many users who
> contribute to an encyclopedia that "anyone can edit" may not want their name
> reprinted on every conceivable medium that their contributions could be
> replicated on. In other words, many users probably don't care even a little
> bit about the attribution requirements of the CC-BY-SA. They contribute
> under the implicit assumption that their work is in the public domain. An
> argument can be made that printing their username all over the place is an
> invasion of their privacy, since with a bit of Googling its often possible
> to relate that to their real identity. I've got a collection of references
> to algorithms that show its possible to link users across social networking
> sites. Some of these methods would apply to a user's edits as well.
>
> My honest intrepretation of the 5 authors or less rule else a hyperlink is
> that it's silly.
>
> On Mon, Feb 2, 2009 at 1:24 PM, Andrew Gray <shimgray@gmail.com> wrote:
>
>> So, whatever way we decide to go with licenses or attribution
>> requirements when this debate has settled, at some point our
>> prospective reader will find themselves confronted with a long list of
>> names, whether printed on the page or at the end of a URL or
>> steganographically encoded into the site logo. :-)
>>
>> On this list, a minority will be real names ("John Smith"); the rest,
>> if we discount the thousand variants on "anonymous" via our IP
>> editors, are pseudonyms ("WikiUser") or modified names
>> ("JohnSmith78").
>>
>> In some cases, users adopt pseudonyms out of a desire for privacy, but
>> in many cases, it doesn't signify much more than a simple decision
>> that a username is a lot easier to work with internally, or a general
>> habit of using some kind of nickname online... or the fact that "John
>> Smith" was taken. And many of *those* people would, no doubt, prefer
>> to be credited by a real name (or at least a real-sounding nom de
>> plume...). Similarly, some of those using pseudonyms who don't want to
>> use real names, may prefer a different pseudonym... etc, etc, etc.
>>
>> It would be helpful to figure out some way of (automatically) being
>> able to have a given username "translate" into a different name when a
>> list of credits is generated - we would have a list which better
>> reflects the attribution wishes of our users, and one which looks a
>> little "neater" for the reuser to put in their Respectable Scholarly
>> Publication. Win-win situation.
>>
>> So how could we do it? At a rough sketch, I'm envisaging:
>>
>> * each user has a "credit" field which they can (optionally!) set
>> through preferences
>>
>> * when we generate the list of contributors to an article, in whatever
>> way we end up deciding to do that, the system can be set to read off
>> this "credit name" rather than simply using the normal internal
>> username, if one is available.
>>
>> I note that MediaWiki already has a user_real_name field - could we
>> use it for this sort of purpose? Would this be technically practical?
>>
>> --
>> - Andrew Gray
>> andrew.gray@dunelm.org.uk
>>
>> _______________________________________________
>> foundation-l mailing list
>> foundation-l@lists.wikimedia.org
>> Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
>>
>
>
_______________________________________________
foundation-l mailing list
foundation-l@lists.wikimedia.org
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l

Re: Attribution made cleaner? [ In reply to ]

removed at example

Feb 2, 2009, 1:45 PM

Post #7 of 22 (2188 views)

* more than a handful of authors

On Mon, Feb 2, 2009 at 2:45 PM, Brian <Brian.Mingus@colorado.edu> wrote:

> Also, has it been discussed that the minimum number of authors rule
> effectually only applies to stubs and some starts? Even these have often
> been edited by many more than a handful of bots.
>
> It would be useful to have an SQL query that output the number of articles
> on en.wp with more than a handful of articles. It's probably fairly small.
>
> So the effectual rule is that attribution is done by a hyperlink.
>
>
> On Mon, Feb 2, 2009 at 2:41 PM, Brian <Brian.Mingus@colorado.edu> wrote:
>
>> Following this line of reasoning in both directions, many users who
>> contribute to an encyclopedia that "anyone can edit" may not want their name
>> reprinted on every conceivable medium that their contributions could be
>> replicated on. In other words, many users probably don't care even a little
>> bit about the attribution requirements of the CC-BY-SA. They contribute
>> under the implicit assumption that their work is in the public domain. An
>> argument can be made that printing their username all over the place is an
>> invasion of their privacy, since with a bit of Googling its often possible
>> to relate that to their real identity. I've got a collection of references
>> to algorithms that show its possible to link users across social networking
>> sites. Some of these methods would apply to a user's edits as well.
>>
>> My honest intrepretation of the 5 authors or less rule else a hyperlink is
>> that it's silly.
>>
>> On Mon, Feb 2, 2009 at 1:24 PM, Andrew Gray <shimgray@gmail.com> wrote:
>>
>>> So, whatever way we decide to go with licenses or attribution
>>> requirements when this debate has settled, at some point our
>>> prospective reader will find themselves confronted with a long list of
>>> names, whether printed on the page or at the end of a URL or
>>> steganographically encoded into the site logo. :-)
>>>
>>> On this list, a minority will be real names ("John Smith"); the rest,
>>> if we discount the thousand variants on "anonymous" via our IP
>>> editors, are pseudonyms ("WikiUser") or modified names
>>> ("JohnSmith78").
>>>
>>> In some cases, users adopt pseudonyms out of a desire for privacy, but
>>> in many cases, it doesn't signify much more than a simple decision
>>> that a username is a lot easier to work with internally, or a general
>>> habit of using some kind of nickname online... or the fact that "John
>>> Smith" was taken. And many of *those* people would, no doubt, prefer
>>> to be credited by a real name (or at least a real-sounding nom de
>>> plume...). Similarly, some of those using pseudonyms who don't want to
>>> use real names, may prefer a different pseudonym... etc, etc, etc.
>>>
>>> It would be helpful to figure out some way of (automatically) being
>>> able to have a given username "translate" into a different name when a
>>> list of credits is generated - we would have a list which better
>>> reflects the attribution wishes of our users, and one which looks a
>>> little "neater" for the reuser to put in their Respectable Scholarly
>>> Publication. Win-win situation.
>>>
>>> So how could we do it? At a rough sketch, I'm envisaging:
>>>
>>> * each user has a "credit" field which they can (optionally!) set
>>> through preferences
>>>
>>> * when we generate the list of contributors to an article, in whatever
>>> way we end up deciding to do that, the system can be set to read off
>>> this "credit name" rather than simply using the normal internal
>>> username, if one is available.
>>>
>>> I note that MediaWiki already has a user_real_name field - could we
>>> use it for this sort of purpose? Would this be technically practical?
>>>
>>> --
>>> - Andrew Gray
>>> andrew.gray@dunelm.org.uk
>>>
>>> _______________________________________________
>>> foundation-l mailing list
>>> foundation-l@lists.wikimedia.org
>>> Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
>>>
>>
>>
>
_______________________________________________
foundation-l mailing list
foundation-l@lists.wikimedia.org
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l

Re: Attribution made cleaner? [ In reply to ]

smolensk at eunet

Feb 2, 2009, 2:29 PM

Post #8 of 22 (2180 views)

On Monday 02 February 2009 22:41:37 Brian wrote:
> Following this line of reasoning in both directions, many users who
> contribute to an encyclopedia that "anyone can edit" may not want their
> name reprinted on every conceivable medium that their contributions could
> be replicated on. In other words, many users probably don't care even a
> little bit about the attribution requirements of the CC-BY-SA. They
> contribute under the implicit assumption that their work is in the public

Do you have anything to back your claims with?

> domain. An argument can be made that printing their username all over the
> place is an invasion of their privacy, since with a bit of Googling its
> often possible to relate that to their real identity. I've got a collection

Are you arguing that we should not have page histories?

_______________________________________________
foundation-l mailing list
foundation-l@lists.wikimedia.org
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l

Re: Attribution made cleaner? [ In reply to ]

removed at example

Feb 2, 2009, 2:45 PM

Post #9 of 22 (2180 views)

On Mon, Feb 2, 2009 at 3:29 PM, Nikola Smolenski <smolensk@eunet.yu> wrote:

> On Monday 02 February 2009 22:41:37 Brian wrote:
> > Following this line of reasoning in both directions, many users who
> > contribute to an encyclopedia that "anyone can edit" may not want their
> > name reprinted on every conceivable medium that their contributions could
> > be replicated on. In other words, many users probably don't care even a
> > little bit about the attribution requirements of the CC-BY-SA. They
> > contribute under the implicit assumption that their work is in the public
>
> Do you have anything to back your claims with?

The operating assumption is that the average pseudo-anonymous user to a
wikimedia project understands and/or cares about the licensing issues and
realizes their name will be printed everywhere that the text they contribute
is printed. Do you have any evidence that this is true? That the average
pseudo-anonymous contributor has a fairly sophisticated understanding of
copyright? Otherwise its quite similar to the ToS at the bottom of every web
page you visit, which you supposedly implicitly agree to, but which you
rarely to never read and is actually a legal grey area.

> > domain. An argument can be made that printing their username all over the
> > place is an invasion of their privacy, since with a bit of Googling its
> > often possible to relate that to their real identity. I've got a
> collection
>
> Are you arguing that we should not have page histories?
>

Just that I am skeptical that people realize their pseudonyms will be
printed on potentially any medium and that they are further aware that this
pseudonym can be linked to their real identity.

The point is that listing the authors is a silly clause.
_______________________________________________
foundation-l mailing list
foundation-l@lists.wikimedia.org
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l

Re: Attribution made cleaner? [ In reply to ]

andrew.gray at dunelm

Feb 2, 2009, 2:54 PM

Post #10 of 22 (2178 views)

2009/2/2 Brian <Brian.Mingus@colorado.edu>:

> Just that I am skeptical that people realize their pseudonyms will be
> printed on potentially any medium and that they are further aware that this
> pseudonym can be linked to their real identity.

I can't say I agree with your general thrust here - I think that if
people contribute to a massively open project, well, they have to
accept "massively open". Bending over backwards to retroactively
provide anonymity gets impractical fast.

However, this proposal could allow an effective opt-out from any form
of downstream attribution - some kind of "NOCREDIT" magic word,
perhaps. This would neatly sidestep the worry of people not wanting
credited downstream...

--
- Andrew Gray
andrew.gray@dunelm.org.uk

_______________________________________________
foundation-l mailing list
foundation-l@lists.wikimedia.org
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l

Re: Attribution made cleaner? [ In reply to ]

innocentkiller at gmail

Feb 2, 2009, 3:04 PM

Post #11 of 22 (2180 views)

On Mon, Feb 2, 2009 at 4:39 PM, Chris Down
<neuro.wikipedia@googlemail.com>wrote:

> Any reason why? I can't seem to find anything on it.
>
> - Chris
>
> On Mon, Feb 2, 2009 at 9:22 PM, Chad <innocentkiller@gmail.com> wrote:
> > It's disabled on WMF wikis afaik.
> >
> > -Chad
>

Not sure. There's no comment in the configuration files, but it
($wgAllowRealName) is set to false for all WMF wikis. You'd
have to ask someone official for the reason. Privacy issues?

-Chad
_______________________________________________
foundation-l mailing list
foundation-l@lists.wikimedia.org
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l

Re: Attribution made cleaner? [ In reply to ]

removed at example

Feb 2, 2009, 3:13 PM

Post #12 of 22 (2176 views)

I advocate a much more flexible attribution scheme than listing the authors
or printing a url to the history page. I think a simple (Wikipedia) is a
sufficient attribution for text. If you have the text it is trivial to find
the original author of that text. It's not so trivial with images, but a
link to the history page of an image can be embedded in its metadata.

On Mon, Feb 2, 2009 at 3:54 PM, Andrew Gray <andrew.gray@dunelm.org.uk>wrote:

> 2009/2/2 Brian <Brian.Mingus@colorado.edu>:
>
> > Just that I am skeptical that people realize their pseudonyms will be
> > printed on potentially any medium and that they are further aware that
> this
> > pseudonym can be linked to their real identity.
>
> I can't say I agree with your general thrust here - I think that if
> people contribute to a massively open project, well, they have to
> accept "massively open". Bending over backwards to retroactively
> provide anonymity gets impractical fast.
>
> However, this proposal could allow an effective opt-out from any form
> of downstream attribution - some kind of "NOCREDIT" magic word,
> perhaps. This would neatly sidestep the worry of people not wanting
> credited downstream...
>
> --
> - Andrew Gray
> andrew.gray@dunelm.org.uk
>
> _______________________________________________
> foundation-l mailing list
> foundation-l@lists.wikimedia.org
> Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
>
_______________________________________________
foundation-l mailing list
foundation-l@lists.wikimedia.org
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l

Re: Attribution made cleaner? [ In reply to ]

rarohde at gmail

Feb 2, 2009, 3:13 PM

Post #13 of 22 (2175 views)

On Mon, Feb 2, 2009 at 1:45 PM, Brian <Brian.Mingus@colorado.edu> wrote:
> Also, has it been discussed that the minimum number of authors rule
> effectually only applies to stubs and some starts? Even these have often
> been edited by many more than a handful of bots.
>
> It would be useful to have an SQL query that output the number of articles
> on en.wp with more than a handful of articles. It's probably fairly small.
>
> So the effectual rule is that attribution is done by a hyperlink.

I can't speak for enwiki, but I did a similar thing for ruwiki (WMF's
10th largest wiki), and a very substantial fraction of articles had 5
or fewer editors. If one condenses IP editors to something like " and
3 anonymous editors" (as some people have suggested) rather than
listing each IP, then the number of articles with 5 or fewer editors
is roughly half of all articles.

Of course there is also a significant fraction of articles with dozens
of editors, which probably includes most of the well-developed and
easily reusable content.

Enwiki, with its larger articles and higher number of edits per
articles, probably has a lower percentage of articles that would fall
under the 5 or fewer rule, but I don't think it would be trivial.

-Robert Rohde

_______________________________________________
foundation-l mailing list
foundation-l@lists.wikimedia.org
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l

Re: Attribution made cleaner? [ In reply to ]

meta.sj at gmail

Feb 2, 2009, 3:14 PM

Post #14 of 22 (2184 views)

I usually agree with the Mingus, but:

On Mon, Feb 2, 2009 at 5:45 PM, Brian <Brian.Mingus@colorado.edu> wrote:
> On Mon, Feb 2, 2009 at 3:29 PM, Nikola Smolenski <smolensk@eunet.yu> wrote:
>
>> On Monday 02 February 2009 22:41:37 Brian wrote:
>> > Following this line of reasoning in both directions, many users who
>> > contribute to an encyclopedia that "anyone can edit" may not want their
>> > name reprinted on every conceivable medium that their contributions could
>> > be replicated on.

True.
One option should be "Include attribution for non-minor edits to all
articles? Y/N".
Another should be "Name for attribution" -- this is not necessarily a
Real Name, it can be a writing alias, which is again different from a
username.

>> < In other words, many users probably don't care even a
>> > little bit about the attribution requirements of the CC-BY-SA. They
>> > contribute under the implicit assumption that their work is in the public [domain]

Some don't. Some do.
Another option should be "Default license for your edits, in addition
to CC-BY-SA : Public domain" (and perhaps others) for people who want
their work to continue to be available to projects/efforts using other
licenses. A good database would track licensing by revision, in
addition to a shared default license.

> The operating assumption is that the average pseudo-anonymous user to a
> wikimedia project understands and/or cares about the licensing issues and
> realizes their name will be printed everywhere that the text they contribute
> is printed.

I wouldn't recommend changing from the current mechanism to one that
blatantly shows all editor's names on every page whever displayed.
But I would aboslutely lay the groudnwork for prpoer attribution by
aggregating and caching the complete author list, without duplicates,
in some reasonable order; tracking and displaying a non-nick name for
attribution, and having the full author list with some basic metadata
about the contributions of each author at most one click away from the
article itself.

Calling the current history interface "attribution by hyperlink" is misleading.

>> Are you arguing that we should not have page histories?
>
> Just that I am skeptical that people realize their pseudonyms will be
> printed on potentially any medium and that they are further aware that this
> pseudonym can be linked to their real identity.

Then make that clear in user preferences and when people sign up for
an account. It's an important part of joining the community. We
could start by turning off the "include me in attribution" by default
in userperfs and seeing what happens -- people who literally want
'full' attribution could still dredge through page histories.

> The point is that listing the authors is a silly clause.

I couldn't disagree more.

SJ

_______________________________________________
foundation-l mailing list
foundation-l@lists.wikimedia.org
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l

Re: Attribution made cleaner? [ In reply to ]

removed at example

Feb 2, 2009, 3:17 PM

Post #15 of 22 (2174 views)

>
>
> > The point is that listing the authors is a silly clause.
>
> I couldn't disagree more.
>

Just to clarify Sam, I am not suggesting the abolishment of the history
page. Its just that if you are willing to agree that a url is sufficient
attribution, I think you may as well follow the reductio on that argument
and ask them to provide the minimum amount of information necessary to find
the authors. In many cases this is just a single word, Wikipedia.
_______________________________________________
foundation-l mailing list
foundation-l@lists.wikimedia.org
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l

Re: Attribution made cleaner? [ In reply to ]

smolensk at eunet

Feb 2, 2009, 3:21 PM

Post #16 of 22 (2180 views)

On Monday 02 February 2009 23:45:29 Brian wrote:
> On Mon, Feb 2, 2009 at 3:29 PM, Nikola Smolenski <smolensk@eunet.yu> wrote:
> > On Monday 02 February 2009 22:41:37 Brian wrote:
> > > Following this line of reasoning in both directions, many users who
> > > contribute to an encyclopedia that "anyone can edit" may not want their
> > > name reprinted on every conceivable medium that their contributions
> > > could be replicated on. In other words, many users probably don't care
> > > even a little bit about the attribution requirements of the CC-BY-SA.
> > > They contribute under the implicit assumption that their work is in the
> > > public
> >
> > Do you have anything to back your claims with?
>
> The operating assumption is that the average pseudo-anonymous user to a
> wikimedia project understands and/or cares about the licensing issues and
> realizes their name will be printed everywhere that the text they
> contribute is printed. Do you have any evidence that this is true? That the

Yes:

* Contributors who do not have any understanding of copyright will usually
attempt to copy copyrighted material to the project, that will then be
deleted and they will be warned. This will lead them to at least
understanding that they can't just copy any material anywhere without the
author's permission, and logically this leads to the conclusion that other
people can't do the same with their work. Similar thing will happen whenever
someone tries to upload an image for the first time.

* When someone is presenting Wikimedia projects, they usually mention free
licences and what do they mean.

* Practically all printed material today is printed with its author(s)' names;
therefore it is obvious to assume that this material will be too, if printed.

* In past, several books and DVDs were made from material from several
Wikimedia projects, containing all the names of the contributors. They were
marketed in and out of the projects, and I expect that a fair number of at
least contributors of these projects know about them, yet I haven't heard of
anyone expressing surprise about it.

> average pseudo-anonymous contributor has a fairly sophisticated
> understanding of copyright? Otherwise its quite similar to the ToS at the

Understanding that your work should be attributed to you requires only the
most rudimentary understanding of copyright, or none at all.

> bottom of every web page you visit, which you supposedly implicitly agree
> to, but which you rarely to never read and is actually a legal grey area.

I'd say it is quite dissimilar, but anyway - a large amount of Wikipedia
marketing specifies that Wikipedia is a free encyclopedia the content of
which can be freely reused under certain conditions and so on. This is much
more than your average website does.

> The point is that listing the authors is a silly clause.

You have not proven your point.

_______________________________________________
foundation-l mailing list
foundation-l@lists.wikimedia.org
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l

Re: Attribution made cleaner? [ In reply to ]

andrew.gray at dunelm

Feb 2, 2009, 3:26 PM

Post #17 of 22 (2177 views)

2009/2/2 Brian <Brian.Mingus@colorado.edu>:
> I advocate a much more flexible attribution scheme than listing the authors
> or printing a url to the history page. I think a simple (Wikipedia) is a
> sufficient attribution for text. If you have the text it is trivial to find
> the original author of that text. It's not so trivial with images, but a
> link to the history page of an image can be embedded in its metadata.

There's two different issues, here, really, and I think you're chasing
a different one to my original suggestion. I'm certainly not saying
that this method for generating names is automatically a mandate to
require they be used to top and tail every article - just that if
someone does attribute that way, it'll help them do it better.

*However* we decide that downstream reused material should be
attributed, be it heavily or as lightly as possible, there's going to
be a step in the process - perhaps only an optional one - where
someone takes a Wikipedia article and tries to shake out some authors.
Figuring out how to make that work efficiently and cleanly and
helpfully is a good thing in and of itself, whatever conclusion the
main debate comes to.

--
- Andrew Gray
andrew.gray@dunelm.org.uk

_______________________________________________
foundation-l mailing list
foundation-l@lists.wikimedia.org
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l

Re: Attribution made cleaner? [ In reply to ]

removed at example

Feb 2, 2009, 3:29 PM

Post #18 of 22 (2182 views)

It's silly because it's arbitrary and only applies to the lowest quality
articles - start and stub. I have a query running on the Toolserver which I
hope to process into a percentage of articles that have 5 or less authors.
But we already know that the large majority of articles are stubs, and
somewhat fewer start. We also know that the quality of these articles is
related to their popularity, and that an article of lower popularity is less
likely to be quoted, by definition.

If you are willing to accept that a URL is sufficient, then there is no
reason to ever show the authors - it's only to accomodate the fact that the
CC-BY-SA contains a clause which isn't really relevant to the projects.
Better to change the CC-BY-SA or the attribution requirements than kludge
this 5 authors or less statement in there, which just makes it harder to use
the content. That is against the aims of the project, so I do consider the
whole 5 authors or less thing silly.

> > The point is that listing the authors is a silly clause.
>
>
> You have not proven your point.
>
> _______________________________________________
> foundation-l mailing list
> foundation-l@lists.wikimedia.org
> Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
>
_______________________________________________
foundation-l mailing list
foundation-l@lists.wikimedia.org
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l

Re: Attribution made cleaner? [ In reply to ]

meta.sj at gmail

Feb 2, 2009, 3:39 PM

Post #19 of 22 (2190 views)

On Mon, Feb 2, 2009 at 6:13 PM, Brian <Brian.Mingus@colorado.edu> wrote:
> I advocate a much more flexible attribution scheme than listing the authors
> or printing a url to the history page. I think a simple (Wikipedia) is a
> sufficient attribution for text. If you have the text it is trivial to find
> the original author of that text. It's not so trivial with images, but a
> link to the history page of an image can be embedded in its metadata.

Could you implement a wikiblame extension?
That would make attribution much cleaner.

SJ

_______________________________________________
foundation-l mailing list
foundation-l@lists.wikimedia.org
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l

Re: Attribution made cleaner? [ In reply to ]

removed at example

Feb 2, 2009, 3:41 PM

Post #20 of 22 (2172 views)

I actually suggested such a thing in another thread on this topic ^_^ It
would require a monster search index (all revisions of all article text),
but it wouldn't get a ton of use so wouldn't use too many resources.

On Mon, Feb 2, 2009 at 4:39 PM, Samuel Klein <meta.sj@gmail.com> wrote:

> On Mon, Feb 2, 2009 at 6:13 PM, Brian <Brian.Mingus@colorado.edu> wrote:
> > I advocate a much more flexible attribution scheme than listing the
> authors
> > or printing a url to the history page. I think a simple (Wikipedia) is a
> > sufficient attribution for text. If you have the text it is trivial to
> find
> > the original author of that text. It's not so trivial with images, but a
> > link to the history page of an image can be embedded in its metadata.
>
> Could you implement a wikiblame extension?
> That would make attribution much cleaner.
>
> SJ
>
> _______________________________________________
> foundation-l mailing list
> foundation-l@lists.wikimedia.org
> Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
>
_______________________________________________
foundation-l mailing list
foundation-l@lists.wikimedia.org
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l

Re: Attribution made cleaner? [ In reply to ]

meta.sj at gmail

Feb 2, 2009, 3:42 PM

Post #21 of 22 (2174 views)

On Mon, Feb 2, 2009 at 6:29 PM, Brian <Brian.Mingus@colorado.edu> wrote:
> If you are willing to accept that a URL is sufficient, then there is no
> reason to ever show the authors - it's only to accomodate the fact that the
> CC-BY-SA contains a clause which isn't really relevant to the projects.
> Better to change the CC-BY-SA or the attribution requirements than kludge
> this 5 authors or less statement in there, which just makes it harder to use
> the content. That is against the aims of the project, so I do consider the
> whole 5 authors or less thing silly.

OK, now I understand better where you're coming from. The '5 authors
or less' is indeed silly. A link to the history page does not satisfy
my concept of 'a URL', and we can do much better than a URL on WP
itself.

For the purposes of reuse we should facilitate simple solutions; a URL
would be sufficient there if a detailed attribution page is preserved,
for that revision, at a permanent link (and any mirrors thereof).

SJ

_______________________________________________
foundation-l mailing list
foundation-l@lists.wikimedia.org
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l

Re: Attribution made cleaner? [ In reply to ]

Feb 2, 2009, 3:50 PM

Post #22 of 22 (2186 views)

On Tue, Feb 3, 2009 at 12:13 AM, Brian <Brian.Mingus@colorado.edu> wrote:
> I advocate a much more flexible attribution scheme than listing the authors
> or printing a url to the history page. I think a simple (Wikipedia) is a
> sufficient attribution for text. If you have the text it is trivial to find
> the original author of that text. It's not so trivial with images, but a
> link to the history page of an image can be embedded in its metadata.

Precisely, and once you have this as a minimum standard you can still
do whatever you like on top of it. As significant bonuses we're not
diluting/drowning out the promotion of Wikipedia and we're avoiding
situations where authors can go after [re]users for infringement;
effectively the power of enforcement would be vested in Wikipedia (but
not the copyright itself so we don't have to worry about WMF turning
evil, only new license releases which must be 'similar in spirit'
anyway).

The attribution instructions could go something like:

"You must attribute Wikipedia, should reference the name of the
article (with hyperlinks where appropriate) and may also credit the
authors which can be found on the history tab."

I've also been thinking more about the possibility of identifying key
contributors for attribution and I've come to my own conclusion that
it's a non-starter. If you start attributing some people but not
others then those who are not attributed (who would otherwise not care
had the attribution have been for Wikipedia) will get justifiably
upset and may well seek to enforce their 'right' to attribution.

The only way "to shake out some authors" reliably (as Andrew just
said) is to do it manually, which is another avenue for conflict and
resource wastage. Summary: author attribution is an all or nothing
thing; either you attribute a boundless list of 'names' in 2pt font or
you attribute nobody.

Anyway I have to get back to writing 'AttriBot' so I can stamp my name
on any article with <5 authors ;)

Sam

_______________________________________________
foundation-l mailing list
foundation-l@lists.wikimedia.org
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l