I posted this idea in the wikipedia village pump proposals, and voice-of-all refered me to this mailing list. So here it is:
A while back, i made a suggestion on [[Wikipedia_talk:Stable_versions#Semi-automation_-_recent_stable_version_detector]] for a mechanism i called a "recent stable version detector". I've now looked at it from a different perspective and realized that it could be used as a simple vandalism filter for anonoymous / logged-out viewers (which will also filter out edit wars). The idea is simple. Each revision of an article is scored according to a simple formula:
:score = [time of next revision - time of revision] - [current time - time of revision] * constant
: = [time before next revision] - [age of revision] * constant
The revision with the highest such score is the revision that the public will see (i.e. non-logged in users).
The effect of this would be to impose a small delay between when a revision is made and when that revision is published publicly. Revisions that lasted a relatively shorte amount of time before being revised again will be "skipped". Thus, vandalism that is quickly caught by a logged-in user or recent change patroller is vandalism that ''the public will never see.''
Similarly, all the quick flips back and forth between two versions of an article in an edit war will be "skipped".
This mechanism wouldn't require any user intervention, wouldn't interfere with any existing processes, and will alway show the public a relatively stable, vandalism free, and ''current'' revision.
-Kevin
Is there anyone that has done any research on how the number of visitors
relates to the article quality? I believe it is related somehow but I'm
not sure how it can be modeled. It works by counting the visitors that
reads a particular segment of the article, and then will accept the
particular segment as correct when a sufficient number of visitors has
been visiting. It can work together with a system for writer grading,
were this system will change the grade from whatever the writer has.
Compared to this a "stable versions" is like having a visitor with
ultimate power to mark the revision as good. This system does not give
the visitors such ultimate power, and in fact will not give give them
more than a small fraction of the power necessary to claim the revision
is free of vandalism. Combined I guess it is possible to make a system
that will be better than anyone of them alone.
Any real vandalism will most likely never be marked as good, because the
limit can be set so high that it will be found by someone long before it
is marked as "patrolled", and then most likely nothing or very little of
the revision will survive so the revision itself will never be marked as
patrolled. If a known good writer contributes a revision, then it will
get a flying start and it will need few visitors ("anonymous
patrollers") before it is marked as "good". If the writer is unknown the
revision will need a lot of visitors before it is marked as good.
Even very seldom read articles have several visitors each week, and
through a year this will add up to a considerable amount of visitors.
John
Dear John and all,
With regard to your discussion on examining Wikipedia article quality,
the idea somehow overlaps with our intuition of measuring article
quality based on the authority of authors/reviewers. We have been
researching from this perspective. Below attached is the abstract of our
research paper, titled "Measuring Article Quality in Wikipedia: Models
and Evaluation", published in CIKM 2007:
"
Wikipedia has grown to be the world largest and busiest free
encyclopedia, in which articles are collaboratively written and
maintained by volunteers online. Despite its success as a means of
knowledge sharing and collaboration, the public has never stopped
criticizing the quality of Wikipedia articles edited by non-experts and
inexperienced contributors. In this paper, we investigate the problem of
assessing the quality of articles in collaborative authoring of
Wikipedia. We propose three article quality measurement models that make
use of the interaction data between articles and their contributors
derived from the article edit history. Our Basic model is designed based
on the mutual dependency between article quality and their author
authority. The PeerReview model introduces the review behavior into
measuring article quality. Finally, our ProbReview models extend
PeerReview with partial reviewership of contributors as they edit
various portions of the articles. We conduct experiments on a set of
well-labeled Wikipedia articles to evaluate the effectiveness of our
quality measurement models in resembling human judgment.
"
We would appreciate your comments and suggestion.
Regards,
Meiqun HU
-----Original Message-----
Date: Mon, 10 Mar 2008 23:00:39 +0100
From: John Erling Blad <john.erling.blad(a)jeb.no>
Subject: [Wikiquality-l] Checking article quality through number of
visitors
To: wikiquality-l(a)lists.wikimedia.org
Message-ID: <47D5AF87.2030501(a)jeb.no>
Content-Type: text/plain; charset="iso-8859-1"
Is there anyone that has done any research on how the number of visitors
relates to the article quality? I believe it is related somehow but I'm
not sure how it can be modeled. It works by counting the visitors that
reads a particular segment of the article, and then will accept the
particular segment as correct when a sufficient number of visitors has
been visiting. It can work together with a system for writer grading,
were this system will change the grade from whatever the writer has.
Compared to this a "stable versions" is like having a visitor with
ultimate power to mark the revision as good. This system does not give
the visitors such ultimate power, and in fact will not give give them
more than a small fraction of the power necessary to claim the revision
is free of vandalism. Combined I guess it is possible to make a system
that will be better than anyone of them alone.
Any real vandalism will most likely never be marked as good, because the
limit can be set so high that it will be found by someone long before it
is marked as "patrolled", and then most likely nothing or very little of
the revision will survive so the revision itself will never be marked as
patrolled. If a known good writer contributes a revision, then it will
get a flying start and it will need few visitors ("anonymous
patrollers") before it is marked as "good". If the writer is unknown the
revision will need a lot of visitors before it is marked as good.
Even very seldom read articles have several visitors each week, and
through a year this will add up to a considerable amount of visitors.
John
Dear Sir/Madam,
I am currently conducting a research survey on Wiki technology. The survey
title is called "Survey on Adoption of Wiki Technology Innovation".
The objective of this survey is to deepen the understanding of the initial
adoption of Wiki technology innovation by individuals in organizations, and
the Wiki technology's diffusion within the organization. It is based on the
theoretical framework of Roger's Diffusion of Innovation. I would appreciate
your expert view and feedback on this survey. Also, I kindly seek your
assistance in forwarding this email to your colleagues, friends, or anybody
you think use wiki in organisational settings. I hope it will not take too
much of your time and apologize for any inconveniences.
Here is a link to the survey:
http://www.surveymonkey.com/s.aspx?sm=zfANP1maT_2ffOLCMljGEMbA_3d_3d
This short survey will take about 10-15 minutes to complete and will greatly
assist in the study of the factors that influence the adoption Wiki
technology innovation.
Thank you very much for your time!
Warmest Regards,
Weizhang
Singapore
Hi
Sorry for the spam like email but this is just an email to let you all know
about the Wikimania 2008 Call for Participation. I have included this
below, and the page can also be found on the Wikimania wiki @
http://wikimania2008.wikimedia.org/wiki/Call_for_Participation. Please do
forward this onto any local project mailing lists or anyone else who may be
interested in this. Also there are translations in the following
languages: de/Deutsch<http://wikimania2008.wikimedia.org/w/index.php?title=Call_for_Participation…>
es/ <http://tr_1203787823173>Español<http://wikimania2008.wikimedia.org/w/index.php?title=Call_for_Participation…>
fr/ <http://tr_1203787823178>Français<http://wikimania2008.wikimedia.org/w/index.php?title=Call_for_Participation…>
ja/日本語<http://wikimania2008.wikimedia.org/w/index.php?title=Call_for_Participation…>
pt/Português<http://wikimania2008.wikimedia.org/w/index.php?title=Call_for_Participation…>
sv/Svenska<http://wikimania2008.wikimedia.org/w/index.php?title=Call_for_Participation…>
zh/ <http://tr_1203787823208>中文<http://wikimania2008.wikimedia.org/w/index.php?title=Call_for_Participation…>.
More translations are always welcomed and appreciated, so please see the
translation page<http://wikimania2008.wikimedia.org/wiki/Template:Translation/Call_for_Parti…>to
help out.
Many thanks
Mark
(User:Markie)
Call for Participation
[image: Image:Wikimania logo with text
2.svg]<http://upload.wikimedia.org/wikipedia/commons/7/7e/Wikimania_logo_with_text…>
Wikimania <http://en.wikipedia.org/wiki/Wikimania> is an annual global event
devoted to Wikimedia
<http://wikimediafoundation.org/wiki/Main_Page>projects around the
globe (including
Wikipedia <http://en.wikipedia.org/wiki/Main_page>,
Wikibooks<http://en.wikipedia.org/wiki/b:en:Main_page>,
Wikisource <http://en.wikipedia.org/wiki/s:en:Main_page>,
Wikinews<http://en.wikipedia.org/wiki/n:Main_page>,
Wiktionary <http://en.wikipedia.org/wiki/wikt:en:Main_page>,
Wikiversity<http://en.wikipedia.org/wiki/v:en:Main_page>,
Wikiquote <http://meta.wikimedia.org/wiki/q:en:Main_page>,
Wikispecies<http://species.wikimedia.org/wiki/Main_page>,
and Wikimedia Commons <http://commons.wikimedia.org/wiki/Main_page>) and for
its editors and users to gather, meet each other, exchange ideas, and report
on research and projects. It is a community event, which is also open to the
public and to researchers. This year's conference will be held from *July
17-19, 2008* in Alexandria <http://en.wikipedia.org/wiki/Alexandria>,
Egypt<http://en.wikipedia.org/wiki/Egypt>at the new Library of
Alexandria (Bibliotheca
Alexandrina <http://en.wikipedia.org/wiki/Bibliotheca_Alexandrina>).
[image: The Wikimedia Foundation
projects]<http://wikimania2008.wikimedia.org/wiki/Image:Wikimedia_logo_family.png>
For more information, please visit the Wikimania 2008 Home page at
http://wikimania2008.wikimedia.org
We are accepting submissions for presentations, workshops, panels, posters,
open spaces, and artistic artifacts. Please carefully follow the submission
guidelines below. Submissions can be sent via the following link:
https://wikimedia.pentabarf.org/submission/wikimania2008
Important dates
- 1 February – 16 March : Submission
- 17 March – 30 April : Review, feedback and notification of
acceptance
- 17 – 19 July 2008 : *Wikimania*
Conference Tracks
Submissions should address one or more of the following themes:
Wikimedia Communities Interesting projects and particularities within the
communities; policy creation within individual projects; conflict resolution
and community dynamics; reputation and identity; multi-lingualism, languages
and cultures; social studies. We explicitly invite you to discuss your local
Wikimedia project's community. Free Knowledge Open access to information;
ways to gather and distribute free knowledge, usage of the Wikimedia
projects in education, journalism, research; ways to improve content quality
and usability; copyright laws and other legal areas that interfere with
Wikimedia projects. Free Content in the Middle-East/Africa. Technical
infrastructure Issues related to MediaWiki development and extensions;
Wikimedia's technical infrastructure; new ideas for development (including
case studies from other wikis or similar projects). Scientific track Academic
papers about massively collaborative work, open and free content creation,
community dynamics, the social or economic aspects of the Wikimedia
projects, and other topics related to Wikimedia projects. Papers submitted
to the scientific track will be peer reviewed by a reviewing committee
regarding their novelty, rigour, and estimated impact, and accepted or
rejected based on these reviews. The papers will be published in proceedings
afterwards, and depending on the number and the quality of the submissions,
a journal special issue may be pursued. Scientific track papers must be in
English, and must not exceed 7,500 words (or 15 pages LNCS).
Your topic must be related either to the Wikimedia projects and their
communities, or to the creation of free content in general.
Types of Submissions
We are seeking submissions for
- presentations (10–30 minute talks with discussion afterwards)
- workshops/open discussions (60–120 minute session with a discussion
leader and more involvement of the audience)
- panels (group of 2-5 speakers to discuss on a specific subject)
- posters (printed presentations or visual displays that can stand on
their own)
- artistic artifacts (plays, competitions, comedy, visualizations, or
other representations of some aspect of the projects)
In addition there will the possibility to give lightning
talks<http://wikimania2008.wikimedia.org/wiki/Lightning_talks>(5
minute short presentations). These will be organized on the Wikimania
2008 wiki without need to submit via the submission system.
Submission Guidelines
Wikimania is organized by volunteers, so please help us minimize wasted
effort by submitting via the
submission<http://wikimania2008.wikimedia.org/wiki/Submission>system
and following these guidelines. All submissions MUST explicitly
include the following:
1. an English "Event title"
2. a short English "Abstract" of your event in 50 to 100 words. The
abstract will be used for the public schedule.
3. the "Track" your submission fits in best (Wikimedia Communities,
Free Knowledge, Technical infrastructure, or Scientific)
4. the "Event type" (presentation, workshop, panel, poster,
artistic...)
5. information about the speaker (full name, email, a short
description of at least 2 sentences...)
6. for submissions to the scientific track: set "Submission of paper
for proceedings" to "yes" and upload a paper instead of the "Description"
below as "Attachment". Papers must be in English, and must not exceed 7,500
words.
In addition you can add some more information like a a subtitle of the
event, an image (will be resized to 128x128px) and private "Submission
notes" for reviewers and conference organisation. In particular you should
give:
- a more detailed "Description" of your event in English or Arabic.
The description is essential for review: please give an overview of the
areas to be covered or taught. The better you describe your submission, the
more likely it will get accepted. State clearly the relevance to the
Wikimedia projects and whether submission concerns a specific wiki project.
You can also include links. The description will later be used for the
public schedule but you can edit it before.
- special requirements (such as equipment for a workshop or panel) if
needed
- the language used for presentation
- whether you want to submit a paper for proceedings
- whether you want to submit presentation slides
- whether the presentation is intended to be a specific length
- the target audience you are going to reach and what previous
knowledge is needed
- images or sketches of the poster or artistic artifact if available
- for panel submissions a suggested moderator and short biographies of
each suggested panelist
In the "Submission notes" you should tell us whether you will attend to
Wikimania (a) surely, (b) probably, (c) only if your submission is accepted,
or (d) only if we provide travel and/or accommodation. You can also add
yourself to the public list of attendees at the Wikimania 2008 wiki:
http://wikimania2008.wikimedia.org/wiki/Attendees
Please note that all submissions must be dual licensed under the GNU Free
Documentation License version 1.2 or later *and* the Creative Commons
Attribution License! By submitting for Wikimania 2008 you agree to this
condition.
For more information see the submission guidelines at
http://wikimania2008.wikimedia.org/wiki/Submission
Submissions
Once you are sure you have included all of the required information, please
send your submission before the respective deadline through our *submission
system*:
https://wikimedia.pentabarf.org/submission
See also
- About the venue: http://wikimania2008.wikimedia.org/wiki/Venue
- Brainstorming page for program ideas:
http://wikimania2008.wikimedia.org/wiki/Program_ideas
- Editable list of attendees:
http://wikimania2008.wikimedia.org/wiki/Attendees
Dear All,
we have just released in open-source format the code of WikiTrust, the tool
we use for the Wikipedia trust coloring.
The project homepage is http://trust.cse.ucsc.edu/ , and from there, you can
follow a link to a live demo.
The code itself is available from http://trust.cse.ucsc.edu/WikiTrust .
The code is suitable to the trust-coloring of a static dump of a wiki; the
code for the coloring of edits in real-time, as they happen, is under
development.
The code is extensible, and it provides a platform over which it is
(relatively) easy to write wiki analysis tools... for instance, we wrote
small analysis procedures that measure the inter-edit time distribution, and
the amount of text contributed by authors of various reputation ranges.
As the text analysis engine and the dump traversal engines are already
built, it is relatively easy to add other analysis modules.
We hope this will be of interest!
All the best,
Luca de Alfaro
(message sent on behalf also of Bo Adler and Ian Pye)
Dear All,
we have just released in open-source format the code of WikiTrust, the tool
we use for the Wikipedia trust coloring.
The project homepage is http://trust.cse.ucsc.edu/ , and from there, you can
follow a link to a live demo.
The code itself is available from http://trust.cse.ucsc.edu/WikiTrust .
The code is suitable to the trust-coloring of a static dump of a wiki; the
code for the coloring of edits in real-time, as they happen, is under
development.
The code is extensible, and it provides a platform over which it is
(relatively) easy to write wiki analysis tools... for instance, we wrote
small analysis procedures that measure the inter-edit time distribution, and
the amount of text contributed by authors of various reputation ranges.
As the text analysis engine and the dump traversal engines are already
built, it is relatively easy to add other analysis modules.
We hope this will be of interest!
All the best,
Luca de Alfaro
(message sent on behalf also of Bo Adler and Ian Pye)
I'm just testing the latest FlaggedRevs code to see what kind of
configurations we should set up for Wikimedia. Thanks to Aaron for
continuing to work on it. FYI, we'll have a face-to-face meeting in
the Wikimedia office next week with Brion, myself, Philipp Birken &
Luca de Alfaro (Aaron can't make it) to discuss some of the open
issues.
>From what I can tell, the current code will still show a link to the
sighted version, even if current & sighted are fully identical. I
consider this highly confusing behavior, and I don't really see that
we can go live with this, even as an experiment.
Is there any option regulating this behavior, or another way around it?
If not: I would really like us to figure out a solution to this. I do
believe the situation where a page is current and sighted (and where
any included templates are unmodified from the sighted state, or have
been edited by trusted users & auto-reviewed), will be quite common,
because this is after all the situation we're trying to optimize
towards. A UI that makes a fully reviewed version look unreviewed
seems like a major problem to me.
--
Erik Möller
Deputy Director, Wikimedia Foundation
Support Free Knowledge: http://wikimediafoundation.org/wiki/Donate
One thing that stood out for me in the small sample of articles I
examined was the flagging of innocuous changes by casual users to
correct spelling, grammar, etc. Thus a "nice-to-have" would be a
"smoothing" algorithm that ignores inconsequential changes such as
spelling corrections, etc. or the reordering of semantically-contained
units of text (for example, reordering the line items in a list w/o
changing the content of any particular line item, etc., or the
reordering of paragraphs and perhaps even sentences.) I think this
would cover 90% or more of changes that are immaterial to an article's
credibility.
