Slashdot asks: why unsigned articles? Why allow any author to change articles?

[Larry Sanger]

Someone claiming to be from Slashdot wrote and said:
> I'm writing an article for Slashdot (that is, I'm an
> employee, not someone
> who's just going to submit something and cross my fingers
> ) and wanted
> to ask someone what the reasons are for (a) having articles not be
> attributed to their writers, and (b) allowing any author to change an
> article.  I've read both the essay and the FAQ but couldn't find that
> specifically addressed.

Here's my answer.


Clearly, we need to add some answers to really fundamental questions to our FAQ.  Our introductory literature generally assume that Wikipedia is a good idea; it's just that its system needs to be improved.  The questions you ask really can be applied to Wikipedia, to most wikis, and to radical collaborations in general.  Bearing all that in mind, here goes:

(a) Leaving articles unsigned (or work uncredited) is essential to radical collaboration.  It makes others feel as free as possible to change our text.  If we sign our work, others will naturally feel that we've taken ownership of it and they aren't welcome, or not as welcome, to change it.  If project organizers announce at the outset, "We're working on everything together, and no one owns any article," that emboldens everyone to work on whatever they want.  That freedom--to work on whatever you want--is precisely what makes successful wikis grow so fast.

Now, if you already grokked that, you might be asking why the Citizendium (CZ), with its "gentle expert oversight," isn't signing articles.  Don't "experts" (professors, researchers, professionals, etc.) want personal credit for their work?  Well, many of them do, and those people dismiss the notion of radical collaboration out of hand.  But you might be surprised at how many experts really love the idea of Wikipedia.  Certainly enough to make CZ a going concern.  Wikipedia's own contributor base has featured, from its very beginning, a fair number of experts.  And CZ's most active contributors, so far, have been expert editors; and we've got over 150 of them and growing rapidly.

(b) Why allow any author to change an article?  That goes to the very heart of radical collaboration.  As I see it, there are two reasons for the policy.  First, maximal freedom maximizes motivation.  The broader the range of topics we can work on, the more emboldened we will be to contribute.  If we are told that we must get permission from someone to work on an article, then we probably won't even try.  Even if some, not all, articles are restricted, then our motivation will be dampened because we know that the articles we are interested in might be among the restricted ones.

But you might ask why CZ in particular allows any author to change an article.  Isn't that a strange policy for an expert-guided project?  Doesn't it threaten to replicate Wikipedia and its problems?  Well, not necessarily.  Indeed we do allow non-experts to work on articles that experts are developing.  More to the point, we make no distinction between articles developed by experts and others; there are only CZ articles.  Indeed, some people might find this puzzling, but that's only because the system itself challenges their assumptions about what Web 2.0 projects are supposed to be like.

We have many ways to keep participation dynamic and quality high.  First, the very presence of experts in the system, in their role as editors working shoulder-to-shoulder with nonexpert authors, raises standards.  Second, we have a goal that Wikipedia doesn't have: we are trying to develop articles to the point where they can be approved by our editors.  Third, we have an excellent group of so-called constables who keep folks in line (there's been very little work for them to do so far, though, other than get new people on board).  Fourth, while we don't sign articles, we do require that people register and use their real names if they want to participate.  Fifth, we require, as a condition of participation, that people digitally sign a sort of social contract that says they'll play by the rules.

Hope that answers the questions!

Dr. Larry Sanger
Citizendium Editor-in-Chief


Feel free to add your own answers (I think he wanted answer from the whole community, not just me)--I'm pointing the writer to this page.

[bennett]

Thank you for your thoughtful answer.  Let me explain the angle from which I'm approaching this.  I think very resolutely in terms of specific goals and results, and how properties of a project help to achieve those goals.  My question is, what specific goal of Citizendium is aided by having unsigned articles and allowing real-time edits?  You did answer; let me come back to that in a second.

It seems that a free online community-build encyclopedia or pseudo-encyclopedia serves these main purposes:
1) Looking up articles about pop culture topics that wouldn't be covered in a normal encyclopedia ("24 Season 6") without necessarily needing to know that it's 100% accurate.
2) Being able to redistribute the content freely and legally (for example, printing free textbooks for schools).
3) Looking up articles and needing to be close to 100% sure that the information is accurate.  (Say, at least as sure as I would be if I were getting the information from a library book.)
4) Getting information and being able to cite it in a paper for school or even a more serious scholarly work.

The unsigned articles and real-time edits don't affect goals #1 and #2, and I think they actually hinder #3 and #4.  I'm less certain of the reliability of the information I'm looking up, if it's not signed by a credentialed author.  Even if I know that the article was once approved by *some* expert, I don't know if someone has made an incorrect edit since then.  And some schools allow you to cite online sources that are attributed to a credentialed person, but not sources that are anonymous.  (Schools discourage students from citing the encyclopedia, or pseudo-encyclopedias, in serious research papers anyway.  But even more so for anonymous articles.)

So balanced against this are the benefits of unsigned articles and real-time edits.  It sounds like you are saying that these two factors are to maximize motivation on the part of users.  This sounds quite likely to be true.  However, it's also a statement that can be tested empirically -- either by taking a survey of volunteers (easy), or creating a branch of Citizendium that utilized signed articles and allowed only editor-approved edits, and seeing if it produced better articles and got more visitors (hard).  Actually, now that I think about it, that wouldn't work with your existing volunteer base, because you'd only be sampling volunteers who have already signed up to work within the Citizendium system, and thus would include those biased in favor of the existing rules.  What I could try, perhaps, is a survey of the general public, asking them if they would be more likely to (a) contribute to articles, and/or (b) read articles, on a system that had editor-signed content and only editor-approved edits.  Even if the articles were signed, a banner at the top of each article could say, "Suggested edits are welcome, please click the 'Edit' link next to a paragraph that you'd like to change."  Perhaps also a tag that said, "The average time for a suggested edit to be approved is 1 hours 30 minutes."  (If multiple experts agreed to handle edits to an article, then any expert could handle any edit in the queue, and the average response time would go down.)

For me personally, I'd rather submit edits if I knew that (a) the edits would be editor-approved, and (b) I'd get some co-authorship credit (even a very tiny one) for approved edits.  Yes, I'd be willing to live with a delay before the edit appeared on the real article -- I'd consider it a trade-off for the fact that my edit wouldn't get overridden by someone *else* making a frivolous edit.  (I think it would also be cool to have a system where I could sign up to receive notification by e-mail if anybody made an edit to a section of the article that I had edited.)  And I'm quite sure there are lots of experts out there who would be more likely to contribute content if it could be attributed to them.  It's generous enough of them to write content for free; it's narrowing the field by asking them to take their name off of it.  You could say that's an "ego" thing, and you'd be right, but ego cuts both ways -- some people take more pride in work when they have their name on it, and besides, many experts have egos precisely because they're the good experts.

Anyway, the bottom line is that real-time edits and unsigned articles may make the wiki more fun to play in, but I don't think the purpose of an information source should be that it's "fun to play in"   I think the purpose should be to provide accurate, possibly citable information, and the question is whether real-time edits and unsigned articles help to achieve that goal.

*Maybe* they do -- perhaps real-time edits and non-authorship get people fired up to contribute, and the extra horsepower contributed by all those fired-up people, results in better articles than you'd get from a more controlled system.  But it's far from obvious that that's true.  (Certainly some people would be more motivated by a system with attribution and only editor-approved edits.)  My point is that it's definitely worth trying -- some sort of poll or experiment to see whether editor authorship and editor-approved edits would make the system better from the readers' and the volunteers' point of view.

[Zachary Pruckowski]

The essential answer to your questions is that real-time edits allow us to build pages much faster.  Real-time edits are only allowed on unapproved pages.  Once an article has become approved, then it will not allow real-time editing, but users can work on the draft of the next version of the article.  This allows us to have articles like "24 Season 6" that are real-time editted and unstable (your #1), and have other articles be more scrutinized and and more stable (your #3 and #4).  So we handle the two types of articles differently.  As Citizendium expands, there will be more and more approved articles, and so there will be a large base of articles that don't change much.

[bennett]

Ah OK.  Yes that hybrid actually sounds like the best solution with regard to real-time edits, you get the best of both worlds.

But then there's still the attribution question.  Why forego the contributions that some people would be willing to make if only they can be properly credited for them?  Larry said people might feel put off about submitting edits to an article if they see someone's name on it.  Some people might.  But the flip side of that is that people might be motivated to submit edits by the prospect of having *their* name listed, even in the "end credits"   And you can always have a note at the top of each article inviting people to submit edits, to counteract any hesitance they would feel upon seeing the name of an "expert" attached.

One feature that might encourage visitors to become part of the community, even while reading a page that they're not an expert in, would be an "ask question" or "request clarification" link -- where users can post about something they think is unclear in the article, or say that a piece of information that they think naturally belongs in the article, is missing.  This allows users to improve the quality of pages even about topics they're not experts in -- in fact, the non-experts would be the ones who are best suited to the task of highlighting things that aren't clear to newcomers.  (I presume you can do this on Wikipedia by going to the talk page for an article, but Wikipedia doesn't make it obvious how to do that.  Having a clear invitation to the user would be a big step.)

[Zachary Pruckowski]

Ah OK.  Yes that hybrid actually sounds like the best solution with regard to real-time edits, you get the best of both worlds.

But then there's still the attribution question.  Why forego the contributions that some people would be willing to make if only they can be properly credited for them?  Larry said people might feel put off about submitting edits to an article if they see someone's name on it.  Some people might.  But the flip side of that is that people might be motivated to submit edits by the prospect of having *their* name listed, even in the "end credits"   And you can always have a note at the top of each article inviting people to submit edits, to counteract any hesitance they would feel upon seeing the name of an "expert" attached.

Yes, the issue is that changes made to the draft page might not be easily attributable.  That's certainly an unintended consequence, and one we may need to correct.  For unapproved articles, the author has the same attribution WP allows - a spot in the history page.

[Stephen Ewen]

Very thought-provoking arguments regarding the psychology of crediting to authors, bennett. I can certainly see how that can be important to major authors of articles. Stated as you have, it seems a matter perhaps deserving of further review.

[Larry Sanger]

Ah OK.  Yes that hybrid actually sounds like the best solution with regard to real-time edits, you get the best of both worlds.

But then there's still the attribution question.  Why forego the contributions that some people would be willing to make if only they can be properly credited for them?  Larry said people might feel put off about submitting edits to an article if they see someone's name on it.  Some people might.  But the flip side of that is that people might be motivated to submit edits by the prospect of having *their* name listed, even in the "end credits"   And you can always have a note at the top of each article inviting people to submit edits, to counteract any hesitance they would feel upon seeing the name of an "expert" attached.

Bennett, well, this is something that not many people in the project know, but there are some plans to that effect.  There are a couple of different "live" options open to us right now, although neither will happen soon.  One is that CZ will join with the Digital Universe Encyclopedia--comprising the Encyclopedia of Earth and an Encyclopedia of the Cosmos, under development--and the latter will take up the option you mention.  The other is something the Revson Foundation is interested in doing with us (please keep this particular tidbit out of your article, please), namely to host a couple of different specialized encyclopedias that they have funded in the past, which are now getting outdated.  The point is that, just as it says in that general blurb at the top of http://www.citizendium.org/ , we do conceive of CZ as "the flagship of a new set of responsibly-managed free knowledge projects."  There is no reason that these other projects even need be collaborative.  What we're mainly interested in is reliable, independent/neutral, and free information.

One feature that might encourage visitors to become part of the community, even while reading a page that they're not an expert in, would be an "ask question" or "request clarification" link -- where users can post about something they think is unclear in the article, or say that a piece of information that they think naturally belongs in the article, is missing.  This allows users to improve the quality of pages even about topics they're not experts in -- in fact, the non-experts would be the ones who are best suited to the task of highlighting things that aren't clear to newcomers.  (I presume you can do this on Wikipedia by going to the talk page for an article, but Wikipedia doesn't make it obvious how to do that.  Having a clear invitation to the user would be a big step.)

Well, we go far beyond that.  Users who are not expert in the topic of an article can actually edit the article on CZ.  It's just that those people are expected to respect the occasional decisions of those editors on hand who are experts on the topic.

But sure, it's a great idea I think to allow people a way to make (moderated) comments on articles without even necessarily being logged in.  It's just a matter of writing the software and setting up the rules for the process.  Give us time.  We've got far more ideas than resources!

[bennett]

Larry,

Thanks for the explanation.  Do you know how the Digital Universe Encyclopedia articles will work?  Will there be main authors listed at the top of the article, and then more minor contributors listed in the "end credits", say at the bottom?

ZacharyPruckowski had replied earlier, "For unapproved articles, the author has the same attribution WP allows - a spot in the history page."  But there are two different issues addressed by giving credit for articles -- one is the name-up-in-lights factor that induces people to contribute, and the other is the citability factor.  If the history page lists the credentialed authors who approved an article, that addresses the citability problem (at least for people who want to cite an article and know to look on the history page).  On the other hand I don't think it really give authors the public recognition that they might want for their work.  Do you know how Digital Universe Encyclopedia plans on doing it, if they have drafts for a format to use?

I agree completely with your sentence: "There is no reason that these other projects even need be collaborative.  What we're mainly interested in is reliable, independent/neutral, and free information."  That's exactly where I was coming from -- the collaborative nature of a project should be a means to an end, not an end in itself, since the people who end up using the articles don't care if they were written "collaboratively" or not!  I think a lot of people miss this point and think that "collaboration" is the raison d'etre of free projects like WP or CZ.  And the question about giving authors credit should also be, whether it serves the ends of providing reliable, neutral and free info.

[Larry Sanger]

Larry,

Thanks for the explanation.  Do you know how the Digital Universe Encyclopedia articles will work?  Will there be main authors listed at the top of the article, and then more minor contributors listed in the "end credits", say at the bottom?

See http://www.eoearth.com/

ZacharyPruckowski had replied earlier, "For unapproved articles, the author has the same attribution WP allows - a spot in the history page."  But there are two different issues addressed by giving credit for articles -- one is the name-up-in-lights factor that induces people to contribute, and the other is the citability factor.  If the history page lists the credentialed authors who approved an article, that addresses the citability problem (at least for people who want to cite an article and know to look on the history page).  On the other hand I don't think it really give authors the public recognition that they might want for their work.  Do you know how Digital Universe Encyclopedia plans on doing it, if they have drafts for a format to use?

I agree completely with your sentence: "There is no reason that these other projects even need be collaborative.  What we're mainly interested in is reliable, independent/neutral, and free information."  That's exactly where I was coming from -- the collaborative nature of a project should be a means to an end, not an end in itself, since the people who end up using the articles don't care if they were written "collaboratively" or not!  I think a lot of people miss this point and think that "collaboration" is the raison d'etre of free projects like WP or CZ.  And the question about giving authors credit should also be, whether it serves the ends of providing reliable, neutral and free info.

I agree entirely.  It's just that experience with wikis indicates that people will not work on each others' productions if those productions are signed; it's perceived as just rude.  Therefore, you can't help yourself to the most robust wiki dynamic if you sign articles.  That doesn't mean that you can't create a great encyclopedia.  No one denies, I think, that Britannica is a great encyclopedia.  What I suspect is that you can't create an enormous encyclopedia on the Wikipedia scale without the wiki dynamic--without what I have called "strong collaboration."  That's because that dynamic is the only way anyone has ever thought of for building something of that scale.

Of course, the Digital Universe Encyclopedia, or some future project of CZ itself, might prove me wrong.

[Zachary Pruckowski]

ZacharyPruckowski had replied earlier, "For unapproved articles, the author has the same attribution WP allows - a spot in the history page."  But there are two different issues addressed by giving credit for articles -- one is the name-up-in-lights factor that induces people to contribute, and the other is the citability factor.  If the history page lists the credentialed authors who approved an article, that addresses the citability problem (at least for people who want to cite an article and know to look on the history page).  On the other hand I don't think it really give authors the public recognition that they might want for their work.  Do you know how Digital Universe Encyclopedia plans on doing it, if they have drafts for a format to use?

The citability problem isn't much of an issue.  A user can establish at least one editor who signed off on approval by mousing over the approval template at the moment.  The wikilink entitled "editor" points to a userpage.  That's a compromise between having someone's name prominently featured (which is unfair, since it's group effort generally), and showing nothing (which hurts us in terms of citability and being "expert-approved").

With regards to "name-up-in-lights", we're still ahead of WP in that regards, since histories in the articles and the drafts do use real names.  This would allow for a person to say "I contributed!  Look!" and point someone to their name in a history.  There's no easy way to give all authors credit on the main article page. 

• For starters, it could get rather lengthy in some of the bigger articles, especially contentious or popular ones.
• Additionally, it would be hard to reflect how much each user contributed.
• To make matters worse, any such plan would also give vandals credit.  A vandal who damages the page and gets reverted would still see his name on the "contributor" list.  This could lead to multiple types of vandalism, ranging from registering an illegal name to get it on the contributor list to otherwise benevolent users making inconsequential edits for the sole purpose of being listed as helping with as many articles as possible.  I can imagine "Editcountitis version 2.0" breaking out when a user manages to get himself credited on ten thousand articles or something.
• Then there's the issue of accounting for upstream work.  If someone borrows WP material (assuming we're under the GFDL), and we credit WP, do we need to credit every WP editor in that contributor list?

[bennett]

Thanks, d'oh, I didn't bother checking if EOEarth.com had gone live

Looking at the articles as laid out on EOEarth, I'd understand a user being hesitant to submit an edit.  But I still think that could be countered by a banner at the top of each article explaining that articles are collaborative efforts and that user are welcome to submit edits, and by having an "Edit" link next to each paragraph like Wikipedia does -- even if your edits get submitted to a queue to be approved by the editor instead of going live immediately.

For being listed in the "end credits", I was thinking to only include that for users who submit edits after the article has gone into approval mode; if the edit is approved by an editor then the submitter gets credit.  This takes care of vandals, people submitting 10,000 edits just to be listed everywhere, people trying to get credits under illegal prank names, etc.  (While the article is still editable in real time in the initial growth phase, then there is less incentive to submit edits because you don't get credited, but your experience shows that people are willing to make those edits during the growth phase anyway without any credit.  The question is whether the credits would result in more and better edits after the article goes into approved status.)

End credits growing unwieldy might be a bigger problem.  You could have a simple rule like: display "end credits" on a different page if they grow too long.  A more complex system might be: editors award a certain number of points for each type of helpful edit that someone submits (one point level for correcting typos, another point level for substantial changes or corrections, another for adding large chunks of new material, etc.).  So you'd still have people hooked on accruing high "point levels", but now they have to come by them honestly -- either through lots of small corrections to many articles, or substantial contributions to a smaller number of articles, but either way, by making some positive contribution.

I know, I know, already you have more ideas than you have time to develop them   But if you assume that there are lots of people who would be happy to contribute to a database of freely distributable information, but would contribute more if they were credited for their work, my vote would be to put these changes close to the top of the list, to harness the work of all the people in that group.  (The option to list people in the end credits, unless the credits get too big in which case list them on a separate page, seems like an easy short-term solution without having a points system.)  If you choose to go this route at all.

[David Goodman]

There is already a bit of a problem in writing live at the development stage of the article--many of the articles so far are being primarily written by one person, and though other people certainly make additions and corrections, anything significant tends o be worded as a polite suggestion to the principal author--more or less the way collaborative writing is often done in other contexts--everyone does their section and then you exchange comments, and then put it together.  Wikis were supposed to find a better way than this, but I am not sure that for the more scholarly articles it actually will probe to be so. I think at the most we may form trusted 2 or 3 person teams in a sort of round-robin. I wouldn't want to rule anything out, but it will be interesting to watch ourselves adapt and see what structure we finally have.

[Chris Day]

Wikis were supposed to find a better way than this, but I am not sure that for the more scholarly articles it actually will probe to be so. I think at the most we may form trusted 2 or 3 person teams in a sort of round-robin. I wouldn't want to rule anything out, but it will be interesting to watch ourselves adapt and see what structure we finally have.

David, The reason we are seeing a more traditional editing environment is that because that is how most of us have experienced editing in the past. I think what we are seeing here is people unfamiliar with the wiki formatt.  As people get more comfortable with drive by editors then i think we will see many more people weighing in on a subject as well as more activity on the talk pages. In fact, I'd go as far as to say we have to adapt because as our numbers increase (and that appears to be quite rapidly judging by all the new user names appearing) there will be no stopping the drive by editors.

[Anthony.Sebastian]

Regarding the issue of crediting authors/editors of CZ articles:

• Consider a non-editable tabbed subpage entitled "contributors".
• Keep an automatically updated list of all contributors, in order of the date of their first contribution, or in alphabetical order.
• Asterisk, or otherwise highlight, 'major' contributors--'major' in respect of volume of work or of importance of contribution.
• Editor(s) who approved the article responsible for who gets asterisked or highlighted.
• Since ongoing process; list of contibutors page should indicate date last updated.

This might satisfy everyone or none.  Others may wish to suggest modifications of this idea, or different ideas.

[Dave McKee]

Then there's the issue of accounting for upstream work.  If someone borrows WP material (assuming we're under the GFDL), and we credit WP, do we need to credit every WP editor in that contributor list?
[/quote]

WARNING: the following post was written at 1:30 am and may be completely and utterly wrong.

(looks at the GFDL and Wikipedia, and cries)

The FSF needs to make a new version of the GFDL to cover Wikis. I don't think they ever really considered them when making it...

Some problems [Assuming the "Title Page" is the History page, and it is individual wikipedia pages which are licenced under the GFDL]

A. Use in the Title Page (and on the covers, if any) a title distinct from that of the Document, and from those of previous versions (which should, if there were any, be listed in the History section of the Document). You may use the same title as a previous version if the original publisher of that version gives permission.

Permission has, AFAIK, not been given. Dates might give unique names.

B. List on the Title Page, as authors, one or more persons or entities responsible for authorship of the modifications in the Modified Version, together with at least five of the principal authors of the Document (all of its principal authors, if it has fewer than five), unless they release you from this requirement.

At which point, a straight fork of a perfect Wikipedia article needs some additional names slurping across. Unless Wikipedia articles have no principal author.

D. Preserve all the copyright notices of the Document.
E. Add an appropriate copyright notice for your modifications adjacent to the other copyright notices.

Bugger. At this point, the change log is the only list of copyright notices. We *must* preserve *all* the change log, otherwise we're breaking the GFDL.
And that's being generous and assuming they're actually copyright notices, otherwise every modification of an article is breaking the GFDL.

I. Preserve the section Entitled "History", Preserve its Title, and add to it an item stating at least the title, year, new authors, and publisher of the Modified Version as given on the Title Page. If there is no section Entitled "History" in the Document, create one stating the title, year, authors, and publisher of the Document as given on its Title Page, then add an item describing the Modified Version as stated in the previous sentence.

And much the same thing again: we must preserve the change log.

G. Preserve in that license notice the full lists of Invariant Sections and required Cover Texts given in the Document's license notice.
Thankfully, Wikipedia has none.

J. Preserve the network location, if any, given in the Document for public access to a Transparent copy of the Document, and likewise the network locations given in the Document for previous versions it was based on. These may be placed in the "History" section. You may omit a network location for a work that was published at least four years before the Document itself, or if the original publisher of the version it refers to gives permission.
Does not appear to negate the need for the History section.

Gah. Does someone want to go poke FSF?

Dave.

dave.mckee at google's fantastic email service. ('Dragon' Dave McKee on CZ)