NOTICE: Citizendium is still being set up on its newer server, treat as a beta for now; please see here for more.
Citizendium - a community developing a quality comprehensive compendium of knowledge, online and free. Click here to join and contribute—free
CZ thanks our previous donors. Donate here. Treasurer's Financial Report -- Thanks to our content contributors. --

CZ:Proposals/Involving authors in approvals

From Citizendium, the Citizens' Compendium
Jump to: navigation, search

This proposal is presently driverless. Why not become its driver?
You can sign up on its proposal record, which may be found on the driverless proposals page.


Driver: None

Complete explanation

1) Anyone should be able to nominate any article for approval (even their own article).

2) To approve articles requires the support of relevant editors. But authors should also be able (and indeed be encouraged) to express their support or otherwise for approval.

Reasoning

1) I see every reason why all contributors to Citizendium should feel encouraged to read articles with potential approval in mind.

2) As Citizendium values fluency and style, then the opinions of (non-expert) readers are very important in an Approval process that properly matches the ambitions of Citizendium.

Citizendium rightly values expert knowledge and understanding. But these things alone do not a great article make. Style, presentation and readability are at least as important, and on these things the opinions of technical experts are no more relevant and often less useful than those of non-experts. We want people to read our articles for goodness sake so let us hear and value those opinions. Obviously how much weight they carry will vary by article - a highly technical article on a specialised topic needs to be clear, concise and accurate, not accessible. Other articles are the reverse - they need to be engaging, accurate and interesting but need not be exhaustive and academic.

Implementation

This will need the following changes to the Approval Processes.

Existing text An editor in the article's workgroup will be needed to nominate the article for approval. If the editor has worked on it herself as an author, she asks another editor to approve it; or, if there are several editors all doing significant work as authors on the article, then at least three of them can agree to approve it. (These rules are to prevent a single person from approving his or her own work without involving review by experts who were not authors.)

Replace with:

Any member of Citizendium, including any of the authors, may nominate an article for approval. The nominator should briefly justify their nomination in a message on the article Talk page, referring to the criteria for Approved articles. The approval date should allow at least 21 days - enough time for editors and others to take a look.

Existing text:(proposed replacements in bold)

So then (one of) the nominating editor(s) (nominator) places a [] template on the article's talk (discussion) page.

If another editor (any editor) in the article's workgroup finds that the article is so objectionable that approval, in his considered opinion, should not be granted,

If the nominating editor (nominator) notes that the discussion on the talk page that has occurred since the nomination for approval template was originally placed brings up important objections, then the editor (nominator) may delay the date for approval on the "to approve" template,to allow for work to continue before the stable version is generated. The nominating editor(s) (nominator(s)')may also change the version nominated for approval on the "to approve" template to an updated draft that is considered superior to the one first nominated.

Addendum: after approval, copyediting may be performed by a nominating editor (...the nominator or by any editor of the relevant workgroup) with the help of the approvals editor. This may occur at any time.

Present text

Who may approve For any given topic, only editors who may be considered experts on that topic may approve an article on that topic, at a minimum, these experts must be editors in the article's workgroup.

Replace with:

Nominators may select either to seek Community approval or expert approval, depending on the nature of the article. If the article is technical in nature, then expert approval will be the only option. However if the article is broadly based, or of general interest, or intended for a very general readership, then Community approval will generally be more appropriate.

Add

Community approval.

Community approval will need the support of at least four members of the community other than the nominator, who should declare and briefly explain their support on the Talk page. If any editor in the relevant workgroup opposes approval then he or she should explain why and may delay or block approval. Opposition to approval from other readers should be expressed clearly (politely and constructively) on the Talk page and the points raised should be addressed. However some dissent will not prevent approval if there is, in the view of the Approvals Manager, a general balance of opinion favouring approval


The approval template will need some adjustments, and placing a template should trigger an alert to workgroups and for community approval, to the whole community - perhaps a call on the Main page?

Discussion

This needs some implementation details, but I think it is rather a good idea, especially given the paucity of active editors in some workgroups. Right now, someone may create really good articles about engineering, but unless the (other) editors are paying attention, or are notified by the authors, they may languish. If the author notifies a few of the editors listed in the workgroup, but chooses only inactive editors, the article will still languish, unchecked, and unapproved.

I see two things which may need technical implementation. First, I think that there should be some way to distinguish non-editor nominations and self-nominations by editors from outside nominations by editors (which currently lead to semi-automatic approval). Second, in the case of author nominations and self-nominations by editors, it might be nice to notify all the editors in the appropriate workgroups of the nomination, so they can weigh in on it. Anthony Argyriou 14:58, 14 February 2008 (CST)

I'm inclined to support both points. It desperately needs a driver and to be spelled out in considerably more detail, however. --Larry Sanger 19:13, 14 February 2008 (CST)

Its a great idea. Denis Cavanagh 08:55, 15 February 2008 (CST)

Now that some details are added, I see that it's a far more significant change than I thought. Especially the community approval, where articles can be approved without any input from experts, worries me. Can we get some more information or examples on what articles are eligible for community approval? Who decides? I'm also awaiting details for how to notify the community / editors. -- Jitse Niesen 09:27, 15 February 2008 (CST)

I must share this concern (as an author) with Jitse. To have a system of "community approval" would undermine the core idea that this wiki is founded upon, that we have "approved articles" reviewed by "experts". I think we would receive a lot of criticism for choosing to go down this road and I believe in the long run it will do us more harm than good. However, I think that it's reasonable that an author should be able to nominate an article for approval and that nomination should be vetted by editors.
I don't share this concern, because some articles, like Bowling, could be approved by pretty much anyone who has ever bowled. David E. Volk 10:53, 27 March 2008 (CDT)
I think the core problem is that there's no system to programautomatically notify editors that articles need reviewing. --Robert W King 09:41, 15 February 2008 (CST)

The mechanism puts editors in pole position in that they and only they can block approval. It doesn't give editors any less power (if they wish to use it), but it does give non-editors a bigger involvement. The proposal is really to address the many articles that are written to be of general interest, and which accordingly draw on many areas of fact and knowledge. One obvious example is Edinburgh, another is Tony Blair, which I mention because both have been recognised (one as Article of the week, one as New Draft of the Week). These both are written for the general reader. Who has the expertise to vet Edinburgh exactly? Should we ask someone to check the geography and the history - do we wait for an academic expert on Edinburgh history and geography - actually the facts as there are verifiable readily by anyone through the links. It doesn't take an expert to verify, just a conscientious reader.

So if the article is written with a general readership in mind, yes it must be accurate but more importantly it should be interesting and readable. It's for these reasons that the article needs to have the opinions of a general readership.

Now any editor is free to say - no this is in such and such a workgroup and really needs expert review. That's fine. But actually, often, articles don't. They just need intelligent review. Gareth Leng 10:13, 15 February 2008 (CST)

On alerting the workgroup, yes I think this should happen automatically with the template regardless of whether its Community approval that's sought.

Who chooses? I think the nominator proposes which path he thinks is appropriate. Obviously if any editor in the relevant workgroup thinks it needs an expert review, then that option will be denied. But of course, obviously an expert can support approval if the Community route is chosen - and in that event the bar to approval will in effect be higher than for expert approval. Indeed, in general, I suspect that expert approval will be very much the softer option for most articles, and community approval will only be sought for articles that really aspire to be readable and accessible as well as good.Gareth Leng 10:31, 15 February 2008 (CST)

So - for Edinburgh especially I would rather put this to the potentially painful test of how my target readers see it than to get expert endorsement merely that it is accurate, when it is based on publicly freely available information.Gareth Leng 10:58, 15 February 2008 (CST)

There's already CZ:Ready for approval for authors to bring articles to the attention of editors. In my opinion a lot of articles on that page aren't really ready for approval. If authors were allowed to add the approval template, Category:Articles to Approve would likely fill up with author approval nominations that editors have not got around to canceling, and would therefore be a lot less useful. The standards required for an expert to stake their real-world reputation on are very high, so it's natural that most of our articles will not be of approvable quality. IMHO, Citizendium has few approved articles due to a shortage of active editors and approval-quality articles, not due to a flaw in the approval process. If you're impatient to get more approved articles, try recruiting editors. --Warren Schudy 16:01, 15 February 2008 (CST)

There are two quite active politics editors: Martin B.E. and Richard Jensen. Richard Jensen has edited Tony Blair and hence cannot approve it, but Martin should be able to approve it. Warren Schudy 16:12, 15 February 2008 (CST)

This is missing the point; it's not about those two articles (I just used them as obvious examples) or any particular articles - the issue is a broad one - who is best qualified to approve articles intended for a broad or general readership (e.g. William McGonagall or P.G. Wodehouse or Douglas Adams). And those will include the articles on popular culture that we have, articles on sport, as well as many articles on people and places and some on wide general issues. On these articles, expert opinion is no more valuable per se than the opinion of any intelligent interested reader. If we want articles to be readable, interesting and to a high standard of presentation then this should be reflected in an approval process. As for the numbers of good articles ready for approval - I'm just not sure. I agree that the CZ:Ready for approval list doesn't have many - but there are others out there not on the list. Some are technical articles suffering from the absence of relevant editors, some are general (cross workgroup), some are of general interest.Gareth Leng 03:51, 16 February 2008 (CST)

I think it's an excellent idea. However, this sentence needs a lot of clarification: "However some dissent will not prevent approval if there is, in the view of the Approvals Manager, a general balance of opinion favouring approval." What went before sounds as if a sole editor can veto the community approval, whereas this sentence seems to allow the Approvals Manager to allow a page to be approved if the number of people favouring approval exceeds the number dissenting -- a very different thing. How about: "A single editor of a relevant workgroup can veto approval. If no editors approve or veto the page, then for the page to be approved, the number of approving authors must be four more than twice the number of dissenting authors: thus if 10 authors oppose approval, 24 authors must support approval for it to be approved."

I see no particular reason why there can't be a variety of levels or types of approval. I think it would be useful to have a category, for example, of pages classified as factually accurate but not particularly well-organized or readable. Author-approved articles would be simply another category of pages, to be used by the readers as they choose, just as developed articles are also available for readers to use. --Catherine Woodgold 20:17, 21 February 2008 (CST)

Thanks, you make good points. I'd be happy with your suggested rewording. Equally happy to let it loose for now and fix rules in the light of some experience.
I think we should be careful with "author approvals" - there should be some thorough checking of the factual content by *someone*, to avoid problems of selective citation, and some review to watch for selective emphasis. For example, I could write an article about Oakland, California which made it out to be a really wonderful place, or a real pit. Oakland has aspects of both, and a good article should reflect that. But if the approvers weren't familiar with Oakland, and merely checked my citations, they'd approve a one-sided article. Anthony Argyriou 23:24, 28 February 2008 (CST)
This is a good point - and one that also holds for expert approvals. When you really get into expert issues there are very few people in the world really competent to judge (believe me I know exactly how hard it can be to find competent reviewers). I guess what I'd like to see is people stating clearly the grounds for their support for approval - e.g. declaring that an article is very clear interesting and readable (but I can't vouch for accuracy), or (from an expert) that the section on molecular biology is fine but I'm not competent to really judge the physiology etc. I think I'd like somehow to see a way to a path whereby everyone can feel that they have a relevant input - as anyone whose read the article should indeed have. I think though that while we can evolve towards that, we can't set rules in advance, we really have to see how it goes and establish the culture of expectation - maybe by setting examples. But what I really want to see is people taking an interest in the quality of Citizendium generally, and feeling involvedGareth Leng 13:32, 29 February 2008 (CST)


I think levels or types of approval might be a good idea - it may be premature just now (adding to the complexity) but as Citizendium grows that I think would be useful.Gareth Leng 03:39, 27 February 2008 (CST)


The "complete explanation" section and the info box in the proposals queue state that "to approve articles requires the support of relevant editors". The implementation section documents an approval procedure that does not require editors. My guess is the implementation section is right and the summary section is wrong; is that right?

Changes in the proposal I'd like to see:

  • Community approval should have a long minimum time frame, such as 15 days, to make up for the potential lack of expertise.
  • Require at least 3 authors or 1 editor to concur that an article is ready for approval before the to-approve template can be placed.
  • Indicate that if the expert track is chosen, at least 3 editors are required unless one of the editors has not edited the article significantly. That rule used to be in the paragraph on nominations, but that paragraph was rewritten, eliminating that rule.

Even if these changes were made, I don't think I would support this proposal at this time. If in a few years we continue to have articles that no editor feels are sufficiently in their area to approve, then we may need to consider allowing non-editor approvals. I'm not convinced there is a problem here worth jeopardizing the credibility of Citizendium approvals. --Warren Schudy 23:48, 4 March 2008 (CST)

timeframe

Forgive me for being so backlogged that I would find 14 days too short a time frame to get through an approval process. Could it be at least 3 weeks? Also, I do think any one editor should be able to veto an approval, but there should be an arbitration process if other editors feel that the one editor's objection is unreasonable.Pat Palmer 19:06, 20 April 2008 (CDT)

Thanks. Point taken, changed to at least 21 days. If other editors disagree then this can be resolved through the normal approval process; essentially this is intended as a process for articles of general or wide interest where the main issues are style, readability, breadth of coverage etc, issues where there is no clear role for expert editors. If editors feel they can be actively involved in approval, the process could take the normal route.Gareth Leng 03:57, 29 April 2008 (CDT)


Proposals System Navigation (advanced users only)

Proposal lists (some planned pages are still blank):