Proposed tweak of approval procedure

[Warren Schudy]

Sometimes an editor nominates an article for approval and then disappears before the process is finished, presumably because of real-life business.  By current procedures, silence on the part of editors is interpreted as continuing acceptance of the approval, even when authors object to the approval.

For example, the editor who nominated [[Sailing]] for approval did not edit the wiki after the nomination. Authors objected to various problems, but no editor showed up to cancel the pending approval. The only reason that the faulty article was not approved was the fact that the nominating editor forgot to specify a date for the approval to take effect.

The pending approval of [[Fossilization]] was recently withdrawn the day after it was scheduled for approval because some remaining minor issues had not been addressed (http://en.citizendium.org/wiki?title=Template%3AFossilization%2FMetadata&diff=100268051&oldid=100262430).  If a constable had visited that page yesterday, my understanding is it could have been approved, despite an editor's objection to approval on the talk page.

It doesn't seem reasonable to expect editors to remove the to-approve template whenever problems are found and add it again when they are fixed. Instead, I propose that constables not be allowed to "do the honors" on their own, but only upon explicit request by at least one nominating editor. This change would ensure that the nominating editors still support approval of the version indicated in the template.

[Matt Innis]

Just FYI, I did visit Fossilization yesterday and did not approve due to the untouched list of open issues written by one of the editors.  However, I did approve Petroleum refining processes.  The first paragraph of the sysop instruction is to look at the talk page and make sure there is agreement among the editors.  So I did not perform the mechanics of approval. 

Sailing, however, was interesting.  If you notice, I put a message on the talk page to get attention that the article did not look appropriate for approval, and a lot of authors stepped in to agree, and even made some changes, but no-one made any efforts to contact the editors that I can tell.  So technically, I would have had to approve the article as it was, but would I?  Well, I'm not sure because the date kept changing :-) so I never had to actually consider my options.  I think I would have probably looked to the editor and seen he had not been here, then emailed him and if he did not respond, I would have notified the Editor in Chief.  But I can't be sure.  I also may have gone ahead and approved it in hopes that someone would step up and fix the article.  I have to be careful to determine whether I am being asked to follow the letter of the rules or to use common sense.  It's those gray areas that get sticky. 

While I have been contacting editors and giving hints 2 or 3 days ahead of schedule, I am not sure this solution works in the long haul.   As things get really busy, I don't think it is reasonabe to expect a constable to track down an editor for a workgroup.  Authors need to understand that if they have a problem with an article, they need to go back to the workgroup and get it fixed before the constable even gets there.  Unless of course we expect to have more constables than authors... which I hope is not the case.

[Warren Schudy]

IMHO, if a decision is likely to be controversial one should lean towards the letter of the law, but otherwise spirit is better.

As you mention, the current approval process asks sysops to:

Examine the talk page. Make sure it's clear that there are three editors who are in agreement about the approvability of the article, or, if it seems there is an individual approval going on, make sure that the person approving the article has not worked much on the article. (For that, examine the article's history.)

I think this should be strengthened a bit to more fully address this problem. How about adding:

Make sure not only that three editors support the approval in principle, but also that at least one editor currently supports the approval of the specific version mentioned in the approval template. If there is any doubt in their support, request clarification before doing the honors. Reasons to doubt their support include the nominated version being out of date or the existence of unanswered author critiques.

I am not asking constables to track down editors. If support for approval is unclear, the constable should simply leave a note on the talk page to that effect and move on. It's the editor's responsibility to read the note. If the editors don't show up to respond, then the article simply stays unapproved until an editor shows up.

[Matt Innis]

Make sure not only that three editors support the approval in principle, but also that at least one editor currently supports the approval of the specific version mentioned in the approval template. If there is any doubt in their support, request clarification before doing the honors. Reasons to doubt their support include the nominated version being out of date or the existence of unanswered author critiques.

I have no problem with that revision, as it seems to be a better explanation of what we already do.

I am not asking constables to track down editors. If support for approval is unclear, the constable should simply leave a note on the talk page to that effect and move on. It's the editor's responsibility to read the note. If the editors don't show up to respond, then the article simply stays unapproved until an editor shows up.

Good news.  Anything that gets more author/editor collaboration to get these things approved will benefit the quality of the article.

[Martin Baldwin-Edwards]

The important underlying principle in all of this should be, when in doubt, don't approve. The default should always be to request clarification, or delay approval, rather than to go ahead. In practice, this seems to have occurred anyway. Formalisation of existing practice is a generally good thing to do, so I have no dispute with any of this.

[Warren Schudy]

Barring any objections, I will implement this change in a few days. (This doesn't seem important enough to be worth getting a more official approval.)

[Larry Sanger]

I'm going to ask that you submit any reasonably significant changes to approval rules to cz-editcouncil@citizendium.org.  Simply clue the Editorial Council in.  It's quite possible that no one will even comment, and if it's so innocuous, certainly no one will want to vote on it.  That's what I do, and I am basically requiring it for all of us: when we're off in some corner of CZ discussing things among ourselves, and make some decision that others don't know is going on, publicize the decision.  This is crucial to our remaining a fully open project.

I'll be writing & publishing a rule to this effect soon.

[Warren Schudy]

I'm going to ask that you submit any reasonably significant changes to approval rules to cz-editcouncil@citizendium.org.  Simply clue the Editorial Council in.  It's quite possible that no one will even comment, and if it's so innocuous, certainly no one will want to vote on it.  That's what I do, and I am basically requiring it for all of us: when we're off in some corner of CZ discussing things among ourselves, and make some decision that others don't know is going on, publicize the decision.  This is crucial to our remaining a fully open project.

I'll be writing & publishing a rule to this effect soon.

I just tried sending a message to the editorial council but apparently only subscribers can post to that list. Can someone on the Editorial Council please bring this to the attention of the Council, e.g. by sending the following message to the list?
------Message ----
The current approval process asks sysops to do the following before executing a
pending approval:
"Examine the talk page. Make sure it's clear that there are three editors who
are in agreement about the approvability of the article, or, if it seems there
is an individual approval going on, make sure that the person approving the
article has not worked much on the article. (For that, examine the article's
history.)"

As discussed in a forum thread
(http://forum.citizendium.org/index.php?topic=1563.msg14042), I propose adding
the following:
"Make sure not only that three editors support the approval in principle, but
also that at least one editor currently supports the approval of the specific
version mentioned in the approval template. If there is any doubt in their
support, request clarification before doing the honors. Reasons to doubt their
support include the nominated version being out of date or the existence of
unanswered author critiques."

Per Larry's request, I am hereby requesting your approval for this change.

Warren Schudy
http://en.citizendium.org/wiki/User:Warren_Schudy
---------

[Martin Baldwin-Edwards]

Warren, I could post it [although probably your email will be accepted: I think Larry moderates the list]. But I think my reaction would be:

(1) where is the actual text you propose to modify?> please provide a link

(2) provide the suggested amendment, and a quick explanation why (not only the Forum chatlink]

Can you email those to me, or post it here? I will then forward it to the EC.

At this time, I am still slightly bemused by the phrasing in the original of "three editors...in principle", since very few approvals have had the support of three editors. My feeling is that we may need to amend more than the small amount you suggest.

[Warren Schudy]

Here's an updated message draft.
----------
The current approval process (http://en.citizendium.org/wiki/CZ:Approval_Process#Sysops_make_it_official_with_the_.7B.7BApproved.7D.7D_template) asks sysops to do the following before executing a pending approval:
"Examine the talk page. Make sure it's clear that there are three editors who
are in agreement about the approvability of the article, or, if it seems there
is an individual approval going on, make sure that the person approving the
article has not worked much on the article. (For that, examine the article's
history.)"

I propose replacing that text with:
"Examine the talk page. Make sure it's clear that there are three editors who
are in agreement about the approvability of the article, or, if it seems there
is an individual approval going on, make sure that the person approving the
article has not worked much on the article. (For that, examine the article's
history.) Also ensure that at least one editor currently supports the approval of the article
version mentioned in the approval template. If there is any reason to doubt editor
support, request clarification before doing the honors. Reasons to doubt editor
support include the nominated version not being the latest one and/or the existence of
unanswered author critiques."

The rationale for this change is to ensure that sysops err on the side of not approving rather than approving when editor support is unclear. For further discussion, see http://forum.citizendium.org/index.php?topic=1563.msg14042.

Per Larry's request, I am hereby requesting your approval for this change.

Warren Schudy
http://en.citizendium.org/wiki/User:Warren_Schudy

[Warren Schudy]

Martin, did you send that message to the editorial council? Did anyone reply?

Thanks,
ws

[Martin Baldwin-Edwards]

Yes. As I recall, there was one query from Matt. I don't remember the content. Otherwise, there was no objection to proceeding without voting.

[Warren Schudy]

Change made: http://en.citizendium.org/wiki?title=CZ:Approval_Process&diff=prev&oldid=100275482 .

ws