Many thanks for that reply - very useful to have the facts out in the open and I hope it helps to build trust.
When oversight or suppression are used, it's book policy that oversighters almost never discuss or disclose anything, beyond what can be seen openly in the public logs.
In many cases, that makes sense. However, in this case, the sensitive material was only sensitive at the time - once the subject was released there was no continuing risk.
As you mentioned, oversight wasn't necessary in this case. However, it's not inconceivable that another case where oversight is used might also be "temporarily sensitive". Perhaps, for instance, if it has been used in a suspected harassment that turns out to be something else.
In that case, it might make sense for the "book policy" to allow disclosure (or even reversal) of the oversight in these cases.
----- "FT2" ft2.wiki@gmail.com wrote:
From: "FT2" ft2.wiki@gmail.com To: "English Wikipedia" wikien-l@lists.wikimedia.org Sent: Thursday, 16 July, 2009 21:20:04 GMT +00:00 GMT Britain, Ireland, Portugal Subject: Re: [WikiEN-l] News suppression: Did it use Oversight or RevisionDelete?
Update: I've now checked the case, and yes I had heard of this matter. But being on a break for the last few weeks to deal with real-world matters, I hadn't made the connection just from the words "Rohde/NYT". I checked which article with Rohde in the title, also covered the NYT as well. Luckily there was only one.
Quick explanation :)
FT2
On Thu, Jul 16, 2009 at 9:06 PM, FT2 ft2.wiki@gmail.com wrote:
A quick answer.
I have no idea which dispute or real-world issue this was about, nor when. I'm assuming following a quick search the page concerned is "David S. Rohde".
When oversight or revision delete are used, it's almost without exception for serious reasons, for example where there is a concern over potential defamation or breach of privacy policy in the post. Not mere offensive comments, and not mere undesirability. A significant number of users cross-check each other on it, and there is an audit committee on english wikipedia to investigate any concerns as well. Privacy issues are taken extremely seriously.
When oversight or suppression are used, it's book policy that oversighters almost never discuss or disclose anything, beyond what can be seen openly in the public logs. The trust required is why oversighter selection is a big deal. The underlying reason for the policy is that sometimes just having confirmation that a person or topic was targeted can be enough to do serious harm, when genuine cases such as stalking and serious harassment etc are intended by someone, if you think about it. (And if some were answered and others weren't then things might be read into a non-answer.)
So the standard answer to all inquiries of this kind by any oversighter is "we don't discuss such matters, but we will look and check nothing untoward has happened, if you would like"
However in this case I have discussed the inquiry and can confirm, that no material was or has ever been oversighted or suppressed (using revisiondelete) from the article I think you're referring to, "[[David S. Rohde]]".
Hopefully that's enough to put your mind at rest. Don't count on such confirmation another time -- it's exceedingly rare to get it :)
FT2
On Thu, Jul 16, 2009 at 7:41 PM, Joseph Reagle reagle@mit.edu wrote:
Does anyone know if during the NYT/Rohde case the Oversight function was used to hide edits? When the story broke, I could see all the edit history, but I presume the function can be deployed against select revisions and then removed? Or maybe it was the new RevisionDelete?
WikiEN-l mailing list WikiEN-l@lists.wikimedia.org To unsubscribe from this mailing list, visit: https://lists.wikimedia.org/mailman/listinfo/wikien-l
WikiEN-l mailing list WikiEN-l@lists.wikimedia.org To unsubscribe from this mailing list, visit: https://lists.wikimedia.org/mailman/listinfo/wikien-l
On Thu, Jul 16, 2009 at 11:47 PM, Andrew Turvey < andrewrturvey@googlemail.com> wrote:
As you mentioned, oversight wasn't necessary in this case. However, it's not inconceivable that another case where oversight is used might also be "temporarily sensitive". Perhaps, for instance, if it has been used in a suspected harassment that turns out to be something else.
In that case, it might make sense for the "book policy" to allow disclosure (or even reversal) of the oversight in these cases.
Its very rare - almost all uses of oversight/suppression are material that isn't time dependent. In the few exception cases commonsense applies. In this case the focus is known and the event that it's hanging on is a publicly known one. In such a circumstance there may be a possibility. But suppose during the incident someone had posted some personal information, defamation, possibly defamatory accusation about the subject or someone involved..... I would then have had to say "sorry, I wont be answering that" and you would not have had a way to know if there was no issue, material relevant to the incident, or completely irrelevant material just happened to be posted to that page.
Unfortunately the problem is that a promise to disclose in some cases implies that conclusions will be read into others. That's got to be a no-no, however much one might wish otherwise. I hope you can understand that; there doesn't seem to be an easy way around it that ensures the system won't get gamed, except trust, commonsense, and understanding of the reasons behind it.
I'm sorry. I can't see a way round it that protects privacy, if there is a norm that disclosure will be given in some cases but not others. if that were a norm, it could too easily be used for probing if there were some privacy issue or harassment case or whatever - and some people would ask because they wanted to know or wanted to "uncover" stuff, not realizing it is for a good reason, and some real live person might be affected who these policies are precisely there to prevent being harmed in these serious ways. If you want to discuss it, a thread on Meta would be the way.
But whatever was decided, there will always be some matters where it just won't work, and I think that's just got to be accepted. It's why oversight exists in the first place.
FT2
On Thu, Jul 16, 2009 at 3:47 PM, Andrew Turvey < andrewrturvey@googlemail.com> wrote:
In many cases, that makes sense. However, in this case, the sensitive material was only sensitive at the time - once the subject was released there was no continuing risk.
As you mentioned, oversight wasn't necessary in this case. However, it's not inconceivable that another case where oversight is used might also be "temporarily sensitive". Perhaps, for instance, if it has been used in a suspected harassment that turns out to be something else.
Discussion here can get to be a bit confusing -- "oversight" can refer to too many things. In the interest of clarity, I'll use "oversight" to refer to the user priviledge, HideRevision to refer to the older oversight tool, and RevisionDelete or RevDel to refer to the new tool -- oversighters have access to both tools and their respective logs. By my understanding, anything removed using HideRevision can't easily be viewed or restored, except I believe with direct intervention by developer(s) with database access; anything removed using RevisionDelete can be restored by any oversighter, and can be viewed by oversighters (and sometimes admins, depending on settings used at the time of suppression).
That covers a few raw technical aspects. In terms of policy and practice, I'm not immediately aware of any cases where something deleted with RevDel has later been restored, nor am I confident at first impression that doing so would ever be wise. I certainly would be loathe to do so, absent perhaps cases of obvious mistakes or egregious abuse (think compromised accounts, say), and definitely not without relevant precedent or prior discussion.
I might be thinking of something you're not, though -- from personal experience with the tools, the majority of cases I see are terribly mundane, "Foo is dumb and his phone number is xxx-xxx-xxxx" types.
-Luna