Talk:Response testing/WMF Projects

I think this should bed in for a while, and I may try and propose something related to the 'mike handel' thing - any thoughts on that? Privatemusings 01:44, 12 April 2010 (UTC)

Peter, I really do think you should drop any idea of doing research on WMF projects for at least a few months. They are just a red rag to a bull for many people. Also I think WV needs to get its house in order about managing research projects that raise ethical issues in general. Australian universities, as you probably know, have Ethics Committees that review all research proposals that involve any interaction whatsoever with people (they also have a similar committee for any research involving animals). WV needs such an Ethics Committee and it needs research projects that come before it to be properly lead by people who have experience in similar research. It is not appropriate for research to be carried out by a loose bunch of amateurs. So, please forget about everything to do with testing WMF projects for the rest of this year and see what happens over that period. --Bduke 03:18, 12 April 2010 (UTC)

This project isn't about encouraging response testing as far as I can see, and repressing observational research because of the past flap over such testing is foolish. I see nothing in the project as it is now configured about organizing or encouraging testing, per se. And in the flap, I saw nothing objecting to observational research, as such, and the implication that officials of the WMF do not want us to understand how wikis, including Wikipedia, operate, which would be beneath an assumption that the objection was deeper than simply to response testing, is offensive. The objection to "loose bunch of amateurs" is also offensive to the open wiki concept. A project here will hopefully involve academic and other professionals and "amateurs," including students at academic institutions and others. If there is something offensive on the page, please point it out. Otherwise, please don't perpetuate that unfortunate dispute here by claiming that Privatemusings is doing so. --Abd 15:50, 13 April 2010

Excuse me, has Privatemusings ever revealed his name on this wiki? Geoff Plourde 03:20, 12 April 2010 (UTC)

(very quick 'cos I only have two secs right now) - I'm fine with folk calling me Peter - but thanks for the concern Geoff - let's de-escalate every chance we get :-) cheers, Privatemusings 03:32, 12 April 2010 (UTC)

Sorry, if I let that out. I thought I had seen you use "Peter" here. --Bduke 09:13, 12 April 2010 (UTC)

(longer reply) I can see what you mean, Brian, but I'd like to respond to a few of your concerns, and raise my own concerns about what they seem to me to imply. First off, I think I may need to make this page clearer that I personally don't intend to actually to any sort of research, I'm just interested in things that have happened in the past, and things that others have done (the newt project, the 'last words' hoax, the mike handel thing etc. etc.) - so given that, I'm happy to commit to not actually engaging in any active research for the forseeable. In the unlikely event that I decide there's something useful I could add to any research, I'll suggest, discuss, and detail what I'm up to openly here. Perhaps this calms a whole raft of concerns, and if not, could you indicate where concerns remain, and why?

Not actually doing research may calm things down, but I would still be inclined to keep away from these issues for quite a while. It will look to some as if you are just itching to start the whole drama up again. --Bduke 09:13, 12 April 2010 (UTC)

I also wanted to respond to the idea of an ethics committee - I'm broadly supportive of this idea, and of course would look forward to presenting something to such a body, should I intend to take any action (which as detailed, I'm not planning on) - but I'm not sure that the concept is entirely congruent with the spirit of wikis in general - in particular, you, like me, have probably heard this line; 'It is not appropriate for ..... to be carried out by a loose bunch of amateurs' in reference to news reporting, biography writing, encyclopaedia editing etc. etc. - I'm not sure on what principled basis you'd feel that wikiversity should be apart from other wmf projects in this sense, nor is it entirely clear to me whether you feel only research into wmf projects should in some way be different to research in other areas - do you feel that?

I do feel research is different. After all, universities employ their staff after a rigorous process yet they still think that research proposals that may affect people have to be looked at by their ethics committee. I do not feel that research on WMF projects is different in this sense. I think all research that would come under an Ethics Committee at the University of Melbourne, for example, should be treated in the same way here. We need to work out how to do it this is our own way, but it does have to involve experts in research. It is not like the isue of experts editing wikipedia. Without experts, the research will not be creditable. However, I do think research on WMF projects is different in another sense. It is just too close to us. It is like a university's Education Department doing research on teaching in the same university. It is better if they research a different university from their own. It raises sensitivities. It might be best if we left research on WMF projects to others and just offered them advice. --Bduke 09:13, 12 April 2010 (UTC)

The argument here, that the subject is "too close to us," would seem to rule out all study of psychology, sociology, politics, etc. Some, indeed, might think this way, but I don't. Obviously, when we study ourselves, we have to be careful. However, how is ethics relevant here? Encouraging response testing, per se, does raise ethical issues. Studying it does not. An Ethics Committee would surely be interested in being advised by a project like this one, but that does not mean that this project raises ethical issues. --Abd 15:56, 13 April 2010 (UTC)

I'm hoping that folk as measured, calm, and with your sort of background, might be up for offering further advice as little bits of work like this unfold - it'd be great from my perspective if you'd be up for sitting on something like a 'review board' - whaddya reckon? cheers, Privatemusings 04:46, 12 April 2010 (UTC)

I am a scientist. I do not research people. I only use computers for my research, so I am a geek, if an ancient one, just like most wikipedia editors. I have never sat on a University Research Ethics Committee or even on a University Animal Ethics Committee, although as dean and head of School I have seen what they have done. I think we need people who have been on such a committee getting involved. --Bduke 09:13, 12 April 2010 (UTC)

I think news agencies also often employ their news reporters after a rigorous process because news reporters often do research too. Yet unless I am missing something Wikinews seems to work just fine without an ethics committee or review board. Wikinews even has a section about Wikipedia. I'm not sure why Wikiversity should be any different in this respect to Wikinews. If research creditability is even something Wikiversity wants, I think following a decent research process should be all that is needed to achieve a level of creditability. -- darklama  12:42, 12 April 2010 (UTC)

Two points. First, I am not sure that "Wikinews seems to work just fine". I do not see it doing rapid cutting edge investigations that the news reporters you mention do. Is is not much beyond how wikipedia handles news items, just rather more inclusive. Second, how do we know that "following a decent research process" has been done? How can we convince people on other WMF projects that the research is sound, worth doing (and therefore worth them helping it) and ethically OK? Is that not what the recent row was all about? Wikipedia editors and admins were not convinced that wikiversity editors were doing something acceptable? The message was clear that wikiversity has to get its house in order on research. --Bduke 21:16, 12 April 2010 (UTC)

I hope you are not comparing Wikinews to TV news. TV news is where I would expect rapid cutting edge investigations. Why do people on other WMF projects need to be convinced? I doubt most news agencies and researchers that have done research about WP have asked for WMF's blissing first. Why should Wikiversity be any different in that respect? I think the recent row had nothing to do with worth doing, but a belief that the research was or could provoke people to vandalize Wikipedia. When I asked Jimbo during the community review if researching Wikipedia is within our scope his response suggested that it was. He even made some suggestions about how to go about doing the research in a way that I assume is intended to eliminate any kind of impression that vandalizing Wikipedia is being encouraged. -- darklama  22:21, 12 April 2010 (UTC)

Let us leave wikinews and concentrate on research here. Where does "asking for WMF's blessing first" come from? I was not suggesting that. I do not see that you are really disagreeing with me. You ask "Why do people on other WMF projects need to be convinced?". Well clearly on WP, they need to be convinced that WV research there is not going to vandalize WP". I am not suggesting that research on WP is outside our scope. I am concerned that it has to be done right. I think we just have to be very careful. I think we have to self-police it. Have folks here seen this group on WP, which is an attempt to help outside researchers interact better with WP and ensure that editors there know the research is OK. I think research from wikiversity on wikipedia would fit that just as it fits research from outside universities. That Group is still controversial. I came across it only when it was up for deletion, but that debate has just been snow closed as keep. --Bduke 00:38, 13 April 2010 (UTC)

I think there wasn't any WV research going on within WP. Where is that coming from? How does Wikinews self-police? I'm looking at WN because I think that WV could learn some things from WN on how research can be done right or at least better. Wikinews is the only other Wikimedia project that is allowed to do some form of original research. I believe Wikinews has already been through this problem itself. I think the solutions that Wikinews has found to work should be looked at to see how they might be adopted for use by Wikiversity. Why reinvent the wheel when another project has tackled similar issues. -- darklama  01:32, 13 April 2010 (UTC)

OK, set up a page to let us explore what we can learn from wikinews. I doubt that I will be able to help. "there wasn't any WV research going on within WP" - is that not what "Response testing on WMF projects" was all about? People here proposing to research on wikipedia? I think we might be confused about what we mean by research. Is this serious research that might be peer-reviewed published? Or are we just talking about finding stuff out in the way WP calls original research? If we are in any way like a university, we should be encouraging serious research, or shall we define wikiversity as a "teaching and learning institution only"? I'm going to leave this now. I think it is clear, after recent events (including the comments on meta in the proposal to close wikiversity down), that wikiversity needs to look very carefully at itself. This is not the place to do that. --Bduke 07:27, 13 April 2010 (UTC)

"Wikinews is the only other Wikimedia project that is allowed to do some form of original research." Minor point, but I'd say Wiktionaries have original research and are allowed to have it. Also, less of a minor point, I would have thought it something the communities were allowed to decide on (perhaps within certain parameters)? Ever wonder 16:33, 14 April 2010 (UTC)

Research policy

Please participate in discussion of research ethics at Beta Wikiversity. --JWSchmidt 06:50, 12 April 2010 (UTC)

Action research

Wikimedia Ethics is usefully categorized as action research. Action research is a way for members of the Wikimedia community to study their community and seek ways of improving the community. Such action research is a normal part of participation in Wikimedia wiki projects and requires no special oversight or review beyond those already provided for in Wikiversity policy. --JWSchmidt 15:52, 12 April 2010 (UTC)

"Action research" seems to imply taking actions to test (what? ethics?). This project doesn't seem to be suggesting that, and it was the idea that there would be testing that seems to have caused all the flap with the former project. Perhaps you can better explain "action research," and, hopefully, it doesn't mean active-interference lab work, where wikis and their editors are treated as experiments, instead of as systems to be studied neutrally through observation. (The wikis can be experimental themselves, but by "experiments" I mean what was called "response testing.") --Abd 01:36, 13 April 2010 (UTC)

See the page about action research at Beta Wikiversity. --JWSchmidt 13:43, 13 April 2010 (UTC)

Cool. Action research is obviously great stuff. However, I was referring to the wikipolitical appearance of the usage of this term in the context of recent history. I requested explanation to defuse this possibility. Perhaps that page on beta is enough. And perhaps not. After all, we are dealing with people who sometimes react instead of researching. --Abd 15:39, 13 April 2010 (UTC)

There is a lot of overlap between action research and the 'wiki way': identifying something that is causing a problem, taking action to remedy that problem, reflecting on the action that was taken, and using this to prompt further action. (That is one characterisation of action research, but a commonly used one; incidentally, I don't know why there is a page on beta as well as the page on action research here.) However, I don't think that we can wave away the problematic aspect of action research as easily as saying that it "requires no special oversight...". I think that all human research, and particularly action research, is a particularly sensitive endeavour, and needs to be very careful about how it does what it wants to do. Beta's guidelines are good and will help us here - but there isn't much specific detail there on what to do in such cases as have been causing so much trouble over the last two years. I would say that some of the principles of action research might help us here - specifically, in involving the people about whom the research is oriented in the designing and doing of the research itself. In other words, if studying Wikipedia, involve the individual Wikipedia editors who would have an interest in building such knowledge. Think of the research as a good faith collaborative endeavour, rather than something attempting to uncover some 'truths' about some group of editors 'over there'. Cormaggio ^talk 08:15, 14 April 2010 (UTC)

"such cases as have been causing so much trouble over the last two years" <-- It is important to understand what caused the trouble. The Wikiversity projects that have been attacked by Wikipedians caused no trouble. The bad behavior of Wikipedians trying to disrupt Wikiversity is what caused the trouble. The Wikipdians failed to assume good faith and acted on false claims about "attack pages" and "trolling". --JWSchmidt 13:44, 14 April 2010 (UTC)

We are talking here about Moulton's ethics project, yes? Isn't it worth asking why people on Wikipedia felt attacked, instead of dismissing their actions as assuming bad faith? I would also include in this discussion your own learning projects around deletionism, e.g. Albanian sea port history. I don't think that these projects (i.e. yours and Moulton's) set an initial tone that was conducive to good faith collaborative work. Furthermore, the concerns were not from some external community (even if that external community is still within Wikimedia), but were also voiced from within the Wikiversity community, including myself (as you well know). My main point here, John, is to try to find ways of promoting good faith collaborative work without provoking acrimony - can you help with that? Cormaggio ^talk 15:07, 14 April 2010 (UTC)

Cormaggion, it is worth, indeed it is necessary, for this project to develop its own ethics guidelines, or to adopt guidelines that are developed on a larger scale. If I imagine this a seminar at a University, I would very much dislike seeing it become a star chamber that prosecutes and convicts "culprits." I added an NPOV tag here, which, if accepted, will make it clear that this is not about ascertaining guilt or blame, it is about deepening our understanding of what actually happens. Thanks for your comment. I will extend this, though, to noting that we are also not about, here, blaming Privatemusings or Moulton or anyone else for prior disruption. It's simply off the point. There is no active resource here, as far as I'm concerned, until one more editor shows up who wants to participate, rather than just debate whether it should exist or not. --Abd 18:00, 14 April 2010 (UTC)

Cormaggio, "Isn't it worth asking why people on Wikipedia felt attacked, instead of dismissing their actions as assuming bad faith?" - Loaded question. Do you think nobody asked? Which parties do you think were more responsible for the break down in communication? Which parties preferred deleting content and blocking instead of talking and resolving their differences? Hillgentleman | //\\ |Talk 16:20, 14 April 2010 (UTC)

It is worth asking, and, with regard to the Newbie project, that would be a fundamental part of the work, to examine the perceptions of the project. However, I caution you against an approach that attempts to assign blame. A sober and neutral examination will bring out evidence that might be used, somewhere, by someone, to make judgments, but that is not our task here. Agreed? --Abd 18:00, 14 April 2010 (UTC)

I would love to agree, but though what you said would be ideal, what in fact would and did happen is that these people would prefer to cry foul and remove your questions instead of adding their own comments from their point of views. Hillgentleman | //\\ |Talk 18:10, 14 April 2010 (UTC)

Let me put it this way. If they do this, then meta-analysis will reveal exactly their agenda. You have a concept of "questions" being removed. I don't see the project as asking questions, at least not primarily. Questions are often designed to further an agenda. Where would these "questions" be asked? (Beyond the basic question, which we will be answering, as I see it, of "What happened?") Further, you have a concept of "comments," coming from "points of view." That's exactly what we will try to avoid. We will be seeking *evidence*. It may start simply by finding and pointing to all the relevant discussions and edit histories, making the data accessible more easily. Right now, it's a lot of work to just figure out what, say, the Newbie thing is about. If "comments" exist in the evidence, to make it easier to approach the evidence, they will properly be rigorously neutral, and if there is objection to them, they can easily be removed. The point is not to "accuse" anyone of anything except having an edit history with a specific set of edits. And this is where, if anyone thinks it's been cherry-picked (rather than selected according to a stated criterion), the addition of the rest of the cherries would be in order. "That was cherry-picked" is a POV assertion, unless backed with evidence, and if the evidence is included, the entire body of evidence is no longer "cherry-picked," it is more complete.

Let me propose an example: Suppose it turns out that some admin made a deletion that looks abusive. First of all, we aren't going to call it an abusive deletion, we are simply going to describe what happened, and "abusive" is a judgment, not a fact, unless very clear criteria have been established. But, even if we do this carefully, someone will complain, you can almost count on it, that this was "cherry-picked," that the editor made 1,000 deletions and that's just one of them. So, fine, anyone can put in evidence of 1,000 deletions. Diffs for all? Sure, if someone wants to go to that trouble and insists on it rather than just giving the statistic. And that would then be a subpage with the detail. The right to include evidence is not a right to bury a page in it! But, of course, this would then more or less invite an examination of the other 999 deletions. Quite a bit of work, but it could be done. Again, a criterion for classification would be decided. Okay, someone trying to disrupt the process -- or with a legitimate concern --will debate the criteria. Fair enough. Anyone can propose a criterion and then apply it and report the results!

Once we understand that agreement on evidence (what it is, not what it means) is a valuable goal, once we develop the habit of inclusion when it comes to evidence, there will still be conflicts, but, I expect, they will not prevent us from conducting a sober examination, that others may then find valuable. We will generate a consensus report, with minority reports allowed. Again, this is all pretty standard process in real environments where collegiality and consensus are important, while freedom of inquiry is also valued. --Abd 19:50, 14 April 2010

"why people on Wikipedia felt attacked" <-- in 2008 a few Wikipedians did not want it revealed how they conspired to violate Wikipedia policy. In 2010 RTG and others interpreted "ethical breaching project" to mean that it was a project designed to breach ethics. I'm not interested in dismissing anything...I'm interested in continued study of people who called a study of Wikipedia policy violations an "attack" and those who mistakenly viewed as an "attack" Privatemusings' attempt to help Wikipedia. "set an initial tone that was conducive to good faith collaborative work." <-- The learning projects that you mentioned did not set the "initial tone". Both of those project were created as a response to the actions of others...in one case a violation of Wikipedia BLP policy and in the other case it was a response to those who prefer to delete Wikiversity content rather than improve it. Both projects were open to collaboration and I certainly felt that in both cases I was able to participate in collaborative editing that was educational. These projects might no suit your learning style, but I think that is your limitation and not a sound basis for questioning the good faith efforts behind those two learning projects. The correct way to question and improve a learning project is to click "edit" and discuss your concerns with other editors. "without provoking acrimony" <-- I think we have to face facts. Some sysops become acrimonious when their actions are questioned. Some sysops don't like it when their bad blocks are questioned. Some sysops think they have the right to treat non-vandalism edits as if they were vandalism and they become acrimonious when their actions are questioned. Some sysops think they have the right to conspire off wiki to impose emergency desysop procedures when no emergency exists. Those are serious problems that I try to protect Wikiversity from. My defense of Wikiversity is not the problem, the problem is abusive sysops. --JWSchmidt 17:51, 14 April 2010 (UTC)

I disagree, JWSchmidt, the problem is not abusive sysops but the lack of effective dispute resolution process that does not involve massive disruption, and that thus cannot handle sysop error. However, that's not the purpose of this resource, to resolve "the problem," but just to develop some information through study. I fear that this project is being used, to some degree, as a coat-rack for long-term dissatisfaction, which I well understand, but this is not the place to address that. The work done here, if anyone is actually going to do it instead of arguing moot or irrelevant points, may help address it, in the end, but I value knowledge for the sake of knowledge, and any strong agenda that we maintain here may easily contaminate our results and make them less valuable. Please help, and, if you have the inclination to at least watch and help keep us on track, sign up for the project. Perhaps part of the value of this project could be in developing procedures that help us to step back and step away from the blaming habits that we can so easily fall into. --Abd 18:11, 14 April 2010 (UTC)

"the lack of effective dispute resolution process" <-- I don't agree. The first part of the "dispute resolution process" is clicking the "edit" button and talking to people. Abusive sysops reach first for the "rollback", "delete" and "block" buttons and conspire in secret to impose emergency desysops when no emergency exists. Until Wikiversity is protected from abusive sysops it will be impossible for this community to function as it should. --JWSchmidt 18:22, 14 April 2010 (UTC)

Sigh. This isn't the place for this, JW. I'll reply on your talk page. See User talk:JWSchmidt#Discussion_on_"the_problem" (perm. link) --Abd 18:45, 14 April 2010 (UTC)

My main point here is around developing a framework for collaboration. In your saying "the problem is abusive sysops", I take it that you are naming me as one of those abusive sysops. That does not invite me into this conversation as I would like; it makes me feel stressed and less inclined to participate. I think that your learning projects around deletionism were taken in more or less the same way; and I absolutely do not accept that this is down to individuals' learning styles. I don't think that every potentially problematic learning project can be solved by clicking the edit button; there needs to be a shared conception of 'what I can get out of this' before someone will feel motivated to participate - which is what I'm calling a framework for collaboration. Back to "abusive sysops" and such labels, every Wikiversity custodian who has dealt with these problematic issues in Wikiversity tried to do so in good faith - and I do not feel it is conducive to a productive conversation to call people "abusive sysops" when it is clear that their/our actions were taken to addess what was widely acknowledged to be highly problematic. I feel that unless we can develop a realistic framework for what is likely to be problematic and how this could be ameliorated, problems such as happened in 2008 and again recently will inevitably resurface in the future. Cormaggio ^talk 18:38, 15 April 2010 (UTC)

I'd like to clarify that when I say we should "develop a realistic framework for what is likely to be problematic", I include policies around defining Wikiversity's scope and legitimate custodian actions so as to mitigate against the kinds of actions that I was involved in, even though I had never wanted for such actions to be taken. I hoped to make that clear at the time, though I don't think everyone in the community either understood or believed that. Cormaggio ^talk 18:46, 15 April 2010 (UTC)

Cormaggio, Privatemusings started this page, but I expect this editor, because of the complications, to take a secondary role. I'm the only other editor signed up. I specifically invite you to participate, and to indicate this by signing up on the resource page as a participant. That doesn't mean that you must take an active role, you could merely watch what comes up to help make sure we don't get into trouble. I'm assuming that the simple collection of data and neutral analysis of it, to develop a consensus report for future study, will not be problematic. Okay? The only project ready for moving to active status, if you sign up for it, would be the Newbie article creation testing that was done by a set of editors, including highly reputable ones, on Wikipedia. I'm waiting for another to join before actively working on it, though I might decide to go ahead just based on Pm's proposal and my acceptance. I'd prefer another who will at least watch the page! Absolutely, I do not want to see a repeat of prior problems. This project, if I have any say about it, and I believe I do, will not define anyone as abusive, that is outside its scope and would require judgment that is not NPOV. It will document what happened, using diffs of edits or other clear evidence relating to the topic, and it will be arranged in the final report simply to make that accessible to anyone wishing to study the incident reaction to it. I believe this will be useful to all who want to improve the projects or to understand how they function in reality, as distinct from theory and "guidelines." Please consider joining. --Abd 18:58, 15 April 2010 (UTC)

Thanks Abd. My feelings are that, while I definitely appreciate many of the comments you've made here, and while I think there's much worthwhile work to be done in this area, I would prefer ethics and other related projects to be based on a set of principles which we first need to develop before setting up any such projects. I tend to agree with Bduke's comments above that we need to get our house in order before engaging in work that has been causing concern within the Wikimedia community. Wikiversity has lost the trust of many; we now need to build or regain trust so that people can see the positives that this work can achieve. And by "positives", I don't mean that conclusions need be positive: we can be critical, challenging, and provocative, but I would like to see such conclusions based on a much more rigourous and broad-based process than they have been in the past. Cormaggio ^talk 22:40, 15 April 2010 (UTC)

Thanks. The problem is that building policy, which would include ethics policy, cannot be done well in a vacuum. Because there were many complications around the original project on response testing, it's hard to tease out what was objectionable from what was not. I support the development of the principles you mention, but I prefer to also begin to actually do a study, with the understanding that it will come to a screeching halt if it again raises hackles as to what it is actually doing, not to what someone did in the past. Then the ethics policy discussion can address the point raised, and this project will only proceed here when issues raised are resolved. Learn by doing is what's been said to me recently. Let's start with what is on the project page now. Is there anything there that should be removed or changed? I have not noticed any objection to it specifically, except general objection here based on what others have done at other times, and to the identity of the editor who started this, which wasn't me.

Strictly, this is not an ethics project as described. It is a study of wiki history when tested by past events (how the wiki "responded," to events that were either deliberate testing or events that had the effect of testing. (A study of response to newbie articles could be done by direct review of a sample of new articles by newly registered editors, and would serve a similar purpose to the kind of deliberate testing that was done by the Newbie project proposed for study). So perhaps it should be moved out of this subspace. Do you have a suggestion for where to put it? I don't know the WV structure.

The information that this resource develops might later be used in an ethics project, that is up to future generations. --Abd 23:19, 15 April 2010 (UTC)

en.wiki project on Newbie treatment

I support the study of this project as proposed by Privatemusings. It is politically sensitive and great caution should be exercised; however it immediately revealed some very serious defects in how Wikipedia operates, as well as raising possible ethical questions, and it is crucial that these not be lost. We should avoid publishing conclusions, but only arrange clear evidence, unless consensus becomes very clear and is based on sufficiently broad participation. After all, evidence is evidence and evidence itself is not biased, though it may be cherry picked, the solution to which is to pick the rest of the cherries.

Because Privatemusings should not lead this study, for obvious political reasons, I will want to see another support before proceeding. I would assume that a project here may clearly begin with two supports and no sustained objection. --Abd 16:27, 13 April 2010 (UTC)

Well, underwhelmed by response, I am changing my mind. Technically, even one "second" isn't an absolute requirement to start a resource, a study, on Wikiversity, so I'm going ahead and creating the resource page for the first study, on Newbie treatment. I saw no specific objection above, just warnings to Privatemusings that this editor shouldn't be the one to "lead" it. Please, if anyone is concerned that issues of research ethics are being raised, watch the project even if you don't join it, and promptly revert any edits that create some risk. These can then be discussed, as appropriate, to resolve the issues. This project, because of the sensitivity, cannot continue on Wikiversity if pushes ahead in the presence of unresolved objections. --Abd 23:05, 21 April 2010 (UTC)

I think you should reconsider. The wise course of action would be to do no research on wikipedia for some time, and I mean months not days. I suggest that the absence of other support is that others, like me, are being cautious, and rightly so. --Bduke 01:22, 22 April 2010 (UTC)

Bduke, I don't see this as a specific objection. What is the harm in "doing research on wikipedia," per se? Is wikipedia a prohibited topic? I understand that some kinds of research can be problematic, particularly where, say, an editor believes he was abused on wikipedia, and then comes here to do "research" on it (which consists of belaboring his particular history, and identifying the abusers, etc.), can cross a line, understandable motives of blame and judgment become involved, and WMF is sensitive to the effect on volunteers of excessive criticism. I had problems at Wikipedia myself, to be sure, but this particular project is not one that is even remotely connected, as far as I can tell, with the problems I had, and I am not banned, for example, from commenting on Wikipedia itself on the issues of this newbie project (I'm not blocked there), I am only restricted with respect to certain specific things on Wikipedia. I understand the reasons for caution, which is why this project is being established with great caution. Please take a look at the opened project page, linked from this resource, and please look at the initial research guidelines suggested there and on the Talk page attached. The first step is purely the collection of evidence without judgment (except the necessary judgment of deciding that something is relevant). This project will come to a screeching halt if anyone makes a specific objection by removing allegedly objectionable material, it will not be replaced without consensus, and "consensus" includes the whole Wikiversity community, if needed, which, of course, includes any WMF project editor who cares to participate.

I do not have Jimbo's specific permission to proceed with this, but neither do I see or expect him to have any objection, the way that the research is planned. I have already received a small measure of support from him, in the sense that he personally confirmed my observations about the situation. I have a sense of his general support; if I'm wrong about that, I'm sure I'll find out. Until then, I'm assuming the best, and I'm positive that this project, as presently constituted, presents no danger at all to Wikiversity. Whereas running scared, as it seems the advice represents, does represent a danger, a serious one. It is as if we believe that the WMF management of this site is malevolent, about to punish Wikiversity for engaging in simple, neutral research. I'm assuming otherwise.

Did you notice, BDuke, that a proposal was floated on meta to close Wikiversity. It went down in flames; at this point the standing !vote is running 38:6 against the proposal. What happened here was that (1) Jimbo intervened, for his own reasons, and (2) a handful of editors, not familiar with Wikiversity protocol and the ways in which Wikiversity differs from, say, Wikipedia, began to agitate for shutting the project down. This did some damage, and part of that damage would be an atmosphere of fear. But we have been assured that there is no plan to shut down Wikiversity. There was, perhaps, overreaction on all sides. So let's get back to the business of this site, developing its unique potential. --Abd 02:15, 22 April 2010 (UTC)

NPOV tag for this project.

Because of the sensitivity of this resource, I have added the NPOV commitment tag to the resource page. Contributions on the resource page should be rigorously NPOV, and not present conclusions, judgments, or the like. Primarily, the page should point to subprojects that are themselves subject to the NPOV commitment, to develop resources, primarily evidence or attributed opinion, for the study of the project topic, and not to promote or publicize conclusions. If evidence (or attributed opinion) has been cherry-picked, the remedy should be the addition of evidence to restore balance, not removal; except that mere assertion of evidence without citation may, of course, be removed. --Abd 18:20, 13 April 2010 (UTC)

mini cull

I removed this bit from the lead of the project description;

Because participants here are not necessarily disinterested persons, being, often, active editors on wikimedia projects, we should be prepared to step aside from our opinions and personal agendas and focus, here, on collecting evidence and developing cogent and neutral analysis of it. For the work here to be truly useful, it must reflect a high level of consensus, but, because we will avoid coming to conclusions that are not fully supported by evidence, such consensus should be possible.

I think it's a little bit redundant, and I think I disagree with some apects (I think mentioning 'consensus' in this vein is a bit of a minefield!) - anywhoo, I took it out - thoughts? Privatemusings 01:30, 16 April 2010 (UTC)

Well, that's a description of what would make the work truly useful. Do you disagree that consensus would make it useful? Do you think that consensus is impossible? What is the objection, i.e., what do you disagree with?

I'm describing a seminar, a study group, that has a purpose and operating principles and that will develop a report. A "consensus" report does not mean that everyone agrees, it means that everyone, to the extent possible, agrees that the report reflects the findings of the study group. If there are conflicting findings, those can be presented as such, though I wouldn't give up so easily on the possibility of full consensus. It's classic NPOV, and if people with Wikipedia experience don't know how to do that, well, I suppose that explains some of the trouble over there.

In any case, this was put in by me to help alleviate the concerns of some about this project. Is it your intention to leave the editors who have expressed misgivings above in a condition of worry? What is your purpose in taking that out? Do you know of any harm it was doing? --Abd 03:04, 16 April 2010 (UTC)

I'm just not really sure that it would be effective in mitigating worry - I'd prefer those with such feelings swing by and discuss it, or edit away. I'm also not (yet!) sure quite how we're going to manage the nature of the collaboration, and 'consensus' - I'm just not really sure that's exactly what I'm up for working towards - I'm more into careful examination of the facts, neutral presentation of them, and then maybe some 'original', personal conclusions. I guess if we all agree then we can say that we have consensus, but I'm not sure that I want to cite 'consensus' as a goal / endpoint, thas all... Privatemusings 06:17, 16 April 2010 (UTC)

• You have not answered with the requested specific objections, rather simply with "I'm not sure." You "prefer" that others behave in a certain way. Isn't that a form of control? You aren't yet "sure" about how we will manage, and your uncertainty about that is natural, and we will not, in fact, know, until we do it, though I have quite a bit of experience with this kind of thing. It cannot be specifically predicted. If it could, we would not be human beings, we'd be robots, primitively programmed. If, Pm, you are actually "not sure" as you say, why not let my contribution stand? I asked you if it was harmful, but you did not present one single proposed harm. How about allowing what you don't understand to stand? Isn't that how we learn, by watching what we don't understand?
• You have now, properly, separated the collection of facts from your conclusions about them. Indeed, consensus on conclusions may be elusive, people may react differently to the same collection of facts, and it sometimes may not, within practical limits, be possible to resolve all these differences, and, in addition, there is the possibility of bad faith participation, another issue that we need take great care about. Accusations of bad faith don't belong in seminar space, but if someone is actually disruptive of our "meeting process," we can call the "campus police," if necessary. I don't expect it.
• So I would think we have agreement on the first part of the process, the collection of fact, over which consensus should be possible, and where it is missing, we would segregate such controversial "facts" for special attention, as well as not mixing them with what we have consensus is fact. I really do expect absolute consensus on this part, and bad faith, if it exists, will normally be too ashamed to confront verifiable fact, it becomes entirely obvious what is going on. Is it necessary for us to agree on the second step and how to take it before we take the first step? That would make all this far more difficult, don't you think?
• There are lots of options if we don't have agreement on the second step. First of all, an initial report can be prepared that is "just the facts, ma'am." This, alone, could be a valuable outcome. Secondly, we can then present an analytical report that contains attributed conclusions. Some of this, indeed, can be a part of the first report, since with attribution and balance it would be NPOV, and that so-and-so wrote such-and-such can be verified. I think what may be more controversial would be going further to any assignment of blame, without having objective standards for blame, as well as recommendations for improvement, which also requires standards of "goodness" or "value." At this point, certainly, I'd prefer to stay away from that, far away, because it is precisely where the objections have arisen, and I'm trying to tease out those objections from possible objections to simply reporting what happened. I do not know that anyone is actually objecting to that, or, alternatively, that anyone is willing to acknowledge that as the basis for their objection. I'll assume the former, that's what AGF means and requires here. --Abd 19:33, 16 April 2010 (UTC)

Mephedrone suggestion moved from resource page

I've moved this from the resource page because my researched developed evidence that there wasn't anything here but apparent media error, though I haven't seen the Private Eye article, it isn't openly available AFAIK. I shouldn't have discussed it there in any case. The proposed project was by Privatemusings.- -Abd 03:08, 16 April 2010 (UTC)

• w:Mephedrone - w:Private Eye (magazine) recently reported that the inclusion of the name 'meow' for this drug in the wikipedia article led to widespread republication, and indeed the adoption of the term in a self fulfilling prophecy. We can look at;

• how wikipedia policies and editing practices were followed in this case
• the more philosophical angle / epistemology and wikimedia etc. Privatemusings 01:27, 12 April 2010 (UTC)

• See w:Wikipedia:Wikipedia_Signpost/2010-04-05/In_the_news and in particular, notice that the term apparently did not originate with Wikipedia, though it is not impossible -- but seems unlikely -- that it spread more rapidly because of Wikipedia. The insertion of "Meow" as a name was not an error. See [1] for inconclusive evidence that the name existed previously (document, apparently December 2009 report, notes geographical distribution of names). Not a good example of anything related to wiki ethics or reliability or testing. I recommend withdrawal of this proposal. --Abd 23:59, 13 April 2010 (UTC)
• I found what appears to be conclusive evidence that the name "meow" was in use in March 2009, by searching googlegroups for "methedrone meow", in a post that indicates some kind of street usage, though it might have been some individual seller. I wasted a perfect good hour on this.... --Abd 00:16, 14 April 2010 (UTC)

hopping back

I hope to have a bit more time to hop back into this - so I'm 'pinging' my watchlist to help me find the right spot! Privatemusings 03:18, 24 July 2010 (UTC)

Why don't you organize a "Response testing" course and materials instead, with some examples from WMF? Focusing a class on WMF, and more importantly not having a parent course / community discussing response testing in general, allows this to become self-referential, unconsciously biased, and not helpful for most of the world (who might benefit from the general course, but not such a specific one). It also may attract only students and fellow teachers with some axe to grind who think that academic 'analysis' of the situation will give them a vantage from which to gripe. On the other hand, if a general course takes off and attracts or produces people versed in the subject area in general, a more focused follow-up course/research effort might be more effective. –SJ+> 08:14, 30 July 2010 (UTC)

Yes. Great idea. One problem, though. People tend to create resources or courses here when they develop some passion about the topic, and it is possible that the passion is more focused or, here, connected with "WMF projects," so we should not prohibit such more specific resources from being developed. There is a problem with all this being classified under "Ethics," which is a high-level discipline, highly abstract or normative, when what we need first is to develop evidence or sources or the like, without making ethical judgments. So I see a need for restructuring here. There is room for a department of Wiki Studies, and this resource could be a part of that, or at least linked from the department. There is, in my view, a crying need for coherent and deep examination of how wikis work, why they work and how they fail, and a too-common habit of blaming the failures on individuals is part of what has caused so much trouble in this work.

In my view, the problems are generally lack of structure, or structure that sets up problem conditions, creating abusive roles which people will then fill (all in a good cause, as they would see it). I've found it impossible to examine this in the open wiki environment, elsewhere, because it requires a depth of consideration that is frequently not tolerated ("Too long." "Tome, TLDR." "What's your point?"). --Abd 18:54, 30 July 2010 (UTC)

that's not a bad idea, sj, but I'm not really sure I'm capable to extend the topic up in that way - I probably should clarify that I really participate here from the point of view of a student / learner (something I wish I had more opportunity to be in rl!) - and am not completely comfortable in organising the learning of others, but I'd love to learn with anyone who's interested.

I'll rejig the pages in the manner you suggest anywhoo, and will welcome further feedback - or even collaboration :) cheers, Privatemusings 11:14, 3 August 2010 (UTC)

I've tried to start the new structure - and think I've moved / fixed most things? - I'll probably start on my two new suggestions soon(ish) - the pace, as ever, will be glacial. Privatemusings 06:52, 6 August 2010 (UTC)

restoration to facilitate linkage

I hope that by now it's not too controversial of me to have created the Response testing/WMF Projects/previous page - which I grabbed from here - I was reviewing some stuff, and it became annoying for me to have to try and work out how to make the wikipedia links work in my browser, so I felt it was worth dropping it in here for ease of use. Obviously there's the additional utility of there being a public record of previous work - there may be issues relating to attribution, but hopefully we can fix them up ok? - any and all feedback appreciated :-) cheers, Privatemusings 04:11, 19 August 2010 (UTC)

It is certainly controversial, and in poor taste considering the circumstances of its deletion. Please don't start poking at this project again unless you have the time and patience to create a version of the project in good taste - as Abd suggests below, or as I suggested above. It's fine to keep this in the page history, but a separate page linked from the header seems uncalled for. –SJ+> 21:44, 7 September 2010 (UTC)

hmmm... well I can only really repeat a) the intention was simply to help me out with some links, and in some discussions and b) I'm not sure I understand why you feel it's controversial - I can sort of see the obvious that it caused a fuss, and obviously has the potential to cause further fuss if done clumsily etc. - but it really hasn't caused any problems that I've been aware of, and you're the first person to raise a concern - 'poor taste' is a bit subjective, I think (and a bit mean - /me sticks his tongue out at sj) - p'raps the link is an easy starting place to compromise on though - I'll get rid of it in a mo. If I think it's appropriate, I'll link it somewhere else - and I'll try to do so in the best possible taste ;-) Privatemusings 05:52, 10 September 2010 (UTC)

I will be so bold as to presume that any perceived controversy has died down / dissipated - I may have a little time this week to make some further (probably pretty small, but never mind!) edits. I continue to aim for the best possible taste, of course :-) cheers, Privatemusings 03:44, 20 September 2010 (UTC)

Are you secretly kidding us Privatemusings? You are going to hijack Response Testing, under the guise of a customer awareness project, and restart the vandalism thing? The only way such "testing" can be carried out reliably is covertly. Can you not even try to formulate response testing without vandalism? You are bound to turn up something of interest. Is that not exciting enough? No big lights in that one? If you "can only repeat", you can do no wrong, right? ~ R.T.G 09:07, 15 October 2010 (UTC)

I mixed up two pages before I wrote this. When I thought I was reading "Response Testing/WMF Projects", I was actually reading the "Ethical breaching experiments" or "Response Testing/WMF Projects/previous" page. I am still wary at the idea that "experiments" may be encouraged on other projects. It is obvious that you should not undelete the page like this. Are you rating yourself as up there with Mike Handel and other topics? Don't you think that page is out of bounds unless, for some reason, it is absolutely essential? If you are going to teach ethics, is it not important for you to be one step ahead of those ethics as the need may arise, rather than undeleting pages as you see fit? If not, you are teaching un-ethics, whatever else it may say on any of these pages. ~ R.T.G 10:02, 15 October 2010 (UTC)

We have a process for page deletion here, RTG, if you believe a page should be deleted, you may follow that process. If you believe that particular content on that page is improper, you have access to an edit button. There is no clear definition of ethics as it applies to Wikipedia, and where individuals have been banned, they are under no contractual obligation. However, by most common-law standards, such as an expectation of fair treatment, ethical violations abound on Wikipedia, and someone discussing testing here is minor by comparison. How about addressing the beam in your own eye before complaining about the mote in another's? --Abd 13:01, 15 October 2010 (UTC)

Don't instruct me without thinking, friend. The page was deleted. There was a big fuss about it. Don't point me to the deletion process as if I don't know all about it and carry on rhetoric as though we were discussing ethics afresh or as though the ethics of Wikipedia are hard to define in relation to Wikiversity, thanks. Don't talk about my eyes either, cheers. If you had something to say about me, that you believed in, what would you have done? So I'll just say clearer again, don't talk about me if you can't attempt to be clear, don't pretend for one second that I am measuring up to Privatemusings, don't ever suggest that the minute differences in policy between Wikiversity and Wikipedia amount to some huge leap in ethical understanding, and thanks. ~ R.T.G 13:46, 15 October 2010 (UTC)

My, my, my. Noted. There was a fuss about deletion but never a confirmation and, in fact, indications to the contrary. Let's not go there, let's seek consensus, as should have been done in the first place. I've suggested a solution on the Talk page. --Abd 14:24, 15 October 2010 (UTC)

But you you you. You went there. "indications"? I become truly skeptical of your wishy washy words and slurs. Suffice to say, you are pushing a side of a dispute you yourself describe as unsolved as though it were straightforward for you. I'll tell you what is straightforward. It's quiet and you're pushing inward. Go back to the ones you hope to predict the results of and make finishing them concievable before making even more. When you run out of ground, start building new platforms. Is that not good? Must you insist to hold a deck of cards before you do the basics? Basics are obscure here. The only basic is that, "adding false information is bad". Why is there not a description, outside the talk pages, of how Wikipedia responds to certain things? Because you're all too busy on the talk pages planning the heist. LoL. ~ R.T.G 14:41, 15 October 2010 (UTC)

Ad. There is a related topic to Wikipedia reactions. The one about the newbie treatment. Take it all on. Go all the way to the silver service as an anon and note how polite they are, careful of personal COI with your normal account. Then you'll probably get special permission for any hoax you want but if you are just hoaxing for hoaxing sake you are going to spend time after time explaining yourselves, the guilty man. You are either an investigative researcher, or an experimental demonstrator. Hats off to all but this is hardly a stage. ~ R.T.G 14:49, 15 October 2010 (UTC)

comments about the en wiki signpost newsletter

Privatemusings, did you see (source): In March, a page on the English Wikiversity about researching Wikipedia by "Ethical Breaching experiments", which included suggestions to vandalize it on purpose, generated controversy (see Signpost coverage.) ? Do you agree that the page includes suggestions to vandalize Wikipedia? --JWSchmidt 09:22, 19 August 2010 (UTC)

No, I hadn't noticed that - and really I think the only bit which you could possibly read as a suggestion to vandalise would be the bit suggested by Gomi? - even that I think is a bit of a stretch. I rather suspect that the author didn't have the chance to review the original material, so sort of went by a bit of word of mouth / chinese whispers. I'll drop a note in over on the signpost to see what they think - but in the grand scheme of things this example isn't a big deal - I would hope for higher standards at the signpost though :-) cheers, Privatemusings 01:16, 20 August 2010 (UTC)

Privatemusings, you might be interested in this. --JWSchmidt 01:34, 20 August 2010 (UTC)

This is a collision between form and substance. Developing guidelines for response testing will need to determine where to draw the line. Specifically, there were some suggestions made (http://www.mywikibiz.com/Directory:Wikiversity/Wikimedia_Ethics/The_Ethics_of_Breaching_Experiment#Suggestions_from_Gomi_.28link.29), and the discussion JWS refers to is on a Wikipedia user page, where the user thinks these would be "vandalism." Which is presumed Bad, by definition. What is lost is the actual effect of the experiments, as described, vs. the possible gain. Wikipedia, in particular, has become a place where ever-tighter rules are in place, many of them not written. But people are quite certain about them. Inserting any kind of false information to the project is Bad. Problem is, it happens all the time, and it happens with registered users who suffer no consenquences, and it happens with IP users, and it is a veritable flood. How much of this is caught by Recent Changes Patrol, and how much is caught later? Obvious vandalism is normally caught quickly, except RCP isn't reliable without flagged revisions or some process that insures that every edit be reviewed by someone responsible. Sometimes, obvious vandalism gets through, it's a toss of the dice, it depends on when it happens, how many editors are doing RCP at the time, and once it's gotten through that initial gauntlet, I've seen, it can last for years, on an obscure article, and sometimes on suprisingly notable topics.

However, obvious vandalism is not terribly interesting or terribly damaging, in itself, because everyone recognizes it. What's more serious is information that is ostensibly sourced, where the source doesn't actually support it. Is there any way for Wikipedia to determine how well it is doing on this level?

It's possible to do it without "response testing," but the labor involved would be substantial. The plus side is that improvement of the project would result, and one might be able to come up with quite a few volunteers as researchers. Here is the idea:

A copy of the database at a particular time is set aside for investigation. Researchers follow a predefined protocol to identify text that is apparently sourced but where the source does not support the usage. The protocol would not allow picking and choosing, and a committee would resolve doubtful cases, classifying them. A body of identified "errors" would thus be developed, and the sampling techniques would allow extrapolation of this to an overall error rate. Then the subsequent history of these errors would be examined. How many were found outside of the study process? If, then, researchers correct the errors, what happens? Researchers would follow a pre-approved protocol for correction, but they would not disclose (perhaps!) that they were affiliated with the project.

Every on-wiki action of the researchers would rigorously follow policies. No false information would be introduced, indeed, all that would be done would be to identify and remove false or misleading information. Yet this would provide information that might even surpass that of "response testing." Basically, it would be testing the response of the wiki to what already happens anyway.

It's tempting, though, to create artificial tests, as was described. That is, to my mind, inferior. It might be more efficient, in some ways, but it requires design and quite possibly isn't a fair test, because the "errors" can be made artificially difficult to detect, as may have happened with some of Greg's tests early this year. (This, by the way, must be a standard criticism of "response testing" as often handled. It can amount to entrapment, sometimes, creating an extraordinary risk, say, for an employee, an employee who might never have otherwise been tempted; thus "response testing" of this kind can exaggerate the level of real problem that exists.)

Now that I'm thinking in this way, the same project could identify unsourced text, then follow a process to determine if reliable source exists for that. Again, this would require a project "process" to resolve disputes that might arise. The point isn't to make a rigorously "correct" determination on each "fact," but rather to examine if text reasonably satisfies RS guidelines, according to some reasonable process. In a sense, on a small scale, this project would set up a review panel to make "content decisions." Later, some of this might find actual use on the project, but, of course, it would all be subject to review, the "review panel" would not be making any actual project decisions. It's only for the study, even if someone later uses some of it for other purposes.

My guess is that a project like this, with proper design, would be approved by ArbComm and the Research Committee. If not, well, because this doesn't involve actual editing -- except for the second stage of the project, seeing what happens when someone, following rigorous guidelines, attempts to fix an error -- we'd want, I'm sure, to know why. And to fix it so that it would be approved, if that is possible within our own ethical guidelines. --Abd 19:07, 6 September 2010 (UTC)

The ethics here, in the experimentation with errors, is that you are, not only uninvited, but specifically prohibited. What you speak of is a demonstration, not simply a research endeavor. Unless you believe strongly that you can demonstrate convincingly why Wikipedia should change one of the specific procedures that prevent certain actions, you should not plot to subvert them.

• "Now that I'm thinking in this way, the same project could identify unsourced text, then follow a process to determine if reliable source exists for that. Again, this would require a project "process" to resolve disputes that might arise." ~ Abd, above.

Identify unsourced text, identify a reliable source... ? It sounds like a good plan.

• "Wikipedia, in particular, has become a place where ever-tighter rules are in place, many of them not written. But people are quite certain about them. Inserting any kind of false information to the project is Bad." ~ Abd, above.

It is if you risk people being fooled by it. Perhaps it could be a good thing for April fools. Perhaps you want to test reflexes, but if all your planning is done here in plain sight, your research is moot anyway. Anyone could just follow you around and clean up the mess for fun and you'd wind up with a false reading.

All that stuff about studying the wiki, all of that can be done without entering more of these false informations. It would be more accurate because you can make target groups to study, 5 yr old FA down to 6 mth old stub, and determine how random you are going to be before selecting. You could dredge up the less notable editors who have inserted false information to targetted or random areas and follow them from scratch. There are hundreds at varying degrees of effect to study without creating new ones.

• " The point isn't to make a rigorously "correct" determination on each "fact," but rather to examine if text reasonably satisfies RS guidelines, according to some reasonable process. In a sense, on a small scale, this project would set up a review panel to make "content decisions." Later, some of this might find actual use on the project, but, of course, it would all be subject to review, the "review panel" would not be making any actual project decisions." ~Abd, above. "My guess is that a project like this, with proper design, would be approved by ArbComm and the Research Committee. If not, well, because this doesn't involve actual editing..." ~Abd, above.

Great. ~ R.T.G 14:24, 15 October 2010 (UTC)

And, now, the problem with what I suggested. It is far more work. Any volunteers? A little response testing can come up, quickly, with far more information. Response testing in business has some serious ethical issues because of possible entrapment, and because people's careers and income can be at stake. On Wikipedia, nobody's career is at stake, and the most damage that can happen is a little embarrassment, i.e., see the "treatment of newbies re speedy deletion" test. In my view, sensible administrators who got "burned" thanked the testers for revealing to them how they'd been insufficiently careful.

Only an admin could research speedy deletion, because non-admins cannot see the articles. But anyone could run a newbie test. It's clear that it is possible to run ethical response testing, so a sounder approach would be to develop guidelines for it. Further, it would be superior if testing plans could be confidentially reviewed before the testing is done. There is now a research committee, I believe it was established by ArbComm, but there are possible difficulties with conflict of interest. To be optimal, a research committee that approves testing (approved testing would be exempt from sanctions, provided guidelines were followed) should be independent. I am on the research mailing list, but I haven't been following it closely.

"People being fooled by it." Wikipedia suffers from a flood of incorrect or unverifiable information. With the response testing done by Thekohser, there was practically no risk that anyone was actually "fooled," except maybe some Recent Changes patrollers. Valuable information about BLP process was gained, it remains to be seen that any sound application has taken place, but one has to start with information. A response test like that of Thekohser was relatively ethical, because the test field was identified, there was no risk that the errors introduced would survive, unlike the normal flood, which survives for years, often. --Abd 14:46, 15 October 2010 (UTC)

a thought

I've read the latest posts from Rtg and Abd above, and have a feeling that perhaps Rtg might have gotten the wrong end of the stick about some stuff - perhaps rather than dive into the above discussions, would you (rtg) like to have a chat here? - would you mind summing up your concerns and clarifying what you feel the best next steps are, and hopefully I can explain and we can reach an understanding :-) cheers, Privatemusings 22:39, 20 October 2010 (UTC)

You shouldn't have undeleted that page. What would you say if I made a page of racist material which was deleted because it was offensive then I undelete it for reference in my "What Wikipedia doesn't like" project? You most likely say that the material was considered unsuitable. You just want noted as causing controversy Privatemusings. I don't see any inroads of study. No plans except to cause. I understand how the topic is interesting but you have gone through the half a dozen most notable incidents and can only see a half dozen more in the making. But are there no roots? Is there no view to practicality in this study? If not it just reminds me of a whole lot of other things which did not become practicable. ~ R.T.G 13:57, 25 October 2010 (UTC)

RTG, were I an admin, I'd consider deleting the restoration (it was not "undeleted") and restoring the original page, possibly with blanking, or not, because then the history would be available. Blanking would depend on the content. But I'm not an admin, and I'm simply not exercised enough about this to ask for an admin to do it. If you have some specific objection to content there, the page appears to be open to editing, last I looked, or you could object to the specific content on the attached Talk page. What's your point here? I see no specific objections to any content on that page, and I'm not about to pour over to try to guess. --Abd 22:05, 25 October 2010 (UTC)

Abd dozens of people discussed that page and Jimbo Wales himself deleted it. What has changed now to reprint it? Were you not part of that discussion? I think I have to check... ~ R.T.G 10:30, 28 October 2010 (UTC)

By all means, check. I found Wikiversity:Community Review/Wikimedia Ethics:Ethical Breaching Experiments. Jimbo deleted the page, indeed, with an edit summary of: (Beyond scope: content was: 'Why do people execute breaching experiments? How do they do it? On this page we discuss, analyze, explore, and study why and how people execute breaching experiments - in particular whether it's possibl)

"Beyond scope" would not be adequate grounds for a steward or functionary -- or Founder -- to delete a page without consensus to do so. "Scope" is determined by the community, and that is especially important at Wikiversity. Now, if a page was offensive and immediately damaging to the Wikimedia Foundation or other WMF wikis or contained illegal content, that could justify it. If there is any such content remaining in the page, the immediate remedy is to delete it with ordinary wiki deletion, and if the information is so damaging that it should be invisible in history, it may be revision-deleted, the page deleted by filing an RfD, or even oversighted. However, I see no claim here that there is such content. If an editor believes there is such content, they should delete it unless community consensus has already been established, and if there is disagreement over this, community comment should be solicited.

That community process was not followed caused a huge flap and waste of time, and led, in combination with a similar incident at Commons, to Jimbo resigning the tools that allowed him to delete and block.

There was no discussion that decided to delete the page. There was unilateral action by Jimbo, accompanied by blocks and removal of sysop tools, followed by restoration of one of the blocked accounts that of privatemusings, and the tools, but with an implied threat to reblock if PM touched "response testing" as a topic.

I am not arguing that that page was proper, nor that the existing content of this page is proper, nor am I arguing that Jimbo was badly motivated or made the wrong decision, given the information that he had and his limited time to investigate. I'm arguing for process.

RTG, you appear to have brought on the entire fuss, by calling in Jimbo, having set up inflammatory text and then pointing to it, and you were active on other wikis in attempting to enforce the block of the other user blocked by Jimbo. What is your dog in this race? Why did you do all this? It caused enormous disruption and waste of time, damaging every wiki you touched in this matter. Advocating your own opinion is fine. Acting as an editor, following proper behavior, is fine, i.e., you could have blanked the page or offensive content at any time and you could have filed a RfD if it was disputed. If an editor is disruptive, there is likewise process for dealing with that. Instead you did what is described in Ban of RTG. And it seems you are now interested in perpetuating the disruption. Please stop. --Abd 20:37, 31 October 2010 (UTC)

< you mention above 'No plans except to cause.' - I think you maybe left out the word 'disruption' or 'trouble' - but what I'm aiming for is 'insight' :-) - I'm glad you understand why the topic is interesting - I'm really just a lazy wiki gnome over here taking a look at it when I get the chance - do you perceive any specific problem with the material as is (do you find it offensive in some way? if so, would you mind explaining how?) - cheers, Privatemusings 09:49, 31 October 2010 (UTC)