24

A large community proposal was very recently closed in Area 51: Neuroscience (update: in fact it has just been deleted and all data permanently erased, see below).

150 early supporters gathered in less than two months, including the heads of major laboratories in experimental Neuroscience across the world. This was a concerted effort across multiple institutions established by top practitioners in the field to create a much needed venue to exchange expertise and technical knowledge for students and researchers. It started as a follow-up to discussions on the open neuroscience summer school (TENSS), which every year gathers a community of technical experts in Europe to teach the next generation of experimental neuroscientists. Lecturers, TAs and students from all over the world agreed to support and promote a proposal that would be created in the Area 51 website.

The proposal gathered a large momentum from day one, even before the main networking outreach efforts were started. Many of the most valued and critical specialists in the field are biologists and non-CS people which are actually not keen on using Stack Exchange, so it was very surprising to see such a large number of experts step forth to support this proposal (as you can infer from user statistics, >50% of supporters were completely new to SE).

Surprisingly, after these first promising few months, users suddenly woke up to a closed proposal. Moderators from Area 51 had decided that the Neuroscience proposal was a duplicate of Cognitive Sciences and referred users to the CogSci website.

This was a source of confusion and surprise to everyone involved. Some users did try to reach out to the Cognitive Sciences community and ventured some experimental neuroscience questions, but they were closed off as out-of-scope, apparently by being too technical in nature. On hindsight, this is not so surprising since the scope and content of the two proposals were worlds apart from their very inception: CogSci clearly has their focus on theoretical, computational and cognitive models of the human mind. They also address many layman and popular questions on human cognition, including philosophical conundrums like the origin of consciousness or free-will.

The Neuroscience proposal on the other hand had the specific goal to address the many technical questions faced by Neuroscience researchers and students while trying to design and run experiments. These included, for example, questions about implantation techniques (e.g. electrode drives, micro-injections, patching, optogenetics, etc.), as well as technical details for getting in-vivo neurophysiology and imaging hardware/software to work. These types of questions are particularly challenging and specific to the domain of experimental neuroscience because they stand at the intersection of multiple engineering fields (optics, electrical engineering, programming, material sciences, biotechnology) and thus will usually challenge even experts in those fields. As advocated in many other technical Stack Exchange websites, these types of questions do not blend well with more conceptual, opinion-based questions which are more likely to invite debate.

Given these differences, it was not surprising that there was a strong contest presented by the experts supporting this proposal against this closure mandate. Requests for justifications were sent personally and publicly to the area 51 moderators. Many users presented rational, extensive counter-arguments, but they were answered with silence and a deserted site.

Sadly, the way the whole process was conducted drained the momentum behind the neuroscience proposal and scared experts away. Most users that were contacted feel like this was a lost chance. There is a high probability now that neither CogSci nor Neuroscience will gain these new experts and questions. In this case, Stack Exchange itself won’t benefit from the interdisciplinary expertise which could have spread to other sites in the network.

Furthermore, as all Area 51 closed proposals get permanently deleted and all data is lost, the entire history of the effort will disappear, and no future initiatives will be allowed to learn from what happened. For this reason we decided to intervene at the level of Stack Exchange meta and make our case that the way Area 51 is organizing new communities could maybe use some improvement. It is concerning in our eyes that a movement that self-organized around real-world experts, a clearly defined technical scope, and had a strong growing user base from day one can be shot down without even having a chance to prove themselves.

We think the only reason neuroscience experts would be interested in a Stack Exchange site is if it tackles the interdisciplinary technical problems neuroscientists face every day. We argue that broadening the scope of questions to the point of including opinion based conceptual or philosophical questions is not appealing to a community looking to share technical challenges. We believe it is better to start with a very technically focused site. Experts will attract experts and eventually regular users who want to get answers from experts.

Ultimately, we believe that the decision to close this proposal was made by people who, despite their good intentions, are not experts in the field and don’t really understand the problems facing our community. This leads us to speculate: how can SE make sure that new communities are created and driven by experts if they don’t listen to them? Taking our failed proposal as an example, it seems that it doesn’t really matter how many experts you gather, or their quality, because you can be shot down by the moderators without having even a chance to voice your argument properly and openly.

Did it make sense to prematurely shut down this proposal? Given the vision laid out by Stack Exchange, who should make that call?

Top example questions and user statistics

Since the original proposal has now been thoroughly deleted, we include here the top 10 example questions and user statistics at time of closure for better appreciation of the points made above.

Representative top example questions

  • I'm doing extracellular recordings with tetrodes and using optogenetics for tagging. How should I handle photovoltaic artifacts in my recordings?
  • I want to synchronize video tracking with my electrophysiology acquisition. I am using Open Ephys. Does anyone have a good configuration to share?
  • I am using solenoids for reward delivery and despite attempts at shielding them they produce a lot of electrical noise in my recordings.
  • How should I store my neurophysiology data for easy sharing with other neuroscientists?
  • I am trying to image synaptic responses in dendritic spines. How can I isolate them from fluorescence arising from back-propagating action potentials?
  • What minimum time window is needed to meaningfully estimate coherence between two LFPs at a given frequency?
  • How to ensure long-term stability of chronic headpost implants?
  • How do you prevent light leak from the optogenetic stimulation light onto the PMT when 2p calcium imaging and stimulating simultaneously?
  • Is there a tangible benefit of synchronising the sample clock to the laser pulses in a resonant scanning 2P microscope?
  • Is there any good automatic open source solution for whisker tracking in freely moving rodents?

Users also following

  • 10.1% Open Science
  • 6.1% Artificial Intell...
  • 6.1% Internet of Things
  • 4.7% Literature
  • 68.9% only this proposal

followers active in

  • 15.5% Stack Overflow
  • 15.5% Documentation Beta
  • 4.1% Meta Stack Exchange
  • 3.4% Super User
  • 3.4% Cross Validated
  • 3.4% Mathematics
  • 2
    As a CogSci moderator, it must be said that even though you might have been aware about Cognitive Sciences, and had discussions about it on Area51, no effort whatsoever was done to reach out to this existing community, to establish whether or not the sites differ in scope. It would have been very beneficial for your proposal to have done this prior to launching the proposal, and in fact it is still beneficial now to do this prior to creating a new proposal. Showing you did have discussions with potentially overlapping communities is a good way to convince them your proposal is different. – Steven Jeuris Sep 7 '16 at 15:29
  • @StevenJeuris we did reach out to the CogSci community as a number of discussions on the (now deleted) proposal and on CogSci meta show. We did not do this prior to creating our proposal because we honestly believed (and still do) that the proposals are vastly different. We were approached early on by moderators of CogSci claiming our proposal was a duplicate and we addressed their concerns in extensive writing. We were closed by an Area 51 moderator who ignored our replies. I think we have echoed our case a number of times already that it should be clear by now exactly what happened. – glopes Sep 10 '16 at 19:46
  • 1
    @StevenJeuris what is more sad is that I keep getting daily personal emails from students and experts on Neuroscience that wished for a space to exchange their very real problems while we are arguing about semantics. The bounty is now over and it seems we got no real answer to the question of which expertise should guide the creation of new proposals... I don't know what stack exchange is about but every single expert I know from neuroscience has been disappointed by the bureaucracy in the network. – glopes Sep 10 '16 at 19:58
  • I only recall one post by Robin Kramer (A CogSci member), who posted about the possible duplicate proposal on Area51. All CogSci meta posts on this topic were written after the proposal got closed. – Steven Jeuris Sep 11 '16 at 9:52
  • You do know what to do: make a clear case why Neuroscience is different from Cognitive Science and Biology (ideally by engaging with these respective communities), and then reopen a new proposal. – Steven Jeuris Sep 11 '16 at 10:19
  • 4
    @StevenJeuris In hindsight, I agree, we probably should have. But not because I think we were not right about opening a Neuroscience proposal. But because apparently SE area51 lives in a very weird bubble and has some weird unwritten mechanics. You should also say that After Robin posted in the Area51, the whole community answered back immediately. – nachosan Sep 12 '16 at 10:08
  • 3
    Stack Exchange does operate along weird undocumented mechanics. :) Your community was also very receptive and constructive to Robin's post, indeed. – Steven Jeuris Sep 12 '16 at 10:17
  • Are the people who have been interested in starting this proposal still around? A solution to this problem would be to restart the site outside the SE network with a similar software. For example OpenScience which was closed in private beta by SE employees has done this successfully and now runs on a basic version of the software PhysicsOverflow uses. In case you are interested in this, please write an email to admin@physicsoverflow.org to further discuss this. – Dilaton Oct 6 '16 at 18:14
  • Maybe going to Neurostars neurostars.org/t/Latest would be an idea too? I have just found this. – Dilaton Oct 6 '16 at 19:57
  • 1
    @Dilaton Thanks for the OpenScience pointer. Regarding Neurostars we are aware, and It might be a possibility. – nachosan Oct 14 '16 at 14:39
3

From a quick look at your example questions and Cognitive Sciences' help center, it's clear why the two proposals were thought to be duplicates. The help center on CogSci says that:

If you have questions about ...

  • Cognitive science
  • Psychology (e.g., cognitive, social, developmental, biological, applied, clinical, organizational, etc.)
  • Psychiatry
  • Neuroscience and neurobiology

... then you're in the right place to ask your question.

Emphasis mine.

Neuroscience is on-topic at Cognitive Sciences. Creating sites that only cover a subset of what an existing community covers is generally a bad idea - they just drain traffic away from the existing site, which isn't a good thing. However, creating a site which covers an area already covered by an existing site, plus some extra areas, may be worth it.

In this case, I think it would be worth trying the proposal again. From the reaction of CogSci to the questions posted there, it seems that the theoretical science is on-topic there, but the applications and technical aspects of neuroscience are not. That's your extra area - the new Neuroscience site could cover the theoretical, plus the application and technical sides of neuroscience.

Given that, my advice would be to try again. Create a new Neuroscience proposal on Area 51, contact your field experts again to tell them you're having another attempt. Once you've done that, one of the first things you should do is to create a discussion question on Area 51 about the proposal, outlining why you think it's not a duplicate (because you also cover X and Y areas, which CogSci doesn't) and why you think it would make a good site (because you've got subject experts committed to making it happen). This should provide anyone looking at the proposal with the necessary information to make their own judgement.

If that fails too, then at the end of the day - oh well. Stack Exchange isn't the only Q&A platform in the world.

  • 2
    We were aware of the CogSci website and a discussion of why we were not a duplicate was indeed created and heavily debated from early on. We thought our point was clear (i.e. that mixing opinion based questions with technically focused questions is not a good idea), but we were still shot down. If we were to create the proposal again, how can we be sure that we wouldn't be shot down in the same way? Be aware that none of the Area 51 moderators participated in these discussions and that they shut us down even though we followed all the rules, with little feedback on how to improve the proposal. – glopes Sep 3 '16 at 10:06
  • @glopes That is, unfortunately, the way of Area 51 - there's plenty of stuff going on, so moderators can't always participate in discussions. Has anything come up on this question to say you're not a duplicate that wasn't mentioned in the Area 51 discussions? I'm thinking particularly about scope - if you can show you take questions about X and Y which CogSci doesn't, then it's more clear that you're not a duplicate. – ArtOfCode Sep 3 '16 at 10:09
  • 2
    Also, simply stating that something is covered by a proposal is not the same as actually covering it. There is plenty of evidence that cognitive science doesn't actually fulfill this role and indeed CogSci itself is currently struggling with this (see here, here and here). – glopes Sep 3 '16 at 10:09
  • If Area51 moderators cannot participate in community discussions, then that simply makes the point of this question all the more relevant, doesn't it? – glopes Sep 3 '16 at 10:11
  • @glopes After spending some time on the Biology SE site I have to say that all this argumentation about cog-sci would be OK if it came from them. The type of questions they have/answer are much closer to the idea behind the Neuroscience proposal than the cog-sci. Indeed expanding it to include more technical questions would be possible. Indeed there is some overlap in Biology SE's case. The fact that Biology didn't even come up demonstrates a lack of attention from area51 moderators that is borderline irresponsible, and points to some vested interest behind the scenes or general incompetence. – nico Sep 7 '16 at 9:49
  • 1
    Wait I lie, it did come up, I brought it up and it was chirp chirp chirp... maybe the guys from cog-sci don't like the Biology guys? In any case I don't think renaming cog-sci would change the core expertise that is lacking right now to answer technical neuroscience type of questions. Maybe you guys should defer to the Biology site whenever questions come up? – nico Sep 7 '16 at 9:52
  • @nico, I reached out to the closed Neuroscience proposal on Area51, and posted specifically Biology is an alternative destination. As a moderator at CogSci, it is not my task to initiate a discussion with Biology. I can only guide the discussion with CogSci. – Steven Jeuris Sep 7 '16 at 15:24
3

Your situation seems very similar to another one I read about.

But first, I want to say that "all data permanently erased", "thoroughly deleted", and the entire section about losing the history of the proposal forever are overstatements. It's not gone at all; it's just hidden to everyone with less than 10k rep:

What is a trusted user?

Trusted users are allowed to perform trusted actions, including:

  • cast delete and undelete votes on proposals
  • view deleted proposals

Anyone can look at the archive.org version.


Now, back to the story I wanted to tell.

There was another proposal that was created years ago. For several months, it gained traction, including commitments from several "big names" in the industry, only to be closed as a duplicate of other sites. However, questions asked on these other sites were likely to be closed.

If you clicked on the link for that proposal, you'd notice that it's deleted now. But the story doesn't end there.

A new proposal was created, and it gained progress very fast as everyone from the old proposal committed. It would go to beta, and then became the fully-fledged Mathematica site you can visit here.

You can read more about this here.


I wouldn't give up hope for your site. After all, a site can come back from deletion, or a new one could take its place.



More information about closing down sites:

Proposals that do not meet these requirements within three days after submission are subject to removal:

  • 5 example questions you would like to ask on this site
  • 5 users who are willing to 'follow' this proposal

Note that abandoned proposals which receive no activity for a period of 30 days may also be removed.

Private beta, which is limited to Area 51 committers, lasts 7 days. It is during this time that the community must demonstrate that it is capable of putting together quality content.

Public beta, which is open to anyone, lasts at least 90 days. It is during this time that the community must demonstrate that it is both capable of maintaining a high level of quality for a duration of time and that it is capable of significant activity.

If at the end of either period, the community has not met its goal, the site will be closed.

More information about deleting sites:

It takes 3 votes, minimum, to delete a closed proposal — and the proposal must have been closed for 2 days. However, the number of delete votes required scales to the number of followers. If you feel a proposal should be deleted despite having lots of votes, please flag it for community moderator attention.

  • Posts may be deleted for being inactive:

Inactive proposals that do not receive any activity for one month are subject to deletion.

  • Moderators have the power to instantly delete (or close) proposals:

Occasionally, proposals may be removed from Area 51 for reasons of moderation: spam, off topic, abuse, etc.

Note that the community may (as far as I know) delete proposals with flags (if the proposal is spam or offensive):

The offensive and spam flags are designed to automatically eliminate truly offensive posts through the collaboration of the community. When a post receives 6 flags, it is locked, deleted, and its owner loses 100 reputation.

  • 1
    Thank you for the archive.org link, that is very useful. Also for the inspiring story, indeed it gives some hope to our proposal. Regarding the permanent erasure of data, we were informed that this was so directly by the Area 51 moderator @Robert Cartaino. He mentioned Area 51 used some old code which had limited, i.e. non-existing, ability to recover deleted data from proposals. If this was incorrect, it's good to know and I will edit my question. – glopes Sep 2 '16 at 23:10
  • Thanks for that, a faint glimmer of light at the end of the tunnel was badly needed. I have to read more carefully the whole process but it is very similar to ours also in the complaints about SE's MO. I wander how much potential was lost with this policy that could have been converted in new sites. I think that the whole bureaucratic-must-check-all-items-in-box attitude is very counterproductive, stifles innovation and can unnecessarily create bad blood. – nico Sep 5 '16 at 11:27
  • Maybe you should restart it with the name "Neuroscience Methods" to clearly separate it – honi Sep 9 '16 at 2:48
  • @honi maybe we should. – nachosan Sep 12 '16 at 10:11
-5

There is no easy solution for your problem. The original cause of the problem was without doubt the poor moderation culture of the cogsci SE.

Unfortunately, SE sites, as they grow, tend to become more and more closed, their rules tends far away from the common sense, they start to implement illogical, meaningless policies what they apply more and more against any rationality, with increasing harshness.

I wouldn't close out, that the CMs, at least partially understand the problem, but even they can't do too many. It is also possible that they simply ignore the problem (and also the site statistics). Sometimes I had a quite clear impression, that also the CMs tried to do something against a poor, sick site culture, but unfortunately it doesn't happen enough often.

It is easily visible, for example, that most of the small but active SE sites have around 5 regular close voters. It is because they are the minimal set of users capable to close questions.

It is not easy to say, what to do. If you restart the neural science proposal, it will be probably closed again as "cogsci clone", while the cogsci will still close the neural science questions. And then they will say it is ontopic. Yes, being ontopic in the site FAQ doesn't mean that it is being ontopic also in the review practice.

Simple solution for this problem clearly doesn't exist, longterm solution can.

First, get to a high rep on the cogsci and try to invest all of your influence to fix its community:

  • Most of the VtC voters simply don't understand what are they taking part. They live in a world where they are taking care the site by closing bad questions. Awake them.
  • Find other, mainly newbie users handled on unfair ways and try to explain this also for them.
  • Between the "caretakers", identify the few really bad guys. Their primary characteristic, that they have a strong wish for power and that they are gaming. I.e. their goal is not tp read and write good questions and answers - although many of them does it, they can be even the best experts of the site. But, their unsaid goal is simply to get to the top in the site hierarchy. It is probably some psychological defect by them. For them, closing 3542 questions is quite similar as if you would play a first-person-shooter with your friend and you would shot his 3542 characters on the screen.
  • 2
    For your last bullet point, can you provide some more substantiation for the idea that closing lots of questions (which aren't conveniently summed up anywhere in the UI, never mind put on a leaderboard) is a high-score game or some sort of power play? At present, it looks like a weird random guess at someone else's secret motivations based on no expertise or personal experience, just armchair psychology. What puzzles me is that I've seen this sort of guess before, but it's never substantiated, and I really can't figure out where the idea even comes from. – Nathan Tuggy Oct 7 '16 at 1:14
  • @NathanTuggy It is always hard to estimate, what are others real reason to a continuous hostile behavior. But the problem, unfortunately, exists. It is not random guess, it is based on years of consultation, watching, investigating, and yes, also on a little bit of applied psychology. I am now quite sure, this is in the background, although I can't provide links or evidence - from the nature of the problem, it is obviously impossible. Of course none of them will come here to admit what are they doing. Possible, that some of them wouldn't even understand the last bullet. – peterh says reinstate Monica Oct 7 '16 at 7:30
  • 2
    "The original cause of the problem was without doubt the poor moderation culture of the cogsci SE" Do you have any proof or reasoning for this? I'm not involved in any of these communities so I have no idea, but I don't see it stated anywhere else that this had anything to do with CogSci and this just looks like a (only tangentially related) attack on CogSci. – Cai Oct 7 '16 at 8:01
  • @Cai Any idea, how could I give you proof for a poor moderation culture? – peterh says reinstate Monica Oct 7 '16 at 8:02
  • @peterh but what is the reasoning for saying the closure of the the Neuroscience proposal was down to poor moderation on CogSci? – Cai Oct 7 '16 at 8:05
  • @peterh: What about the null hypothesis, that these users actually believe such questions should be closed and are trying to close them accordingly? How do you reject that? – Nathan Tuggy Oct 7 '16 at 15:51
  • @NathanTuggy My rejection is based on my direct, yearlong experience (simply I know that the reality is quite far from the idea of the goodstanding, naive close-voter, and that it turns most of the SE sites simply a bad place). I also investigate since years, why is it going so. I learned a lot about the reasons, and I would be glad to share my theories with you in the hope that you will find also my evidences enough. – peterh says reinstate Monica Oct 7 '16 at 16:22
  • @peterh: Yes, that's what I'd like to see in the answer. – Nathan Tuggy Oct 7 '16 at 16:23
  • @Cai The causality chain is the following: poor moderation culture of cogsci -> many question was closed unfairly as offtopic -> their OPs tried to start a new a51 site from these questions -> cogsci power users & CMs suddenly interpreted topicality enough wide to consider this new proposal as a cogsci clone -> the site proposal was closed, deleted, destroyed, and its traces were hinted with salt. – peterh says reinstate Monica Oct 7 '16 at 16:26
  • @Cai I've toke part already in some similar nice tries - mainly to provide concurrents for SE sites where the revieweing mechanism gone mad - and I know the choreography. The SE kills these tries as clone on the spot. They don't bother themselfes to check, why the clone (or, partial clone) was started, and they also won't ever fix the rampaging reviewer "community". This is one of the worst problems of the whole SE network and I suspect it is the most probable reason why changed it from exponential growth to linear some years ago. – peterh says reinstate Monica Oct 7 '16 at 16:32
  • @peterh is there any indication anywhere that any "power users" from CogSci had anything to do with the closure of the proposal? Fair enough if there is but I don't see it anywhere here – Cai Oct 7 '16 at 16:36
  • @Cai Probably no. First, it was originated not surely from the cogsci power users, it could be originated also from the mods or from the CMs. Second, how X influenced Y, particularly in such sensitive situations, in most cases remains hidden from the public view (although sometimes significant infos can be found, not always proof strength but enough to estimate what was going on). But, proposals considered clones are destroyed on the A51, it is well known, and the concept of "rampaging reviewers" is considered unimaginable on the meta SE by definition. – peterh says reinstate Monica Oct 7 '16 at 16:46
  • @Cai This is not the first similar case I see, this is why this problem is well known to me. In my opinion, the best would be if the SE network and the topics would be a "doubly connected graph". It means, actually all question could be asked in at least 2 sites. Extending this, sites with similar topics should be able to import the closed/deleted content of their neighbors. It would result an "internal race" between the communities for the visitors and members, while the unfairly closed-destroyed content wouldn't be lost from the view of the SE network. – peterh says reinstate Monica Oct 7 '16 at 16:58

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .