Proposal: Open Science

This is the second startup for an Open Science site: I'm here to ask, what elements went wrong, and what can we do to improve, or make sure that they don't repeat themselves in a second start?

share
1  
More activity :) – Franck Dernoncourt Aug 17 '15 at 23:29
    
@FranckDernoncourt Obviously, I think my main goal in this post is to try to identify other mishaps or issues that were likely a factor. I have an answer to this already, but I want to see the community first. – Zizouz212 Aug 17 '15 at 23:30
4  
Is there a logic to this process of shutting down a proposal? I think we've pretty much established that the site won't get off the ground without being public/open...so I'm not sure it's really viable without some kind of special dispensation to go into public beta even when "weak". – Thomas Aug 18 '15 at 0:21
1  
@Thomas I can explain the process to you, it would be more efficient to find a chatroom to do so though. Let me get ourselves a chatroom :) – Zizouz212 Aug 18 '15 at 0:34
1  
@Thomas and others... I've created a new chatroom so that particular issues can be discussed more efficiently and thoroughly. It can be found here: chat.stackexchange.com/rooms/27080/discussion-for-open-scien‌​ce – Zizouz212 Aug 18 '15 at 0:38
    
Link to the previous old proposal: Open Science – kenorb Aug 18 '15 at 9:19
3  
Starting afresh sounds quite useless if we don't even know what failed. Two weeks is too short a time to produce any real information. I think we should aim for an extension, even if it's just one week more. – Yisela Aug 19 '15 at 12:16
    
@Yisela Actually, two weeks is plenty of time to at least get about 100-200 questions on the line. Most private betas should be producing 25-50 on the first day, Open Science only produced 19. – Zizouz212 Aug 19 '15 at 15:18
1  
@Zizouz212 Oh, I meant real information around what failed and what should be avoided for a possible next launch. The questions themselves were not that many, but were all really good! – Yisela Aug 19 '15 at 15:29
1  
@Yisela Alright. I'm not a community manager (obviously), but I've watched a lot of sites grow (it's been my new thing to do), so I might be able to provide some guidance into this. I'm going to be posting a pretty large answer with what I thought went wrong. – Zizouz212 Aug 19 '15 at 15:44
    
@FranckDernoncourt: we need only 4 more 10+ questions to go to the next stage - there is a risk that again this will happen in summer... – Alexander Konovalov Jun 11 '16 at 23:10
    
@AlexanderKonovalov Yes, we'll see… – Franck Dernoncourt Jun 12 '16 at 1:43
    
@AlexanderKonovalov I know! So much excitement! – Zizouz212 Jul 10 '16 at 23:12
    
@FranckDernoncourt do you have a link to a place where the main failings of the previous attempt are summarised? Thanks – Alexander Konovalov Jul 22 '16 at 10:58
    
@AlexanderKonovalov sorry, nothing off the top of my head. – Franck Dernoncourt Jul 22 '16 at 14:09

Here are my few suggestions/ideas:

  • define in better way what we call 'open science' and what kind of questions can be asked,
  • increase threshold of Commitment phase (so more people should join in order to start),
  • don't launch the private beta on the vacation season, so more people can participate,
  • people! don't take any vacation when the site is about to launch:) (or install mobile app),
  • re-import previous successful questions again to encourage people in participation.
share
6  
totally agree with not launching in the middle of vacation! I for one was also not aware the beta would be so short-lived, otherwise I would have done more as soon as it launched. – Stephen Eglen Aug 18 '15 at 10:41
    
Excellent. As a note to anyone reading this, though, many of the previous questions have been migrated to other Stack Exchange sites. – HDE 226868 Aug 21 '15 at 21:30

I can't help wondering whether this is a big enough field. Once the obvious "why is open science a good idea" questions are answered - and most of them were at least asked in the last beta - I suspect it comes down to more or less specialised methodology and tools discussion, most of which would fit at least as well on a subject- or tool-specific site.

I'm not saying this can't work; I'm just not convinced.

share
    
I am convinced it is: it is in its infancy, but open science as a concept has large political, academic, societal, global, educational importance. So to be "on the edge" will always feel a little wobbly. I believe, ten years from now, we might look at these tethering problems as silly. – Benteh Aug 19 '15 at 13:02
2  
I didn't say that open science isn't important - just that I'm not sure that it has a wide scope of questions for a SE site, that wouldn't be better answered elsewhere. – Simon W Aug 19 '15 at 13:25
1  
In 10 (20?) years I hope that open science is simply part of general scientific method and not deserving of its own site at all! But let's concentrate on now, not decades hence. – Simon W Aug 19 '15 at 13:26
    
Good point. It might turn out that this is not the time, or that it is better suited in other formats. I do not think so, though. Nothing like the present! – Benteh Aug 19 '15 at 13:30

I waited for a little while before I would go about posting my thoughts about why the site was closed. I don't want to talk to much about scope for right now (I believe it didn't have a focus, but that likely wasn't a major factor in the closure of the site). I want to talk about scope, so I have raised a separate discussion post about it: What's the purpose of the site?

I'm not a community manager (obviously), but I've watched a lot of sites grow (it's been my new thing to do), so I might be able to provide some guidance into this. Don't expect me to know anything, or that I'm speaking for Stack Exchange either. They've already told you on the Open Science meta.

Here are the biggest factors that likely closed the site:

  1. Number of questions

    I'm going to be including photos of graphs of other SE sites that have gone through beta. I'm unable to other a graph on Open Science, or Computer Graphics: Data SE doesn't offer data for sites in closed beta. I'll see if I can get something with the SE API though, but that might take time. I'm providing these graphs for comparison, don't start arguments about them. You can find the query here: Number of Questions Per Day.

    Graph 1: Open Source

    Open Source Questions

    With Open Source, they had 79 questions within the first two days, which is more questions than Open Science had during the entire closed beta. Did the activity slow down? Yes, but that is perfectly normal. Private betas are supposed to be getting tons of questions within the first week: During the entire length of the private beta, they should generate 100-200 questions. A lot of these questions are artificial, and not always the best. Keep that in mind, I'll be talking about it later.

    Graph 2: Law

    Law Questions

    Law was also off to a good start. They got roughly 68 questions within the first two days.

    What am I trying to get at here?

    That's an excellent question. Look at something in common between the two graphs above. You had a spike in the beginning, then it went to nearly nothing: That's normal behaviour for successful sites. Sites that go into public need lots of questions so that the public community can find content, and ask questions that match a certain level of quality... Which brings me to my next point:

  2. Question and Answer Quality

    There should be tons of questions in a private beta. Yes, they are artificial questions... Many people are asking for the sake of asking, and not always asking because they want the answer (many have a sense of the answer anyway). As such, there is a bit of a smack in quality's face, the quality decreases slightly, because no one asks questions that are about real-life situations or problems, there's no questions that will always apply to people. As a result, at least 90% of questions in private betas are mostly theoretical questions, and at least 70% of questions are the theoretical solutions to these theoretical questions.

    However, the quality could've arguably been much better on Open Science. Browsing again through the questions, many questions were asking for links to external resources. This is not what Stack Exchange is about: the goal is to create a sort of Wikipedia on the subject, not redirect users to other places to do things, or learn about things.

    Many questions shouldn't be like this. You want lots of content to be produced in private beta, not "list" answers, or "stopovers" (answers that redirect people through links). Remember, the questions produced in a private beta is what is going to make the site in the future. Ask horrible questions now, you'll get horrible questions in the future. Ask great questions now, and you'll great questions (and answers) in the future. Case Study: Project Development on Open Source. Many questions that are now asked aren't the best, also because in this section of scope, we asked overly broad questions, such as "How do I attract new programmers?".


I wanted the site to succeed. However, the site was right to try to close: It just didn't have the resources to sustain itself. There's nothing wrong with that, and you can always keep going! For those of you who put a lot into the site, thank you!

While you're at it, define the scope here as well: What's the purpose of the site?

share

I'm new to StackExchange. I came here simply for the Open Science Beta. Yesterday. What are the advantages of StackExchange over similar alternatives? (I'm not quite sure what those alternatives are.) One feature that I'd appreciate, in addition to questions, is the ability to post content (i.e., here's something interesting to the open science community) followed with discussion.

Edit: The comments below illustrate what I think is a problem with the StackExchange format. I have used StackOverflow before to solve problems, but these are precise problems with answer that are right or wrong, or perhaps work or fail. However, a lot of open science questions require more discussion. For example, here was a good question, but I think it requires more of a back-and-forth discussion.

share
    
Could you elaborate on the kind of discussion? – Zizouz212 Aug 17 '15 at 23:48
1  
@Zizouz212 Good question. I guess I'm not really sure. I suppose that any content (e.g., personal thoughts, new article, new tool, etc.) could be posted in the context of a question that would help intimate the discussion. – carlislerainey Aug 17 '15 at 23:52
2  
Alright: So then that would be a disadvantage. Stack Exchange isn't designed for discussion (they're not a forum), but more like a concise Q&A. Questions and Answers, no chit-chat is how it goes. What SE does have, is chatrooms, to enable further discussion. They can be linked with particular questions and answers as well. I guess Stack Exchange is a proven system for Q&A, and it's got a good reputation with multiple communities. – Zizouz212 Aug 17 '15 at 23:55

I second @kenorb - but i also think we all have to spread the word! Any social media, faculty meeting, notice board; whatever. Twitter should be a-twitter with this, by now.

share
    
This is a good idea, however, you can't really do that in Private Beta's, since no one really has access to it. What you can do there, is invite people to the private beta and then they will have access to it. What we prefer however, is for more experts and experienced SE users to start in the private, so they can make sure that they start a quality standard, experience in moderation, community handling... etc. – Zizouz212 Aug 19 '15 at 15:20
    
Surely people can join here? A good deal of people – at least in my "circle" have StackOverflow accounts. – Benteh Aug 19 '15 at 19:58
    
I don't completely understand what you are trying to say – Zizouz212 Aug 19 '15 at 19:59
    
Hm, perhaps we just misunderstand each other. My point is that information for this proposal should be spread widely regardless of state. People can commit to this like you and I did. I do not see the problem of spreading the word even when the site is in private beta (which is only a period of two weeks) when only "we" can contribute. It is first and foremost about awareness. – Benteh Aug 19 '15 at 20:05
    
For sure, but the role of a private beta is far different then the operation of a normal site, or public beta for that matter. – Zizouz212 Aug 19 '15 at 20:05
    
Yes, but private is only for two weeks. So anyone can join now, and after those two weeks when it goes public. I do not see why "recruiting" should be postponed. – Benteh Aug 19 '15 at 20:07
1  
Why don't you hop over to the chatroom - chat.stackexchange.com/rooms/27080/discussion-for-open-scien‌​ce - That way, we're not spamming the comments :) – Zizouz212 Aug 19 '15 at 20:11

You must log in to answer this question.

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