OpenWetWare talk:Publishing group/Reddit idea


 * Reshma 08:48, 11 August 2006 (EDT): In thinking about the Reddit idea further, I think some of the concerns we raised at yesterday's steering committee meeting are quite significant. I realize that the Reddit idea is quite attractive since it is a nice out of the box solution to the page ranking and commenting solution but I am not sure that that outweighs the cons. The cons being that ...
 * Reddit is based on closed source software. Do we really want to rely on closed source software for OpenWetWare?  Now the primary counter argument to this is that we include google searches in OpenWetWare.  And google search is closed source.  But still, I am reluctant to rely on a software feature that we can't release to others.
 * Again, to reiterate from yesterday's meeting, since Reddit is a startup and likely eventually looking to be absorbed into a larger company ... I am not sure it is wise to use their software. Realistically, I don't think they can make any guarantees about what happens in 6 months or 12 months.  So I'd be pretty nervous about entering into a relationship with them.
 * Reshma 08:48, 11 August 2006 (EDT): Note that nothing prevents there from being a section of Reddit devoted to OpenWetWare. I am just not sure we want to tie Reddit into OpenWetWare formally via either software or a formal agreement (like user accounts).
 * Who owns/controls reddit and does it matter? Endy 21:04, 8 August 2006 (EDT)
 * Jasonk 21:34, 8 August 2006 (EDT):Yeah, good question - I can't find their license, who owns the comments?
 * Lucks 21:40, 8 August 2006 (EDT): I know one of the Reddit founders (Chris Slowe), who is a grad student in the physics department at Harvard. The company is based in Somerville and is the big success of Paul Graham's and Robert Morris's (MIT Computer Science) VC firm Y-Combinator.  I have been involved with them in creating arxiv.reddit.org, and at that time they were very enthusiastic about helping out science with their technology.  They were also flexible in tayloring their technology for science needs, and aware of the concerns about a company involved with open content.  Chris and I also talked with John Wilbanks at Science Commons about some of these issues with arxiv.reddit, and making all comments that appear on the cite covered by an appropriate creative commons license.  I think they would be up for adopting OWW's policy.  In the end they are a company, but one that wants to help science.
 * Do we know if mediawiki is developing an equivalent system built in to their source base? Endy 21:04, 8 August 2006 (EDT)
 * Lucks 21:44, 8 August 2006 (EDT): It would be great if we could add a mediawiki extension and do everything ourselves from the view of not involving a company. However, reddit has designed a nice technology to deal with common problems of such a system (spam, posts on the 'hot' list don't stay up there forever but change with time, etc.)  The drawback is that last time I looked arxiv.reddit had adds.  I do know that they also license their product out.  Perhaps a publishing house would like to pay for us to license their technology?  These might be long-term thoughts.
 * Austin 22:23, 8 August 2006 (EDT): Look at the development OWW site. I just installed a Review extension which appears to have been developed for wikipedia but was not used for some reason. Each page under the toolbox has a per-user rating.
 * Lucks 22:29, 8 August 2006 (EDT):What is the url for the development site?
 * Lucks 22:33, 8 August 2006 (EDT): http://dev.openwetware.org
 * Lucks 22:38, 8 August 2006 (EDT): Very interesting. But if every page has the option of being reviewed, that brings up Jason's point below.
 * Jasonk 21:47, 8 August 2006 (EDT):This is in the "things to think more about" section now, but I think it's something we'd have to address before rolling this out: Can anyone on OWW submit any page? If so this presents two problems: (1) I can get "credit" (reddit karma) for submitting someone else's good protocol (2) The actual author of the protocol may not have wanted to have it rated by users, or put into a DOI final form. (e.g. they still consider it a work in progress).  I don't want to discourage contribution to OWW because people are afraid of having their rough drafts torn into.
 * Jasonk 21:49, 8 August 2006 (EDT):p.s. I very much like the idea of trialing a system that involves ratings, I just want to make sure we think about the way it might affect users who don't want to use it.
 * Lucks 21:56, 8 August 2006 (EDT): Seems like there are 2 levels of control. One on OWW that might limit who can make posts (i.e. we could ask reddit to restrict posting to a certain button displayed on OWW only for those people who are authors on the current page), the other on how reddit assigns karma.  As it is now, there is only one user associated with each post - perhaps they could change that to be multiple users - the authors and the poster.  I would almost say keep the one user assignment, but only let authors post content.  That way it is only posted if people feel like the work is complete.  All users would be free to post news items or interesting papers like on arxiv.reddit.
 * Kathleen 23:29, 8 August 2006 (EDT): I'm a bit concerned about who is "allowed" to post pages to a journal. For instance, even if you limit it to just authors of a particular page, there is nothing to prevent me from making a minor edit on a page, thereby becoming an author of that page, and then posting it. Additionally, what about a situation in which two people contribute equally to a page yet only one person can post it? I'm also not sure that there is a simple way to determine who is the author of a page without taking a detailed look at what contributions have been made by whom to a page. For instance, if you are a person who excessively saves while editing, you could have 20 changes to a page and have only added one sentence. There is definitely a difference between posting something on a wiki and "publishing" it. I think we should tread cautiously on this one.
 * Lucks 06:39, 9 August 2006 (EDT): Very good point.
 * Johncumbers 20:42, 9 August 2006 (EDT) I think that a trial is a good idea to see how things pan out. Although author contribution is something regularly fought over now, I don't think that it will be in the future.  I agree that we might see issues of who came up with "the main" idea on a page, but this only becomes an issue when history credits one particular person with an idea and not somebody else.  This problem could be eliminated if you allow people to reference particular authors from the history file, instead of the whole page.  The community then can decide who's idea it actually is, and if they decide it is the idea of 3 people, then those 3 people get referenced.  I think a trial is the best way to find the best solution to this.   Also, I'd suggest if you don't want your idea referenced, then put a 'NOT READY FOR REFERENCE' template on that page, or 'ASK ME BEFORE REFERENCING PLEASE' again, I think it is something that would happen rarely and could be sorted out if and when it does happen by the community itself.
 * Jasonk 00:08, 10 August 2006 (EDT):I don't think the community wants to be in the business of deciding who gets authorship, that's a pretty big can of worms. An alternative solution might be to say - whoever started the page gets to be the final say, that way if you were going to contribute to a page you'd want to sort out the authorship situation ahead of time.  If you didn't and things didn't work out how you wanted, then it's just your fault and there is no need for a community vote or something on every authorship dispute. (not advocating this exactly, but it would avoid community decisions on authorship)
 * Jasonk 00:08, 10 August 2006 (EDT): Also, I wouldn't want to put a 'NOT READY FOR REFERENCE' flag on every page I put up on OWW - it's a hassle, might make the page look ugly, etc, and I shouldn't have to do it. The current publishing system says websites aren't really referencable objects, so why should I go through the trouble of pointing that out to people?  Not to mention that a big part of OWW unique content is rough, rapid-time scale information (e.g. this weeks experiments, my latest verion of a protocol, etc) -- it sort of goes against that whole philosophy to turn around and say every page is a polished, referencable object.  Maybe the deafult should be that pages aren't referencable and you should put a 'ITS OK TO REFERENCE THIS' tag on pages that are finished.


 * Kathleen 23:29, 8 August 2006 (EDT): I also wanted to bring up a point about further editing once something is "published". Are authors allowed to continue editing a publication after it has been posted to the OWW journal (say, in response to comments made)? Should there be an "in review" section where things are pre-published to allow for comments for a period of time? Then, things could be revised, potentially posted to a "post review" section and then finally "published"? This largely takes the current publishing model and makes all steps open (free access to information), but maintains some of the good aspects of traditional scientific publishing, such as critical peer review before things "go live". Just some things to think about.
 * Lucks 06:39, 9 August 2006 (EDT): That idea is covered in Nature peer review debate: Erik Sandewall. It seems like it has many benefits, but  I think we can implement unmoderated 'review' through a voting system like the one proposed.  If this works well enough, I don't think there would be a need for their second step of closed-door refereeing.


 * Kathleen 09:15, 10 August 2006 (EDT): Because this is a new concept for publishing, I would like to see some investigation of how publishing on OWW will impact the ability to publish things in more traditional ways in the future. For instance, if I have a project that has 3 major parts that need to be completed in order to generate a "minimal publishable unit" for a more traditional journal, can I publish each of those parts on OWW as they become available and then later publish the ensemble in a traditional journal? Figuring this type of stuff out may lead to more people wanting to try out this new publishing forum.