Proposal talk:Wikiscience

Some interesting work is already happening in this area, especially in the biomed/bioinformatics community.

• OpenWetWare, networking bio labs and groups, using customized MediaWiki
• Registry of Standard Biological Parts for Synthetic Biology, using customized MediaWiki,
• IGEM synthetic biology competition uses MediaWiki heavily
• WikiProfessional uses heavily customized MediaWiki for protein annotation (not sure it's still being actively developed)

I'm not as sure about other scientific domains but it would be good to get an environmental scan of what's happening, both using MediaWiki and other platforms.--Eloquence 22:41, 24 September 2009 (UTC)

There are others, too. However, all I have seen are area-specific, typically narrow-area. I don't think they will create a central data storage, common for different sciences, and facilitating the interdisciplinary researches.

Also, their popularity is in most cases low to average. A Wikimedia project will be many times more popular than most of them, even in their areas, due to the Wikipedia connotation. And a wiki needs popularity to thrive.

They can be useful source of MediaWiki extensions, procedures, general ideas etc. Also, these who join will make a good start for the project. In return, a Wikimedia project will be a powerful free hosting for them. Also, its far greater popularity will give them a signifficant boost. In total, if our project gives them the praise they deserve for being the first, creating some great tools etc., I think many will be inclined to join. -- Григор Гачев 23:53, 26 September 2009 (UTC)

Wouldn't a wiki for all sciences be too messy? OpenWetWare is very focused, and still quite big. I think that a namespace for each science or perhaps sub-discipline would be better. As in the incubator:, where we have lots of sub-projects with subpages, but less ugly. Otherwise, people won't be able to find out what is happening in their project (we don't have a "watch category" o "category recentchanges" feature, neither...). Maybe a subdomain for each project would be event better. We can also think of a specific Wikia-like wiki ecosystem, with lots of interconnected wiki created at request. Such a wiki could be or not be a "Wikimedia project" like others; maybe "only" a different, separated thing hosted and supported by Wikimedia but managed by some scientific network (UNESCO?).

I added two questions. Such a wiki can't function with a reputational systems: research wouldn't join it. Also, I don't think that scientists would accept to be under the "authority" of WMF board, to simplify the problem. We should explore how existing wikis are functioning, yes. --Nemo 11:24, 27 September 2009 (UTC)

Such a wiki will be messy, and should be. Science creation is most effective when almost chaotic; a good order requirement will cripple it. Compare with the Wikipedia problem: the better the instruments for keeping the articles polished, the less the incentive for editing (this may be non-obvious at first glance, but if you delve into the details, you see it everywhere). I think science namespaces, and domains even more, will make harder most interdisciplinary research, which is a key strength of such a project, and will scare away a lot of non-IT geeks.

Yet the messiness should not limit the effectiveness. What about encapsulating every project within its page (each page can have unlimited amount of subpages), and a good category system? The encapsulation is easy and natural; the good category system is trickier, but completely within the abilities of a typical scientific / wikipedian community. As categories will include mostly project main pages, they won't need extra watching. A "recentchanges for this page and its subpages" tool will be very handy, and is easy to make.

An attempt to answer both your questions: I propose the wiki as a whole to be under the aegis of WMF, which provides hosting and requires free license for what is developed in the wiki. The scientific projects, however, are not WMF projects: it merely provides some infrastructure, and requires no scientific credit for it. Except for the license, the workers in a project have complete control over it. They admin their projects (if they lack qualified person, WMF may provide a volunteer task force that works on their behalf, without participating in the project otherwise, and without taking scientific credit). They define their rules (participation, writing in the wiki, editors hierarchy etc). The project rules are enforced by its admin(s), and by technical means where possible. Admins are technically limited, or if that is not possible, required to not intrude with their rights on other projects: if they do, they are desysoped. (Will add this proposal to the main page.)

Providing a reputational system may mean burdening the wiki with extra functions: it is meant to be platform for working together, not a platform for determining who is greater. Let's explicitly stay away from the second function, at least for now. -- Григор Гачев 23:02, 29 September 2009 (UTC)

«A "recentchanges for this page and its subpages"» is not so easy to create. And you'll need to explain people how to use subpages, fix pages not added to the correct basepage, etc. The only existing tool is now "related changes", but you'll need an updated list of all subpages or pages in the same category (i.e. not new to-be-fixed pages). --Nemo 07:38, 3 October 2009 (UTC)

A "recentchanges for this page and its subpages" tool looks to me like several hours of work, at most (and if you are not into MW database structure, a week or two to learn it) 10 minutes to add several lines to SpecialRecentChanges.php, and at most a couple of hours to implement the needed parameters as links and form fields. Explaining things is always hard, but most people seem to get MW skills quickly enough. Scientists probably will not be much more stupid. -- Григор Гачев 12:17, 4 October 2009 (UTC)