go_gentle: (Default)
just a girl who's afraid of the dark ([personal profile] go_gentle) wrote2008-03-13 05:32 pm
Entry tags:

More Wiki Musings

Wait, you actually want me to make my crazy ideas happen? (I had a thought today - the ideal finished product I have in mind is not that far off the HP Lexicon in content. But if I get as out of hand as the Lexicon has been in recent months, someone please take my internets away. (Although I suspect that publishing this hypothetical wiki - if one could transform it to book format - would be much less legally problematic than publishing the Lexicon. But I digress.))

Someone once told me that there are two types of people when it comes to projects. One prefers to start and figure out the details later, one wants to get all the details straight from the get-go. Clearly, I am the latter.

Okay, disclaimer: I have no idea what I'm doing. That doesn't mean I'm not going to try to make it happen - I'm well familiar with blind fumbling in the dark, but if you do know something, feel free to chime in. Most of this is based on research and not actual experience. Also, it's mostly me thinking out loud, so please tell me where my thinking is going wrong. This got more epic than I expected - feel free to comment on as much or as little as you want.


Technical Details

Format

MediaWiki is what runs wikipedia (also the fan history wiki, the F_W wiki, etc.) I think its popularity means more people are going to be familiar with how to edit it, etc. I know I found editing MediaWiki sites very easy to learn, but I'm not sure if that's universally true. I have a deep hatred of WYSIWYG editing - I prefer to stay closer to the code, which is one reason I like the way MediaWiki sites are edited. Will that be off-putting for newcomers to wikis?

My personal preference would be to use a wiki that runs MediaWiki.

Hosting

Someone else

There are sites that are dedicated to hosting wikis (called 'wiki farms'), often for free or supported by ads. A list can be found here. A comparison of features is here.

Browsing through and looking at offered features, I like wikilot, wiki-site, scribblewiki. (ETA: Wikispace has been pointed out in comments, and I'm adding it to the list of options worth considering.) Maybe wikia. I have no experience with any of these sites, and I don't know how long they've been around/will stay around. Thoughts?

Commerical Hosting

Wikipedia is about 3GBs, according to LifeHacker. Presumably, anything we build would be orders of magnitude smaller than that, which I think is a very workable size with just about any commercial webhosting. With the amount of statistics put out by the FanHistory wiki people, I should be able to find details on size/traffic, but two minutes on google doesn't turn anything up.

The choice of commercial hosting would probably mean we'd have to install MediaWiki (or another wiki software) ourselves. The instructions looks doable, but not trivial, but I have no experience to base this on.

DreamHost provides instructions on how to install MediaWiki on a site hosted with them. A number of other hosting sites advertise one-click installation of a wiki, although not all of them are MediaWiki.

I am not familiar with the pros and cons of various hosting services, and I don't know where to start looking. I also don't know if I have missed an option.

My thoughts are that at this point, free hosting on a wikifarm is preferable over the other option. My only concern is reliability/long term viability. Agree? Disagree?


Content

Things I Want:
1. Band Bios!
2. Band Member Bios!
3. Lyrics! (the principle is 'why the fuck not?')
4. Media Indices! (I suspect that hosting video/scans/whatever may be beyond at least the initial stages, but something like 'link to such and such interview on youtube, in which these things happen and the band touches on these topics' would be kind of neat.)
5. Tour dates!
6. Citations! Seriously. Citations are awesome.

Things I Don't Want:
1. Fic/Art/fanmade media. Just not the place for it.
2. History of fandom. First, this is an impossible topic; second, there are other places already working on it.
3. Wildly unsourced rumors.
4. Wank. Because, really, who does?

Yes? No?

I suspect that the wiki will require a small team of moderators (along with an owner) to ensure that pages remain reasonable. Some pages may also need to have editing privileges restricted, depending on vandalism.


Setup and Organization

Clearly, at startup we have huge backlogs of info that needs to be added. Suggestions for ways to make this as easy as possible for users? One way would be to have a small group of users create a massive amount of stubs, which then could be filled in later by the general public - sort of creating an outline for the eventual document.

I don't know how it would be best to organize the pages. I suspect this is something it would be nice to get mostly right from the beginning, because re-organization is always a pain in the ass. Suggestions?


Social Organization

Another thing I have no experience in! Things I am thinking about:

-Do we need an LJ comm?
-Should I post to BSM about this? Tangentially, feel free to pimp this to your flists/comms/wherever - I'd like as many opinions as possible so that the final product is as useful to as many people as possible.
-How should mods be chosen?
-How to avoid stepping on toes - what toes should I be worried about stepping on? [livejournal.com profile] bandfandom_ref? [livejournal.com profile] bandom_primers? Something I don't know about?


What am I forgetting? Any other thoughts or suggestions you have?

[identity profile] olivia-circe.livejournal.com 2008-03-13 11:37 pm (UTC)(link)
I have nothing useful to add on the hosting/editing/kind of wiki front, but everything you are saying makes sense to me. The WYSIWYG editing thing might not be an issue if there are people and/or guidelines available to help the more coding-illiterate among us.

Content and Organization:

4. Media Indices!
Or, in my language, annotated bibliographies! Including videos, interviews, pictures, etc., yes? I am 100% behind annotated bibliographies. For that matter, everything you mention in terms of content sounds good to me, both the wants and not-wants. I would also add comprehensive timelines, which might be easier to assemble once we fill in everything else. Because what we're essentially trying to do here is create a canon encyclopedia, right?

I think the annotated bibliographies, the citations, and avoiding unsourced rumors all go together, and could potentially be handled by a team of citation-tracking and fact-checking mods who keep tabs on whether or not the information people submit is, in fact, substantiated, cited, and linked in the annotated bibliographies/indicies.

Probably starting out by creating a bunch of well-organized stubs and then letting people slowly fill in the information is the way to go. As people start filling in information, however, I imagine you'll needs mods checking their citations and checking their code - the researchers and the techs, if you will.

(The citation-checking process probably goes like this: x submits a bunch of tour dates without a citation, mod says "hey, x, do you have a citation for this information?" x says, "oh, whoops, here it is" or "I got it from here, where someone posted it but didn't cite it," in which case the mod gets to do detective work. If the dates can't be substantiated by the mod team and posters, they get taken off the wiki. Or something, I don't know, we'd have to work on the rules.)

I guess this is all a roundabout way of saying that, should it be deemed a good idea, I volunteer to be on the citation/fact-checking mod team. All the reference librarians in bandom should also be on this team. :)

Social Organization

Ideally nobody's toes get stepped on, because we are all working on it together! Um. Wishful thinking? Okay. But I don't know, I mean, I imagine the [livejournal.com profile] bandfandom_ref and [livejournal.com profile] bandom_primers people could potentially want to help out, since our shared goal is more canon for everyone. That's just me, though, and I know nothing about the politics.

I think an LJ comm might not be a bad idea, if only as a place to gather people who want to help out/submit information/talk about how this will work. Of course, doing that in your LJ is fun, too...
ext_3467: a path from the forground to the background, through a yellow and green field (Default)

[identity profile] go-gentle.livejournal.com 2008-03-14 02:38 am (UTC)(link)
YES! I love basically everything you've said in this comment. Also, research librarians. So much love for reference librarians.

I really really like that process for dealing with citations - it looks like a good way to optimize getting as much info as possible while still making sure it's reliable. I also like the idea of researchers and techs to correct things like formatting and coding, to make the barrier to entry as low as possible for people who have info to contribute.

I am full of wishful thinking too. I just don't want to piss anyone off before the project even gets started. I suppose I should just email the mods of those comms and give them a heads-up/ask to advertise in their comms.

[identity profile] olivia-circe.livejournal.com 2008-03-14 07:15 am (UTC)(link)
Reference librarians are hot, dude. :)

I was talking with [livejournal.com profile] sweetvalleyslut about this yesterday, and she said "Why hasn't anyone thought of this before?" and I think that's really the point. Obviously you don't want to piss anyone off, but honestly, we really do need some kind of comprehensive canon directory, and I can't imagine anyone not being on board with that somehow, you know? (Which is all basically to reiterate that you are awesome, and thanks again for doing this.)

My basic problem with wikis - wikipedia included - is that they aren't always well-cited, and consequently not always reliable. I mean, I guess that's the point, and my roommate the techie tells me that ideally wikis are supposed to eventually gather people who make it their business to track citations, but...I don't know, I'd just rather have those people in place to begin with. That's the point of mods in this situation, I think: to keep the entry threshold low and the ultimate quality high, right from the get-go.

Speaking of the get-go, though, how to organize the material? Searchable directory...categories...categorical structure...ahh, I have no idea. I think I used up my good ideas for the day. I will keep thinking, though, because I am so very much on board with this! :D
ext_3467: a path from the forground to the background, through a yellow and green field (Default)

[identity profile] go-gentle.livejournal.com 2008-03-15 12:26 am (UTC)(link)
Good point on the role of mods. I still like everything you're saying.

I have been thinking about organizing this all day. Any sort of organization system is not going to be at all trivial to set up. Grrr. Arrgh.

[identity profile] olivia-circe.livejournal.com 2008-03-15 12:35 am (UTC)(link)
Thank you!

I...yeah. I am looking forward to whatever you end up saying about it later tonight? Hah, sorry. It seems like it should be simple, and yet it really isn't. I'd be inclined to have BANDS at the top of some sort of ladder/umbrella structure, or possibly in the middle of one of those brainstorming webs from middle school, but I have no idea. Are the subcategories of BANDS, then, PEOPLE, TOUR DATES, MAJOR EVENTS, ALBUMS, LYRICS? We can always cross-list? Make it possibly to search by PEOPLE or TOURS? I bet this is one of those things it's easier to do with a whiteboard and a lot of diagrams.

Wow, that was totally useless. You still rock, though!