Right... so what should we do with the wiki?

[Milnivri(imported)]

Well, i'm still here and i'm willing to restart the entire Wikifang project all over again. I agree with Kimbles that the pages aren't really that structured, and that's partly my fault because I've always been a list-kind of person (a person who likes organizing things into lists and stuff), and I agree the pages aren't interesting to read. I imagined Wikifang as a database of sorts, instead of a real information source, and well, I guess if you guys are still willing, we can work together to start it over again, rewrite the articles and stuff, reorganizing all the categories and restructuring templates.

Moving to our own server will allow a lot of customization and stuff, you can add in whatever Mediawiki plugins you want, instead of finding alternatives through javascript and whatnot, perhaps breaking compatibility in other browsers. On that note, I didn't really think anyone these days still used IE6, and such I didn't consider compatibility with IE6. Anyway, compatibility with IE6 is really a pain in the ***, so I apologize to Kimbles and other IE6 users. Oh and if RacieB could lend us the server space for the wiki, that would be great If not i'll try finding free webhosts, which might work too.

I really want to see this work, since I've actually spent lots of time on Wikifang, and I'm sure the others have too. It's really quite sad if after 3 years it just dies because Wikia made every wiki change to the new skin, even though I don't think that's the real reason.

So, if enough people are still motivated enough we can do this all over again, like how it started a little more than 3 years ago. So who's up for it?

[Kimbles(imported)]

I'd definitely be up for it. XD One question is whether we'd want to import/export some of the old pages to preserve the histories and stuff, or start completely over... Personally, seeing as the old site isn't going down any time soon, I don't think it really would be necessary if we're planning to rewrite a lot of it anyway. *shrug*

Regarding the templates and categories and stuff, from working on Medabots Wiki I found that it's pretty nice to keep things non-automatic. In most cases it would be just as fine to have people fill out the details for a template on a Denjuu's page instead of having it automatically filled in based on its number. Since the person person putting the template on the page probably knows what they're doing, and it gives you much more freedom to change things when you're editing each individual page. It also really reduces the complexity of the template code, so that people will have an easier time tweaking them if they want to. (I had a really hard time editing the templates on Wikifang when I was starting out, because they were so full of stuff and I honestly couldn't make heads or tails of the code.) This is easily my biggest issue with Wikifang right now, and it would be simple to fix if we were starting the wiki over... Also, less complicated templates have the side effect of working better in different browsers, which is something I always consider important. I don't use IE6 anymore (I use Opera), but I know some people who still do who would definitely be using the wiki. Even if things don't work perfectly, keeping it less broken is always a good thing. That's just my 2 cents, anyway. XD

[Sanqui]

Hi, Milnivri! Nice to hear from you again! (I would have noticed sooner but i'm forgetting to mark forums as read)

About moving, I still need to work on that with RacieB. If that doesn't work out for whatever reason, a friend of mine would be willing to host us too.

Other than that, both Milnivri and Kimbles have good points. I'd love to have the site structured, with a database from the game and lists and pages being generated from the database, instead of having manually entered duplicate data everywhere, something many online Pokédexes do. But that does not work with a Wiki from principle. For our purposes, we can either have:
- An application written in a programming language language (Python, PHP) specifically designed to work as a dex for sorts. veekun is a great example. There's always the problem of writing it, though. On the other hand, Denjuu data is /static/. Anything other then translations isn't getting updated anytime soon. So one a database is finished, it's just the frontend.
- A database-esque Wiki, i.e. Wikifang. As we've learned, this doesn't work that well. There's a lot of confusing templates going on, and they don't solve the problem -- if i view, i.e., List of Denjuu in Telefang 1 by base stats, there's still a lot of duplication going on: stats are both on the Denjuu pages and on the list. Not to mention things like the Japanese name, for which there are switch templates. It enables us to do some cool stuff, but at a cost of being confusing and needlessly complicated. And all the Denjuu pages were written by hand, anyway! However, this is what we have right now, it works, and we want to keep it. Redoing everything, unless we pick the first opinion, is not a good idea. I think we should import the old pages.
- A wiki with templates filled in manually, a la Bulbapedia. This works well for Bulbapedia, because they have bots to do all the tedious stuff like new Pokémon pages. I don't think it's feasible to manually enter every single Denjuu, move and item from both games. We failed at that once, and without any automated template support, it would be even harder. I'm willing to try and automatize it.

These are the choices.

As for the rest:
Compatibility will always be better than Wikia, as i find it absolutely unusable in a text browser now. Monobook is a good skin that doesn't require much changing to look good - just the banner, the background and some colors.

Also, i think we should do namespaces for the series. This isn't too good of an example, but right now we have [[Fungus]] and [[Fungus (T2)]]. With namespaces, we'd have [[Telefang 1:Fungus]] and [[Telefang 2:Fungus]]. That's obviously too long, so as an (automated) redirect, [[T2:Fungus]] would suffice.
If you enter Fungus in the search box, would you be redirected to the T1 page, or to a disambig?
This wouldn't work too well for articles like Chiru (probably Telefang:Chiru, or the main namespace?), and for other stuff like games and companie (main namespace).

Also, it sure would be great if we could move the forum out of zetaboards too! Copying the old posts over isn't such a problem (i did it once as a proof-of-concept), and unlike Wikia, there wouldn't be a ghost forum left behind. Also some moderation cleanup, LiamBrady and Sora444 obviously don't go here anymore (last visits 2009), yet are still admins/moderators.


WHOA LONG POST

Also hey you guys there's still this irc thingy over at irc.badnik.net #telefang

[Kimbles(imported)]

I dunno, wikis that use namespaces to seperate their content have always seemed weird to me, so I'm not sure I agree with that idea. The one advantage would be that you can have two articles with the same name and different content (like Fungus) but I don't think the few returning Denjuu in T2 would really justify setting the whole wiki up like that. There's also the issue (with Mediawiki, anyway) that the "go" search only works with the main namespace, so you'd need to find an extension to get around it in that case.

I don't think a manually-written wiki would be out of the question. I disagree that the reason Bulbapedia works is because of bots and automated stuff... It's more that they have a huge number of dedicated people, which we don't really. I enjoy organizing stuff and doing tedious work like making lists, so I would totally help out with that stuff if you didn't want to. That said, I think it's a bit misguided to assume that every single aspect of the games would require its own page... The items in Telefang are a much more minor feature than in Pokemon, for example. I think it would be silly to make a page for each one when it would be easier and more helpful just to have a good "list of items" page.

If you want to make a fully automatic non-wiki database for the games, I think that would be awesome, but then I couldn't really help out anymore. :O

As much work went into the old wiki articles, I would love to see the new wiki start over from scratch and not have to worry about fixing what's already there... I think it would probably turn out a better product in the end, with much less headache. That's just me though~

[Sanqui]

I dunno, wikis that use namespaces to seperate their content have always seemed weird to me, so I'm not sure I agree with that idea. The one advantage would be that you can have two articles with the same name and different content (like Fungus) but I don't think the few returning Denjuu in T2 would really justify setting the whole wiki up like that. There's also the issue (with Mediawiki, anyway) that the "go" search only works with the main namespace, so you'd need to find an extension to get around it in that case.

If the namespaces are set up as content namespaces, a "go" search works for them. But i agree, just throwing the idea there.

I'm not saying a manual wiki is out of the question! I'm pretty sure I could stub out the Denjuu articles automatically, because the data is all over the place..

But are you sure about starting from scratch? It... could work, i suppose. But the wiki is usable as is, how fast can we rebuild it?

EDIT: Well now i'm contradicting myself. Oops.

[Milnivri(imported)]

In my opinion, I don't see what's wrong with templates being complicated. Any widely-used and common template on Wikipedia is already complicated seeing that they are dependent on other subtemplates and is extremely frustrating for anyone to edit if they don't know what they're doing. (e.g. infobox templates, took me a full hour to understand the whole thing when I first started out) I think it's an inherent problem with MediaWiki, seeing that editing must be done on the browser, as compared to normal HTML pages which you can simply edit in Notepad and see everything update, whereas MediaWiki you have to keep 1 million tabs open if you're editing several templates simultaneously, and wait for cache to reload and stuff, which is pretty annoying.

A fully automatic wiki would probably not be a wiki any more, and probably makes use of MySQL databases as the main source and space for the information storage, and would be like what veekun is. veekun is extremely complete and easy to look for the information you want immediately. I had envisioned Wikifang to be like that once, but I agree that a wiki is a much better platform if you want users to be able to contribute at all. I could create an "automatic wiki" with the ability for users to edit the data, but again not everyone can be allowed to edit since there may run the risk of vandalism, and if not everyone is allowed to edit there is no way we are going to get new members to help out with the project. I personally feel that this is not a bad idea for organising the data and stuff, but again, it has many drawbacks as well.

If we are going to do the original way and start again from scratch, I feel Bulbapedia has a pretty good structure in which their pages are built. Namespaces for each game is a good idea IMO. (I think you can change the way the namespaces look, instead of Telefang 1:Crypto it can appear as Crypto (Telefang 1), with some PHP editing) Because when I started out with Wikifang I was still learning and experimenting, many things could have been improved on should I do them again now. Starting again from scratch may prove to be better than the current setup. Also, the issue of time should not be a problem. I think if we all planned things out instead of diving headfirst into it again, it would be much better and save much more time in the long run. Though I dunno where to start...

Oh and I was always here, just didn't feel like posting. I wanted to ask Sanky about this on MSN for damn long but he was always on idle mode...

[Sanqui]

(About idle, i'm idle all the time! Just ping me and i'll respond when i'm awake/on the computer.)


Anyway, i'm okay with both automated/manual templates or anything, really! I feel pretty confident about doing all the Denjuu articles with a bot, too. (Of course, that would just be stuff like stats and moves, not translations or descriptions).

[RacieB]

I'm still willing to host on meowcorp, I have to figure out how to handle the FTP stuff though. Right now I just use one password for everything, and while I know you guys for example wouldn't go through and eat all my uploaded files I really need a more secure option XP That way I could potentially host other subdomains as well (hey it's unlimited space and bandwidth basically so why not?)

---

Okay cool I just played with it a bit and I think I got it figured out~ wikifang.meowcorp.us now leads to a placeholder page, anybody who needs the login information please PM me here or catch me on MSN and I'll give it to you.

[andwhyisit]

My hosting is also available if you wish. Although you may not want wikifang.andwhyisit.com or andwhyisit.com/wikifang as your url. Maybe I should get the domain wikifang.com? It's only another $9-12 a year. Alternatively wikifang.info is just $2 a year. *shrug*

EDIT: I started writing this before Racie edited her own post btw.

[RacieB]

Guess it just depends on what url is most desirable? Though I dunno if it really matters so much.