State of the wiki

[Sho]

95% of the wiki is a pointless waste of resources.

Current wiki content can be divided into two major categories: gameplay and data. The gameplay - Your First Cantr Day, Interface Guide, Known Bugs - belongs on the wiki, because it is opinion and experience from the community at large. The data - manufacture requirements, machinery uses, vehicle data - does not, because it is top-down information, and because it requires a publication mechanism completely different from a wiki.

A wiki is ideal for a situation where knowledge is spread widely among a large community, perhaps in a fragmented state. Everyone adds their piece to the quilt, a few moderators patrol the community, and the resulting mesh of knowledge far surpasses what any one person could have produced. Everyone who needs something has something else to give. People don't just read; they participate, and they do this because they know something the rest of the community doesn't know. Wikipedia is the best-known example of this - other examples are support and troubleshooting wikis for open-source software.

A wiki is the worst possible solution for a situation where knowledge is concentrated in a group of a few people who know everything there is to know, and needs to be distributed to a much larger group that has little or nothing to share. The insiders are too few and too busy to spread the information to the community. They don't actually see the wiki - it's got nothing they need. The outsiders are shut out and reduced to reading the wiki. They don't really care about the wiki - there's nothing they can do for it. If they know something that's not in the wiki, it's usually something that most of the others know as well - why bother? The people who don't know it, the real outsiders, are left in the cold.

All Cantr data is already in one central location: the RD database. It would be relatively simple to set up basic read-only pages to display all current data on machinery, weapons, foods, etc. This would be essentially a one-time job - no updates or moderation needed.


I have a few ideas about how this current situation arose. I know that at its inception, the wiki was intended to be very much like what I suggest that it become. I quote from Jos's original post announcing the wiki (http://www.cantr.net/forum/viewtopic.php?p=97757#97757):

It will be a lot of work if you want to describe all machines, resources, manufacturable things, etc. in the Wiki. I intend to program as soon as possible (next week?) a system that allows you to browse through that information automatically, inside the game. So the very technical information - what resources are there, what they used for, how long does it take to dig them, what machines produce what - will be made available soon enough. So you are allowed to add that to the Wiki, but you might want to wait, especially since it can become outdated. For more special items - locks, vehicles, seals, storages, etc. - all those things that have functionality that is not common among 'objects' in general, you might want to describe them.

But I would also expect to see a new interface guide, explanation of buttons, introduction to the combat system, etc. I will add an overview of skills at some point.

I couldn't agree more. This is what the wiki should be, preceding my rant by over 7 months. Everything I put above this quote was written before I ever read it, and they disagree in no significant respects.

What happened? I have trouble finding anything on the forums after this post that shares its vision - it's all people asking for information to be put on the wiki, people discussing the format, people arguing over RD disclosure of this data that should have been made public. Then it dies down to a slow simmer of me asking or being asked for information. Finally I see Jos and Chris Johnson releasing heaps of RD data on the forums - nice of them to release it, but what's it doing on the forums or on the wiki?

It may be that after the staff saw how much information was going on the wiki, they decided that the wiki could run itself, and that the misgivings about data in the wiki were unfounded. I don't know when it started, but right now I see a complete misunderstanding of the principles of wiki development and management. I'm told that in a wiki, the community is the editor. I'm told wikis moderate themselves. I'm told wikis generate their own content. I'm told that the whole point of a wiki is that nobody is responsible or has to oversee things.

Well, here's the truth. A wiki can't run itself. It can't even be run by a small group of people. The wiki system is an extremely inefficient one for content management ("write once, edit thrice"? on a wiki, it's "write twice, edit ten times, append, edit twice, spam, revert, edit more . . .") that succeeds purely by brute force of numbers. WIthout those numbers, a wiki is just a really big sheet of paper.

Having made over a third of the edits (2957 out of 8513, give or take) in the wiki and done most of the out-of-wiki pleading for data, I bear a good share of the responsibility for keeping up the sham that the wiki was a productive, vibrant knowledge base. That's over. Barring a miracle, the wiki is never going to be that.

The best we can hope for is that we fill in the most obvious gaps in the wiki, and keep it updated, until the database interface gets built. If we're not getting the database interface, we'll need a whole new department to manage the wiki - and I do mean that, not just "oh, the CD can moderate the wiki" or "oh, we'll give you one guy to liase with staff." Wiki syntax and editing is a skill separate from forum moderation, and if we're to complete the wiki we'll need multiple people working ProgD or RD hours on nothing but wiki editing.

Either way, I need more direct access to data. I'm completely fed up with having players "find out in game" the things that need to go in the wiki - they hate doing it, I hate herding them around to do it, and the community hates it because the information is patchy and unreliable. It's nice when RD gives me the heads-up on new stuff and tells me everything I need to put in the wiki, but that's still a time-consuming process. If you want more data on the wiki, it's going to have to be given to me. Recognition wouldn't be a bad thing, either. I know I'm an ego whore and all that, and I also know I'm not exactly the most trustworthy staffer in Cantr history, but it would give me a nice warm feeling to know someone realizes that I built half the wiki and moderate all of it. And I need that. The IRC lurkers can testify on how cynical I've been getting about it.


P.S. And can we please get that language split? It's been sitting on Flyspray for months - it's got nothing to do with any of this, but it would be neat. Took me five minutes to install a wiki on my computer - I know there are special considerations and all, and that the database isn't in the best of shape right now, but it's doable, and quite due.

[Phalynx]

Sorry you feel unloved Sho...

It must be hard to admit the Wiki is shagged when you have done all the hard work...

I can agree with you fully, recently the damage stats for all the weapons was released on the forum and you and your wiki minions (if you have any) vitually immediately transferred them into the wiki, but if the wiki is an official source for the game you have to ask why it wasnt put directly into the wiki in the first place.

There are massive hole in the wiki which is certainyl not your fault, your ..ahem.. obsessive approach to the wiki means that if info comes your way it goes striaght in the wiki, but the info has to come your way in the first place...


Maybe the info in the wiki would be best transferred to a simple support website which can be editted/controlled by you and abandon the wiki. It wouldn't be impossible to have a link in the inventory - like everyone could spawn carrying a 'basic manual to life' (jeez man if only you could have that IRL)


I also can't help but feel that just as you have admitted there is an ego component, other staffers derive pleasure from being the one to disclose (or not disclose) a juicy detail about game mechanics... I don't have a porblem with this for a new feature or something someone is proud of doing, but perhaps a week later it needs to be added to a reliable source for the game.

Anyway heads up to Sho, and don't lose heart friend!

[olaf]

That must be difficult for you, Sho - you've been getting a lot of (IMHO) unwarranted flak, and I've always been amazed at how quickly any info that is released is put into a very usable format on the wiki... (Although I've been playing almost a year, I use it multiple times a day as a reference or to plan things).

I've always seen you as someone who is sustaining things hugely for all of us, but yeah... never quite communicated that, sorry.

[Peanut]

Hmm?

I thought the wiki was used alot.
I always suggest clueless new newspawns to visit the forum and wiki.
And i use it alot to make economical formula's for my ingame company.

I find the wiki quite usefull. Keep up the good work imho.

[formerly known as hf]

I use the wiki a lot to find out a whole lot of information.

But, like Sho, I have thought that it does seem a really inefficient way of disseminating this information.

I'll admit that I've never had much input in the Wiki. Mostly because I don't know what the wiki should include.

So much of the 'database information' - what can dock to what, how is eaten, what projects a certain machine has etc etc etc - is just that - information that is held within the database.
The only 'added value' the wiki provides is easy and ordered access.

I have thought before (and mentioned it someplace) that surely it must be, in the long term, less time consuming to provide some form of access directly to the values in the database?

The only reason I can imagine that the wiki works, is becuase there are more people who have the skills to edit a wiki than have the skills and access to create some form of (easy) public interface with the relevant parts of the database.

Although, saying that, I am well aware that Sho seems to be the only one really doing the bulk of that. I can imagine, most people, like myself, see that all they are doing by keeping the wiki 'up to date' is copying from in game things that anyone with database access could give them.

It doesn't make sense that the access we have to this information is to have it repeated, second hand, with a delay, in the wiki.

All that is happening at the moment is that the database is being copied - by hand! and via indirect means.

[Marian]

Actually the wiki has been very helpful to me, even if there are a lot of gaps in it. I go there whenever I'm wondering about something, although it may be that some of the things I wonder abotu are things older players already know, or just trivial stuff that no body bothers to include because they figure everyone else already knows it.

But all in all Sho has done a really good job, though I agree that it shouldn't be him that's expected to do all the work, seeing input from other people more often would be nice too, since that's what a wiki's all about in the first place. I'll try to browse around there and add stuff a little more often myself, if I come across something I know about.

[Nalaris]

I like the wiki. I don't have much to do with it right now, but I did when I first began attempting to produce things. Now I'm trying to make steel rapiers starting with nothing but my hands and dust. The weird thing is it's working. I used the wiki when I needed to figure out how to make the rapier, but not so much anymore. I don't visit it enough to notice these gaps, I guess...

[colonel]

I understand how much work and time it takes do something like this and I really love the wiki, I read from it on the air all the time. It is very useful and a very big part of cantr now.

I never really added much because I was scared I was going to screw something up.

[Nick]

I don't think the higher ups realise what the wiki would be like without Sho. I think they assumed that if he didn't do it, somebody else would have.

Not that I agree. I was quite involved in the initial creation stage of the wiki, editing almost as much as Sho. But I stopped editing the wiki quite a while ago, and he has been doing all of this work. Keep in mind, not only has he no RD database access... he doesn't even have access to the page you get when you hit the manufacture button, or objects or locations button. He has no account. This makes things even more difficult, and what he's done, even more of a feat.

Did you know, most of what's on the Wiki (outdated now, though) was already put into a database publically available by Anthony Roberts, former RD chair, like over a year ago?

Sho is right, I think.

However, I think we still need the wiki. But it needs to be separated from the RD data. Let's see that database navigation tool that Jos had proposed a while ago. And actually Sho, I had read and quoted what you quoted from Jos a while ago, as well. That would require a signifigantly less ammount of work.

[nateflory]

I'm newish to the world of Wikis, but I had assumed the Wiki was built and maintained by teh same folks who created teh game, (ProgD & RD, etc).
Finding out it is not goes a long way towards understanding why some info is just not available (Placeholder pages).

I'll try to find out the missing info and add it myself if I get a chance. I'm back at zero minutes playtime so will haveto wait. Must...play...Cantr!

[wulf]

It is a useful resource... but, assuming all these figures are stored in the game database in a relatively straightforward way, it would seem sensible to provide a lot of the material automatically, making it accurate, complete and a lot less work to maintain.

Wulf

[Jos Elkink]

I very much like the title of this thread, btw - forgot to mention that

[Jos Elkink]

Oh wait, that sounds odd without any prior post. I replied to Sho in private mail.

I still agree with what I originally said and what was quoted, so I don't disagree much with Sho, except for the necessity of having staff organising the Wiki. I do, still, believe in player-based Wiki development. And yes, an automatic index of goods would be very good and useful. It's just a bit of programming nobody got to so far. ProgD has many, many ideas currently on its plate.

[Joshuamonkey]

Well, I am one of Sho's wiki minions. I use the Wiki a lot and I've added everything my characters had the ability to find out. Even if all the information from the wiki is put on the IC pages, the wiki is a lot easier to compare different items, and there's a search.
And I agree that the players should do a lot of the wiki editing, just like they create the Cantr world. And I had a character who was creating something like the wiki on a note. Now that there's a wiki, what's the point. lol, though sometimes I even use that note to add information to the wiki. And Sho, please don't get mad about me just trying to help. o:)

This topic is probably why there's a bunch of new editors. If everyone adds what they know and can find out, then the wiki will be a lot better.

[Sho]

Even if all the information from the wiki is put on the IC pages, the wiki is a lot easier to compare different items, and there's a search.

First of all, the search is broken. Second, the proposed RD database access system could have a basic search tool of the sort that is used in MediaWiki.

This topic is probably why there's a bunch of new editors. If everyone adds what they know and can find out, then the wiki will be a lot better.

What new editors? I am getting more RD involvement, and that is a good thing (shiny thanks to you guys!), but for the most part that help is in the form of data being sent to me to enter into the wiki - it's the best way for it to be done, since it lets me format the data as I like, but it's not editing.

Also, much of the information in the wiki was taken directly from the RD database. Of the information not provided directly by RD, most was grabbed from the manufacturing menu, which is essentially part of the database. Very little of the information was added by lay users just posting what they knew - and in fact, ideally all such information would have to be checked by RD. I have tried using the community to add information as you suggest, but that's over.

As far as the wiki goes, the ideal system would be for all data to come from RD. Data entry and formatting would be done by a small group of regular editors - anyone could be in this group, and there would be no defined organization to it, but the amount of experience and work needed to do this task would make it essentially a permanent team. The rest of the Cantr community would be relegated to proofreading, making suggestions and adding subjective commentary - the non-data of the sort that (as Nitefyre says) could go in the webzine.

If this plan looks nothing like a normal wiki's organizational tree, and more like a privately managed bridge between the community and the RD database, that's the idea. The traditional community-effort model was tried, and failed. If the wiki is to continue to be used in its current capacity, it will only work if it becomes more like the RD database gateway that it should be replaced by.