Not adding to the Wiki?
admin
I think our wiki is great, but I feel like it could be so much more than it is.
We have a lot of editors now, but people don't do much. Why is this?
There is so much great content here on the forums, but it gets lost over time as it scrolls down into the back pages.
Is the wiki too slow to open up? Our wiki software has been a server hog in the past, maybe it's still slow for people.
How difficult is it to search for things? We have FindPage, but maybe we could add Google search, or re-organize to make it easier to discover things.
Does the wiki software get in the way? We could change to another wiki backend, if that would help. Like maybe MediaWiki, what they use for Wikipedia. Our current software does have an easy editor, but you kind of have to seek it out.
Any ideas to make things better?
Replies
I tried to get write access once, years ago though, i think I had to ask a fella named 'Don' or something. Nothing ever came of it.
Xoliul, I can easily set you up as an editor, just send me your info.
For example:
http://wiki.polycount.com/CategoryTexturing
Coming from the front page, a little time could be spent introducing the concept of textures/maps, defining them, and their relation to materials and different techniques and programs to create them. Instead, everything looks a bit placeholder and is a little hard to scan: "CategoryXxxxxxx" AndOneWordLinks Makereadinghard. Is the wiki software hardcoded to always use the page name? If links could be made more legible and lists could be logically grouped (ex. Pages in this Category) and in multiple columns, it would help a ton. As always, look at the pages from the eyes of a new user working on their first or second model.
As a hub page, I think most of links in the 'related pages' would be better served as category headers here, instead of spread out over several pages a user needs to drill through. The content pages are fine, EdgePadding is great, it's just a matter of getting there.
To me, these are big deterrents. Categories seem to have a good tree structure, but finding the right places for things feels like a big messy chore. Sign me up as an editor, I'll send you a PM.
Don't worry about hurting my feelings or anything. I'm totally open to improving things, not attached to the current setup in any way.
Yeah, the Category function is ugly. Hard-coded that way in MoinMoin unfortunately
Happy to sign anyone up as an editor. I just need two things: your email address (never shared, for password recovery) and the name you want to use (real name in FirstnameLastname format, or your Forum name). If you don't want your email out here on the Forum, then a PM is probably best.
I think it would benefit from a much simpler visual organisation - something bright and easy to read through, like the pages of the Unreal Wiki for instance :
https://wiki.unrealengine.com/Dynamic_Depth_of_Field
I hope this helps !
Another thing. Earthquake's post history as a wiki page, discuss.
Also
Another concern I had was noobs adding stuff that is just plain wrong. But it's probably better to just open the gates wide, see what happens.
EQ definitely has a high signal-to-noise ratio. :poly106:
Honestly the best way for noobs to learn is by helping others or writing tutorials. I got my start in modding knowing little about 3d graphics and grew a lot by writing articles, it forced me to learn more about the finer points before committing anything to final, I'm sure you can attest to this.
At least with a wiki (as opposed to video), discussion can be had corrections can be made, and everyone wins. Worst case scenario is that the page needs to be nuked.
I just need two things: your email address (never shared, for password recovery) and the name you want to use (real name in FirstnameLastname format, or your Forum name). If you don't want your email out here on the Forum, then a PM is probably best.
On that note - Eric can you reset my password? I can't seem to do it...
Optinium, I don't think so. The wiki would have to be linked with the forum database. I think opening it up wide would probably be the best thing, as a start.
It has to be converted first though, will have a chat with the mod team.
The organization is also very odd. The fact that it is so slow makes me not what to learn its organization.
What about having all of the Wiki content on a 3rd party site, like Wikia?
Also, one thing that makes all other Wikis that I have frequented great (like Wikipedia, video game Wikis), is that everyone has the ability to edit the Wiki.
I know it sounds terrifying to allow the control open to the general public, but you will get a much more up-to-date Wiki if it allowed.
Maybe I'm paranoid, but without that ad bots or trolls could get their hands on the wiki and all of a sudden the wiki is flooded with fake links to sites with malware, surveys and ads or just offtopic stuff.
A couple of months ago I wanted to add a bunch of animation and rigging related tutorials to the wiki but once I saw that I had to contact the mods first to get an account, I kinda shied away, because I felt that since I am not an old and trusted member with 1k + posts I wouldn't get access anyway (I am not saying of course that it IS like that, I just thought so). So I think that ultimately a sort semi-open access system, could bring a lot of new entries to the wiki.
+
maybe link the account to the forum account and only let users post that are registered for longer than 6 months and have 10 posts (for example)
so you would avoid all the spam you would have to delete otherwise
MediaWiki – This is the most popular wiki software on the net, and runs wikiHow, Wikipedia, and many other wiki sites. Many of the most popular wiki farms use the MediaWiki software as well.
TikiWiki - This is the second most popular wiki software available, and runs a large number of wikis and wiki farms. TikiWiki has strong plugin support, allowing you to add features such as forums, image galleries, calendars, and more.
UserPress – This is a wiki plugin for WordPress. It has most of the functionality of MediaWiki and other standalone wiki software, but it's much easier to use.
DokuWiki - This is a smaller wiki software program that is growing in popularity, especially in enterprise-level implementations. It is designed first and foremost for team and workgroup collaboration, and has multiple levels of user access.
Some copypasta above but it saved me re-writing it ;')
The categorising
could use some improvement. For example, this page
Could be merged with this page (if it's not already?)
While the wiki is an amazing resource and has saved my ass so many times,
Some of the pages need more content. and some pages have no content at all. Maybe move some of the legacy content to a separate page, as it's probably not relevant to most people any more.
Personally I never really thought of contributing because I considered myself too much of a noob but I reckon I could contribute in some specific fields. HDRis for example, if there's any interest. Or even just cleaning up links that don't work, adding tutorials, resources etc.
I think 82 is not kind of "a lot", and many of them are "inactive". So, open access sound good, but to prevent spamming, trolling,... we need to limit who can access, may be based on years on polycount and post count , anybody else can still have access through the "old way" ( send request to the mods ). Of course new user's post will be considering before approve.
Definitely this. I've used the wiki a ton in the past and still do, the biggest issue I have is how slow some pages are. Especially when you click a link and it doesn't open a new tab, just simply goes to the link and then when you want to go back, it takes just a long to reload the page. I agree also with kitty that is indeed a bit messy on some pages, his example is a pretty decent example as well.
It's a great resource for finding different tutorials but just a bit slow.
Not sure if you know this, but middle mouse button is a quick way to open links in a new tab.
We're looking into new wiki software.
In the mean time, totally open to re-org ideas. If you would like to try something, go for it!
To reiterate...
http://wiki.polycount.com/3%20Dimensional%20Tutorials/UVW-Mapping
http://wiki.polycount.com/CategoryTexturing
Maybe that's a sign of an easily identifiable bug somewhere ?
because duh.
+1 for non-hierarchical organization
I've always been of the mind that a mediawiki implementation would be vastly superior. I think one of the biggest strengths there is that there is NO strict hierarchy and lots of flexibility to organize information without a hierarchy. quick-linking pages and categories is huge as well.
When people talk about 'deep-diving' into information, there is an intuitive process that involves moving from one topic to another organically, being linked deeper and deeper into a topic and moving around between information organically. The current wiki does NOT enable this, you more or less move into one topic, then hit a dead end. We need to build a web of interconnected data, not a hierarchy of roads that lead to dead-ends. This promotes learning and makes it feel easier to find what you want.
Strict hierarchies may be perceived as useful for helping find info in the first place, but meta-tagging and 'categories' as loose concepts can be just as good without the hierarchy.
+1 for open access
I've had to battle spam on wiki's I have run as well, but having the doors open is often a big deal and a way to get editors in. Even if having the doors open for a while THEN locking it down later would help. With the current setup you could advertise open registration for ONE WEEK ONLY and then clean up the mess afterwards, and it would likely be way more successful then telling people "just ask nicely and we'll let you in!"
I get very excited about the idea of a rebooted, open, faster, polycount wiki. I think there is some huge potential there.
1. Search for the page name, see if any pages link to it.
2. If so, delete the link from those pages.
3. Finally, delete the page (More Actions dropdown).
pior, yeah it's because Macros in MoinMoin are crap. We could remove the macros, but then we'd lose a lot of links, which promote the non-hierarchical navigation like what cman2k is talking about. Which is a big plus, much agreed.
I'll try open access, see what happens.
But I will take the time to say that I utterly love having the polycount wiki as an available resource. I don't add to it, but I've used it many, many times in the past to try and educate myself a bit more before giving critique, and it's come in really handy the couple times I've dabbled in simple 3D-related work.
I've never had any issues with the organization and have always been able to find what I was looking for very easily, but I'm not opposed to seeing it improved (because that would be silly of me.)
http://wiki.polycount.net/CategoryWikiMembers
I changed the macro from FullSearch to FullSearchCached.
Searching is still slow though, not much we can do about that yet. Organization is definitely a problem as well, we'll tackle this at some point.
I'm pretty excited about moving to MediaWiki, I'm all for whatever manual work that needs to be done during the change.
We have scripts to convert with, but they're each broken in some way. So it's sure to be a bit of a chore. But once it's done, there will be teh rejoice.
+1
I know how much the wiki have helped me when I went to school. I'll definitely be down to do a push to get things up and running! Eric, what ever you think I/we can help with, just shoot an email. Can't promise I'll have time, but I'll try!
Thanks for all the hard work Eric!
[FONT="] . So . here we go .[/FONT]
[FONT="]+1 The thing that always seemed to put me of using he wiki was finding what I wanted as mentioned before I thing using a combination of hierarchy, tagging and Google searching would be the bests as it caters to everyones search methods [/FONT]
[FONT="]+1 For opening the wiki up to everyone but I agree the spam and accuracy of posts needs to be moderated I suggest maybe a 4 week period before allowing posts on the wiki to stop bots and a flagging system that everyone can use to alert Admins of the site for spam, outdated info, broken links and such to give everyone to method of contributing to the quality of the wiki, especially if its wick and easy.[/FONT]
[FONT="]+1 To adding a discussion section to tutorials like on Wikipedia to help expand and deepen the information as mentioned. Maybe even give posts in the forum the quick re-post option to add to a tutorial discussion as really useful information gets added in passing on threads that is easily missed so providing a way to ad such posts to a tutorial or a more central resource section that could refer back to the thread would be amazing! [/FONT]
[FONT="]+1 To think having the white background is easier on the eyes for tutorials than the colour combination (it was mentioned earlier, but I also find the darker colour combination a little more of an effort to read)[/FONT]
[FONT="]+1 Bookmarks / subscriptions to tutorials would also be a cool idea, especially if discussions and tags are added then people can get email updates if a tut is updated or new tutorials are posted with tags that a subscriber likes[/FONT]
[FONT="]+1 Tutorial summaries viewable before looking at the main tut for making searches easier so people can get an idea of what the tutorial is about before opening the whole thing[/FONT]
[FONT="]+1 A rating system for tutorials to help find popular tutorials?[/FONT]
[FONT="]On a bit of a side note is it possible also to make this thread a sticky? So people can keep giving suggestions and not just in one big burst for this revamp?[/FONT]
[FONT="]I would love to help with the transition as well, I still consider myself a noob but I hope helping converting tutorials across will increase my knowledge base XD[/FONT]
crestas69, I think most of your suggestions are there in MediaWiki, so we'll just inherit them for free. Login to MediaWiki to check them out.
We talked about other ideas, maybe a WordPress kind of site where people would use GitHub to submit content, to be curated, something like code reviews. I thought that might be a bit more hassle than artists would be willing to put up with. Anyone here use GitHub? Maybe you can evangelize.
One thing we've been able to do so far (most of the time) is to moderate the wiki content to make sure you get accurate info. I see the wiki as a filtering of forum threads, not the same kind of freewheeling content. The Forum is where people hash out ideas, thrash around until they figure out what's true. The wiki IMHO should be where we put the distilled knowledge.
I don't know how that will work in the future, unless experienced users are involved more frequently to police the content. We will just have to try it out, see how it goes.
As for summaries and ratings, that will be up to the users to make. There is a Mediawiki extension for ratings, we might install it. Here's a wiki that uses it... http://www.wiki4games.com/F.E.A.R. (rating is down on the right side)
I won't make this thread a sticky though. Too many of those already! Just search for "wiki".
Have you had much vandalism on the Polycount wiki?
We used to have a huge spammer problem, before we gated registrations. I would love to connect editing privileges to one's status on the Forum, but that would require a bit of custom coding to tie the Forum db to the wiki system. Anyone up for this task?
So I've long known how terrible the wiki performs from watching the server stats but thus far any attempts to "fix" MoinMoin have been met with terrifying amounts of frustration. Furthermore, MoinMoin is a steaming pile of shit in every way imaginable. So.....
We're gonna burn it down.
Ok, so not entirely raze and burn since we can do an import of most of the content. The downside is that lots of links and all images/attachements are pretty much broken. I could invest time trying to unfuck the mess or perl that the conversion script is written in or we could just move on and in the process of fixing up the wiki take care of lots of cruft while we have some momentum.
I've built us a lovely little cookbook (Chef for those sysadmin/devops minded) that stands up MediaWiki in under 7 minutes and have been playing around with the import but we are about ready to push this up to the internet alongside the current wiki, something like http://newwiki.polycount.com, and then we can go nuts on transferring content.
Sound good? If so stop editing content - there isn't an elegant way to lock the old wiki - and I'll make the machinations to get the newwiki live and stuff. I can create admin accounts for what I'm calling "Wiki Team Alpha" and we can start dragging ourselves into the 21st century.
Benefits of mediawiki:
- Not a steaming pile of shit
- PLUGINS!!
- Creative Commons Licensing on pages (I like licensing so shutup)
- Easier to scale than Python (php/mysql/apache/memached)
- will generally be easier to integrate with EVERYTHING
Anything GitHub based, while prettier and nicer from an editorial standpoint is not really attractive for our audience. I mean I think it would be hot shit but I also don't make art daily and think git is sensible so MediaWiki, while still a wiki, is the best tool for not just the job but our community.
Do it!!