Jump to content
  • Advertisement

Archived

This topic is now archived and is closed to further replies.

Oluseyi

GameDev.net Wiki

This topic is 5540 days old which is more than the 365 day threshold we allow for new replies. Please post a new topic.

If you intended to correct an error in the post then please contact us.

Recommended Posts

quote:
Original post by JohnBSmall
Wikis are very powerful, but you cannot say that they do not have their problems, even when used in the domain that they work best in.

Pages can easily become unfocused and rambling. In fact, the entire wiki easily becomes unfocused and rambling, producing a complex web of pages with no hierarchy to support it, and leaving a text search as the only feasible option for finding information (and although this is adequate for many things, it is not ideal for all).
Information can get split up or duplicated over multiple pages, because someone may create a new page, without realising that there is an existing one that covers the topic.

However, all of these problems apply equally to forums, which have other problems of their own. Wikis are far better than forums for mid to long term storage and retrieval of information. Perhaps the more interesting and useful question is: Is there something better than a wiki (that would be appropriate in the same, or at least similar, situations)?

John B


a response to this would probably best be a link to the wikipedia. they are the largest wiki target out there for griefers (because of constant links on slashdot, etc) and they have put in place a very effective system to deal with them. true, the format isnt perfect, but as you point out, nothing is.

i guess as i envision it, the gdnet wiki wouldnt be so much a replacement for the forums, but a supplement to them and a far superior replacement to the faq. i''m being idealistic, but it could become the game devolpment community''s (non-software) answer to OSS...something you dont have to pay for but is superior (or at least broader and sometime more in depth) than many graphics books out there. but like i said, i''m being idealistic.

the upshots, though, are that you eliminate many repeat questions, that you actually get more people involved in the community, and that the forums become what they should be: answering questions on implementations, discussing your new release, and flame wars about a hypothetical perfect programming language.

Share this post


Link to post
Share on other sites
Advertisement
quote:
Original post by JohnBSmall
Pages can easily become unfocused and rambling. In fact, the entire wiki easily becomes unfocused and rambling, producing a complex web of pages with no hierarchy to support it, and leaving a text search as the only feasible option for finding information (and although this is adequate for many things, it is not ideal for all).
Information can get split up or duplicated over multiple pages, because someone may create a new page, without realising that there is an existing one that covers the topic.
I think I''ve stated enough times that the Wiki wouldn''t be traditional; not all visitors would be able to create/edit pages. It''d have more of a CVS structure where contributions constitute branches and administrators merge selections into the trunk.

quote:
Original post by seanw
For example, I used to frequent comp.lang.c++. If you read a few of the threads, pretty much the first reply to certain threads are that the question is in the FAQ, followed by a link to it, then _nobody_ else replies. You just see thread after thread of one post replies pointing to the FAQ. Before I ever posted anything, I would check the FAQ because it was so apparent you would simply be told to go and look for it was answer there. Their FAQ is actually quite comprehensive and contains many pedantic topics, so the questions that get discussed on the newsgroup tend to be interesting ones that aren''t in it.
Yes, comp.lang.c++ is self-moderating. But consider the members, the pedigree, the heritage and the force that convention exerts there. That newsgroup is decades old! Many of its members - regulars - are the foremost professionals in the application of the language. GameDev, in contrast, is a hobbyist site, many of its members being under 18 years of age.

I''d say there''s no point making that comparison.

Share this post


Link to post
Share on other sites
quote:
Original post by justo
a response to this would probably best be a link to the wikipedia. they are the largest wiki target out there for griefers (because of constant links on slashdot, etc) and they have put in place a very effective system to deal with them.

Well, I agree that a wiki can cope with malicious users, if implemented well, although I'd be interested to know what component(s) of wikipedia's implementation you think help it deal with onslaughts so effectively.

But I think retaining well written and well organised articles is a harder problem to solve. I have noticed that many pages on the first WikiWikiWeb are long, and contain a lot of ThreadMode, instead of having information integrated into a single article. Although there is generally a main article at the top, there is often useful information in the chat below that has not been distilled into the main text. You can counter this with wikipedia too, of course, since it doesn't seem to suffer from that problem. But what is it about wikipedia that prevents (or perhaps just efficiently corrects?) thread mode? And is it something to do with implementation, that can be included in a GDnet wiki? Or is it something else (the strong emphasis on wikipedia being a "proper" encyclopedia, perhaps) that could not be translated to the context of this site?
And in fact, does the prevention of thread mode also prevent good input from people who don't feel they have the time to write up an article (or properly integrate their knowledge into an existing article)? Is it possible to prevent thread mode without preventing the free flow of information?

Edit: I suppose this post is mostly moot, given Oluseyi's stance on letting everyone edit everything. If "normal" users can only post separate additions to pages, then thread mode will naturally follow.

John B

[edited by - JohnBSmall on February 13, 2004 12:01:41 AM]

Share this post


Link to post
Share on other sites
quote:
Original post by JohnBSmall
Why would a wiki be overkill?
If GDnet became a wiki (excepting the parts that would remain forums, of course) then it would be the FAQ, and it would automatically be updated to answer not only the most frequently asked questions, but any question that comes up. People would read it, because on a wiki, the emphasis (at least as I see it) is on searching/reading first, and then editing later if you''ve got something to add. The only way to create a new page is to link to it from an existing page, and unless you''re being stupid or malicious (both of which are dangers, I admit), then you probably won''t create a link from somewhere totally unrelated. Therefore, by the time you try and create a new page, you''ve already searched and found the related information, and you will have found any existing answers.



I just think that contributing to a forum is easier than to a wiki, they aren''t really comparable because it isn''t the same kind of experience. It wouldn''t be hard to imagine people getting bored of updating it and it stagnating.

Share this post


Link to post
Share on other sites
quote:
Original post by Oluseyi
Yes, comp.lang.c++ is self-moderating. But consider the members, the pedigree, the heritage and the force that convention exerts there. That newsgroup is decades old! Many of its members - regulars - are the foremost professionals in the application of the language. GameDev, in contrast, is a hobbyist site, many of its members being under 18 years of age.

I''d say there''s no point making that comparison.


I would say there is. Some common questions here are "how do I make a fps/rts/rpg?", "is stl fast?", "is c++ better than c?". Create a good FAQ that covers the majority of these questions and make sure all the regulars know about it. When one of these questions is posted, link to the answer in the FAQ. If enough people did this, more people would do it to. Soon, you''d have even new people knowing about the FAQ and the fact that if they ask a FAQ they will just be linked to it. I''d say that closing threads of the really frequent questions (like how fast is STL) would be perfectly acceptable as these threads are so repetitive and obvious.

I don''t see how the people on comp.lang.c++ being older means there is no point making a comparison. It''s just encouraging people to act in a certain way.

If you make a Wiki, you''d still have to get people to read it in some way, and the above methods would work. Otherwise you''re just going to get the same threads over and over again regardless of their inclusion in the Wiki.

Share this post


Link to post
Share on other sites
quote:
Original post by clum
Whatever happenned to this idea? Is it being implemented?
Yes, but I don''t make announcements until there is a demonstrable product. I''ll post in Your Announcements when it''s usable.

Share this post


Link to post
Share on other sites
I haven''t read the entire thread, so I''m sure these points have already been mentioned, but I''ll voice my support for them anyway:

I''m all for the Wiki, but the Staff of course must implement some safety features:

First, I think the Staff (and maybe Moderators) should be responsible for providing logical, well-organized categories under which Wiki pages would be deposited. I think general users should be able to Edit pages IF they
a) have a sufficient approval rating (like a karma system, I think you guys are implementing something like this) AND
b) have a significant number of posts.

I think the standard should be set higher for users to be able to create new pages. Moderators would always be able to create new pages (though it seems likely most moderators would have enough posts and a high enough rating to do so anyway).

Naturally, Staff and Moderators should be able to delete pages that aren''t useful or present duplicate information. ALL users should be able to report such content.

I also think there should be a limit to the number of pages one can add per week, and the number of pages one can edit per day, in order to prevent excessive, frivolous posting.

Finally, it goes without saying that the Wiki should have a VERY robust search engine.

Share this post


Link to post
Share on other sites
We''re no longer going with a "wiki" in the traditional sense of the word. Here''s a brief overview of the current approach:

The HyperFAQ (new name) is a tiered set of indexes to static and dynamic content on GameDev.Net. The top tier breaks subjects into broad categories such as Programming, Asset Creation, Management, etc. The second tier then tackles various domains within those categories - Graphics Programming, Artificial Intelligence, Sound Design, Coordinating Distributed Internet Teams, etc.

Beneath these domains are topic lists, tackling specific individual issues. As much as possible, each topic will start with an article or summary of the current consensus of the issue, with links to external resources, archived GDNet discussions and a live "thread" (link right into "canonized" forum topic). Whenever things changed markedly, one of the site administrators (staff, mods and selected Community Editors) will modify the summary/article or write a new one (causing the old one to be archived and linked from the page as well), which automatically closes the current topic thread, adds it to the links on the entry page and creates a new thread for updated discussion.

I think that''s a fair compromise between the various approaches.

Share this post


Link to post
Share on other sites

  • Advertisement
×

Important Information

By using GameDev.net, you agree to our community Guidelines, Terms of Use, and Privacy Policy.

GameDev.net is your game development community. Create an account for your GameDev Portfolio and participate in the largest developer community in the games industry.

Sign me up!