NathanLee

Thoughts on Wikipedia

I love the idea of wikipedia: online content, community contributions, ability to create articles on niche areas that would not get research effort in traditional encyclopaedias. I think wikipedia should have an article on everything in the world, so long as someone's prepared to go to the initial effort to add some information.

Contributions

When I find things missing from wikipedia I try to add them in. It is rather time consuming creating new pages from scratch. I have been doing a bit of random page fix-ups. Things like spelling, grammar, lack of consistency and too, many, damned, commas, e,v,e,rywh,ere. A couple of pages that I've decided to dig deeper and add more content. Pages of ongoing editing interest or ones I've created:

• Triumph Daytona 675 - my bike. Added bunch of content, picture.
• Triumph Street Triple 675 - the naked version of my bike. Created.
• Paddlepop - can't believe this didn't exist. What aussie hasn't eaten a paddlepop icecream when growing up?? Created.
• Ellsworth Handcrafted Bicycles - a mountain bike company (and I own one of their bikes). Created, added pictures.
• Triumph Tiger 955i - Noticed it was missing as I had taken some pictures out at a bike shop and figured I'd add it. Instead created the page, will fill it out gradually.
• People for the Ethical Treatment of Animals, Factory farming, Intensive farming, Industrial agriculture - first encounter with wikipedia Point Of View pushers. Classic moments include arguing against dictionary definitions and encyclopaedia Britannica in favour of some bizarre definition of terms. Some people couldn't accept there was more to agriculture than their bias would permit.
• webMethods, webMethods Integration Server, webMethods Developer, webMethods Glue, SAP Business Connector - added info or created. Products I worked with in a past life.
• Bulahdelah, New South Wales - Created.
• Blastolene Special Noticed it needed a lot of tidy up and references.
• Triumph Tiger 1050 Created.
• Triumph Daytona 650 created.
• Triumph Daytona 600 created and took/uploaded a pic for it.
• Ducati Monster 696‎ created.
• Mondo spider created (saw this weird beastie on a youtube vid and did some investigating).
• Religion in Australia - added the no religion section
• Islam in Australia - reasonable cleanup, add of material
• Melbourne Shuffle - major cleanup, frequent vandalism/self promotion
• Irreligion in Australia - created
• Islamic mythology, Christian mythology - cleanup
• Noah's ark, Searches for Noah's Ark - islamic and science view. Some don't know what "mythology" means.

Problems with Wikipedia

Like most things: wikipedia in the real world has some issues

• casual vandalism
• subtle vandalism
• use of rel="nofollow"
• skewed content based on the heavy reliance on online sources
• bias against use of any non-mainstream source
• anonymity
• excessive sensitivity of community
• lack of consistency
• inability to separate out different English varieties (e.g. "colour"/"color") which makes wikipedia essentially USELESS in my opinion.
• abuse of power by longer term admins (e.g. "The Untouchables")
• lack of referencing

Thoughts on Fixing wikipedia

Just some random thoughts about how to counter some of the problems I've outlined above.

Casual Vandalism

Require all edit abilities to sign in as a user. I really don't think there is ANY advantage longer term to having non-login editing. Signup is easy, doesn't result in spam, doesn't cost money etc. By far the most casual vandalism is from IP addresses (e.g. non logged in users).

Subtle vandalism

This one is linked in with another complaint I have about "abuse of power". The concept of gradually steering an article to suit POV by either reverting work of others citing reasons they do not apply to their own edits. Hiding of editing changes under non-descriptive edit comments. Page refactoring that removes sections without replacing them. Replacing words with euphemisms etc. Disallowing criticism sites or anything deemed "private".

The Untouchables

Like any system, there are those who have been around long enough to fall under the banner of "established editors" who are then above examination or who simply block those who disagree or point this out. There doesn't seem to be any formal process for raising complaints about a user with elevated status

These users, in my opinion can end up with a mindset that shows little respect for constructive additions or improving wikipedia, but rather attempt to enforce their will with little regard for whether or not they are merely pissing off new users and thus reducing the pool of enthusiastic contributors.

A solution might be to have limited terms for higher positions with reviews and voting process to retain "office" as per normal democratic process. Or a policy that stipulates those users should withdraw from editing any articles on contentious issues.

Skewing of content to reliance on online sources

Some topics are never going to get much exposure on the web unfortunately. Although there's the ability to add in references to offline sources: there's really not much of a process for validating such a source. I'm a bit at a loss as to how to deal with this. If it is possible to take a digital picture of the reference in the source and if this constitutes "fair use" to have a library of such sources attached to wikipedia. I think this is achievable given the increased availability of digital cameras/scanners etc.

Bias against use of non mainstream source

A tricky one also, particularly on contentious topics: citing of "no personal sites" or "no attack sites" allowed. Now this then leaves only "mainstream" sites such as news outlets. This really means that some things are never going to be verifiable or able to be added to wikipedia. For example:

• fan sites on movies/tv shows are usually the only references for transcripts or trivia.
• the arbitrary declaration of a site as "an attack site" means that you can pretty much reduce the available sources on a contentious group or topic to just one side as honestly you're then disregarding critical comment on a topic as allowed.

I'd also think that if a blog is well referenced then it should probably get treated as a reliable source. Original research from someone other than a newspaper or encyclopaedia is then possible (as who is going to get an article on the gender inequality balance of the smurfs? Or country based variations in foam mattress design?).

English varieties

Wikipedia needs an en/US, en/GB, en/AU etc ability to be added for any words which differ. It is simply ridiculous to force redirection for different varieties to "whatever was created first". This just tends to mean a skew towards en/US spelling as a lot of the original authors (and indeed the site's creator) are from the USA. Taking "colour"/"color" as an example:

• Colour should be a separate page to Color or the page should be named "variety1/variety2/variety3" e.g. "Colour/Color" and depending on user's preference should filter to show just the spellings that apply
• a simple syntax for in-line content should help things along e.g. each reference to the word is placed as a reference to the title e.g. "articleName" or "ArticleName" type tag to allow for capitalisation etc

Or

• the page that satisfies most language varieties should be the main one, the other should be a subpage

Until this is fixed, I believe wikipedia is near useless as it is flat out incorrect to have wrong location based spelling or a mix.

Lack of consistency

Even amongst pages that are on similar topics (e.g. cities/towns, technologies, variants of something) they have whatever structure or template that happened to get thrown in there by the original author. I suggest that there needs to be some default populating of pages and a more UI oriented way to choose templates. Currently every man and his dog has created different template/structures. A wizard to create new pages would help I think. Or simply pre-populating with by default at least an auto filled "references" section with the auto fill reference tag to encourage proper use of references.

Lack of references

Think any new paragraphs should require an extra step to check that there's a reference cited. User can override, but if someone is adding in extra paragraphs or significant amounts of text: it should have ref tags as a rule.

Use of rel="nofollow"

This is one of the stupidest changes ever made to wikipedia. Not only does it go against the concept of rewarding the original sources: it means that wikipedia is only ever a selfish site from a web-search ranking sense. Vandalism to promote commercial sites is pretty quickly stamped out, but refusing to give any ranking boost to relevant decent sites is an appalling feature of wikipedia. Until that is fixed, I really think that no other site on the web should ever give a search engine relevant link to wikipedia. Every single link to wikipedia should be rel="nofollow" as currently wikipedia treats every primary, good research reference the same way. We should not have wikipedia ending up higher up the search rankings than primary sources for everything.

Excessive sensitivity of community

The impression I get from wikipedia by looking at various discussions is that any criticism directed at a particular user is discouraged and can quickly result in a ban. Any good democracy HAS to allow dissent and criticism. I think this excessive need to be "wikipedia politically correct or else banned" has lead to the rise up the ranks of people who really need to have been called out on their bias/nasty behaviour. An effective central location for lodging complaints about administrators (e.g. abuse of power or accusation of bias) is sorely needed.

Anonymity

A blessing and also a curse of wikipedia. I've chosen not to disguise myself on here, but many have and have obtained sysop or admin rights with no verification or independent review of who they are or if they work or are affiliated with an organisation. This could lead (may/has already? Who knows) to infiltration of the site with editors who work for political parties, companies or simply people who want to take down wikipedia. Although stalking is an issue that perhaps is a justification, it does seem that some degree of transparency might be a good idea. The head representative of wikipedia is certainly a visible character (does the talk show rounds).. But would wikipedia benefit from having some faces to the anonymous editor masses.. To address longer term reliability of data concerns I think that that almost has to be a requirement.